Evaluation of

the National
headspace
Program
Final Report

Department of Health

June 2022

Evaluation of the National headspace Program – Final Report

June 2022

This evaluation was undertaken by a consortium led by KPMG, and including the Social Policy Research
Centre (SPRC) at the University of New South Wales, and batyr.
This report was written by members of the KPMG and SPRC consortium teams.
KPMG would like to thank SPRC and batyr for all their work across the evaluation, especially engagement
with young people across Australia.
Disclaimer
Inherent Limitations
Appendix A: Evaluation scope and method outlines the approach and limitations of the engagement. The
services provided as part of this engagement are advisory and therefore are not subject to assurance or
other standards issued by the Australian Auditing and Assurance Standards Board and, consequently no
opinions or conclusions intended to convey assurance have been expressed.
No warranty of completeness, accuracy or reliability is given in relation to the statements and
representations made by, and the information and documentation provided by the Department of Health’s
stakeholders consulted as part of the process.
KPMG has indicated within this report the sources of the information provided however those sources
have not been independently verified unless otherwise noted within the report.
KPMG is under no obligation under any circumstances to update this report, in either oral or written form,
for events occurring after the report has been issued in final form.
The report findings have been formed on the above basis.
Third Party Reliance
This report is solely for the purpose set out in Appendix A: Evaluation scope and method and for the
Department of Health’s information, and is not to be used for any other purpose or distributed to any other
party without KPMG’s prior written consent.
This report has been prepared at the request of the Department of Health in accordance with the terms of
KPMG’s contract dated 29 June 2020. Other than our responsibility to the Department of Health, neither
KPMG nor any member or employee of KPMG undertakes responsibility arising in any way from reliance
placed by a third party on this report. Any reliance placed is that party’s sole responsibility.
Accessibility
To comply with the Commonwealth Government’s accessibility requirements for publishing on the
internet, two versions of this Report are available: a KPMG-branded PDF version and an unbranded
Microsoft Word version. The KPMG-branded PDF version of this Report remains the definitive version of
this Report.

KPMG | iii

©2022 KPMG, an Australian partnership and a member firm of the KPMG global organisation of independent member firms affiliated with KPMG International Limited, a
private English company limited by guarantee. All rights reserved. The KPMG name and logo are trademarks used under license by the independent member firms of the KPMG
global organisation. Liability limited by a scheme approved under Professional Standards Legislation.
Evaluation of the National headspace Program – Final Report
June 2022

Contents
Contents ii
Glossary 1
Index of Tables 3
Index of Figures 6
Acknowledgement 9
Executive Summary 10
1 Introduction 28
1.1 Evaluating the headspace model 28

2 Understanding headspace 32
2.1 Overview of the headspace model 32

2.2 What success looks like for headspace 35

2.3 Components of the headspace model 44

2.4 headspace in context 50

2.5 Services currently available at headspace 59

2.6 Changes to services available over the last five years 69

2.7 Understanding headspace – in conclusion 76

3 Effectiveness of headspace in achieving program outcomes 77

3.1 Measuring outcomes of the headspace model 78

3.2 Effectiveness of the headspace model 85

3.3 Overall effectiveness of the headspace model 105

4 Cost-effectiveness and value of headspace 108

4.1 The full cost of delivering headspace 108

4.2 Economic evaluation of services provided by headspace 118

5 Factors affecting the future implementation, sustainability and enhancement of

headspace 127
5.1 Barriers and enablers to headspace meeting its objectives 127

5.2 External factors that have impacted or will impact headspace objectives being

delivered 131

5.3 Changes required to the design of headspace to enable it to meet its objectives 132

5.4 Changes required to the implementation of headspace to enable it to meet its

objectives 133

5.5 Changes required to the funding arrangements of headspace to enable it to meet its

objectives 142

5.6 Broader system changes that would support headspace to meet its objectives 143

5.7 Evaluation conclusion 145

Appendix A : Evaluation Scope and Method 147

Appendix B : Consultation 160
Appendix C : headspace services as at 30 June 2020 203
Appendix D : Effectiveness in achieving intermediate outcomes 209
Appendix E : Effectiveness in improving mental health and wellbeing outcomes 275
Appendix F : Inclusion and exclusion criteria 318
Appendix G : Definitions of K10 distress levels 322
Appendix H : Factors affecting the likelihood of completing the follow up survey 323
Appendix I : Extrapolation of the follow up K10 outcome measure 326
Appendix J : Costing assumptions 329
Appendix K : Economic evaluation parameters and inputs 330
Appendix L : Reference List 339

KPMG | iii

©2022 KPMG, an Australian partnership and a member firm of the KPMG global organisation of independent member firms affiliated with KPMG International Limited, a
private English company limited by guarantee. All rights reserved. The KPMG name and logo are trademarks used under license by the independent member firms of the KPMG
global organisation. Liability limited by a scheme approved under Professional Standards Legislation.
Evaluation of the National headspace Program – Final Report
June 2022

Glossary
Acronym Meaning
ABS Australian Bureau of Statistics
ACCHS Aboriginal Community Controlled Health Services
AIATSIS Australian Institute of Aboriginal and Torres Strait Islander Studies
AIHW Australian Institute of Health and Welfare
AMHS Adult Mental Health Service
AOD Alcohol and other drugs
ASGC Australian Standard Geographical Classification
CAMHS Child and Adolescent Mental Health Service
CYMHS Child and Youth Mental Health Service
DFV Domestic and Family Violence
DID Difference-in-Differences
DSS Department of Social Services
EOC Episode of Care
ED Emergency Department
EMHSS Enhancing Mental Health Support in Schools (Victorian Government)
EPYS Early Psychosis Youth Services
FTE Full Time Equivalent
GPs General Practitioners
hMDS headspace Minimum Dataset
hMIF headspace Model Integrity Framework
hAPI headspace Applications Platform Interface
ICER Incremental cost-effectiveness ratio
IPS Individual Placement Support
LGBTQIA+ Lesbian, Gay, Bisexual, Transgender, Queer, Intersex, Asexual
K10 Kessler Psychological Distress Scale
KPI Key performance indicator
MAT Minimum Adequate Treatment
MBS Medicare Benefits Schedule
MLT MyLifeTracker
OOS Occasion of Service
NDIS National Disability Insurance Scheme
PBS Pharmaceutical Benefits Scheme
PCYC Police Community Youth Centre
PHCRIS Primary Health Care Research and Information Service
PHNs Primary Health Networks
PMHC-MDS Primary Mental Health Care Minimum Data Set
QALY Quality Adjusted Life Year
RTM Regression to the mean
SOFAS Social and Occupational Functioning Assessment Scale
SPRC Social Policy Research Centre
TMHS Tertiary Mental Health Service
TMLD Trade Mark Licence Deed
UNSW University of New South Wales

Term Definition
headspace Refers to the headspace program
headspace National headspace National Youth Mental Health Foundation
headspace services operating in accordance with the headspace Centre
headspace centres
Model
headspace satellites or satellite
Alternative headspace model providing a reduced range of services
services
headspace network, or headspace Refers to the national collection of headspace services, including
services headspace centres and headspace satellites
headspace model, the model The headspace Centre Model as described in the hMIF

Index of Tables
Table 1: Recommended changes to the implementation of the headspace model 22
Table 2: Recommended changes to funding for the headspace model 25

Table 3: headspace objectives and impacts 39

Table 4: Overview of K10 psychological distress levels 43

Table 5: Overview of headspace services and population by jurisdiction as at 30 June 2020 59

Table 6: Overview of headspace services by remoteness as at 30 June 2020 60

Table 7: Overview of headspace services by service type as at 30 June 2020 60

Table 8: Overview of headspace services by PHN as at 30 June 2020 61

Table 9: Overview of professions of headspace service staff during 2019-20 62

Table 10: Wait times by service rurality 67

Table 11: Growth in the number of headspace services between 2015-16 to 2019-20, by PHN 71

Table 12: Surveys on hAPI 78

Table 13: Reports available on the headspace dashboard tool 80

Table 14: Effectiveness findings in summary 105

Table 15: Summary of responses to questions on indirect service contributions 110

Table 16: Range of out-of-pocket costs by service provided during 2019-20 115

Table 17: Total costs of delivering headspace 116

Table 18: Overview of economic evaluation 118

Table 19: Results of incremental cost-effectiveness analysis 122

Table 20: One-way sensitivity analysis of selected evaluation parameters 123

Table 21: Sub-group analysis of headspace service ICERs 125

Table 22: ‘Hard to reach’ groups recommendations 135

Table 23: Service integration recommendations 137

Table 24: Governance and commissioning recommendations 139

Table 25: Monitoring and reporting recommendations 140

Table 26: Funding arrangements recommendations 142

Table 27: Evaluation Questions 148

Table 28: Geographic regions of focus for the evaluation 154

Table 29: Stakeholders consulted from headspace services 160

Table 30: Stakeholders consulted from Primary Health Networks 162

Table 31: Stakeholders consulted from Indigenous organisations 163

Table 32: Stakeholders consulted from tertiary mental health services 163

Table 33: General Practitioners consulted 164

Table 34: Stakeholders consulted from headspace National 164

Table 35: Stakeholders consulted from Commonwealth Government 165

Table 36: Stakeholders consulted from State and Territory Governments 165

Table 37: Stakeholders consulted from peak bodies 166

Table 38 Stakeholder engagement themes from young people who use headspace 167

Table 39 Stakeholder engagement themes from young people who do not use headspace 175

Table 40 Stakeholder engagement themes from headspace service providers 179

Table 41 Stakeholder engagement themes from Primary Health Networks 190

Table 42 Stakeholder engagement themes from other service providers 194

Table 43 Stakeholder engagement themes from Commonwealth Government, state and territory

governments and peak bodies 199

Table 44: headspace services open at 30 June 2020 203

Table 45 Overview of mental health literacy objectives of headspace 209

Table 46 Overview of early help seeking objectives of headspace 214

Table 47 Overview of access to service objectives of headspace 220

Table 48: Average number of young people accessing headspace per year 221

Table 49 Overview of objectives of headspace for ‘hard to reach’ groups 224

Table 50: Share of young people accessing headspace who are Aboriginal and Torres Strait

Islander 240
Table 51: Share of young people accessing headspace with culturally and linguistically diverse
backgrounds 241

Table 52 :Share of young people who identify as LGBTQIA+ 241

Table 53 Overview of objectives of headspace for stigma reduction 243

Table 54 Overview of objectives of headspace for service integration and coordination 245
Table 55 Overview of objectives of headspace for culturally appropriate and inclusive

services 262
Table 56: Probability young person respond ‘agree’ or ‘strongly agree’ to satisfaction domains across
episodes created from 2015-16 to 2019-20 263
Table 57 Overview of objectives of headspace for appropriate, accessible and youth friendly

support 267
Table 58 Overview of objectives of headspace for young people’s participation in the design and delivery
of services 272

Table 59 Overview of mental health and wellbeing objectives of headspace 277

Table 60: Regression to the mean effect by outcome measure 280

Table 61: Average intake, final and change in K10 measurements in young people accessing
headspace 281
Table 62: Average intake, final and change in SOFAS measurements in young people accessing
headspace 282
Table 63: Average intake, final and change in MLT measurements in young people accessing
headspace 282

Table 64: Reliable significant change index by outcome measure 283

Table 65: Clinically significant change index by gender and age 283
Table 66: Improvement in average outcome measures in young people accessing headspace (per completed

episode) 284
Table 67: Average Improvement in raw outcome measures in young people accessing headspace (per

completed episode) 284

Table 68: Logit regression of receiving two or more OOS 286

Table 69: Linear regression of mental health improvements 291

Table 70: Shaply decomposition by patient, service, regional components 295

Table 71: Correlation matrix between services fixed effects 301

Table 72: Linear regression of service-specific components on service fixed effects. 302

Table 73: Shaply decomposition by service and regional components 308

Table 74: Average improvement in K10 outcome measures in young people completing the follow up
survey 309

Table 75: Number of follow up survey responses 310

Table 76: Outcome measures 312

Table 77: Difference-in-Difference analysis of the impact of number of headspace services on area-level
measures of mental health 314
Table 78: Difference-in-Difference analysis of the number of headspace clients per 1,000 young people on
area-level measures of mental health 315
Table 79: Difference-in-Difference analysis of headspace intensity on area-level measures of mental
health 316

Table 80: Follow up survey responses from 2015-16 to 2019-20. 320

Table 81: Definitions of K10 distress level 322

Table 82: Logit regression of completing the follow up survey 324

Table 83: Extrapolation of the K10 score at the follow up 326

Table 84: Volume-weight average of equivalent MBS benefit fees 329

Table 85: headspace OOS cost determination 330

Table 86: Determination of mental health and substance abuse hospitalisation costs 331

Table 87: Average QALY gain for closed episodes in 2019-20 334

Table 88: Average costs and QALYs gained per episode in the world without headspace 335

Table 89: Input values used in the economic evaluation 336

Index of Figures
Figure 1: Summary of the headspace program logic 11

Figure 2: Overview of the evaluation design 12

Figure 3: Findings at a glance 16

Figure 4: Outcome areas in scope for this evaluation 18

Figure 5: Overview of the evaluation design 29

Figure 6: Summary of the headspace program logic 34

Figure 7: headspace governance structure and partners, as at January 2022 52

Figure 8: High level summary of mental health supports available for young people 56
Figure 9: Proportion of headspace service and lead agency respondents indicating the profession was
difficult to access for their local service 63

Figure 10: Services provided across every headspace OOS during 2019-20 64

Figure 11: Services provided across headspace services in 2019-20, by service type 65

Figure 12: Services provided across headspace services in 2019-20, by remoteness of services 66
Figure 13: Percentage of episodes of care where young people say 'yes' to having waited too long to be

seen at headspace 68
Figure 14: Growth in the number of headspace services between 2015-16 to 201920, by
jurisdiction 69
Figure 15: Growth in the number of headspace services between 2015-16 to 2019-20, by
remoteness 70
Figure 16: Changes in the mix of services provided during each headspace OOS between 201516 and
201920 72
Figure 17: Changes in the mix of services provided by headspace services between 201516 and 201920, by
headspace service type 73
Figure 18: Changes in the mix of services provided by headspace services between 201516 and 201920, by
service remoteness 74

Figure 19: Service delivery modality by month from January 2020 to June 2020 75

Figure 20: National headspace grant funding by service by volume during 2019-20 109

Figure 21: Distribution of OOS funding source during 2019-20 110

Figure 22: MBS funded OOS by provider type 111

Figure 23: Proportion of OOS funded by the MBS by headspace service 112

Figure 24: MBS funded OOS vs total OOS 112

Figure 25: Histogram of in-kind contribution 113

Figure 26: Histogram of PHN funding contribution 114

Figure 27: Histogram of out-of-pocket share by service 115

Figure 28: Distribution of the total cost per OOS by headspace service during 2019-20 117
Figure 29: Mental health service use by target population with or without the headspace

program 120

Figure 30: The model structure for comparing the world with or without headspace 121

Figure 31: headspace service-specific incremental costs and effects 124

Figure 32: Distribution of responses to “I feel that I know more about mental health problems in general
because of attending headspace” from 2015-16 to 2019-20 210
Figure 33: Proportion of episodes of care for young people who agreed their mental health literacy had
improved after using headspace services, based on the number of OOS accessed during their episode of
care 211
Figure 34: Responses from lead agency and headspace survey representatives on how effective headspace
services are in increasing mental health literacy 212

Figure 35: Distribution of age by young person from 2015-16 to 2019-20 215

Figure 36: Mental risk status on initial OOS for all episodes of care during 201920 216
Figure 37: Stage of illness during initial OOS for all episodes of care between 201516 and 2019-
20 217
Figure 38: Responses from lead agency and headspace survey representatives on how effective headspace
services are in increasing early help seeking 218

Figure 39: Barriers to access for young people 223

Figure 40: Survey responses about whether the headspace model is less effective for particular cohorts

compared with the general population of young people 226

Figure 41: Responses from service and lead agency survey: how well does your centre provide services

that support mental health literacy for young people from priority cohorts? 227

Figure 42: Distribution of age by young person during 2019-20 230

Figure 43: Mental health risk status on initial OOS for all episodes of care during 201920 232

Figure 44: Stage of illness during initial OOS for all episodes of care during 2019-20 234
Figure 45: Responses from lead agency and headspace services survey to ‘how well does your centre
provide services that support early help seeking for young people from priority cohorts?’ 236
Figure 46: Responses from the lead agency and headspace service survey to ‘how well does your centre
provide services that support access for young people from priority cohorts?’ 239
Figure 47: Young people responses to other services their GP referred them to (young peoples’
survey) 248
Figure 48: Barriers to supporting pathways to care identified by service and lead agency
representatives 250
Figure 49: Young people responses to other services their GP referred them to (young peoples’

survey) 259

Figure 50: Summary statistics on young people’s ratings of service at headspace 265
Figure 51: Responses from service and lead agency survey: how well does your centre provide services
that are youth friendly, appropriate and accessible 270
Figure 52: Responses from service and lead agency survey: how well does your centre provide services
that are youth friendly, appropriate and accessible 271
Figure 53: Young people’s experience of being at headspace, interaction with the staff and the service
received 273

Figure 54: Primary issue during initial presentation per episode during 2019-20 278

Figure 55: Distribution of OOS per episode during 2019-20 279

Figure 56: Distribution of K10 improvements by headspace service 288

Figure 57: Distribution of SOFAS improvements by headspace service 288

Figure 58: Distribution of MLT improvements by headspace service 289

Figure 59: Responses from service and lead agency survey: how well does your centre provide services
that improve the mental health and wellbeing of young people? 293

Figure 60: Average improvement in the K10 by young person, OOS and service-level factors 296
Figure 61: Average improvement in the SOFAS by young person, OOS and service-level

factors 297

Figure 62: Average improvement in the MLT by young person, OOS and service-level factors 298

Figure 63: Distribution of the K10 fixed effects 299

Figure 64: Distribution of the SOFAS fixed effects 300

Figure 65: Distribution of the MLT fixed effects 300

Figure 66: Average K10 improvement by service-level factors 305

Figure 67: Average SOFAS improvement by service-level factors 306

Figure 68: Average MLT improvement by service-level factors 307

Figure 69: Distribution of follow up survey completion time 309

Figure 70: Exclusion pathways 319

Figure 71: Exclusion criteria for sustained outcome analysis 320

Figure 72: Additional exclusion criteria for cost-effectiveness analysis 321

Figure 73: The average K10 outcome measure for episodes with at least three OOS 332

Figure 74: Mean QALY change for episodes with at least three OOS 334

KPMG | 3

©2022 KPMG, an Australian partnership and a member firm of the KPMG global organisation of independent member firms affiliated with KPMG International Limited, a
private English company limited by guarantee. All rights reserved. The KPMG name and logo are trademarks used under license by the independent member firms of the KPMG
global organisation. Liability limited by a scheme approved under Professional Standards Legislation.
Evaluation of the National headspace Program – Final Report
June 2022

Acknowledgement
The Evaluation Team would like to acknowledge those individuals who gave up
their time to be consulted as part of this evaluation, including those young people
with lived experience of mental ill-health. It would not have been possible to
complete the evaluation without their considered input.

It is acknowledged that there is no single set of terminology that suits all situations
and people. No exclusion or harm of people is intended by the terms used in this
report. The report endeavours to use inclusive language, while acknowledging the
evidence base and the experiences of young people.

KPMG | 3

©2022 KPMG, an Australian partnership and a member firm of the KPMG global organisation of independent member firms affiliated with KPMG International Limited, a
private English company limited by guarantee. All rights reserved. The KPMG name and logo are trademarks used under license by the independent member firms of the KPMG
global organisation. Liability limited by a scheme approved under Professional Standards Legislation.
Evaluation of the National headspace Program – Final Report
June 2022

Executive Summary
The headspace program has been evaluated twice before, in 2009 and 2015. The current evaluation, the
subject of this report, builds on these prior analyses to explore how the model operates today, and the
impact of ongoing changes in its design, reach and priorities on the availability of highquality, effective
mental health care for young Australians.
This evaluation is focused on headspace service provision, as provided by individual services around
Australia from 1 July 2015 to 30 June 2020. Various developments in Australia’s mental health landscape
within the period are taken into account for this evaluation, which is intended to help inform policy and
investment decisions about the future direction of the headspace model.

The headspace model

headspace is often referred to as the Australian Government’s flagship mental health program for people
aged 12 to 25. Since 2006, it has played an important role in efforts to tackle mental ill-health, self-harm,
and suicide among young Australians. Delivered as a network of community-led and governed centres
across Australia, headspace services support young people and their families to access clinical and
community mental health supports and interventions.
headspace provides services across four core streams to provide holistic support for young people. The
four core streams are:
• mental health and wellbeing;
• physical and sexual health;
• work and study support; and
• alcohol and other drug (AOD) services.
These services can be delivered in-person at headspace services and through telehealth, to help ensure
young people are able to access mental health supports when they are needed, particularly for those young
people who live in regional and remote areas.
The headspace program’s service cohort is young people aged 12 to 25 with mild to moderate mental
health conditions and those experiencing episodic or situational need. The headspace model is designed to
meet the mental health needs of young people who are deemed at risk of, or who are experiencing, the
early stages of a mental health disorder or who are facing common co-occurring situational stressors or
difficulties. It is intended that young people with more intensive needs who present to headspace are
supported to access other services through partnerships and service system linkage.
Figure 1, below, provides an overview of the headspace program logic, outlining the relationship between
elements of the headspace model. The model is designed to achieve a range of short-term impacts,
including improved mental health literacy, increased early help seeking behaviours, the promotion of a
positive experience of service for young people, and improved psychosocial outcomes. These then lead to
medium-term impacts for the functioning, wellbeing and quality of life of young people and their families
and friends, as well as improvements to the identification and treatment of mental health problems for
young people and improved pathways to care through service integration and accessibility. In the long-
term, the model is intended to drive enhanced service provision and access, to improve health outcomes
and to increase social and economic outcomes for young people over their life course.
Figure 1: Summary of the headspace program logic

Source: KPMG 2022, adapted from headspace National

Evaluation domains of
inquiry
This evaluation targeted four domains of inquiry. For each domain, a range of evaluation questions were
specified and have been answered through this evaluation. These questions fall broadly into three
categories of evaluation, being process evaluation, economic evaluation, and outcome evaluation.
Statistical methods, rather than an experimental design, have been used for the evaluation, due to project
timeframes and the absence of pre-existing data linkage arrangements.
Figure 2: Overview of the evaluation design

Source: KPMG 2022

Domain 1: Understanding headspace

This domain of inquiry utilises process evaluation methods, focusing on program design documentation
and administrative data and literature review to test alignment to need and model fidelity.
The following evaluation approach was applied:
• Exploring the design of the headspace model, and evaluating the model against the mental health and
wellbeing needs of young people in Australia.
• Detailing the program’s reach and take-up over the five year period, including analysis of who accessed
support through the headspace model, what support they received and who provided the support.
• Examining variation in geographical spread and the characteristics of young people accessing the service.

Domain 2: Effectiveness of headspace in achieving program

outcomes
This domain of inquiry utilises outcome evaluation methods, with a ‘pre-post’ design to explore the
difference the model makes for young people in each outcome area, looking at a comparison of each
outcome before and after they engage with headspace. This domain focuses on the self-reported
improvements of young people against each outcome area, service providers’ observations of the model’s
success in improving these outcomes, and clinical data reported by service providers and young people as
part of accessing the headspace service.
The focus of this domain is to test the effectiveness of the headspace model.
The following evaluation approach was applied:
• Exploring the evidence that the model achieves intermediate outcomes, such as increased mental health
literacy and early help seeking, and increased access to mental health support.
• Evaluating the extent to which the model achieves its intended outcomes in being appropriate, youth
friendly and accessible.
• Assessing the extent to which these outcomes apply equally to ‘hard to reach’ groups, comprising young
people with demographic characteristics associated with reduced help seeking, often due to
experiences of stigma, discrimination and systemic racism. For this evaluation, ‘hard to reach’ groups
include:
o Aboriginal and Torres Strait Islander young people;
o young people from culturally and linguistically diverse backgrounds;
o young people who identify as LGBTQIA+; and
o young people with disability . 0F

• Evaluating the extent to which the model is effective in improving pathways to care for young people.
• Evaluating the clinical evidence that the model is effective in improving mental health and wellbeing
outcomes for young people.

Domain 3: Cost-effectiveness & value

This domain of inquiry utilises economic evaluation methods and draws on the clinical outcomes analysis
conducted in Domain 2, along with cost data estimates obtained through interviews, surveys and analysis
of administrative records.
The focus of this domain is to test the value of the headspace model.
The following economic evaluation approach was applied:
• Defining the program’s target population to young people with predominantly mild to moderate mental
health needs that fall within the scope of services provided by headspace.
• Defining the comparator as ‘the world in the absence of the headspace program’ in which some, but not
all, young people would access mental health treatment.
• Designing an evaluation framework that has the capacity to capture two key effects of headspace
presence: the benefits of mental health treatment it provides, and the improved accessibility of
treatment relative to the comparator.
• Estimating the cost of delivering headspace services, including direct and indirect costs funded through
the core grant, Medicare Benefits Schedule (MBS) funding, through Primary Health Networks (PHNs)
and other sources, with the goal of establishing the cost per occasion of service (OOS) of headspace
mental health treatment.
• Converting the mental health outcomes as observed in the headspace Minimum Dataset (hMDS) to
quality-adjusted life years (QALYs) for the calculation of an incremental costeffectiveness ratio
(ICER) which is the standard outcome for expressing value for money of health policies and
interventions.
• Extrapolating QALYs gained from treatment over 12 months after the last observed health outcome data
point.
• Defining the consequences of not accessing mental health treatment (or achieving Minimum Adequate
Treatment (MAT) levels of treatment) and estimating the associated costs.
• Assessing the cost-effectiveness of the headspace model, estimating the value of the treatment services
provided by headspace services and using clinical outcomes to estimate improvements in quality of
life associated with seeking support through the headspace model.

Domain 4: Future enhancement

This domain of inquiry brings together findings across the evaluation activities, and considers them in the
context of broader recent analysis of the mental health services sector. Drawing on literature review and
program design documentation, and considering qualitative and quantitative findings across the first three
domains of inquiry, this domain is focused on testing the sustainability of the model.
The following evaluation approach was applied:
• Reviewing the components of the model against the findings of the effectiveness and value analyses to
identify barriers and enablers associated with the headspace model.
• Exploring external factors that have impacted the headspace model and its overall performance in
achieving its intended outcomes.
• Assessing whether introducing changes to either the design or implementation of the headspace model
could improve its associated outcomes and value.
• Considering broader system changes that would support the headspace model to better meet its
objectives.

Data and methodology

considerations
Evaluation methodology
As described above, this evaluation focused on headspace services for the period from 1 July 2015 to
30 June 2020. A mixed methods approach was used to collect data across the evaluation period with the
following data collection activities undertaken:
• review of program documentation;
• consultation with policy owners and the mental health sector;
• deep dive consultations with six headspace services and their local stakeholders;
• a survey of young people who have and have not used headspace;
• a survey of headspace services and their lead agencies;
• focus groups and interviews with young people who have and have not used headspace services, school
and university counsellors, and General Practitioners (GPs);
• analysis of the hMDS;
• an area-level effectiveness analysis; and
• an economic evaluation of headspace cost-effectiveness.
Each of these data collection activities are detailed in Appendix A.

Data considerations and limitations

There are several important data considerations and limitations that should be considered in conjunction
with findings documented in this report.

Time period for the evaluation

This evaluation specifically considered the period from 1 July 2015 to 30 June 2020 for headspace
services. The evaluation period concluded on 30 June 2020 to allow for collection of data relating to full
financial years, when data collection and extraction activities commenced in the first half of 2021. There
has been significant change both during and following the evaluation period with ongoing mental health
reform, increasing numbers of headspace services being established, and the start of the COVID-19
pandemic. These changes are important contextual considerations and have been referenced where relevant
within this report. While the evaluation focused on the period between 1 July 2015 and 30 June 2020, the
period in which the evaluation was undertaken extended from July 2020 to May 2022, and different
evaluation activities took place within different timelines. For example, consultation with headspace
service stakeholders took place in the first half of 2021, and was supplemented with an additional survey
of headspace services and lead agencies in late 2021, while consultation with young people took place in
the second half of 2021. It should also be noted that stakeholders consulted did not always have
involvement with, or knowledge of, headspace services for the full period of the evaluation from 1 July
2015 to 30 June 2020, which may have impacted reflections from some stakeholders consulted.

Approach to analysis of hMDS data

Analysis of the hMDS specifically considered all episodes of care which commenced between 1 July 2015
and 30 June 2020, and which had been completed by 9 December 2020. This was the base dataset used for
the evaluation, with different inclusion and exclusion criteria applied for different analyses, where
required. For example, analysis of clinical outcomes for young people using headspace services considered
a subset of this full dataset, where young people had received at least two occasions of services within
their episode of care, to ensure that both pre-treatment and posttreatment clinical scores were available.
Where a subset of the base dataset was used for specific analysis, this is highlighted within the report
(Appendix F provides further detail).

Key data limitations

There are several key data limitations documented within this report and detailed in Appendix A. Key data
limitations include:
• Data linkage was a preferred evaluation method, to compare outcomes of young people using headspace
services to those of young people who have not used headspace services. However, whether personal
data collected from young people can be used to support data linkage within current consent processes
has not been investigated. In addition, to undertake data linkage for this evaluation, data would have
had to be collected from individual headspace services for linkage. It was estimated by the data
linkage authority that this type of data linkage would take approximately 18 months to complete,
which was not feasible for evaluation completion. The arealevel effectiveness analysis was undertaken
in place of direct data linkage with other key datasets.
• There is variable compliance with data collection for the hMDS. This variation occurs between different
headspace services, data items, and young people. COVID-19 also reduced completion rates for
surveys provided to young people for the last period of the evaluation.
• The hMDS has been updated over time, with new data items collected and the definition of data items
changing during the evaluation. Some data is not comparable across the full evaluation period or is
only available for the last financial year within the evaluation period.
• Completion rates are very low for the follow up survey provided to young people three months after an
episode of care is completed, at approximately four per cent, and young people who have experienced
better outcomes from headspace services completing the survey at higher rates.
• There is no consistent collection of data across headspace services related to the cost of delivering
headspace supports. For example, the cost of MBS items is not identified, and there is no data
collection for other indirect and in-kind costs incurred. While the evaluation sought this information
directly from headspace services through deep dive consultations and the lead agency and service
survey, very few services were able to provide cost breakdowns.

Evaluation findings
Figure 3: Findings at a glance
Source: KPMG 2022

Figure 3, above, presents the key findings across the four domains of inquiry. Further commentary on each
follows.

Domain 1: Understanding headspace

The design of the headspace model has been well articulated and is in line with international standards for
the provision of youth-friendly care. There is evidence of high levels of demand for mental health services
for young people, and different levels of need from young people across different demographic groups.
The broader literature supports the headspace model’s identification of a number of priority groups for
active engagement, and the design of the model aligns to the mental health and wellbeing needs of young
people in Australia. Stakeholder perceptions of the value and intent of the headspace model are well
aligned to the intended outcomes and objectives of the model, which are clearly defined in the program
logic underpinning the headspace Model Integrity Framework (hMIF).
The most significant changes to how headspace services are implemented were the introduction of local
PHN commissioning in 2016, along with a complex distributed governance model, and the introduction of
new delivery models for headspace services, including satellite services and outreach models.
Over time, the reach and take-up of the model have improved. With increased investment from
government, there has been significant growth in headspace services, from 98 in June 2016 to 118 in June
2020 and 154 services in operation by 1 May 2022 . The number of headspace service locations per
1F

jurisdiction also broadly aligns to the population size for young people.
At a national level, mental health services (57.5 per cent) provided through the headspace model between
1 July 2015 and 30 June 2020 greatly outweigh alcohol and other drugs (AOD) services (0.4 per cent),
vocational support (2.2 per cent) and sexual and physical health (1.8 per cent). This mix of services
provided through the headspace model has remained largely consistent over time, with the exception of
outer regional and remote services where a greater proportion of vocational (8 per cent) and sexual and
physical health (13 per cent) support services have been provided compared to other regions, and fewer
mental health services (38 per cent). In all cases, however, mental health services comprise a greater
proportion of services provided. A significant proportion of OOS are also made up of intake and
assessment (23.7 per cent).
Most services provided through the headspace model are provided to an individual young person, rather
than to families or groups (74 per cent of OOS in 2019-20), and most services are provided face-to-face
(60 per cent of OOS in 2019-20). The proportion of services delivered face to face was considerably lower
in 2019-20, due to the impacts of COVID-19 and the resultant shift to online and telehealth service
delivery. In the months from July 2019 to February 2020, face-to-face sessions made up 79 per cent of
OOS delivered (noting that 16 per cent of OOS had missing service mode information).
Overall, the headspace model is well designed, aligned to the mental health needs of young people, and
has a reach and take-up which has increased over time, in line with government investment and increased
demand.

Domain 2: The effectiveness of headspace in achieving program

outcomes
As part of this domain of inquiry, the effectiveness of measuring outcomes through the headspace model
was evaluated. The evaluation found:
• data is collected and disseminated across a broad range of activities;
• gaps in activity data prevent measurement of some elements of the headspace model, including
community engagement and services integration; and
• longer-term outcomes associated with the model are not measured.
Each of the outcome areas set out in Figure 4 below were separately evaluated.
Figure 4: Outcome areas in scope for this evaluation

Source: KPMG 2022

Intermediate outcomes – summary of findings

The evaluation found the following with respect to intermediates outcomes:
• The headspace model is effective in supporting intermediate outcomes for the general population of
young people. These outcomes include mental health literacy, early help seeking and increased access
to required services, which in turn improve the likelihood that young people will seek support with
their mental health and achieve improved psychosocial outcomes in the longer term.
• The headspace model achieves more mixed success in supporting these intermediate outcomes for ‘hard
to reach’ groups.

Service system outcomes – summary of findings

The evaluation found the following regarding service system outcomes:
• The headspace model is effective in supporting youth mental health through advocacy and promotion
activities, and stigma reduction activities undertaken as part of the headspace model are also effective.
The model is also recognised as providing a range of additional contributions to local communities
that are highly valued by those communities.
• The headspace model has mixed effectiveness in areas related to the broader service system in which it
operates. The implementation of the model is often impacted by the broader service system in which it
operates, particularly in regard to:
o improving pathways to care;
o providing a localised service offering; and
o supporting a ‘no wrong door’ approach that assists young people to access the most appropriate
support.
• These outcomes are constrained by the capacity of other services, workforce shortages, and difficulty in
attracting MBS billing staff, which is exacerbated in regional and remote areas.
• These challenges lead to increased wait times for services and reduce the generally high levels of support
the model receives from other primary care and mental health providers.

User experience outcomes – summary of findings

The evaluation found the following regarding user experience outcomes:
• The headspace model provides a highly appropriate mental health service approach for young people
with mild to moderate, high-prevalence conditions.
• The model successfully supports the participation of young people in the design and delivery of
headspace services, which is associated with strong, positive views as to user experience.
• The model has mixed success in providing culturally appropriate and inclusive supports for young people
from culturally and linguistically diverse backgrounds, and Aboriginal and Torres Strait Islander
young people.
• The model is reasonably effective in enabling young people to access support where, when, and how they
want it, however opening hours and waiting times detract from this.

Psychosocial outcomes – summary of findings

The evaluation found the following with respect to psychosocial outcomes:
• Young people benefit from more engagement and treatment through the headspace model, which is
associated with greater improvements in mental health and wellbeing.
• For young people who access six OOS or more, headspace is associated with similar improvements in
mental health and wellbeing as comparable psychotherapy treatments.
• The largest proportion of young people accessing the headspace model only attend once (36 per cent of
episodes of care within the data period were a single OOS), and only 19 per cent of episodes of care
are for six or more OOS.
• The model is associated with positive psychosocial outcomes for young people, however, for those young
people accessing the service who met clinical thresholds (moderate or above), the majority do not see
a clinically significant change to their outcomes.
• Young people who present with high levels of mental distress and who go on to access multiple sessions
(at least six to eight sessions) achieve the greatest improvement in outcomes.
• Clinical outcomes, although positive, are not as strong for LGBTQIA+ young people as they are for the
general population of young people accessing the headspace model.
• Clinical outcomes for Aboriginal and Torres Strait Islander young people are not so obvious. When using
the K10 outcome measure, this cohort achieved statistically similar outcomes as the general
population of young people accessing the headspace model. However, outcomes are not as strong for
Aboriginal and Torres Strait Islander young people when using the Social and Occupational
Functioning Assessment Scale (SOFAS) and the MyLifeTracker (MLT) outcome measure.
• There is some evidence that headspace has a positive effect on some area-level outcomes, such as
reducing substance abuse hospitalisations and the number of self-harm hospitalisations. However,
these impacts are not consistent when looking at alternative measures of the headspace treatment
effects, such as the number of headspace clients per 1,000 12 to 25 year olds and the ratio of MBS-
funded mental health services provided by headspace to MBS-funded mental health services provided
outside headspace. These results should not be considered conclusive regarding the impacts of
accessing headspace services.

Overall effectiveness in achieving outcomes

As a set of objectives, these outcomes represent key outcomes across the headspace program logic which
drive engagement, service experience and clinical improvements in mental health.
Overall, analysis of extensive qualitative and quantitative evidence demonstrates that the headspace model
is effective in achieving many of these intended outcomes. There is some inconsistency, however, in
outcomes for different groups, and across some aspects of the model program logic.
Intermediate and user experience outcomes are well supported for young people in the general population;
however, this is not the case for young people from ‘hard to reach’ groups. Psychosocial outcomes
improve for young people from the general population accessing the headspace model, however, in most
cases, these are not clinically significant. The short episodes of care most young people experience may be
a factor in these modest clinical improvements, given that more engagement and treatment through the
headspace model is associated with stronger outcomes.
The quality of user experience for young people accessing headspace services is reduced where there is
high demand for services and challenges in attracting a multi-disciplinary workforce, which increase wait
times. Similarly, service system outcomes are not as well supported by the model, however, this is in large
part due to pressures felt across the broader mental health services sector.
Domain 3: Cost-effectiveness & value
Estimating the cost of delivering headspace
The full cost of delivering headspace includes the national headspace grant; any additional funding that a
PHN, state or federal government may provide to deliver core services, activity-based funding of services
through the MBS, in-kind contributions; private donations, and any out-of-pocket payments made by
young people or their carers.
No single source captures these ranges of costs of delivering headspace. The Department of Health (the
department) records the national headspace grant costs but does not have oversight of the division of the
grant between service provision and indirect costs, such as rent and utilities, office expenses and
community awareness expenses, as this is held at the service level. The hMDS identifies the funding
source for each OOS provided by headspace but does not capture the value of funding for that OOS. Any
in-kind contributions to headspace services, for example free use of physical space, can only be provided
by the service itself and may be prone to a range of data quality issues (e.g., definition and quantification
of in-kind support may vary).
In 2019-20, 112 headspace services included in the cost analysis delivered 401,325 OOS. The average cost
per OOS was approximately $307. The average direct cost per OOS was $230 under the assumption that
the direct service costs account for 75 per cent of the total cost. This is twice as much as the MBS fee (and
any out-of-pocket costs) for a typical mental health session . However, the average direct cost per OOS is
2F

slightly lower than the Australian Psychological Society’s recommended fee for a 46 to 60 minute session
of $260 .
3F

How cost-effective is headspace?

Results of the cost-effectiveness analysis show that, over an 18-month time horizon and after adjusting for
regression to the mean (RTM), the ICER was $44,722 per quality adjusted life year (QALY) gained. While
Australia does not have an explicit cost-effectiveness threshold for public healthcare funding decisions,
experience shows that this result is cost-effective when compared to thresholds considered for other
,
similar healthcare services .
4F 5F

Given the parameters required for performing the headspace economic evaluation, many of which are
unknown or uncertain, the base case ICER is based on conservative assumptions. On the one hand, there
are a number of considerations that indicate headspace may be more cost-effective than what the base case
suggests. These considerations are generally associated with greater benefits of mental health treatment
than in the base case modelled. For example:
• Allowing for the treatment benefit to last longer, for up to five years, results in an ICER of $20,205 per
QALY gained.
• Removing the RTM adjustment (i.e., assigning all observed benefit to headspace treatment) results in the
ICER dropping to $32,567 per QALY gained
• Allowing for a partial benefit from an incomplete treatment consisting of two OOS produces an ICER of
$35,713 per QALY gained. On the other hand, the base case evaluation uses the available data to
support an assumption that only three OOS are sufficient for a course of treatment to meet the MAT
requirement. This assumption favours headspace in light of the literature that suggests that at least four
OOS are required. Changing the assumption to match the literature results in the ICER of $56,894 per
QALY gained.
Furthermore, the full cost of providing an OOS by headspace could not be determined within the
evaluation, as there is no data on the actual cost of MBS-billed services, in-kind and indirect funding. If
additional costs were incurred by headspace services, this would result in the ICER increasing. Sensitivity
analysis conducted indicates this may increase to $54,693 per QALY gained, when additional costs are
accounted for.
Sensitivity analyses have notably shown that the key unknowns of the economic evaluation (the proportion
of young people not receiving care in the ‘no headspace’ scenario, the relative effectiveness of treatments
provided outside of headspace, and their cost) are not key drivers of the model outcomes. When explored
within their plausible value ranges, these parameters had only minor impacts on the ICER.
There is a large variation in cost-effectiveness across services. This stems both from the variation in cost
per episode of care and the variation in outcomes. As discussed in the cost analysis section, under the
current funding model, all services receive relatively similar annual funding amounts, regardless of the
volume of services they deliver. Even assuming outcomes are similar across services, this alone can lead to
a large variation in average cost per OOS (larger services would be more cost efficient than smaller
services). The effectiveness and the cost-effectiveness analysis further show there is also considerable
variation in outcomes and QALYs gained across services. This may be due to the extrapolation of benefits
beyond the last observed outcome at the follow up time, which amplified QALY gains in services with
better treatment outcomes and exacerbated the variation in cost-effectiveness across services.

Domain 4: Future enhancement

In reviewing the components of the headspace model against the findings of the effectiveness and value
analyses, consistent barriers and enablers to the success of the model have been identified in relation to:
• community awareness and engagement;
• providing four core streams of services;
• service integration;
• the national network model;
• attracting and retaining a multi-disciplinary workforce;
• the blended funding model; and
• monitoring and evaluation.
Challenges associated with these areas of the headspace model interact to increase wait times for services
and to reduce outcomes for ‘hard to reach’ groups.
At the same time, a range of external factors put pressure on how the headspace model works in practice.
Limited referral pathways available in many areas, broader mental health workforce shortages, high
demand for services and complexity of presenting need all drive increased wait times and reduce access to
service. Stigma and discrimination in the community against those with mental illness continue to impact
early help seeking, particularly affecting young people from culturally and linguistically diverse
backgrounds and Aboriginal and Torres Strait Islander young people, reducing their early help seeking
behaviour. Within the headspace model, the role of service providers requires diligent effort to compensate
and adjust for these external factors and to ensure the objectives of headspace are met.
On balance, however, this evaluation has not found any evidence to suggest that changes are required to
the design of the headspace model in order to enable it to meet its objectives. Despite challenges in
meeting the needs of some cohorts, and constraints and limitations brought about by broader mental health
system issues, headspace is achieving its intended outcomes with its current design. However, given the
challenges, enablers and barriers faced by the headspace model, and the low cost-effectiveness of the
model overall, there are several areas where implementation of headspace services could be enhanced to
enable it to meet its objectives more efficiently.

Recommended changes to the implementation of the headspace model

The evaluation findings point to several key areas in the implementation of the headspace model which
require further development to optimise the model’s ability to meet its desired objectives.
As discussed above, while the headspace model is effective overall, the needs of ‘hard to reach’ cohorts of
young people are not as effectively met by the headspace model as those of young people in the general
population attending headspace. Wait times are also an area of criticism for the model and the complex
governance arrangements are burdensome. Given that psychosocial outcomes are strongly associated with
engagement and treatment through the headspace model, there are opportunities to improve user
experience and clinical governance arrangements to support longer episodes of care for young people
where appropriate.
Table 1: Recommended changes to the implementation of the headspace model

Recommendation - ‘Hard to Reach’ Groups

1. The headspace model has had mixed success in reaching and supporting young people from ‘hard to
reach’ groups. Enhancing representation of these groups within the workforce may support
engagement and ongoing support for young people who identify as part of ‘hard to reach’ cohorts.

Lead agencies and headspace services should draw on PHN needs analyses to prioritise their
workforce needs, and implement strategies to diversify the headspace workforce to be representative
of the local community and to lead engagement with relevant ‘hard to reach’ groups.

Recommendation – Service Integration

2. There is a need to further enhance integration with headspace services and local mental health and
other service providers. This should build on the current service integration piloting and evaluation
activity underway through the IAR and the PHN regional commissioning role. It should also consider
the National Agreement, and bilateral agreements developed with each state and territory in relation
to specific strategies to support service integration.

This would support access to more appropriate initial connections to services for young people and
provide greater clarity for referrers locally. It would also support regional service connections and
providers’ understanding of services and supports available during and following a young person’s
episode of care (EoC) with headspace.
Recommendation - Governance and Commissioning
3. This evaluation has identified tension between different stakeholders regarding the agility of the model
to address local needs, and constraints on the capacity to tailor headspace services locally.

Government should work with PHNs and headspace National to undertake a refresh of roles and
responsibilities across the network. This should focus on clarifying the scope of roles in planning,
commissioning, delivering and tailoring headspace services.

4. There is a high degree of consistency of service mix across headspace services, with AOD, physical
and sexual health and vocational support representing a very low proportion of services provided.
Stakeholder feedback has suggested this may not always reflect local or regional need, and that
headspace service planning inconsistently draws on PHN needs analyses to inform and update the
local headspace service mix of the four core streams. It would be expected, for example, that a region
with significant substance misuse issues for young people may need a greater mix of AOD support
services at the local headspace service, or similarly where there are areas with higher rates of chronic
health issues in younger populations, physical and sexual health services should be appropriately
prioritised.

Government should consider investing in an implementation refinement project to explore how the
PHN local lens could be better used to commission a model consistent with the hMIF that responds to
 identified regional need. This could allow greater capacity to reflect the PHNs’ local needs analysis
and the local service landscape, including areas of high need. The project should consider the
potential risks of reducing the consistency of costs and outcomes across headspace services and
ensure mechanisms are in place to maintain a level of fidelity to core elements of the headspace
model.
5. Whilst there was overall improvement in mental health outcomes for young people accessing
headspace services, reliable improvement and clinically significant change results were lower than
expected. This suggests that clinical governance and the quality control of the delivery of evidence-
based interventions could be enhanced.

PHNs should take an active role in ensuring that headspace lead agencies prioritise clinical
governance which ensures quality service provision and adherence to evidence-based approaches.
With support and monitoring from PHNs, lead agencies should formalise processes to regularly
monitor efficacy, performance against outcomes benchmarks and evidence-based approaches, where
these are not already in place. This could be achieved through mechanisms such as: ensuring
interventions meet recommended practice guidelines; setting and achieving clear benchmarks for
outcomes; regularly monitoring service outcomes data; and supporting staff to access focused
training and supervision.
• Recommendation - Monitoring and Evaluation
6. Despite extensive reporting undertaken across activities within the headspace model, a number of gaps
in data collection were identified through the evaluation. Filling these gaps could support better
monitoring and evaluation of outcomes associated with the headspace model.

The following data should be collected by headspace National to inform future evaluation and
continuous improvement processes:
• outreach and engagement activity data – including activity type, duration, and number of young
people participating;
• outcomes data beyond 90 days post EOC – with a particular focus on episodes involving a single
OOS;
• reason for closure data – to differentiate between unplanned exits and planned exits;
• referral data – service type referred from and to, stage in care at point of referral (e.g., intake, mid-
treatment, exit), whether referral onwards was taken up;
• demographic data – enabling service users to identify as having disability, and to identify as
neurodiverse;
• funding data – capturing ongoing, in-kind support and specific MBS items claimed through
headspace services in hAPI; and
• workforce data – capturing more detailed workforce information including full-time equivalent
workforce available and their characteristics.
The extent to which the needs of young people are being met at an area-level, as estimated through
PHN local needs analysis, should be considered a priority monitoring activity by PHNs.
7. While data is collected extensively across activities within the headspace model, the longer term
impacts of headspace are not measured.

Data from headspace should be collected in a way that allows it to be linked to other datasets, so that
outcomes over time of young people who access headspace can be better understood when compared
to those who do not access headspace. Ethical considerations should also be prioritised, for example
to ensure that individuals cannot be identified in the data. The administrative burden of additional
data collection activities for providers and young people accessing headspace should be balanced
against the benefits provided through enhanced reporting. Linked data sets might include:
• self-harm hospitalisations;
• substance abuse hospitalisations;
• suicide deaths;
• MBS mental health services accessed;
• PBS usage;
• mental health related emergency department presentations;
• education and employment outcomes; and
• income support use.
Data linkage should be supported by government, and should be complementary to data linkage
being conducted under the National Agreement.
8. A number of areas across the headspace program logic could benefit from further evidence to
understand the best implementation approach to support improved outcomes for young people.

Data linkage should be supplemented by studies using experimental or quasi-experimental designs so

that outcomes can be rigorously measured and attributed to headspace. Where this is not achievable
through control or comparison group analysis using linked data, government should allocate funding
for one-off experimental studies. Priority examples include:
• exploring differences between centre and satellite headspace services;
• research into single session interventions, given that approximately 36 per cent of episodes of care
 have a single OOS, and wait times lead to disengagement of young people before treatment;
• examining how AOD, physical and sexual health and/or vocational assistance support mental health
and wellbeing, both in the short and medium-to-long-term;
• exploring the most appropriate intake and assessment approaches when engaging with Aboriginal
and Torres Strait Islander young people;
• exploring the most reliable measures of mental health and wellbeing in Aboriginal and Torres Strait
Islander young people, for use within headspace;
• examining the extent to which young people and families experience more streamlined and less
fragmented pathways of care in the medium-term.
• Detailed logic documents should be developed to support the collection of appropriate data.

Source: KPMG 2022

Recommended changes to current funding arrangements

Services are currently funded through a blended funding model, including core grants received from the
department, through PHNs as the commissioning body, use of MBS billing by practitioners providing
supports through headspace services and other funding sources, including additional grant or project
funding from PHNs.
Currently, there is no specific funding model used to determine the grant contributions made by the
Commonwealth to headspace services. A model was previously used; however, this has been moved away
from in recent years, and all headspace services now receive similar volumes of grant funding, according
to the type of headspace service, with little variation. One-size-fits-all approaches to providing funding to
headspace services are not cost-effective, and this is demonstrated by the significant variability in cost-
effectiveness between individual headspace services.
In addition, headspace services have varied success in making use of the blended funding model. Some
services provide considerably more OOS than other services while receiving similar grant funding, as a
result of MBS-billed services from private practitioners. In other services, a model that relies heavily on
MBS billing is not viable or sustainable as there are local workforce shortages, which impact the ability for
these headspace services to deliver MBS-based clinical services.
To address these issues, a new funding model should be developed to guide funding for all headspace
services moving forward. The funding model should be flexible and consider the individual characteristics
of each headspace service.
Table 2: Recommended changes to funding for the headspace model

Recommendation – Funding Arrangements

9. headspace services do not currently collect or report the full costs of operation, with in-kind
contributions and indirect costs not captured under funding agreement requirements. Without
accurate data regarding the full costs of operating a headspace service, the cost-effectiveness of the
headspace model can only be estimated, as has been done through this evaluation.

Government should prioritise the collection of full and accurate data to inform a more detailed review
of current cost information across all headspace services. This could be done through individual
engagement with headspace services, or compulsory survey of all headspace services. This would
confirm current costs of delivering the headspace model, including in-kind contributions provided to
services and other indirect costs. This would also support the identification of differences in costs for
different headspace services based on location, and other service-specific factors. The official count
of headspace services should also be revisited to improve clarity of funding arrangements, e.g., the
count of headspace services could be updated to reflect the number with a Trade Mark Licence Deed.
10. While the headspace model is broadly effective in achieving its intended outcomes, a number of areas
related to funding are challenging for services providing headspace. Difficulty in attracting and
retaining a multi-disciplinary workforce varies across regions, as does the need to undertake
extensive community engagement activities with ‘hard to reach’ groups. At the same time, across the
headspace services included in this evaluation, the number of OOS funded each year varies widely,
while funding levels within the core headspace grant are relatively consistent across services. This
variation in demand and service provision leads to considerable differences in the estimated
economic efficiency across headspace services.

Government should develop a variable funding model based on demand and regional need which
accounts for differences in location, population and service delivery modes and volumes. This should
consider core funding components, such as administrative costs and management costs, as well as
more variable cost components which may include:
• location of the headspace service, including regionality and areas of workforce shortages, with
increased allowance for salaried staff where access to MBS-based staff is challenging;
• the size of the population to be supported by the headspace service, including the number of young
people within the headspace service catchment and geographically proximate communities to be
supported by the service, and associated required service FTE; and
• the headspace service type to be implemented, including whether the service is a headspace centre,
satellite service or outreach service.
A separate funding model, or specific element, should be considered for establishment costs required
for a new headspace service.
Government should consider how a revised funding model may apply to established services, in
 addition to new services established going forward.
Source: KPMG 2022

Broader system changes required

There are also a range of broader system-level changes that are currently underway across sectors that
would support headspace to meet its objectives going forward. These factors are not within the remit of
individual headspace services, or the headspace program overall, to control but would benefit headspace as
part of the broader mental health service system. These changes include:
• increased prevention and early intervention services;
• improved service integration and pathways; and
• development of national mental health workforce.

In conclusion
This evaluation has examined the headspace model across several criteria. A range of data and evidence
has been analysed to assess the model's alignment to need and the fidelity of the model in practice,
including in terms of take-up and reach of service provision. The effectiveness of the headspace model has
been assessed against intermediate outcomes, service system outcomes, user experience outcomes and
psychosocial outcomes achieved. The economic value of the headspace model has also been assessed,
alongside the model’s ongoing sustainability.
Through the range of methods and analyses applied, this evaluation concluded that the headspace model
provides a comprehensive and complete set of components to address the mental health needs of young
people. The model incorporates components which are designed to prevent mental illness, through mental
health literacy, early help seeking and stigma reduction, and to treat mental illness whatever the presenting
need. While the model is intended to support young people with mild to moderate high-prevalence mental
health conditions, through the 'no wrong door' approach and as a result of capacity pressures across the
mental health service sector which constrain referral pathways, every young person presenting at a
headspace service, including those with more severe mental health conditions, receives support of some
kind.
When outcomes are examined, young people from 'hard to reach' groups continue to be less well served
through the model, across outcome areas. The model achieves its intended outcomes for the general
population of young people across domains, and the cost-effectiveness of direct services provided through
the headspace model is on par with established benchmarks on cost-effectiveness ratios. When longer-term
benefits are included in analysis, the headspace model may be cost-effective, but more data is required to
substantiate this.
While the model is associated with positive psychosocial outcomes for young people, the majority do not
see a clinically significant change to their outcomes. In general, associated psychosocial outcomes only
become comparable to other psychotherapies once six or more sessions have been accessed.
There are opportunities to improve the efficiency and effectiveness of the model, through targeting the key
areas of 'hard to reach’ groups, service integration, governance and commissioning and monitoring and
evaluation. Pressures and reforms in the broader mental health services sector currently, and will continue
to, affect the headspace model. In its role as a national program to support the mental health and wellbeing
of young people, there is an opportunity to greater leverage the headspace platform for broader reform in
the sector.

2 Introduction
2.1 Evaluating the headspace
model
2.1.1 Overview
KPMG and its research partners, the Social Policy Research Centre at the University of New South Wales,
and batyr, were commissioned by the Commonwealth Department of Health (the department) to evaluate
the national headspace program, as delivered through headspace services.
The headspace program has been evaluated twice before, in 2009 and 2015. The current evaluation – the
subject of this report – seeks to build on these prior analyses to explore how the model operates currently,
and the impact of ongoing changes in its design, reach and priorities on the availability of highquality,
effective mental health care for young Australians.
Funding for headspace services and supports has grown as mental health investment has been prioritised in
recent years by the Australian Government. In this context, evaluating the model represents an important
opportunity to take stock of what is being delivered at individual services and across the headspace
network, and how this aligns with the core intent and expectations of the headspace program.
This evaluation focused on headspace service provision, as provided in individual services around
Australia, during the period from July 2015 to end of June 2020. Several aspects of the broader program
were explicitly out of scope, including the operations and performance of headspace National and
eheadspace. Other programs were also excluded from this evaluation, including the Individual Placement
Support (IPS) trial funded by the Department of Social Services (DSS), and the Early Psychosis Youth
Services (EPYS) Program provided at selected headspace services. These initiatives have been the subject
of separate evaluations.
It should also be noted that, while the evaluation primarily considered the period from 1 July 2015 to 30
June 2020, there were challenges associated with ensuring all stakeholders relate their views only to this
period. Stakeholders, who are described in more detail in Appendix A, were engaged following ethics
approval for the evaluation being granted in May 2021 through to December 2021. There may be
differences between these views and the data captured through headspace services between 1 July 2015
and 30 June 2020.

2.1.2 Current environment impacting the evaluation

It is also important to consider the context in which this evaluation was completed. The period between
2020 and 2022 was impacted by a range of factors, including the black summer bushfires in 2019-20,
which created increased demand for mental health services. This was also exacerbated by the COVID-19
pandemic, which has had significant impacts on service delivery, and increased focus on mental health.
The evaluation scope was also directly impacted by COVID-19, with the last four months of the evaluation
period from March 2020 to June 2020 being part of Australia’s first pandemic wave, with lockdowns and
restrictions in place. During this period, services including headspace services, were required to shift
service modalities to provide telehealth and virtual services. In addition, the number of occasions of
service delivered by headspace may also have been impacted, with fewer young people able to access
services.
Since the opening of the first headspace service in 2007, there has also been broader sector reform. Some
of the significant recent changes and developments include:
• the establishment of the National Mental Health Commission (2012) and its review of mental health
services in 2015;
• the endorsement of the Fifth National Mental Health Plan in 2017, committing all Australian
Governments to eight priority areas ; 6F

• the Productivity Commission’s review into Mental Health (with the final report publicly released in
November 2020);
• the Victorian Royal Commission into Victoria’s Mental Health System (final report delivered in February
2021);
• the National Mental Health and Suicide Prevention Plan (released in May 2021) and the work of
Australia’s National Suicide Prevention Adviser;
• the Select Committee on Mental Health and Suicide Prevention’s Inquiry into Mental Health and Suicide
Prevention (Final Report released in November 2021); and
• the National Mental Health and Suicide Prevention Agreement . 7F

These developments in Australia’s mental health landscape are important factors to take into account for
this evaluation of the headspace model. At the same time, this evaluation will help inform policy and
investment decisions about the future direction of the headspace model. The evaluation outputs will also
feed into the reform agenda shaping mental health service delivery in Australia for the next decade and
beyond.

2.1.3 Evaluation domains of inquiry

This evaluation is targeting four domains of inquiry. For each domain, a range of evaluation questions
were specified and have been answered through this evaluation. These questions fall broadly into three
categories of evaluation, being process evaluation, economic evaluation, and outcome evaluation, using
statistical methods rather than an experimental design, which is unfeasible within the project timeframes in
the absence of pre-existing data linkage arrangements.
Figure 5: Overview of the evaluation design

Source: KPMG 2022

Domain 1: Understanding headspace

This domain of inquiry utilises process evaluation methods, focusing on program design documentation
and administrative data and literature review to test alignment to need and model fidelity.
The following evaluation approach was applied:
• Exploring the design of the headspace model, and evaluating the model against the mental health and
wellbeing needs of young people in Australia.
• Detailing the program’s reach and take-up over the five-year period, including analysis of who accessed
 support through the headspace model, what support they received and who provided the support.
• Examining variation in geographical spread and the characteristics of young people accessing the service.

Domain 2: The effectiveness of headspace in achieving program outcomes

This domain of inquiry utilises outcome evaluation methods, with a ‘pre-post’ design to explore the
difference the model makes for young people in each outcome area, looking at a comparison of each
outcome before and after they engage with headspace. This domain focuses on the self-reported
improvements of young people against each outcome area, service providers’ observations of the model’s
success in improving these outcomes, and clinical data reported by service providers and young people as
part of accessing the headspace service.
The focus of this domain is to test effectiveness of the headspace model.
The following evaluation approach was applied:
• Exploring the evidence that the model achieves intermediate outcomes, such as increased mental health
literacy and early help seeking, and increased access to mental health support.
• Evaluating the extent to which the model achieves its intended outcomes in being appropriate, youth
friendly and accessible.
• Assessing the extent to which these outcomes apply equally to ‘hard to reach’ groups, comprising young
people with demographic characteristics associated with reduced help seeking, often due to
,,
experiences of stigma, discrimination and systemic racism . For this evaluation, ‘hard to reach’
8F 9F 10F 11F

groups include:
o Aboriginal and Torres Strait Islander young people;
o young people from culturally and linguistically diverse backgrounds;
o young people who identify as LGBTQIA+; and
o young people with disability.
• Evaluating the extent to which the model is effective in improving pathways to care for young people.
• Evaluating the clinical evidence that the model is effective in improving mental health and wellbeing
outcomes for young people.

Domain 3: Cost-effectiveness & value

This domain of inquiry utilises economic evaluation methods and draws on the clinical outcomes analysis
conducted in Domain 2, along with cost data estimates obtained through interviews, surveys and analysis
of administrative records.
The focus of this domain is to test value of the headspace model.
The following evaluation approach was applied:
• Defining the program’s target population to be young people with predominantly mild to moderate
mental health needs that fall within the scope of services provided by headspace.
• Defining the comparator as ‘the world in the absence of the headspace program’ in which some, but not
all, young people would access mental health treatment.
• Designing an evaluation framework that has the capacity to capture two key effects of headspace
presence: the benefits of mental health treatments it provides, and the improved accessibility of
treatment relative to the comparator.
• Estimating the cost of delivering headspace services, including direct and indirect costs funded through
the core grant, MBS funding, through PHNs and other sources, with the goal of establishing the cost
per OOS of headspace mental health treatment.
• Converting the mental health outcomes as observed in the hMDS to QALYs for the calculation of an
ICER, which is the standard outcome for expressing value for money of health policies and
interventions.
• Extrapolating QALYs gained from treatment over 12 months after the last observed health outcome data
point.
• Defining the consequences of not accessing mental health treatment (or MAT levels of treatment) and
estimating the associated costs.
• Assessing the cost-effectiveness of the headspace model, estimating the value of the treatment services
provided by headspace services and using clinical outcomes to estimate improvements in quality of
life associated with seeking support through the headspace model.

Domain 4: Future enhancement

This domain of inquiry brings together findings across the evaluation activities and considers them in the
context of broader recent analysis of the mental health services sector. Drawing on literature review and
program design documentation and considering qualitative and quantitative findings across the first three
domains of inquiry, this domain is focused on testing the sustainability of the model.
The following evaluation approach was applied:
• Reviewing the components of the model against the findings of the effectiveness and value analyses to
identify barriers and enablers associated with the headspace model.
• Exploring external factors that have impacted the headspace model and its overall performance in
achieving its intended outcomes.
• Assessing whether introducing changes to either the design or implementation of the headspace model
 could improve its associated outcomes and value.
• Considering broader system changes that would support the headspace model to better meet its
objectives.

Further detail
Further detail regarding the evaluation scope and method, including the project governance, data collection
activities and data sources, are provided at Appendix A.
This report provides a synthesis of key findings against each of the four domains of inquiry, with detailed
data analysis and reporting provided in the appendices.

3 Understanding
headspace
In order to understand the headspace model, a range of factors need to be considered alongside the design
of the model itself. As a starting point, this chapter sets out an overview of the model and its primary,
intended outcomes. The target user group of the model is discussed, along with a high-level overview of
the needs of young people attending headspace, the intended outcomes, and objectives of the model for
young people, and how these align to the program logic of the headspace model.
The key features of the model, including its core and enabling components, the support services that are
provided through the model, and the types of headspace services operating around Australia, are presented
at a high level, before the presentation of an overview of how the model has changed over time with
government investment.
The full range of stakeholder relationships of relevance to the headspace model is also described, as a key
aspect of understanding the headspace model in context. This provides a sense of the complexity and
challenge involved in the day-to-day operations of the model within the mental health service sector. The
extent of the stakeholder landscape also demonstrates the breadth of perspectives to consider in evaluating
the effectiveness of the model in achieving its outcomes.
This chapter then provides a detailed breakdown of the support services currently available at headspace
services, and how these have changed over the last five years, since the model was last evaluated.

3.1 Overview of the

headspace model
headspace is often referred to as the Australian Government’s flagship mental health program for people
aged 12 to 25. Since 2006, it has played an important role in efforts to tackle mental ill-health, self-harm,
and suicide among young Australians. Delivered as a network of community-led and governed centres
across Australia, headspace services support young people and their families to access clinical and
community mental health supports and interventions.
headspace provides services across four core streams to provide holistic support for young people. The
four core streams are:
• mental health and wellbeing;
• physical and sexual health;
• work and study support; and
• alcohol and other drug services.
These services can be delivered in-person at headspace services and through telehealth (including online
and telephone services). The provision of these services in multiple formats is intended to help ensure
young people are able to access mental health supports when they are needed, particularly for those young
people who live in regional and remote areas. In addition to these services, separate support is also
provided through eheadspace, a national online and telephone support service delivered by headspace
National. However, as eheadspace is not delivered through headspace services, it did not form part of this
evaluation.
The program supports young people aged 12 to 25 with mild to moderate mental health conditions and
those experiencing episodic or situational need. Young people with more intensive needs who present to
headspace are supported to access other services through partnerships and service system linkage.
Figure 6 below provides an overview of the headspace program logic, outlining the relationship between
elements of the program. The model is designed to achieve a range of short-term impacts, including
improved mental health literacy, increased early help seeking behaviours, the promotion of a positive
experience of service for young people, and improved psychosocial outcomes. These are then intended to
lead to medium-term impacts for the functioning, wellbeing and quality of life of young people and their
families and friends, as well as improvements to the identification and treatment of mental health problems
for young people and improved pathways to care through service integration and accessibility. In the long-
term, the model is intended to drive enhanced service provision and access, to improve health outcomes
and to increase social and economic outcomes for young people over their life course.

Figure 6: Summary of the headspace program logic

Source: KPMG 2022

3.2 What success looks like

for headspace
3.2.1 headspace service users
The headspace model is designed to meet the mental health needs of young people who are deemed at risk
of, or who are experiencing, the early stages of a mental health disorder or who are facing common co-
occurring situational stressors or difficulties .12F

Young people
Target age group
Young people accessing headspace services are between the ages of 12 and 25 years. While this age range
contains very different life stages and required treatment models, the headspace model is designed to
support young people throughout this period, avoiding transitioning them out of the service and into adult
mental health services in a disruptive way .
13F
Presenting need
Prevalence rates for mental illness and psychological distress
Mental illness remains prevalent across all life stages for Australians, however it is most prevalent for 15
to 24 year olds, with rates falling as people age . The prevalence rate of mental illness for 15 to 19 year
14F

olds was 24.4 per cent in 2017, and 23.9 per cent for 20 to 24 year olds, with this rate decreasing into
adulthood. Almost three-quarters of adults with mental illness first experience mental ill-health before the
age of 25 .15F

In the Mission Australia Youth Survey 2016, for the first time, mental health was listed as one of the top
three issues affecting young Australians . Since that time, it has remained an ever-present concern for
16F

young people (aged 16 to 25 years). The Mission Australia Youth Survey 2021 reported that 41.9 per cent
of young people were extremely or very concerned about mental health . 17F

There are some groups of Australians who are more likely to experience mental ill-health. These include
young people, unemployed people, Aboriginal and Torres Strait Islander people and single parent families.
However, mental ill-health can affect anyone, at any stage of life and can be a single episode, episodic or
persistent throughout the person’s life. There are several factors that can also adversely affect mental
health, including biological, environmental, and social factors. Examples include trauma and stress, social
conditions associated with the COVID-19 pandemic, and/or recent natural disasters . 18F

The prevalence of moderate or greater psychological distress has increased over time for young people,
similar to the general population. This has risen from 38 per cent in 2011-12 to 44 per cent in 201718 . 19F

Early reporting from the Intergenerational Health and Mental Health Survey also indicates that younger
Australians (aged 16 to 34 years) were more likely to experience high or very high levels of psychological
distress in 2020-21, with 20 per cent prevalence, compared to 15 per cent for 35 to 64 year olds and 9 per
cent for 65 to 85 year olds . 20F

Implications of mental ill-health for young people

The impact of mental ill-health for young people is profound. The period from 16 to 24 years is an
important transition point for young people, with participation and outcomes significantly affecting
economic and social participation later in life. Mental ill health is the leading cause of disability in people
aged 10 to 24 years , and accounts for almost 50 per cent of the burden of disease in people aged 16 to 24
21F

years . Suicide is the leading cause of death in people aged 15 to 24 years .

22F 23F

Young people experiencing mental ill-health are also at higher risk of disengaging from education or
employment. For example, the 2015 evaluation of headspace found that 20 per cent of headspace clients
were disengaged from employment, education, and training, compared with 11 per cent of the comparable
general youth population . This trend has continued, with approximately 17 per cent of headspace clients in
24F

2018-19 disengaged from employment, education and training at the time of their first OOS with
headspace, compared to 8.4 per cent of all young people aged 15 to 24 years as at May 2019 . 25F

headspace, as an early intervention and prevention model, is designed to assist young people in managing
mild to moderate, high-prevalence mental health conditions . Depression and anxiety are the most
26F

frequently reported mental health conditions for the headspace target cohort, while situational or
contextual stress, such as that associated with family breakdown, with school and work, and related to peer
group dynamics, are also frequently reported for this group. By encouraging early help seeking and mental
health literacy, the model aims to support young people to be able to better manage their emerging mental
health needs and, where possible, prevent their mental health from deteriorating into more acute
conditions. With the headspace model’s ‘no wrong door’ approach , headspace services work with young
27F

people who have a range of presenting needs to assist them to access appropriate care.
Demographic characteristics
There is clear data indicating that mental health outcomes and mental illness prevalence vary with different
demographic characteristics. These can vary greatly as a function of the young person's gender, geographic
location, and cultural background.
Submissions to the Productivity Commission Mental Health Inquiry highlighted that LGBTQIA+ young
people are especially at risk of mental ill-health. Same-sex attracted young people are six times as likely to
have attempted suicide compared with their heterosexual peers . Similarly, almost half of young trans
28F

people had attempted suicide and 80 per cent had self-harmed . Homophobic abuse experienced by young
29F

people has been linked to substance-use, self-harm, and suicide attempts.

Young females are twice as likely to engage in self-harming behaviours than young males, and eating
disorders are the second most common cause of mental ill-health for young females. Young females are
more likely to consider taking their own lives, however young males are more than twice as likely to die
by suicide . 30F

There are also particular challenges for young people living in regional and remote areas accessing mental
health services. Submissions to the Productivity Commission highlighted that there are limited services in
these areas, if any, and this results in long waiting times for support, unsuitable services to match the needs
of people, and the need to travel significant distances to access services . As a result, use of mental health
31F

services is also lower in regional and remote areas. People located in major cities and inner regional areas
use mental health-related MBS services through General Practitioners (GPs) at a rate of 152.2 and 151.5
per 1,000 people respectively. This compares to 118.9 per 1,000 people for outer regional areas, 71 per
1,000 people for remote areas and 33.3 per 1,000 people for very remote areas . 32F

At the same time, in some culturally and linguistically diverse communities, there is significant stigma and
taboo associated with mental illness. This can mean that people from culturally and linguistically diverse
communities are not prepared to share their experiences of mental illness or seek support due to feelings of
shame experienced from family and others around them . It has been suggested that young culturally and
33F

linguistically diverse people are particularly exposed to environmental and social risk factors which can
negatively affect their mental health . 34F

Nationally, headspace National has identified a number of priority groups with which headspace services
are required to demonstrate active engagement. The headspace priority groups are:
• young men;
• sexuality and gender diverse young people;
• Aboriginal and Torres Strait Islander young people;
• young people from culturally and linguistically diverse backgrounds;
• young people with alcohol or other drug issues;
• young people experiencing homelessness;
• young people from rural and/or remote communities; and
• other local populations that are under-represented within the headspace service . 35F

These groups have been identified as less likely to seek support for their mental health needs and more
likely to have barriers to access, such as lacking access to transport or being subject to parental attitudes
discouraging mental health help seeking . 36F

Key components of the model are intended to assist in achieving positive outcomes for these groups. For
example, activities associated with community awareness and engagement, to work with the local
community to increase mental health literacy and reduce stigma, and the focus on providing appropriate
care, which includes the identification and consideration of sociocultural factors, are both designed to
improve outcomes for ‘hard to reach’ groups.

3.2.2 Positive outcomes for young people utilising headspace

services
Intended outcomes of the headspace model
The concept underpinning the headspace model was initially developed through research collaboration led
by Orygen, the National Centre of Excellence in Youth Mental Health, in 2006 . The headspace model has
37F

been described in detail in the academic literature, and various evaluations have been published in peer
reviewed articles that focused on different aspects of the model .The model aligns with the World Health
38F

Organization’s youth-friendly health services framework and protocol for establishing quality standards for
adolescent-friendly health services, which emphasise the need for services to be equitable, accessible and
acceptable to young people, appropriate to their needs, and effective, supplying cross-sectoral, evidence-
,,
based services .
39F 40F 41F

The headspace model is set out in the headspace Model Integrity Framework (hMIF) . A detailed program
42F

logic sets out the aims and objectives of the model and a number of short and medium-term impacts.
This presents outcomes across a number of areas, with the following areas explored in detail in this
evaluation:
Intermediate outcomes
• increasing mental health literacy;
• increasing early help seeking; and
• increasing access to required services.
Service system outcomes
• increasing advocacy for, and promotion of, youth mental health and wellbeing in their communities;
• reducing stigma associated with mental illness and help seeking for young people, their families and
friends, and the community;
• improving pathways to care for young people, including through:
o providing a localised service offering;
o other contributions to the local community;
o providing a ‘no wrong door’ approach; and
o securing support for headspace from other primary care and mental health providers.
User experience outcomes
• ensuring young people can access the help they need in an appropriate, accessible and youth friendly way
- providing an accessible, welcome, inclusive and non-stigmatising service, including through:
o providing an appropriate service approach for young people with mild to moderate, highprevalence
mental health conditions;
o providing culturally appropriate and inclusive services;
o enabling young people and their families to access support where, when and how they want; and
o participation of young people in the design and delivery of headspace.

Psychosocial outcomes
• improving mental health and wellbeing outcomes, considering clinical outcomes for young people; and
• improving psychosocial outcomes through providing alternative service delivery models.
Each of these objectives is associated with a range of intended impacts, as detailed in Table 3. In assessing
the effectiveness of the headspace model later in this report, the evidence for the short and medium-term
impacts is explored.
Table 3: headspace objectives and impacts
Objective Short-term impacts Medium-term impacts

Intermediate outcomes

Increasing mental health • Young people accessing • Young people are better able to
literacy - knowledge about headspace services manage their mental health in
mental health, how to seek help improve their mental the medium- to long-term,
and how to manage mental health literacy including identifying when
health (knowledge about they need to seek help and
mental health, how to support
seek help, and how to
manage mental health)

Increasing early help seeking - • Young people and • Young people, their families and
at an earlier age (e.g., under 21 families accessing communities are better able to
years); at relatively low mental headspace services identify when someone needs
health risk status; or when have increased help, and support appropriate,
assessed as at less than the knowledge about, and early help seeking
threshold stage of illness willingness to, seek
• Earlier identification and
help
treatment of emerging mental
health problems for young
people
• Young people increase help
seeking behaviour for mental
health and wellbeing issues

Reducing stigma associated • Young people, their •

with mental health and families and
mental illness - the fear or communities (living
embarrassment of seeking help near headspace centres
for mental health and wellbeing, and satellites) have
and the negative judgment of, improved attitudes
and lack of empathy for, those towards mental health
who do and mental illness
(stigma reduction)

Increasing access to required • Young people from a • Young people receive appropriate,
services - the number of young diverse range of evidence-based treatment early
people accessing headspace backgrounds access
and engage with
headspace services
• Young people and
families can access
headspace services in a
timely manner, and at
low or no cost

Service system outcomes

Improving the pathway to • headspace services • Young people and families

care through service deliver services across experience more streamlined
integration and coordination - and beyond four core and less fragmented pathways
bringing services together to streams (mental health, of care
function as one, providing a physical health,
seamless service experience for • The local service system for youth
alcohol and drug use,
a young person mental health is better
vocational programs)
integrated and coordinated
• headspace services
deliver integrated/
coordinated care

Ensuring young people can • Young people feel • headspace services operate
access the help they need in listened to and flexibly as appropriate to the
an appropriate, accessible and involved in decision- community needs and profile
youth friendly way - providing making
an accessible, welcome, • Local service system provides
inclusive and nonstigmatising • Young people and more youth-friendly, accessible
service families feel their and inclusive services as a
needs and interests are result of learning through
understood and partnerships, shared
reflected in their local professional development, etc
headspace service
(participation
outcomes)
• headspace services meet
the expectations of
friends and family and
Youth Reference
Group
 • Young people from a
diverse range of
backgrounds access
and engage with
headspace services

Psychosocial outcomes

Improving mental health and • Young people accessing • Young people accessing
wellbeing outcomes for young headspace services feel headspace services experience
people aged 12 to 25 years - more hopeful for the improvements (or stability) in
improvements in K10 SOFAS future social and occupational
and MLT outcome measures functioning
• Young people accessing
headspace services feel • Young people accessing
better able to cope headspace services experience
improvements in their quality
• Young people accessing
of life and wellbeing
headspace services
gain skills to better • Family and friends accessing
manage their mental headspace services have
health and wellbeing increased capacity to support
issues their young person
• Young people accessing • Young people report sustained
headspace services improvements in mental health
experience a reduction
• Young people who receive
in symptoms and
work/study, alcohol or other
levels of psychological
drug, and/or physical health
distress and increased
assistance are better able to
wellbeing
manage these aspects of their
• Young people accessing life in the medium- to long-
headspace services term
start to experience
improvement to their
day-to-day lives
• Young people accessing
headspace services
receive appropriate
support for physical
health, alcohol and
substance use and
work and study needs
• Young people who
receive work/study,
alcohol or other drug
and/or physical health
assistance, gain skills
to better manage these
aspects of their lives
Source: KPMG adapted from headspace Program Logic 43F

Stakeholder consultation through this evaluation was used to explore views on what might be considered
‘positive outcomes’ for young people attending headspace. Interviews from across the stakeholder groups
consulted (listed in Appendix B: Consultation) elicited broad support for the objectives and outcomes set
out in the headspace model program logic. Stakeholders from all groups recognised the importance of
providing easily accessible, free services for young people to support their mental health, and spoke of the
important role of services in community engagement and stigma reduction as key enablers to this ultimate
goal. headspace service providers also spoke of other indicators of success, such as when young people
refer their friends to the service and contribute as Youth Reference Group members, as being strong
indicators that a young person’s experience at headspace had been positive.
This stakeholder consultation provides validation of the conceptual design of the headspace model and the
extent to which its objectives are valued by the mental health services sector, policy makers and the
community.

3.2.3 Clinical outcomes

Improvements in the mental health and wellbeing of young people attending headspace are measured in an
ongoing way. headspace services collect a number of clinical measures of mental health and psychosocial
functioning throughout each client’s engagement with headspace. These are collected using consistent
tools, and form part of the headspace minimum dataset (hMDS) held by headspace National.

The Kessler Psychological Distress Scale

The Kessler Psychological Distress Scale (K10) is a 10-item questionnaire intended to yield a global
measure of distress based on questions about anxiety and depressive symptoms that a person has
experienced in the most recent four-week period . The questionnaire asks people to use a 5 point response
44F

scale, ranging from “none of the time” to “all of the time” in response to the following:
“In the last 4 weeks (or since your last visit to headspace), how often did you feel...
1. tired out for no good reason
2. nervous
3. so nervous that nothing could calm you down
4. hopeless
5. restless and fidgety
6. so restless that you could not sit still
7. depressed
8. that everything was an effort
9. so sad that nothing could cheer you up
10. worthless.”
The K10 measure is a sum of all responses to the 10 items, producing a value ranging from 10 to 50, with
higher values indicating higher levels of distress. K10 measures are grouped into four levels of
psychological distress : 45F

Table 4: Overview of K10 psychological distress levels

Total K10 levels Outcome category

10-15 Low
16-21 Moderate
22-29 High
30-50 Very high
Source: 4817.0.55.001 - Information Paper: Use of the Kessler Psychological Distress Scale in ABS Health Surveys

Social and Occupational Functioning Assessment Scale (SOFAS)

The SOFAS is a global rating of current social and occupational functioning from zero to 100, with lower
values representing lower functioning. It is a single-item assessment of current functioning, independent of
the severity of the young person’s psychological symptoms , conducted by the service provider each time a
46F

young person attends an OOS. The response scale used for SOFAS is as follows:
• 91-100: superior functioning in a wide range of activities;
• 81-90: good functioning in all areas, occupational and socially effective;
• 71-80: no more than a slight impairment in social, occupational, or school functioning;
• 61-70: some difficulty in social, occupational or school functioning;
• 51-60: moderate difficulty in social, occupational or school functioning;
• 41-50: serious impairment in social, occupational, or school functioning;
• 31-40: major impairment in several areas, such as work or school, family relations;
• 21-30: inability to function in almost all areas;
• 11-20:occasionally fails to maintain minimal personal hygiene;
• 1-10: persistent inability to maintain minimal personal hygiene; or
• 0: inadequate information.

MyLifeTracker
headspace National has also developed a measure called MyLifeTracker (MLT) which supplements the
other measures being used in their data collection system. This was developed and validated by headspace
National as there were no routine outcome measurement tools available that targeted those individuals
aged 12 to 25 years or that were appropriate across a diverse range of mental health presentations. MLT
measures current, self-reported quality of life in five different areas of importance to young people:
general well-being, day-to-day activities, relationships with friends, relationships with family, and general
coping. The measure enables clinicians working with young people to gain a quick, regular snapshot of
overall client progress and provides a valid measure to assess service effectiveness . 47F

MLT was developed specifically for use in headspace with the purpose of providing a quality of life
measure that better reflects the important areas of life for young people. The MLT is a five-item measure,
where each item is rated on a zero to 100 scale, with 100 representing the highest level of wellbeing in that
domain. The MLT takes the average value of the five responses.

Additional clinical outcomes

Clinical outcome scores are collected using these three key measures – the K10, SOFAS and MLT – along
with a range of others, including a measure of the young person’s mental health risk, as rated by the
service provider, their stage of mental illness, and the young person’s own assessment of the number of
days in the previous month in which they were totally or partially unable to participate in work, study or
day-to-day activities due to their feelings of distress. These measures are undertaken at first presentation,
throughout the young person’s clinical engagement and, where possible, at a further follow up point after
the young person has completed their episode of care. By collecting clinical outcomes data at various
points in time, the model provides clinicians and evaluators with measures of its impact on psychosocial
outcomes.

3.2.4 Success for headspace – in summary

The headspace model is designed to facilitate improvements in psychosocial outcomes, as measured
through clinical tools, and is also intended to improve intermediate outcomes, such as increased mental
health literacy and early help seeking, reduced stigma associated with mental health and mental illness and
increased access to required services. These outcomes, along with improved pathways to care through
service integration and coordination, are intended to ensure young people can access the help they need in
an appropriate, accessible and youth friendly way, which in turn aims to contribute to improved mental
health and wellbeing outcomes for young people aged 12 to 25 years.
A review of recent literature illustrates that young people in Australia are experiencing high prevalence
rates of mental illness and psychological distress, exacerbated in recent years by natural disasters and
events, such as widespread bushfires and the COVID-19 pandemic. Evidence further supports the
identification of priority groups for the headspace model, as a number of demographic characteristics are
associated with reduced help seeking and for mental illness and psychological distress.
When tested with a range of relevant stakeholder groups, the key aims and objectives of the headspace
model, summarised into the six focus areas for this evaluation, were strongly validated and considered
relevant and important to the provision of mental health services for young people in Australia.
The conceptual design of the headspace model is aligned with best practice, and its intended outcomes are
clear and considered valid by relevant stakeholders. With this in mind, in order to effectively evaluate the
model, the activities associated with its key components need to also be understood. The following section
presents each element of the model and the associated activities.

3.3 Components of the

headspace model
3.3.1 Service design
Each component of the headspace model is intended to contribute an essential feature to aid in ensuring
that young people are able to receive accessible, appropriate, effective and sustainable services at a time in
their lives when they are most vulnerable to the emergence of mental health problems . 48F

The headspace model is comprised of 16 components, to which headspace services must demonstrate
ongoing commitment and alignment in order to hold a Trademark Licence Deed (TMLD) and to operate
under the headspace name . The details of the model are set out in the hMIF, including standards and
49F

guidelines for each component of the model.

Service components
The 10 service components required to provide the four core areas of focus are defined in the hMIF as
follows.
• Youth participation – the central and continuous involvement of young people in their own care, and in
the governance, design, development, delivery, evaluation and continuous improvement of headspace
services.
• Family and friends participation – the central and continuous involvement of family and friends in the
care of a young person, and in the governance, design, development, delivery, evaluation and
continuous improvement of headspace services.
• Community awareness and engagement – the ability of the service to work with the local community
to increase mental health literacy, reduce stigma, encourage early help seeking and promote access to
headspace services, while building strong relationships with young people, their family and friends,
other local services and the broader community.
• Enhanced access – meaning that headspace services are engaging, youth-friendly and set up to minimise
the barriers young people typically encounter when seeking professional help. This component
includes the ‘no wrong door’ approach, where no young person is turned away without connection to
appropriate internal or external services. This enables early and easy access to services and supports
effective help seeking behaviour.
• Early intervention – the identification and provision of intervention and support services as early as
possible in the development of mental health difficulties to prevent or delay the onset of mental ill-
health or reduce the impact associated with mental ill-health and improve outcomes.
• Appropriate care – the provision of evidence-based interventions for each individual young person by
matching the type, intensity, frequency, duration, location and mode of treatment to their presenting
need. This includes identification and consideration of factors, such as risk and protective, stage of
illness, psychosocial complexity, and developmental and sociocultural.
• Evidence-informed practice – the use of the best available evidence to guide service development,
delivery, evaluation and continuous improvement. Sources of evidence include clinical guidelines,
peer-reviewed literature, expert opinion, centre-based research and service evaluation; and the unique
knowledge, skills and expertise of service providers, young people and their families and friends.
• Four core streams – the provision of an enhanced primary care platform with four core service streams
– mental health, physical and sexual health, alcohol and other drugs, and vocational and educational
support – to holistically address the main mental health and wellbeing needs of young people within
 the local community.
• Service integration – bringing services together to function as one, providing a seamless service
experience for a young person, particularly if they require care involving multiple service providers
and supports.
• Supported transitions – the process of formal handover that proactively and personally transfers a
young person’s care to any other service provider in a way that supports the ongoing engagement of
the young person and continuity of care between service providers. This includes both transition
between service providers within headspace and exit from the headspace service. Transition can occur
for a number of reasons, including a young person’s preferences, age, need for more specialised
service or geographic location.

Enabling components
Six enabling components sit around these core components in the model:
• National network – the network of headspace services across Australia that collaborates to share
learning, innovation and best practice and, in turn, facilitates continuous improvement of services to
enhance youth mental health and wellbeing outcomes. It is composed of all headspace centres,
satellites and other services, headspace National, PHNs, lead agencies, consortia, and Youth and
Family and Friends Reference Groups.
• Lead agency governance – the people, systems, processes, policies and procedures through which
responsibility and accountability for corporate, clinical and cultural governance is assigned and
exercised in order to ensure the delivery of safe, high-quality and inclusive headspace services for
young people and their families and friends.
• Consortium – a collaborative advisory group comprising local service providers and organisations that
partner with a lead agency to provide partnership opportunities, strategic direction and resources to
enhance the headspace service’s capacity to meet local community needs. headspace services also
form partnerships in the community beyond the consortium to further enhance the wellbeing of young
people in their communities.
• Multi-disciplinary workforce – the clinical and non-clinical workers required from a range of
disciplines and backgrounds – with the right knowledge, skills and expertise – who work together to
holistically meet the mental health and wellbeing needs of young people, and their families and
friends, within the local community.
• Blended funding – the use of multiple funding streams and in-kind contributions to increase income
diversity, flexibility and the sustainability of the service in accordance with the needs of the headspace
service, young people and their community to ensure access to no or lowcost services.
• Monitoring and evaluation – the continual collection and review of comprehensive information to
facilitate service planning, delivery, evaluation and continuous improvement for headspace services,
PHNs and the national network.
A further element of the headspace model is that headspace services are required to be consistent in their
branding and street presence. The internal décor of each service should meet headspace branding
requirements, customised by local Youth Reference Groups to provide connection/relevance to the local
community, with white walls and lime green accents. Services are also to be located in centrally accessible
street frontage.

3.3.2 Support services provided under the model

As described in the hMIF, headspace centres provide services across four core components: mental health,
physical and sexual health, alcohol and other drugs, and vocational and educational support. The early
intervention model is designed to tailor services and interventions to match the needs of the young person,
and the centre-based model is designed for the provision of multiple services and supports from a single
location. Through the consortium of local providers, young people can gain access to services beyond the
four core streams as well, and most centres provide a range of psychosocial supports to supplement the
core streams.
The four core components of support are divided into eight broad categories of service provided directly to
young people, while centres also provide activities and engagement programs outside of these four core
components. The services provided by centres are described using hMDS data in Section 2.6.2 below, to
illustrate the way the model operates in practice.

Intake and assessment

This involves initial engagement and screening as part of a young person’s first contact with the service,
focused on assessing alignment between the needs of the young person and the supports on offer at the
service. Time spent with a young person to build rapport and level of comfort with the headspace service is
also part of this category of services provided.
Conducting a psychosocial assessment of the young person using the HEADSS (headspace) assessment
tool is part of this category of service. This is designed for any service provider within a headspace service
to be able to use, asking screening and assessment questions across 10 domains. The domains are:
• home and environment;
• education and employment;
• activities;
• alcohol and other drugs;
• relationships and sexuality;
• conduct difficulties and risk-taking;
• anxiety;
• eating;
• depression and suicide; and
• psychosis and mania . 50F

Other services provided in this category include review or outcome-based assessments, or assessments
using other tools at intake . 51F

Mental health – medical intervention

This includes support provided by GPs, psychiatrists and mental health nurses, and includes activities such
as the development of a mental health treatment plan, medication related screening, monitoring or advice,
and metabolic screening or monitoring. It includes specific care provided by psychiatrists, and referrals to
specialists . 52F

Mental Health – psychological intervention

There are a wide range of services and supports provided in this category, including:
• Cognitive Behavioural Therapy;
• Cognitive interventions (e.g., Cognitive Analytic Therapy);
• Interpersonal Therapy;
• Acceptance and Commitment Therapy;
• Dialectical Behaviour Informed Therapy;
• psycho-education;
• lifestyle factors (e.g., Sleep, dietary or exercise advice);
• skills training (social and communication skills, anger management)
• behavioural interventions (including general counselling, crisis intervention and mindfulness and
relaxation strategies, among others); and
• Psychodynamic Therapy . 53F

Physical health
GPs, psychiatrists and nurses provide the following service categories through headspace services, along
with other physical health services as required:
• vaccination;
• acute physical illness;
• chronic physical illness; and
• injury .
54F

Sexual health
Within this category of services, GPs and nurses assist with:
• sexual health testing;
• contraception;
• counselling and advice;
• pregnancy management;
• gynaecological symptoms;
• pap smear; and
• other sexual health services . 55F

Vocational
Vocational services are provided within headspace services by specialised workers, who work with the
young person to provide assistance with work and study in an effort to keep them actively engaged in
meaningful activity. Other providers within the headspace service may also provide services or supports
within this category as part of their engagement with the young person.

Alcohol and/or drug specific intervention

Services in this category can be provided by specialist Alcohol or other drug (AOD) workers, or by other
providers within the service. The services include:
• motivational interviewing or enhancement;
• psycho-education (including harm minimisation); and
• Cognitive Behavioural Therapy.

General assistance
The final component of services provided directly to young people at headspace covers the work
undertaken to support the young person through case management or care coordination. This is an
important stream of work to ensure seamless transition between services and that the holistic needs of the
young person are met, both within and beyond the headspace service.

Service modality
The various supports offered at headspace services can be provided in oneon-one individual settings, in
groups, with family or carers of the young person or in small group sessions with young people with
similar needs or interests in skills development. Some services can also be conducted over the telephone or
in online video settings, particularly for screening and check-in contact. The majority of services are
delivered face-to-face through headspace services, with the exception of the period following the COVID-
19 outbreak, which saw a significant shift to online and telephone-based services. This is discussed further
in Section 2.6.3.

3.3.3 Other activities undertaken by headspace services

Broader community engagement
Along with the supports provided directly to young people within headspace services, staff at headspace
also undertake a range of community building and awareness raising activities. These differ across services
and are intended to focus on the needs and issues of the local community, for example in coordinating
suicide postvention protocols and support. Engagement activities can include running local competitions
and award programs, running workshops at schools, holding a headspace stall at community events, and
holding information sessions at meetings of cultural, religious or community groups to raise awareness
about headspace and youth mental health and wellbeing. Similar to the impact on service modality
discussed above, COVID-19 also had impacts on these engagement activities, limiting information
sessions, community events, and other activities through lockdowns and event restrictions.

Family and youth programs

Each service also offers a range of programs for young people outside of the core streams of support
offered. There is a wide array of activities on offer across services, including arts and crafts based
activities, movie screenings, dance classes, song writing, fitness groups, trivia nights, youth groups for
particular cohorts such as LGBTQIA+ young people, and education and training programs to meet local
interest. The aim of these programs is to build connection between young people or family and carers of
young people in a safe setting, and in so doing to build awareness of youth mental health and wellbeing
issues and reduce stigma and other barriers to help seeking behaviours.
The specifics of the supports provided by services are described in Chapter 3 below.

3.3.4 Types of headspace services operating under the model

The headspace model is delivered within five broad types of services, which differ by size, physical setting
and service offerings. These variations in operating model are intended to adapt the model to suit local
need, with funding for each service aligned to the type of operating model in place. Some of these service
types have been introduced in recent years as a result of Commonwealth Government funding
announcements.
The types of service are described below, based on headspace National documentation : 56F

headspace centre
• Full-service facility operating in accordance with the headspace Centre Model.
• All four core streams are delivered (preferably on site).
• A dedicated facility to accommodate all services.

headspace satellite
• Provides a reduced range of services and is linked to a parent headspace centre, operating in accordance
with the headspace model.
• A satellite is located in an area of need surrounding a headspace centre.
• Minimum of three of the four core streams delivered as follows:
o Mental Health (mandatory component, onsite delivery);
o Physical Health (onsite delivery or access via local provider);
o Alcohol and other drugs (onsite delivery or access via local provider/online service); or
o Vocational (onsite delivery or access via local provider/online service).
• A dedicated facility to accommodate a reduced range of services.

headspace outpost
• Provides a reduced mental health service that must be linked to a parent headspace centre, which is
operating in accordance with the headspace model.
• Outposts are established by exception and as determined by government.
• The core stream of mental health must be delivered onsite.
• Usually co-located with an existing service provider but may be in a stand-alone setting.

headspace outreach
• Refers to the range of services delivered outside the parent headspace service in youth friendly settings.
• Outreach activities can be any of the range of services offered by the parent centre, which may include
clinical sessions, psycho-education and community awareness activities.
• Usually visiting, mobile or co-located with existing services.

Hub and spoke model

The hub and spoke model is an informal term used to describe some headspace service types which, while
similar to satellites, provide a considerably reduced suite of services in rural and remote locations, under
the auspices of a headspace service in the nearest regional town. Different to a headspace outpost, there are
typically multiple ‘spokes’ attached to a parent hub.

3.4 headspace in context

Supporting people with mental ill-health is a key public health priority in Australia. The term ‘mental
illness’ covers a range of conditions, including anxiety disorders, depressive disorders, personality
disorders, bipolar disorder, and schizophrenia. The severity, impact of and treatment for these conditions
varies significantly. Mental ill-health affects all Australians at some point in their lifetime, either directly
or through relationships with family, friends, colleagues, and others who are living with mental illness.
headspace operates as a high-profile element of the Australian mental health service system, which is a
complex mix of public and private services, delivered by a range of organisations, and funded by the
Commonwealth, State and Territory Governments, individual service users, and private insurers. The
Commonwealth Government and all states and territories share responsibility for mental health policy, and
provision of supports. A range of different mental health supports and services are provided at each level
of government and by private organisations.

3.4.1 Growth and evolution of the headspace program

The Commonwealth Government has significantly expanded its funding of the headspace program in
recent years , with the network growing from 98 centres in 2015-16 to 118 centres by 30 June 2020 . This
57F 58F

growth has occurred over time in line with government announcements, often through annual budget
measures. This expansion has made dedicated youth mental health services and supports available in more
communities across Australia, often for the first time.
Since the first 10 centres were opened in 2007 and 2008, there have been successive funding rounds by
government leading to the rapid expansion of the network. For example, in the 2019-20 Federal Budget,
funding was announced to support the establishment of 10 new centres and 20 satellite services, which was
then further expanded through 2019 Federal Election commitments to establish a further eight services. As
at 1 May 2022, there were 154 headspace services operating across Australia.
59F

These commitments have also introduced significant changes to delivery models across the network, with
satellites and outreach models becoming more widespread to enable the network to reach young people
living in smaller communities.
Further funding was announced in the 202122 Federal Budget of $278.6 million over four years, which is
targeted at:
• expanding the national headspace network by establishing 10 new headspace services and upgrading five
satellite services to headspace centres, and introducing one new satellite service, bringing the total
number of open and planned headspace services across Australia to 164 , 60F

• boosting clinical capacity at existing headspace services; and

• funding improved coordination, system navigation and referral pathways, and improving access to
culturally safe and accessible services.
The Commonwealth Government contributed approximately $101 million in headspace grant funding to 61F

headspace services established by 30 June 2020 in 2019-20, with an additional $16.6 million provided to
headspace National to support the headspace Network . In addition to this funding, PHNs and other
62F

organisations, including state and territory governments, make further financial and inkind contributions to
delivering headspace services. The provision of MBS funded services by independent medical and allied
health practitioners operating from headspace services is also a significant funding source. Collectively,
these income streams make headspace the largest and most comprehensive youth mental health program in
Australia. A more detailed analysis of the costs of providing headspace services is contained in Section
4.1.

3.4.2 headspace governance stakeholder groups

headspace is delivered through a distributed governance model which involves participants at the national,
regional and community levels. Figure 7 provides a high-level illustration of how key partners work
together to deliver headspace. The role of each key partner is described in greater detail in the following
sections.
Figure 7: headspace governance structure and partners, as at January 2022
Source: KPMG 2022

Commonwealth Government
The Commonwealth Government funds a range of services related to mental health through the MBS and
Pharmaceutical Benefits Scheme (PBS), as well as mental health services through PHNs (such as
headspace). These include providing MBS funding for mental health support with specialist medical
practitioners, psychiatrists, GPs, psychologists and other allied health professions. The Commonwealth
Government also funds other related services that can be critical for people living with mental ill-health,
including income support, social and community support, the National Disability Insurance Scheme
(NDIS), workforce participation programs and housing support.
The Commonwealth Government provides the principal source of funding for headspace through the
Health portfolio, and provides core funding for the operation of each headspace service in the network, as
well as to headspace National. In its role as principal funder, the Commonwealth Government works with
headspace National and PHNs to:
• provide policy and program oversight to youth mental health initiatives, including the National
headspace Program;
• improve access to mental health services for young people through the development and implementation
of new policy measures;
• determine funding levels for headspace services;
• manage the grants arrangements in place to support the headspace program, including those with each
PHN and headspace National; and
• contribute to broader mental health sector reform activities.
Commonwealth and state and territory levels of government also provide support to population mental
health support services, such as Lifeline, Beyond Blue, and Kids Helpline.

Primary Health Networks

Since 2016, Commonwealth Government funding for headspace services has been delivered through
PHNs through grant agreements, in a local commissioning model. PHNs are responsible for
commissioning headspace services in line with the hMIF, have a contractual relationship with the lead
agency running the service, and work with services to ensure the focus of each is aligned to local need.
PHNs also work with headspace National to commission new headspace services.
PHNs are more generally responsible for conducting local needs analyses, assessing the health care needs
[2]
of their community and to commission health services to align with those needs . PHNs also have a key
role in assisting services to connect with each other, and to support shared care and seamless service
transition for clients who need to access more than one provider for their health care needs.
Through this commissioning relationship, PHNs and headspace services work to provide localised
offerings and staffing that respond to the presenting needs of the local community, and connections,
partnerships and referral pathways across the local service landscape. More discussion of the impact of the
introduction of PHN commissioning is contained in Section 5.4.3 and Appendix D.8.

headspace National
The headspace National Youth Mental Health Foundation (headspace National) was designed and
developed in 2005 as a national program of reform, aimed at enhancing access, coordination and quality of
services in youth mental health. The founding consortium of what is now headspace National was led by
Orygen Research Centre in partnership with the University of Melbourne, The Brain and Mind Research
Institute (BMRI) at the University of Sydney, the Australian General Practice Network and the Australian
Psychological Society . headspace National is a company limited by guarantee, classified as a health
63F

promotion charity . With the shift to a local commissioning approach through PHNs, headspace National’s
64F

role changed substantially from commissioner to one focused more specifically on the headspace model
itself. headspace National continues to support PHNs to commission headspace services, in accordance
with the model, to support model fidelity.
headspace National holds the TMLD for the headspace model, as set out in the hMIF. All headspace
services must undertake accreditation every three years with headspace National. The accreditation process
is similar to a detailed performance audit, where documentary evidence is submitted to headspace National
demonstrating that the headspace service is operating in line with each component of the hMIF.
headspace National is funded to provide national coordination and support for the headspace network of
services, including in fidelity assessment and accreditation of headspace services under the model, in
workforce training, education and development, in data collection and evaluation, and in monitoring and
reporting to the department and other funders. Alongside these activities, headspace National provides a
range of services directly, such as eheadspace and digital work and study services, and delivers national
community awareness campaigns and other enabling activities.

Orygen
Orygen is a not-for-profit company limited by guarantee and an approved research institute. The company
has three members: the Colonial Foundation, The University of Melbourne and Melbourne Health . 65F

Orygen continues to be closely affiliated with headspace through its role as a lead agency in the delivery of
a number of headspace services, as well as in its ongoing relationship with headspace National, with whom
it works closely to design, measure and promote resources and interventions aimed at improving youth
mental health and wellbeing. Orygen is also contracted directly by the Commonwealth Government to
provide guidance and support to PHNs in their commissioning of youth mental health services, particularly
for the Early Psychosis Youth Services Program, which operates on the headspace platform, however it
does not have this role for core headspace services.

State government partners

State and territory governments are typically responsible for funding and delivering public sector mental
health services that provide specialist care for people experiencing mental illness. Some state and territory
governments contribute funding to headspace services, largely through ad-hoc, targeted grants. They also
provide core funding to other local providers in the service system, working alongside headspace services.
Commonwealth and state and territory levels of government also provide support to population mental
health support services, such as Lifeline, Beyond Blue, and Kids Helpline.

Lead agencies
Each headspace service is run by a lead agency, commissioned by a PHN to deliver the headspace service
within a specific geography. As part of the local commissioning model, lead agencies are health or social
services providers that are legally, operationally and clinically responsible and accountable for the service.
Lead agencies also contribute staff time to enhancing the capacity of the headspace service, and are
selected by PHNs through a competitive tendering process. headspace National is consulted through this
tendering process to ensure the lead agency delivers headspace in accordance with the model, but is not
responsible for selecting the preferred tenderer.

headspace services
Each service is run by a centre manager who reports to the lead agency. Services are centrebased,
comprised of a multi-disciplinary workforce with staffing profiles which vary across services depending
on available funding. Each service has a team of core staff, with roles including clinical lead, intake and
case coordination, community and youth engagement and administration and practice management.
headspace services also provide onsite availability of services from other providers, such as GPs, private
allied health workers and other specialist services in areas such as crisis accommodation, domestic and
family violence or eating disorder treatment. These services collaborate within the headspace service to
provide integrated care for young people and their families.

Local headspace community consortium

Each lead agency establishes and maintains a collaborative advisory group of local service providers. This
group meets in a regular forum to drive the strategic focus and partnership opportunities within the region
in relation to youth mental health. Consortium members enter into formal memoranda of understanding
with the headspace centre, detailing time, full time equivalent (FTE) staff and other resources contributed
by each party.
The consortium approach aims to promote service integration and to strengthen local relationships
between service providers. It is also designed to help ensure the activities of the headspace service are
aligned with localised need and that the services provided are responsive to the health needs of young
people and the social determinants driving presentation in that location . Consortium members may include
66F

local GPs, primary and tertiary mental health services as well as community service providers such as
housing and homelessness and domestic violence support services.

Youth Reference Groups

Under the headspace model, the local Youth Reference Group is a forum intended to enable young people
to contribute to strategic planning and oversight of the service, as well as to participate in the development,
delivery and evaluation of supports on offer to clients. Local Youth Reference Groups must meet at least
six times a year, and involve at least three members.
The headspace Youth National Reference Group, a separate forum comprising 20 young people with lived
experience of mental ill health from across Australia, also meet on a regular basis to provide advice and
their perspective to headspace National.

headspace Family and Friends Reference Group

The headspace model requires routine involvement of family and friends in the care of a young person, as
well as engaging this cohort in the broader development, delivery and evaluation of headspace services.
headspace services have family and friends contribute to:
• strategic planning and oversight through local consortium meetings;
• service development, delivery and evaluation through a range of local and national consultation
mechanisms; and
• the care of an individual young person through Family Inclusive Practice.
Alongside this local engagement, the headspace Family and Friends Reference Group brings together 10
people with lived experience as carers to young people experiencing mental ill-health, to provide advice
and insight to headspace National.

Other local providers

The headspace model operates within a local service system, with partner services and providers which
vary by region and can vary over time. These partner services include other early intervention and
prevention supports, social services providing housing, employment services and other services, and
mental health providers across the service spectrum.
Funding and governance arrangements vary across the service system. The private sector provides
admitted patient care in private psychiatric hospitals, and private services provided by psychiatrists,
psychologists and other allied health professionals. Private health insurers also fund treatment costs in
private hospitals, public hospitals, and out of hospital services provided by health professionals.
The non-government sector delivers supports through both government and private funding. These
services often focus on wellbeing programs, providing support and assistance to people who live with
mental illness, rather than assessment, diagnostic and treatment supports provided by clinically-focused
services.
Many services from the private and non-government sectors form part of each headspace service’s local
referral pathway. This includes tertiary mental health services provided in public acute and psychiatric
hospital or bed based settings, specialised community mental health services, and residential mental health
services.
There are also non-specialised supports provided, such as emergency department and non-specialised
admitted units, mental health-specific community-based services such as supported accommodation and
social housing programs. headspace services must operate alongside, and integrate with, services across
this spectrum to achieve client outcomes.

3.4.3 Key stakeholders in headspace service pathways

The headspace model requires headspace services to be integrated according to the needs of young people,
and to work with other local services to holistically identify and address their clients’ risk and protective
factors. The model requires that headspace services ensure the coordination and integration of services to
provide seamless care for young people and their families and friends. The service system is complex with
high variation across localities and regions, requiring headspace services to actively engage in networking
and service mapping on an ongoing basis.

Tertiary mental health services

Integration with the local tertiary mental health services (TMHSs) includes actively engaging with Child
and Adolescent Mental Health Services (CAMHSs) and Child and Youth Mental Health Services
(CYMHSs), as appropriate.
These services are part of a national network of providers, funded through state and territory health
departments, to meet the needs of individuals experiencing severe mental health problems. headspace
services take referrals from TMHSs where a young person is able to ‘step down’ into the psychosocial or
mild to moderate clinical support that headspace provides.
Figure 8: High level summary of mental health supports available for young people
Source: KPMG 2022

headspace services also refer young people to TMHSs to ‘step up’ into more intensive or specialised care,
or where their mental health needs require them to be admitted into a hospital setting. TMHSs and state
government policy agencies are not typically part of the headspace governance model, however a small
number of lead agencies running headspace services are statefunded area health services, or CYMHSs. In
general, the integration of TMHSs with headspace is dependent on relationships at the service level, along
with work done at the PHN level, to facilitate integration with local hospital and primary care providers.

Other early intervention and prevention services

Alongside headspace, there are a range of prevention and early intervention services available to support
young people to improve their mental well being and improve resilience.
Such services include online, self-guided wellbeing and relisience programs for young people and
interactive online programs targeting prevention and early intervention for young people and their parents,
including information and skills building. These services are provided by a range of organisations,
including headspace National, Black Dog Institute, beyondblue, BRAVE, and Reachout, along with a
range of related initiatives.

Supports through schools and tertiary education institutions

Mental health support is also provided to young people through schools and tertiary education institutions.
For young people at school, support is provided at an individual school or institution level, through school
counsellors, guidance workers and psychologists employed to work with students, and through broader
system-wide programs often delivered by external organisations.
For young people accessing education through tertiary institutions, there are fewer formal relationships
with external organisations, with the majority of support provided directly through the institution itself.
However, the Productivity Commission identified that the level and types of mental health-related supports
provided by tertiary institutions to students varies between education providers.
Key supports and services provided by external organisations through schools include support educators
from early learning services which develop positive, inclusive and resilient learning communities, as well
as a range of programs within schools and universities using lived experience facilitators to reduce stigma
associated with mental ill-health and enhance help seeking. Some examples of these programs include Be
You, delivered by Beyond blue, Early Childhood Australia and headspace and BeingHerd and
67F

batyr@school delivered by batyr .68F

Engaging with schools and tertiary education institutions is a key role for headspace services, to build
brand visibility and reduce stigma around mental health help seeking.

Helplines and online forums

There are also a range of online and telephone-based services for young people, which provide prevention-
based resources for young people, through to counselling and crisis support. These services include:
• Kids Helpline – a free, 24/7 online and telephone counselling service for young people aged five to 25
69F

years. Support is also provided for parents and carers of young people, and schools and teachers
educating young people.
• Beyond Blue – provides a range of information, advice and support services for all Australians, with
specific services for young people. The Beyond Blue website provides access to information and
resources to support people to manage their mental health and wellbeing, and there are 24/7 telephone
support services and online forums for those who need immediate support. Email and chat services are
also provided.
• Lifeline – provides free, 24-hour crisis support and suicide prevention services for all Australians,
70F

including young people, through online chat, text and telephone.

• Reach Out – provides online self-help information, peer support and referral tools to young people aged
71F

12 to 25 years, with services tailored to the young person’s level of need at the time.
• eheadspace – headspace National also provides additional services in this category through eheadspace.
eheadspace is designed as a ‘digital ecosystem’, not only providing young people with web chat, email
and telephone support, but also access to other services, including group forums and online resources
they can access from home to support their mental health and wellbeing.
headspace services operate in this context of early intervention and prevention focused services and
supports, and encourage young people to access these services as relevant to their situation and needs.

Current challenges in the mental health service system

headspace services face a range of challenges in providing mental health support to young people as part
of the mental health service system in Australia. These challenges are well documented, particularly in the
recent Productivity Commission Inquiry Report on mental health. The report summarises the current
barriers, gaps and challenges facing Australia’s mental health service system. These challenges often
extend past mental healthcare, to the interaction of mental healthcare with physical health care and other
sectors and services beyond health that support recovery. The barriers and gaps include :72F

• A narrow view of people seeking treatment and support: There is often incomplete information of the
types of support people are seeking, with a focus on a person’s symptoms, rather than the broader
support an individual may need to recover and remain well, and how this can effectively be delivered.
• Under-investment in prevention and early intervention: Compared to treatment and crisis services.
This means many people become more ill with time, which may have been prevented, or addressed
earlier, shortening the period they may experience mental ill-health.
• Disproportionate focus on clinical services: There is a heavy focus in the Australian service system on
clinical services, with more limited consideration of other determinants of, and contributors to, mental
health. Contributions from family, kinship groups and carers, and broader social support services all
play an important role in recovery and mental wellbeing.
• Difficulties in finding and accessing suitable support: At times, there are limited services available
within particular regions that are appropriate, relevant or culturally appropriate for people who need
support. There are long wait lists, limited access to information on availability and outcomes, and
challenges with services needed being appropriately linked to support coordinated care for people,
especially as their needs change.
• Supports that are below best practice: A lack of measurement and evaluation of whether a service
works, and a "culture of superiority” means clinical interventions are prioritised over other services,
consumers, families and their carers.
• Stigma and discrimination: There remain challenges with how people with mental illhealth see
themselves, and how others view those who have a mental health problem, and those who care for
them.
• Dysfunctional approaches to the funding of services and supports: Creating poor incentives for
service providers to deliver quality outcomes, and increased and inefficient costs to people with
mental ill-health and the broader public.
• A lack of clarity across the tiers of government about roles, responsibilities and funding: This leads
to overlaps in services provided, gaps between services that exist, and limited accountability for
services at all levels.
The components of the headspace service model, as discussed in Section 2.3 above, form a set of design
features intended to break down these barriers to service access and to provide early intervention and
prevention . 73F

3.5 Services currently

available at headspace
3.5.1 Summary of the headspace network
As described in Section 2.4.1, the headspace program has grown and adapted since its inception, with 154 74F

headspace services open as at 1 May 2022. At that time, there were also an additional 15 headspace
services for which funding had been announced and where commissioning was underway. This includes
funding announced within the 2021-22 Federal Budget on 11 May 2021 for an additional 10 headspace
centres and one satellite service, the locations of which are being determined. Analysis of headspace
services set out in this section has been completed on headspace services open by 30 June 2020, in order to
present comparable data for full years. For changes over time, analysis is presented over the last five
financial years (from 2015-16 to 2019-20), since the last evaluation of headspace.

Services by jurisdiction
There were 118 services opened by 30 June 2020 . Each state and territory across Australia had at least one
75F

established headspace service by this date, with New South Wales (NSW) and Victoria making up more
than half of the overall headspace network. A full list of headspace services included in this analysis is
contained in Appendix C. Table 5 below compares the total number of centres per jurisdiction, with their
share of the Australian population aged 12 to 25 years. In most jurisdictions, the number of headspace
services broadly represents its share of the population of young people aged 12 to 25 years. The biggest
difference is five percentage points, with the proportion of headspace services in Western Australia (WA)
above its share of population. However, as the largest Australian state by size and one with a
geographically disparate population, this over representation is not unexpected.
Table 5: Overview of headspace services and population by jurisdiction as at 30 June 2020

Jurisdiction Number of Proportion of Total population Proportion of those

headspace services headspace services aged 12 to 25 as at 30 aged 12 to 25 in total
June 2020 national population
76F

as at 30 June 2020
NSW 36 31% 1.3m 29%
VIC 29 25% 1.1m 25%
QLD 22 17% 0.9 19%
WA 13 11% 0.4m 6%
SA 11 9% 0.3m 9%
NT 3 3% 0.04m 2%
TAS 3 3% 0.1m 1%
ACT 1 1% 0.1m 2%
Total 118 100% 4.5m 100%
Source: KPMG analysis of the hMDS and headspace funding data
Notes: Percentages have been rounded to the nearest whole number and may not sum to 100 per cent.

Services by remoteness 77F

More than 50 per cent of headspace services established by 30 June 2020 were located in major cities, with
one service, Pilbara, located within a very remote region of Australia.
Table 6: Overview of headspace services by remoteness as at 30 June 2020

Remoteness Number of services

Major cities 61
Inner regional 35
Outer regional 17
Remote 4
Very remote 1
Total 118
Source: KPMG analysis of the hMDS and headspace funding data

Services by service type

The types of services making up the headspace network have been outlined in Section 2.3.4. The
introduction of alternative models has been a recent development, and by 30 June 2020, there were only
nine services established that were not classified as headspace centres. In addition, all except one of the
satellite and outpost services were established after May 2018. This means the majority of services making
up the headspace network are headspace centres, with very few headspace satellites or other models.
However, emphasis continues on diversifying the headspace model by the Commonwealth Government,
with the government announcing funding for additional satellite services prior to the commencement of
this evaluation, to ensure young people in smaller communities are also able to access face-to-face
services. It should be noted that headspace services which fall into the ‘other’ category have been omitted
from some charts. This is to ensure that these services are not identifiable given their small number.
Table 7: Overview of headspace services by service type as at 30 June 2020

Service type Number of services

Centre 109
Satellite 7
Other, including outpost 2
Total 118
Source: KPMG analysis of hMDS and headspace funding data

Services by Primary Health Network

As outlined in Section 2.3.1, the introduction of PHNs saw responsibility for commissioning for headspace
services shift to PHNs across Australia. Each PHN commissions at least one headspace service, with some
PHNs commissioning up to seven headspace services.
Table 8: Overview of headspace services by PHN as at 30 June 2020

Primary Health Network Number of services

ACT 1
Adelaide 4
Brisbane North 4
Brisbane South 4
Central & Eastern Sydney 5
Central QLD, Wide Bay & Sunshine Coast 6
Country SA 7
Country WA 6
Darling Downs & West Moreton 3
Eastern Melbourne 3
Gippsland 3
Gold Coast 1
Hunter New England & Central Coast 5
Murray 5
Murrumbidgee 2
Nepean Blue Mountains 2
North Coast 5
North Western Melbourne 6
Northern QLD 3
Northern Sydney 2
Northern Territory 3
Perth North 3
Perth South 4
South Eastern Melbourne 7
South Eastern NSW 5
South Western Sydney 3
Tasmania 3
Western NSW 4
Western QLD 1
Western Sydney 3
Western Victoria 5
Total 118
Source: KPMG analysis of hMDS and headspace funding data

Service staffing mix

Services provided within each headspace service are delivered by a range of staff who differ based on their
specified team role, age, gender and other characteristics. In 2020, 74 per cent of staff providing services
recorded in the hMDS were female, with 25 per cent male, and one per cent identifying as non-binary. The
average age of staff providing services was 48 years, while the median age was 35 years. In the same year,
2.4 per cent of staff delivering services identified as Aboriginal and Torres Strait Islander.
Services are provided by a range of professions across the headspace network. Table 9 below summarises
the breakdown of service providers who delivered at least one OOS through headspace services in 2019-
20.
Table 9: Overview of professions of headspace service staff during 2019-20

Service providers who delivered at least one occasion of service Proportion of total service providers
Psychologist 50%
Social worker 20%
Medical Practitioner 8%
Counsellor 7%
Occupational therapist 6%
Dietitian <1%
Peer Worker <1%
Aboriginal or Torres Strait Islander health/wellbeing worker <1%
Management <1%
Other qualifications 8%
Source: KPMG analysis of hMDS

Lead agencies and headspace service representatives were also asked to complete a survey as part of the
evaluation. More information on this survey, including the sampling approach and respondents, can be
found at Appendix A.3. Lead agency and headspace service representatives who completed the survey
provided additional context to their current staffing mix. Aside from management and administrative staff,
these respondents most commonly reported psychologists, counsellors, GPs, and nurses as part of their
workforce. Other staff reported included social workers, dietitians, occupational therapists, community
engagement workers, youth workers, peer workers, AOD and vocational specialists, youth access workers,
exercise physiologists, paediatricians, new access coaches, support coordinator and specific cultural
wellbeing workers. There were some differences between professions reported between headspace services
in different locations. Eightysix per cent of metropolitan services reported having a psychologist,
compared to 94 per cent of regional services and 54 per cent of rural and remote services. All four satellite
service respondents indicated their centre either had a psychologist or psychiatrist on staff.
The survey also asked representatives to indicate the specific professions where there is a shortage of
workers, and where they cannot access sufficient staff. Figure 9 below demonstrates responses to this
question. The professions where respondents most consistently indicated they have challenges accessing
staff are psychologists and GPs, followed by psychiatrists. Around 78 per cent of respondents from
metropolitan services, 77 per cent of respondents from regional services, and 61 per cent of respondents
from rural and remote services reported challenges accessing psychologists. This differed to responses
regarding GPs, where respondents indicated greater challenges in rural and remote services (85 per cent),
compared to regional services (71 per cent) and metropolitan services (57 per cent).
Figure 9: Proportion of headspace service and lead agency respondents indicating the profession was difficult to
access for their local service

Source: KPMG analysis of the survey of headspace services and lead agencies

3.5.2 Services provided by headspace services

Services provided by headspace overall
In 2019-20, headspace services provided support to 90,110 young people, over 103,082 episodes of care
consisting of 403,497 OOS . Each episode of care consists of a set of consecutive OOS. As demonstrated
78F

in Figure 10 below, mental health services make up the majority of supports provided by headspace
services, followed by intake and assessment services. Intake and assessment services typically include the
initial visit a young person will make to a headspace service, where service staff and service providers will
undertake an initial assessment of their support needs. The majority of these intake services relate to young
people seeking mental health supports, however a small proportion will also relate to young people
seeking physical and sexual health, alcohol and other drug, or vocational supports. Vocational supports
provided within headspace services and recorded in the hMDS include those provided through the IPS
Program, and delivered through headspace services as there is no separate flag for IPS services.
For some headspace services, single session therapy is being increasingly used as a dedicated strategy to
manage wait times for young people. Under this strategy used by some headspace services, young people
receive one session of clinical support. This approach to single sessions of therapy was implemented
during the last year of this evaluation (2019-20), and sessions are not captured in a dedicated way in the
hMDS, therefore it is not possible to determine which OOS relates to single session therapy.
For episodes of care created between July 2019 and June 2020 that had only one OOS recorded, 65 per
cent recorded an intake and assessment service type, rather than a clinical intervention. Almost 14 per cent
of these single OOS were recorded as mental health services.

Figure 10: Services provided across every headspace OOS during 2019-20
Source: KPMG master dataset
Notes: See Appendix F for a description of how the master dataset is derived. The sample includes 403,497 occasions of service,
103,082 episodes and 90,110 young people. Vocational services include those provided through the IPS Program.

In 2019-20, new data collection was also introduced to record to whom services were being provided,
including young people, families and friends, or young people in group scenarios, and in what mode
services were provided. In 2019-20, 74 per cent of services were provided to an individual young person.
This was in comparison to seven per cent of services provided to young people with a family member or
friend present, less than one per cent of services to family or friends alone, and 3.3 per cent of services to
young people in group settings. Fifteen per cent of services did not have data recorded for whom the
services were provided.
The majority of services provided in 2019-20 were provided face-to-face in headspace services (60 per
cent), with an additional two per cent provided at headspace satellite or outpost centres. Seventeen per cent
of services were recorded as being provided over the telephone, with an additional six per cent of services
provided online or through video. The remaining three per cent of services were provided face-to-face at
another site, for example through outreach, home visits, or other external services, noting that 13 per cent
of services provided did not have data recorded. However, the impact of COVID-19 should be considered
for this year, with a significant shift to telephone-based and online services provided from March 2020. In
the months from July 2019 to February 2020, faceto-face sessions made up 79 per cent of occasions of
service delivered (noting that 16 per cent of OOS had missing service mode information).

Supports provided by service types

The types of supports provided differ across each of the service types described above. In 2019-20,
headspace centres provided 396,825 OOS, compared to 6,665 OOS within satellite services . headspace 79F

centres and satellite services provide a similar proportion of mental health services (57 per cent and 58 per
cent respectively), general assistance, vocational services and alcohol and other drug services, however
satellite services tended to provide a lower proportion of some additional services than headspace centres,
including physical and sexual health services and group work. Satellite services also provided a higher
proportion of intake and assessment and vocational services. These service profiles are in line with the
hMIF requirements that mental health services are the primary focus of satellite services . The breakdown 80F

of services provided across centres and satellites is provided in Figure 11 below.

Figure 11: Services provided across headspace services in 2019-20, by service type
Source: KPMG master dataset
Notes: See Appendix F for a description of how the master dataset is derived. The sample includes 403,497 OOS, 103,082 episodes
and 90,110 young people. A total of 109 services are included as headspace centres, and seven services are included as satellite
services. For clarity purposes, data labels are not included for categories with less than 0.5 per cent. Vocational services include those
provided through the IPS Program.

Services provided by service location

Services provided also differ depending on the remoteness of the individual headspace services. Services
in major cities across Australia provide the most mental health services. As headspace services become
more remote, other service types become more prevalent. In remote areas of Australia, physical and sexual
health, and vocational supports are more commonly used by headspace clients. This is, in part, driven by
two specific remote services, which offer specific GP clinics as part of the overall headspace model, and
the Pilbara service, which trialled a dedicated outreach model, with its supports largely focused on mental
health, intake and assessment, and group work. Group work supports are also more common in headspace
services outside of major cities.
Figure 12: Services provided across headspace services in 2019-20, by remoteness of services
Source: KPMG master dataset
Notes: See Appendix F for a description of how the master dataset is derived. The sample includes 403,497 OOS, 103,082 episodes
and 90,110 young people. A total of 61 services are located in major cities, 35 services in inner regional areas, 17 services in outer
regional areas, and 5 services in remote and very remote areas. For clarity purposes, data labels are not included for categories with
less than 0.5 per cent. Vocational services include those provided through the IPS Program.

Wait times at headspace services

Time taken for a young person to be able to access the service they require is an important measure of the
availability of headspace services. Wait times are measured at two points in the user journey of a young
person accessing headspace, at their first OOS, indicating how long they have waited from when they first
made contact with a headspace service to when they were seen for screening and assessment (wait time
one (WT1)). The subsequent wait to see the recommended service provider to meet their needs is also
measured (wait time two (WT2)), however during this time, young people generally receive access to a
range of supports, including through family and youth programs offering education and support groups.
Wait time data from the period April to October 2021 indicates that the average WT1 was 16.3 calendar
days (across 35,771 episodes of care). Within the same period, the average WT2 was 41.2 calendar days
(across 11,317 episodes of care).Wait times within the period were reviewed for variation by service
rurality, for WT1 and WT2, as seen in the table below. Within the period, headspace services located in
outer regional and remote areas have longer wait times compared with their inner regional and major city
counterparts for time between first contact and intake or assessment. Wait time between assessment and
accessing recommended support is similar in services across all ruralities, except those in remote
Australia, which were consistently shorter.

Table 10: Wait times by service rurality

Centre rurality Average wait Average wait

to WT1 (days) to WT2 (days)
Inner regional Australia 16.2 37.8
Major cities of Australia 15.7 43.2
Outer regional Australia 19.5 44.4
Remote Australia 19.1 28.3
Total average 16.3 41.2
Source: headspace National analysis of administrative data for the period April to October 2021, across 35,771 episodes of care for
WT1, and 11,317 episodes of care for WT2.

Young people attending headspace are asked to provide feedback during every episode of care about
whether they feel they have waited too long to be seen by headspace. Across all episodes of care
commenced in the period 1 July 2015 and 30 June 2020, and concluded by 31 December 2020, the
majority of young people generally indicated that they felt they had not waited too long for headspace
services, in the period to 2019-20.
While this indicates that wait times are not a primary concern for many young people, anecdotal feedback
indicates this has continued to worsen over time, since the conclusion of the data collection period for the
evaluation. It is also important to note that this feedback is only received from young people accessing
headspace services. There is no feedback mechanism or data captured for young people who do not go on
to receive support through headspace, and the extent to which wait times were a barrier to their service
access is unknown.
A range of activities are currently underway to address wait times across headspace services as part of the
headspace Demand Management and Enhancement Program (hDMEP).
Figure 13: Percentage of episodes of care where young people say 'yes' to having waited too long to be seen at
headspace

Source: KPMG analysis of hMDS data.

Notes: The sample includes 381,195 episodes with at least one OOS in the hMDS, the proportion of episodes with missing responses
during their first OOS range from 15 per cent to 31 per cent.

3.5.3 Services currently available at headspace – in conclusion

The headspace network has services across all states and territories, with these locations largely mapping
to the populations across the states and territories in Australia. The majority of headspace services operate
in metropolitan and inner regional areas and more than half of services are located within Victoria and
NSW. The number of service locations per jurisdiction broadly maps proportionately to population
distribution of young people.
While variations of the headspace centre model have been introduced as a result of government policy
decisions in recent years, over 92 per cent of services are operated as headspace centres. This proportion
will change following 2019-20 as the number of satellite services increases.
The role of PHNs is to oversee the commissioning of headspace services in their local regions. For the
evaluation period to 30 June 2020, 28 of the total 31 PHNs commissioned more than one headspace
service.
Services are delivered by a multi-disciplinary staffing team, with psychologists, social workers and
counsellors making up more than three-quarters of the staffing profile, reflecting the strong emphasis on
mental health and wellbeing in the model. The services provided directly reflect this staffing profile, with
mental health services greatly outweighing other services on offer, also consistent with the model. At the
same time, there are identified gaps in accessing key professions, particularly with regards to
psychologists, GPs and psychiatrists.
Services are generally provided directly to young people one-on-one, however some group and family
sessions are also conducted with the young person present. Occasions of service involving family and
friends of the young person, without the young person present, are rarely seen in the headspace data.

3.6 Changes to services

available over the last five
years
3.6.1 Summary of the headspace network
Services by jurisdiction
Figure 14: Growth in the number of headspace services between 2015-16 to 201920, by jurisdiction
Source: KPMG analysis of the hMDS
Note: Only headspace services that had commenced operations by 30 June 2020 are reflected in this figure. Services opened after 30
June 2020 are not included.

The headspace network grew over 20 per cent in size between 30 June 2016 and 30 June 2020, with 20
additional services added across Australia, taking the total number of services from 98 to 118. Most
jurisdictions, with the exception of Tasmania and the Australian Capital Territory (ACT), saw new
services established in this period, with the majority of these services established in NSW and Victoria.

Services by remoteness
The location of headspace services has shifted over the last five years, with the addition of more services
outside of major cities to support the reach of the network into regional and rural communities. Fourteen
services were established in inner regional and outer regional areas, with only four added in major cities.
This period also saw the first very remote service established – the Pilbara Regional Trial.
Figure 15: Growth in the number of headspace services between 2015-16 to 2019-20, by remoteness
Source: KPMG analysis of the hMDS.
Note: Only headspace services that had commenced operations by 30 June 2020 are reflected in this figure. Services opened after 30
June 2020 are not included.

Services by Primary Health Network

The additional 20 services that were added to the headspace network between 30 June 2016 and 30 June
2020 have been concentrated in a few specific areas of Australia, with 13 of 31 PHNs responsible for
commissioning the new services.
The largest increase was seen for Country South Australia PHN, with three additional services
commissioned by this PHN.
Table 11: Growth in the number of headspace services between 2015-16 to 2019-20, by PHN

Primary Health Network Number of services 2015- Number of services 2019- Increase in
16 20 services
Central Queensland, Wide 4 6 2
Bay and Sunshine Coast
Country SA 4 7 3
Country WA 5 6 1
Gippsland 1 3 2
Nepean Blue Mountains 1 2 1
North Coast 4 5 1
North Western Melbourne 5 6 1
Northern Territory 2 3 1
Perth South 3 4 1
South Eastern Melbourne 5 7 2
South Eastern NSW 3 5 2
Western NSW 3 4 1
Western Victoria 3 5 2
Source: KPMG analysis of hMDS and headspace funding data

3.6.2 Services provided by headspace services

Services provided by headspace overall
Across all headspace services, there have been some small changes to the proportion of types of supports
provided to young people. Mental health services provided to young people as a proportion of all services
have decreased by five percentage points between 2015-16 and 2019-20. A range of other services have
had small increases in proportion over time, including group work, general assistance, and vocational
services. For vocational services, this is possibly attributable to services delivered through the Individual
Placement Support (IPS) Trial, funded separately by DSS and delivered through 26 headspace services in
2019-20 . Intake and assessment services have remained mostly consistent over time, with a small increase
81F

in 2019-20. AOD services have consistently made up less than one per cent of headspace services
delivered. Group work services as a proportion of total services have increased over time.
Figure 16: Changes in the mix of services provided during each headspace OOS between 201516 and 201920

Source: KPMG master dataset

Notes: See Appendix F for a description of how the master dataset is derived. The sample includes 1,830,876 OOS, 474,977 episodes
and 426,152 young people covering 2015-16 to 2019-20. For clarity purposes, data labels are not included for categories with less
than 0.5 per cent. Vocational services include those provided through the IPS Program.

Services provided by service types

Figure 17 below demonstrates the changes in services provided by headspace services over time, based on
whether they are a headspace centre or a satellite service. It should be noted that headspace services which
fall into service categories other than centres and satellites have been excluded from this chart, due to the
small number of OOS and to avoid identifying the services.
Figure 17: Changes in the mix of services provided by headspace services between 201516 and 201920, by headspace
service type
Source: KPMG master dataset
Notes: See Appendix E.9 for a description of how the master dataset was derived. The 2019-20 sample includes 403,497 OOS,
103,082 episodes and 90,110 young people. The 2015-16 sample includes 290,834 OOS, 77,833 episodes and 70,940 young people.
A total of 95 services are included as headspace centres in 2015-16, and three services are included as a satellite service in 2018-19. A
total of 111 services are included as headspace centres, and seven services are included as satellite services in 2019-20. For clarity
purposes, data labels are not included for categories with less than 0.5 per cent. Vocational services include those provided through
the IPS Program.

Services provided by service location

Figure 18 below demonstrates the changes in services provided by headspace services across different
locations between 2015-16 and 2019-20. There are typically distinct trends visible, depending on the
location of services. headspace services in all locations have seen a decrease in the proportion of mental
health services provided between 2015-16 and 2019-20. The proportion of group work services provided
has increased in all locations except major cities, with the largest increases in outer regional and remote
locations. The proportion of physical health services has decreased in all locations, with the most
significant decrease being in remote locations, down 14 percentage points. This decrease is, in part,
attributable to the introduction of the Pilbara outreach trial, with its unique service model, and the majority
of supports provided split between mental health, intake and assessment, and group work for young
people. Remote services have also seen significant increases in vocational and group work services
provided to young people.
Figure 18: Changes in the mix of services provided by headspace services between 201516 and 201920, by service
remoteness
Source: KPMG master dataset
Notes: See Appendix E.9 for a description of how the master dataset is derived. The 2019-20 sample includes 403,497 OOS, 103,082
episodes and 90,110 young people. The 2015-16 sample includes 290,834 OOS, 77,833 episodes and 70,940 young people. A total of
57 services were located in major cities, 27 services in inner regional areas, 11 services in outer regional areas, and three services in
remote and very remote areas in 2015-16. A total of 61 services were located in major cities, 35 services in inner regional areas, 17
services in outer regional areas, and five services in remote and very remote areas in 201920. For clarity purposes, data labels are not
included for categories with less than 0.5 per cent. Vocational services include those provided through the IPS Program.

3.6.3 Impacts of COVID-19

In early 2020, the effects of the COVID-19 pandemic began to impact Australia. The pandemic
necessitated a rapid pivot of service delivery across the health and community sector to ensure significant
infection control measures, and that the health of service users and staff were prioritised. Many services
were required to swiftly mobilise telehealth service provision, something which was new to many
providers and service users.
Preliminary analysis of the response of headspace services to the COVID-19 pandemic shows that total
OOS remained relatively stable during the January to June 2020 period . As demonstrated, in Figure 19,
82F

April saw a significant substitution of face-to-face services with the use of telehealth services in the form
of online, video and telephone modes of delivery, making up 82 per cent of all OOS. By June 2020, face-
to-face services had increased but telehealth still remained the major mode of delivery.
Data for new episodes post June 2020 were not included in the evaluation, therefore further analysis to
examine the ongoing effects of COVID19 on treatment modality or effect was not possible. However,
consultations with service providers, conducted after the data period, frequently raised the pandemic and
bushfires as two recent social conditions with widely felt negative impacts on communities. Providers
described these as important for them in service planning, with the need to have therapeutic and treatment
approaches that support young people presenting with the trauma and stress from these events over the
short to medium-term.
With regards to service modality, some providers commented that young people prefer face-to-face
support, and that this should always be prioritised. Further analysis would be needed to investigate if there
will be a more significant and permanent presence of telehealth delivery, due to established infrastructure
and processes. Further analysis should also be considered to determine any variance in young people’s
presenting needs as well as any potential impact of COVID-19 on outcomes for young people accessing
headspace.
Figure 19: Service delivery modality by month from January 2020 to June 2020
Source: KPMG analysis of the hMDS.
Notes: The sample of analysis included 194,983 OOS from 58,958 completed/ongoing episodes and 54,680 young persons observed
between the 1 January 2020 to the 30 June 2020.

3.6.4 Changes to services available over the last five years – in

conclusion
The headspace network has grown substantially during the evaluation observation period, largely outside
of major metropolitan areas.
The proportion of mental health services has also varied in this time, based on the location of services,
with outer regional and remote services providing proportionately more non-mental health related services.

3.7 Understanding headspace

– in conclusion
This domain of inquiry has focused on exploring the design and implementation of the headspace model,
to evaluate its alignment to the needs of the Australian community as well as the level of fidelity seen in
the services provided through the model, examining whether the reach and take-up of the service aligns
with its intended design.
The design of the headspace model has been well articulated and is in line with international standards for
the provision of youth-friendly care. There is evidence of high levels of demand for mental health services
for young people, and different levels of need from young people across different demographic groups.
The broader literature supports the headspace model’s identification of a number of priority groups for
active engagement, and the design of the model aligns to the mental health and wellbeing needs of young
people in Australia. Stakeholder perceptions of the value and intent of the headspace model are well
aligned to the intended outcomes and objectives of the model, which are clearly defined in the program
logic underpinning the hMIF.
The most significant change to how headspace services are implemented was the introduction of local
PHN commissioning in 2016, along with a complex distributed governance model. Also in 2016, the
Individual Placement Support Program was introduced into some headspace services, to increase the
emphasis on vocational support and trial an evidence-based model linking vocational assistance with
traditional clinical mental health support . 83F

Over time, the reach and take-up of the model have improved. With increased government investment,
there has been significant growth in headspace services, from 98 in June 2016 to 118 in June 2020 and 154
services in operation by 1 May 2022. The number of headspace service locations per jurisdiction also
broadly aligns to the population size for young people.
At a national level, mental health services (57.5 per cent) provided through the headspace model greatly
outweigh AOD services (0.4 per cent), vocational support (2.2 per cent) and sexual and physical health
(1.8 per cent) . This mix of services provided through the headspace model has remained largely consistent
84F

over time, with the exception of outer regional and remote services providing a greater proportion of
support services other than mental health services.
Most services through the headspace model are provided to an individual young person, rather than to
families or groups (74 per cent of OOS in 2019-20) and most services are provided face-to-face (60 per
cent of OOS in 2019-20), noting that COVID-19 had a substantial impact on face-to-face services in 2019-
20.
Overall, the headspace model is well designed, aligned to the mental health needs of young people, and
has a reach and take up which has increased over time, in line with government investment and increased
demand.
4 Effectiveness of
headspace in
achieving program
outcomes
The following chapter examines the effectiveness of the headspace model in achieving intended outcomes.
It firstly reviews the evidence collected throughout the operation of the headspace model to assess the
extent to which this supports the model’s operational effectiveness and contributes to improved outcomes.
Evidence is then examined to assess the extent to which the model is effective in achieving its intended
outcomes. Evidence collected through the operation of the model, alongside evidence from key
stakeholders collected throughout evaluation fieldwork, provides a strong indication as to the degree of
success of the model against each of the following outcome areas:
Intermediate outcomes
• increasing mental health literacy;
• increasing early help seeking;
• increasing access to required services; and
• differences in these outcomes for ‘hard to reach’ groups.
Service system outcomes
• increasing advocacy for and promotion of youth mental health and wellbeing in their communities;
• reducing stigma associated with mental illness and help seeking for young people, their families and
friends, and the community;
• improving pathways to care for young people, including through:
o providing a localised service offering;
o other contributions to the local community;
o providing a ‘no wrong door’ approach; and
o securing support for headspace from other primary care and mental health providers.
User experience outcomes
• providing an appropriate service approach for young people with mild to moderate, highprevalence
mental health conditions;
• providing culturally appropriate and inclusive services;
• enabling young people and their families to access support where, when and how they want; and
• participation of young people in the design and delivery of headspace.
Psychosocial outcomes
• improving mental health and wellbeing outcomes, considering clinical outcomes for young people; and
• improving psychosocial outcomes through providing alternative service delivery models.
These comprise key outcomes across the headspace program logic which drive engagement, service
experience and clinical improvements in mental health.
Each of these areas is presented in summary in the chapter below, with detailed analysis provided in
Appendix D: Effectiveness in achieving intermediate outcomes, and Appendix E: Effectiveness in
improving mental health and wellbeing outcomes.

4.1 Measuring outcomes of

the headspace model
The headspace model, with its clearly articulated outcome areas, provides a strong opportunity for robust
evaluation, subject to the quality and appropriateness of data and of measurement activities undertaken
across the model. In order to understand the impact of headspace, and the extent to which it is contributing
to positive outcomes for young people, the impact areas identified in the headspace program logic all
require examination. The extent to which each impact area can be linked with engagement, treatment or
activities at a headspace service also need to be measured.

4.1.1 Overview of current measurement and reporting of

headspace performance
headspace services are required to undertake data collection, reporting and evaluation activities as part of
the monitoring and evaluation component of the headspace model . headspace National uses the data
85F

collected by service to “build an evidence base to support continuous quality improvement, guide service
innovation and inform future directions in youth mental health through advocacy and policy reform” . 86F

Each of the multiple stakeholders with which headspace services interact has different reporting
requirements and activities. An overview of the current measurement, evaluation and reporting activities
for each of these stakeholder groups is set out below.

headspace National measurement and reporting activities

Data collected
Each headspace service collects data from young people and service providers through question sets in an
electronic data collection tool called the headspace Applications Platform Interface (hAPI). hAPI was
introduced across the headspace centre network in January 2013, with a second version deployed in July
2019. This tool is maintained by headspace National and was developed after extensive stakeholder
consultation with headspace providers, headspace clients and their family and friends. It aims to capture
information about outcome data and client and family satisfaction data.
hAPI is used to collect information about OOS. It asks clients questions about themselves, why they have
chosen to use a headspace service and their level of satisfaction with the service provided. Staff members
who see the young person also complete questions about the service they provide. A follow up survey is
sent to clients 90 days after their last OOS to gather information from the client about their mental health
and wellbeing after attending headspace. There is also a survey for service providers to complete to
capture information about group therapy as well as to capture information from family and friends of the
young person. A full list of surveys hosted on hAPI is presented in Table 12 below:
87F

Table 12: Surveys on hAPI

Young Person Surveys Service Provider Surveys

• Young Person Profile • Service Provider Acknowledgement

• Young Person First Visit • Service Provider Profile
• Young Person Wait times • Registration Record (Create New Young
Person)
• Young Person Every Time
• Episode Question Set
• Young Person Outcomes
• Service Provider Phone Intake
• Client Satisfaction (optional)
• Service Provider Clinical Status Survey
• Young Person Follow Up Survey (90 days)
(this survey requires young people to opt in • Service Provider Every Time (Standard
at the commencement of the episode of Occasion of Service survey)
care, and when receiving the survey)
• Family and Friends
• Service Provider [Therapeutic] Group Survey
• Telehealth Occasion of Service Survey
• Service Provider Extra K10+ Survey
(optional)
• Service Provider Closure Survey
Source: headspace Primary Program Minimum Data Set Dictionary

Data captured through hAPI feeds into the hMDS. The data is used in three main ways:
• for capturing service provision across the headspace network;
• for evaluating and reporting on the headspace service; and
• for other agencies (such as PHNs and the department) to monitor and evaluate headspace services . 88F

Information captured through hAPI is collected and reviewed by headspace National to:
• provide a local and national perspective of service usage, trends and comparisons;
• inform local and national service planning, coordination and continuous improvement; and
• enable local and national evaluation and research relating to headspace services . 89F

In addition to hAPI, a separate online survey is available on the headspace website for family and friends
of the young person to complete (the headspace Family and Friends Satisfaction Survey ) to gather 90F

information on how to improve headspace services for family and friends supporting young people
attending headspace.
headspace National data dissemination
headspace National analyses all data collected through hAPI and reports it back to headspace services,
PHNs and lead agencies through an online, real-time dashboard tool using Tableau. The reports available
on the dashboard are outlined in Table 13 below . 91F

headspace services can see benchmarking reports comparing their performance to other services in their
peer group, which vary depending on a national baseline, rurality, operational maturity, remoteness, and
priority populations (Aboriginal and Torres Strait Islander young people, young people from culturally and
linguistically diverse backgrounds and LGBTQIA+ young people). These reports were created by
headspace National to be part of routine clinical care and support the continuous quality improvement of
service delivery, and to allow services to have a more meaningful comparison to like services outside of
the national average.
Detailed quarterly reports are developed by headspace National using the hMDS and provided to each
headspace service. These reports contain a snapshot of the service’s performance over the period as
compared to national performance. The information is related to service access (service activity, nature of
service, wait times, client demographics), effectiveness (reasons for attendance at headspace, presenting
issues, stage of illness, outcomes of K10, outcomes of SOFAS, client satisfaction, clinical diagnoses),
awareness (what influenced the client to come to headspace), sustainability (services provided by funding
stream) and integration (referrals ‘in’ and referrals ‘out’).
The hMDS is uploaded by headspace National to the Primary Mental Health Care Minimum Data Set
(PMHC-MDS). This dataset provides the basis for PHNs and the department to monitor and report on the
quantity and quality of service delivery, and to inform future improvements in the planning and funding of
,
primary mental health care services funded by the Commonwealth Government . 92F 93F

Table 13: Reports available on the headspace dashboard tool

Report Group Heading Types of Reports

Main reports (main reports used • Centre Snapshot
for headspace service reporting)
• Reporting Suite – Summary (high level summary of Key Metrics
& Demographics)
• Reporting Suite – Detail (data at the individual hMDS item level)

Operational reports (provide • Wait Times

operational insights based on
data) • Operation Report (Survey Completion rates and hAPI data entry
issues)

Outcomes (insights on clinical • headspace services – Outcomes (reporting on outcomes such as

outcomes, wait times and client K10, SOFAS and MLT)
satisfaction)
• Client Satisfaction

Specialised reports • headspace services – hMIF (hAPI data for hMIF assessments)
• Peer Groups and Benchmarking (compare service against other
services in peer group)
• PMHC-MDS Summary (high level summary of data that is
uploaded to PMHC-MDS)

Topic reports (present data based Examples include:

on different themes and topics)
• Family and Friends 2016-2021 (report from Family and Friends
Survey 2016-21)
• headspace services – Funding Source (reporting based on the
funding source hMDS item)
• headspace services – Vocational (reporting on vocational services
provided)
Source: Screenshot of dashboard tool provided by headspace National

headspace National evaluation activities

headspace National updates its detailed program logic and research and evaluation strategy every three
years. headspace National also has a partnership with Orygen, which reinforces research and evaluation
activities with a focus on understanding the mental health needs of young people and the most effective
interventions and systems of care to meet their needs . 94F

The headspace National Strategy, Impact and Policy division consults internally and draws from the
organisational strategy to develop the Evidence Building Strategy. This sets the internal agenda to create
new knowledge, evaluate headspace services, comprehensively monitor progress, and support the use of
,
evidence through effective knowledge transfer . 95F 96F

A program logic and monitoring and evaluation framework is developed by headspace National for each of
the headspace programs. Each of the programs are evaluated internally by headspace National using these
frameworks to determine whether the program or services are meeting their implementation and outcome
objectives, with the aim to improve program and service delivery and outcomes.
For each evaluation, hMDS data is augmented as required using bespoke surveys to collect data from
young people or headspace service staff. Internal evaluations and research are shared with the headspace
centre network. headspace National also commissions other entities to collect data and conduct research to
support its evaluation activities. Selected evaluations and research projects are published on the headspace
website or in peer-reviewed journal articles . Evaluation activity also allows headspace National to
97F

demonstrate and report on program delivery and effectiveness to the department and other funders.

headspace services measurement and reporting activities

headspace services actively collect and apply the data in the hMIF as part of their operations. Under the
hMIF, services are also required to use the information provided through the dashboard tool and quarterly
reports to undertake their own evaluations to improve performance and engage in a cycle of continuous
quality improvement. Each service is assessed on its ability to demonstrate this as part of the hMIF
accreditation process. Services are also required to participate in broader headspace National evaluations,
where relevant to them.
The hMIF accreditation process, as described in Section 2.4.2 above, provides an opportunity for services
to learn from best practice and service innovation and allows headspace National to disseminate
knowledge across the service network. At the same time, centre managers work on an ongoing basis with
their clinical leads, consortium partners and PHN commissioning staff to respond to trends in the data,
using these to drive staffing and scheduling decisions and to guide service offerings and program
responses to changes in presenting need.

PHN measurement and reporting activities

In their commissioning role, PHNs review six monthly performance reports from services. They also
require services to provide regular financial reporting under their funding agreements. Each PHN conducts
its own regional level needs analysis and determines its key priorities. This translates into each PHN
having its own priorities and focus for its region and this, in turn, means that performance reporting
requirements on services differ by PHN. In general, reports to PHNs are qualitative in nature, with a focus
on outcomes and achievements against work plans. Typically, these reports include information on
consortium arrangements, service improvement activities, staffing profiles and partnerships and networks
across the local community sector.
PHNs have access to the data collected through hAPI through the online dashboard tool provided by
headspace National, as well as the hMDS data uploaded to the PHMC-MDS. PHNs can review this data
for all headspace services in their region.
Alongside these activities, PHNs also report on headspace service key performance indicators (KPIs) to
their PHN Boards. Some of these KPIs are set by government, and some are created by PHNs reflecting
regional priorities. To share knowledge, PHNs also foster continuous improvement activities through
holding regular joint meetings with commissioned service providers in their regions. The frequency of
these meetings depends on the PHN.

Lead agencies’ measurement and reporting activities

Each lead agency, as the entity operating a headspace service, has its own internal measurement and
reporting requirements. Lead agencies are legally, operationally and clinically responsible and accountable
for the service, and must report to their clinical governance boards or committees on matters pertaining to
care and client related issues, who then report to the lead agency Board. In general, reporting covers
operations, budgeting, and clinical outcomes. The content of this reporting is determined by the lead
agency and varies across lead agencies.
Lead agencies also have access to the dashboard tool maintained by headspace National, giving visibility
of all services that they operate. They actively engage with headspace service management to guide
decision making at the service level about priority actions and focus areas emerging from themes in the
service data.

Consortium measurement and reporting

Some headspace services report to their consortium members, often on a quarterly basis. These reports will
differ depending on the headspace service’s arrangements, but often provide information on what the
service has focused on in the quarter, clinical services data, summaries of program activities and
community engagement activities.

Measurement and reporting for other organisations

One of the enabling components of the headspace model is blended funding, to help support the
sustainability of the model. In practice, this means headspace lead agencies bid for funding from a range of
sources, including through state and territory government programs, as well as philanthropic channels.
Funding secured through these sources generally requires a level of reporting for acquittal purposes.

4.1.2 Appropriateness of measurement and reporting activities

Strengths of the current approach
Monitoring and evaluation activities undertaken by headspace National are extensive and contribute to
continuous improvement of programs and services, as well as contributing to the evidence base of youth
mental health care more broadly. Involving young people and their family and friends in service evaluation
is also a strength, ensuring views on the performance of headspace services are gathered from those with
lived experience of service usage.
Evaluations undertaken by headspace National provide important insights into the effectiveness of
headspace programs, into how and where programs and services can be improved, and contribute to the
current evidence base regarding early intervention in youth mental health. The method by which
headspace National prioritises its evaluation activity ensures these align with broader headspace priorities
and respond to emerging areas of need as they arise. headspace National’s partnership with Orygen to
undertake research projects also demonstrates commitment to sharing information and to collaborating on
best practice to support positive outcomes in youth mental health.

Limitations of the current approach

Data collection, measurement and monitoring activities undertaken by headspace National through the
hMDS capture changes in outcomes, however there is no control group or study design to facilitate a more
robust evaluation of the impact of headspace. This is not a contractual requirement of headspace National
and, across the complex governance landscape for this program, the responsibility for measuring program
impact is not identified. At present, data collection activities are also largely focused on direct service
delivery and do not measure activity undertaken by headspace services in stigma reduction and community
engagement and partnership activity.
During consultations, service providers raised other limitations of current data collection activities. Some
limitations raised relate to challenges with capturing data through hAPI for certain groups, such as
Aboriginal and Torres Strait young people, and that data collected through the hMDS does not allow
important local insights to be captured. Additional targeted and qualitative approaches would supplement
client level minimum data captured via hAPI. The cultural appropriateness of the K10 measure of distress
for Aboriginal and Torres Strait Islander young people was also raised in consultations as an area where
further research could be undertaken, to ensure outcome measurement is reliable for Aboriginal and Torres
Strait Islander young people attending headspace.
headspace providers also report that current monitoring and reporting processes, via many reporting
channels requiring different information, is administratively burdensome and onerous. It also results in no
single organisation having a full picture of the service. Additionally, when the evaluation team consulted
peak bodies and jurisdictions, many noted it is difficult to get a clear picture of outcomes in information
presented to them about headspace, which they would value in their roles as advocates and potential
funding partners, respectively.
A further limitation noted by stakeholders is the lack of centralised economic data to identify and improve
the cost-effectiveness of headspace and maintain accountability for value for money against public funds.
There is a lack of centrally collected, reliable and complete costs data, and shared responsibility across the
distributed governance model for ensuring services are cost effective in providing support to their local
community, with no one party solely responsible for Commonwealth funding.

Assessment of measurements of objectives and impacts

The evaluation team reviewed evidence to identify if the short- and medium-term impacts presented
against headspace’s objectives in Section 2.2 are measured through existing data collection, reporting and
evaluation methods. This process involved a review of hAPI surveys presented in the hMDS, the
headspace National dashboard tool, evaluations and other publications published by headspace National on
their website, other research commissioned by headspace National and reports from services to PHNs and
lead agencies.
Both short- and medium-term impacts were mostly well measured through client satisfaction surveys (both
young people and family and friends), follow-up surveys and studies, client data on cohorts and wait
times, K10 and SOFAS surveys, research commissioned by headspace National, and information contained
in reports to PHNs on consortium arrangements, partnerships and networks. However, there were areas
where it was difficult to identify specific data that measured particular impacts, including:
• how AOD and/or physical and sexual health assistance impacted on young people in gaining skills to
better manage these aspects of their lives in both the short, and medium- to longterm; and
• a direct measure of how young people and families experience more streamlined and less fragmented
pathways of care in the medium-term.
These are areas where headspace National conducts one-off, qualitative evaluations on identified priorities.
While short- and medium-term impacts are mostly measured and reported, consistent measurement of
headspace’s longer-term impacts for young people is not in evidence. Outcomes such as enhanced service
provision and access, improved health outcomes and increased social and economic outcomes for young
people over their life course are part of the headspace service model program logic, and strategies to
measure impact in these areas could be implemented over the long-term. While headspace National has
undertaken a one-off study about longer-term impacts , there is opportunity to develop this further and
98F

sustain studies over time.

Opportunities for refinement

There are a number of potential opportunities for refinement in data collection and measurement activities
that could allow headspace to better demonstrate its positive outcomes. The methodology used to measure
objectives and impacts does not generally include the use of experimental or quasiexperimental designs.
The lack of a control group or counterfactual view of the outcomes in cases where headspace is not
accessed prevents a rigorous test of the impact of the service. A funded cohort study of headspace clients
comparing the pre and post mental health outcomes to comparable non-headspace clients should be
undertaken to highlight the causal impacts of headspace services . Additionally, data linkage should be
99F

established to allow comparisons between young people who have attended a headspace service with
young people who have not. Accessing and linking MBS, PBS, hospitalisations and emergency department
presentations data for these individuals is possible, however, the evaluation recognises this may be difficult
in practice. Also, the evaluation recognises that data from the MBS or PBS can answer some, but not all,
questions. For example, the difference in frequency of mental-health MBS items before and after a
headspace episode.
Data linkage to other datasets should be prioritised, such as those held by the AIHW (for example, AOD,
Emergency Department (ED), other mental health datasets) and administrative data held by Services
Australia (employment and study), that would help facilitate the development of a control group against
which to compare headspace outcomes.
As the hMDS does not collect data by individual MBS item, data linkage is currently not feasible.
headspace National commenced negotiations with the NSW State Government to develop a data linkage
project and, in consultations with the evaluation team, other jurisdictions have expressed their interest in
undertaking a similar exercise, indicating a view that there is strong support for data linkage activities
across the service system. To date, this linkage exercise has not been pursued.
Consideration should also be given to broaden the coverage of what is captured in hAPI data. Current data
collection only captures effort made in relation to a young person from intake onwards. This data capture
does not take into account community engagement activities, ongoing general enquiries from surrounding
services, stigma reduction and mental health literacy activities. headspace National has explored this issue
as part of the latest redevelopment of the hMDS, following feedback from services that there should be a
method to collect information on other work they undertake to support young people outside of direct
service provision, for example secondary consultations with family, other health providers including
referrers, and engagement with a young person in the service that has referred them to headspace.
In addition, headspace services indicated that data capture should also extend to understanding community
engagement, outreach activities and other non-clinical measures. Concerns regarding the burden of this
data capture have prevented implementation to date. This trade-off between data collection and utility must
be assessed. However, capturing this information in a systematic way could be a valuable opportunity for
headspace to demonstrate the work that is undertaken beyond direct service provision, and whether and
how broader engagement and outreach activities achieve positive outcomes.

Measuring outcomes of the headspace model - in conclusion

While a broad range of data is collected from headspace services and users and is made available to
internal stakeholders, this does not extend to community engagement or service integration. Introducing
approaches to collecting this data on a regular basis would improve visibility of the value of these
activities, but would need to be balanced against an impost on providers. For example, a program of short,
periodic data collection on a rolling schedule could be developed to address these gaps.
The effectiveness of the headspace model is evaluated in an ongoing program of one-off studies.
Investment in data linkage across data sets would enable a quasi-experimental approach to evaluation of
the effectiveness of headspace in impacting outcomes for young people, strengthening the evidence and
introducing the potential to analyse long-term impacts for young people of engaging with headspace. This
would require additional resources to undertake this or similar research activity.

4.2 Effectiveness of the

headspace model
As outlined in Section 1.1, a key aim of this evaluation is to assess the effectiveness of the headspace
model in achieving its key objectives. This section discusses the effectiveness of headspace services in the
following outcome areas:
• Intermediate outcomes;
• Service system outcomes;
• User experience outcomes; and
• Psychosocial outcomes.
These outcomes are described in more detail at the start of Section 3 above.
As a set of objectives, these represent key outcomes across the headspace program logic which drive
engagement, service experience and clinical improvements in mental health.
The next sub-sections provide analysis and findings from the detailed evaluation of headspace services
against each of these areas, with Appendix D and Appendix E presenting the full analysis of the evidence
base reviewed.

4.2.1 Improving intermediate outcomes

Increasing mental health literacy
In the headspace context, mental health literacy refers to knowledge about mental health, how to manage
mental health and how to go about accessing support with mental health concerns. Through improving
mental health literacy, the headspace model supports the medium-term impact that young people are better
able to manage their mental health in the medium- to long-term, including identifying when they need to
seek help and support. Ultimately in the model, improved mental health literacy contributes to long-term
impacts of improved health outcomes for young people and increased social and economic participation
outcomes for young people over their life course.
As the world experiences unprecedented challenges in the face of COVID-19, good mental health literacy
in young people and their key support people may lead to better outcomes for those with mental ill-health,
either by assisting early help seeking by young people themselves, or by their support people identifying
,
early signs of mental disorders and seeking help on their behalf .
100F 101F

Previous evaluation work undertaken by headspace National indicates that the headspace model is
considered to be effective in building mental health literacy for young people. For example, Colmar
Brunton conducted a survey which found consistent feedback across stakeholder groups that headspace
supports better understanding of mental health, ill health and seeking help, which all contribute to
increased mental health literacy .
102F

In order for the current evaluation to examine the extent to which the headspace model is effective in
increasing mental health literacy, a range of data and evidence was reviewed from across the fieldwork
activities conducted for this project. These are described in Appendix D.1 and include analysis of young
people’s views collected through the hMDS, interviews with young people who are current or former
headspace service users, and survey responses from service and lead agency staff.
Young people using headspace and staff working within the headspace model have strong, positive views
about the effectiveness of headspace in increasing mental health literacy, and consider that the more
contact a young person has had with headspace, the stronger their positive views of headspace’s impact on
their mental health literacy. Young people interviewed highlighted useful strategies headspace had given
them in identifying and managing their mental health issues, while they also indicated that finding a staff
member at headspace with whom they could build a rapport was an essential enabler of improving the
mental health literacy of young people.
Staff from headspace services and lead agencies consider increasing mental health literacy to be a strength
of the headspace model. They describe clinical, community and information related activities across the
hMIF as integral to this success, indicating that improving mental health literacy is embedded across the
headspace model. They also highlighted that workforce pressures, wait times at headspace services and
limited capacity across the local service system to engage during COVID-19 restrictions were key barriers
to successfully increasing the mental health literacy of young people.
Evidence from young people using headspace and headspace service staff indicates that the
headspace model is effective in increasing the mental health literacy of young people engaging with
its services.
The key enabling elements of the headspace model which support mental health literacy are:
• community awareness and engagement; and
• multi-disciplinary workforce.

Increasing early help seeking behaviour

In the headspace program logic, increasing early help seeking is key to improving short-term impacts for
young people and families in increasing their knowledge about, and willingness to seek help with, mental
health issues. It is also associated with having improved attitudes towards mental health and mental illness.
These, in turn, relate to a range of medium-term impacts around help seeking, early identification of
emerging mental health problems and increased help seeking behaviour. In the headspace context, these
are all identified as contributors to the long-term impacts the headspace model is seeking for improved
outcomes over the life course.
The headspace model includes ‘early intervention’ as a core service component, defined as “the
identification and provision of intervention and support services as early as possible in the development of
mental health difficulties to prevent or delay the onset of mental ill-health or reduce the impact associated
with mental ill-health and improve outcomes” . 103F

Mental illness for young people usually manifests before the age of 21, indicating the importance of
,
treatment and assistance provided early in life, early in illness and early in an incident . Early intervention
104F 105F

programs assist a young person by identifying risk factors early or providing timely treatment for problems
that can alleviate the potential harm from mental illness. Treating risk factors and symptoms early is seen
as not only improving the social and emotional wellbeing of young people, but also as a cost-effective
approach to improving lifelong outcomes for them . 106F

A key evaluation question for this project examines the extent to which the headspace model is associated
with increased levels of early help seeking from young people. For the purposes of this evaluation, early
help seeking is defined as a young person engaging with headspace when they are:
• under 21 years of age;
• at relatively low mental health risk status; and/or
• assessed as at less than the threshold stage of illness.
To examine the extent to which the headspace model is succeeding in contributing to increased early help
seeking behaviour, relevant data and evidence was reviewed from across the fieldwork activities conducted
for this evaluation. These are described in detail in Appendix D.2, and include analysis of the hMDS,
interviews with young people, interviews with Youth Reference Group members, interviews with
university and school counsellors and survey responses from service and lead agency staff.
Evidence reviewed from a range of sources indicates that early help seeking is an area of continued focus
for the headspace model, with generally good results despite barriers. Administrative data in the hMDS
regarding the age, mental health risk status and stage of illness of young people presenting in the period
for an OOS indicate that around three-quarters of young people presenting were aged under 20 years.
Just under half of those presenting (46.1 per cent) in the period had either ‘no risk factors or symptoms of
mental health problems’ or ‘risk factors present’, indicating the presence of one or more situational factors
making them vulnerable to developing a mental health problem. Furthermore, just under half (41-48 per
cent) of young people presenting at headspace in the period had ‘no symptoms of mental health problems
or disorder’ or ‘mild to moderate general symptoms of mental health problems and/or high risk
psychosocial stressors’ (e.g., bullying or relationship problems). These data provide a broad indicator of
‘early help seeking’, where the young person is presenting at a young age, at an early stage of illness or
with low risk factors. They show that a substantial proportion of young people attending headspace meet a
broad definition of ‘early help seeking’.
Data also demonstrates that staff at headspace services are confident that their service drives increases in
early help seeking behaviour, with 87 per cent of service and lead agency survey respondents selecting
‘Very Well’ or ‘Well’ in response to this question. Enablers of this were identified as community
engagement activities and strong brand recognition, while wait times, workforce limitations and
misconceptions of headspace as being for high-needs young people were identified as barriers, and also
raised by school and university counsellors and young people who do not use headspace.
This data provides a range of lenses through which to examine the question as to whether headspace is
improving early help seeking in young people. The model is working well, however pressure points in the
system around demand and workforce issues, as well as perceptions of headspace’s intended purpose,
remain barriers to the promotion of early help seeking. With increased access by those from younger age
groups, and relatively stable levels of access by young people with low mental health risk factors and in
the early stage of illness, data indicates that almost half of those attending headspace are seeking help at a
young age or for mild mental health conditions, and therefore are engaging in ‘early help seeking’.
This evidence suggests that headspace is effective in improving early help seeking behaviour in
young people.
The key enabling elements of the headspace model which support early help seeking behaviour are:
• community awareness and engagement; and
• multi-disciplinary workforce.

Increasing access to required services

Young people accessing required services through headspace is central to the headspace program logic.
Young people and families being able to access services in a timely manner at low or no cost leads to
young people receiving appropriate, evidence-based treatment early and increased help seeking behaviour
into the medium-term. Again, this leads to long-term impacts in the headspace program logic for improved
outcomes over the life course.
As supported by the Royal Commission's work into Victoria's mental health system, there are a number of
barriers many young people come up against when seeking care. Demand has overtaken capacity,
community-based services are under-supplied, unsuitable or driven by crisis, services are poorly integrated
and families, carers and supporters are often left out .
107F

Given the increased level of funding the headspace model has received in recent years, and the expansion
in the number of services around the country, a key measure of its effectiveness is to also examine the
extent to which the increased number of services and service funding is associated with an increase in the
overall number of young people accessing headspace.
Data from a range of sources, including hMDS data on access rates, interviews with young people and
interview and survey data from headspace service providers (as detailed in Appendix D.3), indicates that
headspace is effective in increasing access to required services, but that workforce and demand pressures
continue to constrain the volume of young people able to access support.
At the same time, hMDS data demonstrate that, over time, the number of young people accessing support
through headspace has increased steadily with the increase in number of services. Young people and
headspace staff value the ‘soft entry’ approach to accessing headspace through GPs or schools, mostly
face-to-face but with flexible options during COVID-19. Flexibility in opening hours was also valued, and
features of the physical site were identified as improving access, for example a central location close to
public transport.
Barriers to increased access raised by various stakeholders were long waiting times for the young person
between the intake session and when they are assigned to a counsellor or psychologist. Insufficient funding
for salaried staff was also raised, including community engagement staff, and the costs of an accessible
site.
Data indicates that headspace is effective in increasing access to required services.
The key enabling elements of the headspace model which support increased access to required services
are:
• community awareness and engagement;
• enhanced access (minimising barriers to seeking professional help); and
• multi-disciplinary workforce.

Increasing mental health literacy, early help seeking and access for ‘hard to reach’
groups
Mental health literacy, early help seeking and access are important precursors to further engagement with
the supports young people need to assist with their mental health and wellbeing and, overall, this
evaluation has found that headspace services are effective in supporting these outcomes.
When examined in terms of how effectively the headspace model supports these outcomes for ‘hard to
reach’ groups, the findings are more mixed. Based on feedback from stakeholders across the headspace
landscape, as well as on broader academic and grey literature regarding stigma and service access, for this
evaluation, ‘hard to reach’ groups include:
• Aboriginal and Torres Strait Islander young people;
• young people from culturally and linguistically diverse backgrounds;
• young people who identify as LGBTQIA+; and
• young people with disability .
108F

Engaging groups considered to be marginalised from mainstream health services can be difficult due to
ongoing perceptions and experiences of stigma and discrimination . Groups already at high risk of stigma
109F

include Aboriginal and Torres Strait Islander young people, young people who identify as LGBTQIA+,
young people with disability and young people from culturally and linguistically diverse communities,
where there may be significant stigma and taboo associated with mental illness. The Mission Australia
Youth Survey 2021 highlighted that, among young people participating in the study, 51.5 per cent cited
mental health as their top obstacle to success, compared with 83.2 per cent of young people of gender
diverse backgrounds who reported poorer mental health on numerous measures. In this study, although the
majority of Aboriginal and Torres Strait Islander young people engage in education, value their family and
friends, and feel positive about the future, they also report greater challenges than their peers who do not
identify as Aboriginal or Torres Strait Islander, including being less likely to feel happy or very happy with
their lives .
110F

Other research highlights enduring issues with systemic barriers and unconscious bias within the health
system, where young people from diverse backgrounds are less likely to have their needs met, due to
,,,
factors such as lack of cultural competence and misdiagnoses .111F 112F 113F 114F
It is important to note that young people within these groups are diverse and have a variety of experiences
and perspectives on issues associated with mental health. At the same time, exploring the evidence for how
well the headspace model supports members of these groups allows for consideration of potential systemic
factors which may reduce its efficacy for young people across the spectrum of potential life experiences.
The evidence for how well the headspace model supports young people from ‘hard to reach’ groups across
key objectives of the model is detailed in Appendix D.4. While there is a high degree of similarity in
feedback from these groups across the key objectives, there is also some evidence to suggest that meeting
the needs of some stakeholder groups is a greater challenge than for others. In particular, workforce
shortages of key staff reduce the ability for services to make young people from ‘hard to reach’
backgrounds feel welcome.
As detailed in Appendix D.4, findings show that ‘hard to reach’ groups do not see comparable increases in
mental health literacy, early help seeking or increased access to required services compared to the general
population of young people.
Mental health literacy
Self-reported mental health literacy improvements are similar across all cohorts of young people attending
headspace, however these data do not include an indicator of ‘with disability’, preventing further insight
into the mental health literacy of young people who identify as having a disability.
headspace service providers indicated that Aboriginal and Torres Strait Islander young people, culturally
and linguistically diverse young people and young people with disability all fare below the general
population of young people supported by headspace in terms of the service's impact on their mental health
literacy. Results from staff also indicate that engagement with LGBTQIA+ young people result in better
mental health literacy than for other groups of young people.
Early help seeking
On measures of early help seeking, there are a number of variations across groups.
Aboriginal and Torres Strait Islander young people are significantly more likely to be under the age of 21
compared to the general population of young people attending headspace. They are similarly likely to be
presenting with low mental health risk as young people from the general population, but significantly less
likely to be presenting in early stages of a disorder than the general population of young people attending
headspace.
Culturally and linguistically diverse young people are significantly more likely to be older than 21 years of
age when attending a headspace service, but are equally as likely as young people from the general
population to be presenting with low mental health risk factors or early stages of a disorder.
LGBTQIA+ young people are significantly more likely to be older than the age of 21 when attending a
headspace service, in line with general patterns of help seeking for this group . They are also significantly
115F

less likely to present with low levels of risk factors, but are equally likely as young people from the
general population to be presenting in the early stages of a disorder.
The hMDS does not ask the young person whether they identify as having a disability, preventing similar
analysis of early help seeking to be undertaken for this group.
Overall, these indicators of early help seeking show mixed results for young people from these ‘hard to
reach’ groups undertaking early help seeking for their mental health and wellbeing. Stakeholders reported
the importance of outreach as a key part of the headspace model to improve young people’s willingness to
seek help.
Increased access to required services
On measures of access, data from the hMDS shows that, over time, access rates have slightly improved for
‘hard to reach’ groups, however those working within headspace indicate the service is less effective in
supporting the access rates of Aboriginal and Torres Strait Islander young people, culturally and
linguistically diverse cohorts and young people with disability. LGBTQIA+ young people were again
perceived to be better supported, with higher perceived rates of access than all other groups, which is
upheld by administrative data on young people attending headspace.
Feedback from young people highlighted the continued importance of having staff from the young
person’s cultural group, and the need for ongoing work to reduce stigma and build trust in order to support
access for ‘hard to reach’ groups.
Achieving intermediate outcomes in support of ‘hard to reach’ young people
There are wide variations between groups on perceived improvements of mental health literacy, as
reported by headspace service providers. Young people from ‘hard to reach’ groups are also less likely to
be undertaking early help seeking when attending a headspace service. While access rates have improved
over time for these groups, access rates of Aboriginal and Torres Strait Islander young people, culturally
and linguistically diverse cohorts and young people with disability are not as well supported as for other
young people.
The headspace model does not achieve the same results for ‘hard to reach’ groups compared to the
general population of young people.
The key enabling elements of the headspace model which support intermediate outcomes for ‘hard to
reach’ groups are:
• community awareness and engagement; and
• multi-disciplinary workforce.

Improving intermediate outcomes – in conclusion

The headspace model is effective in supporting important intermediate outcomes, mental health literacy,
early help seeking and increased access to required services, which in turn improve the likelihood that
young people will seek support with their mental health and achieve improved psychosocial outcomes in
the longer term. When considering the model’s effectiveness in supporting ‘hard to reach’ groups with
these intermediate outcomes, overall, this evaluation has found that headspace services achieve mixed
success in supporting these outcomes.
The key enabling elements of the headspace model which support these outcomes are:
• community awareness and engagement;
• enhanced access (minimising barriers to seeking professional help); and
• multi-disciplinary workforce.
These elements are the focus of potential recommendations to support better outcomes for ‘hard to reach’
groups, detailed in section 5.4.

4.2.2 Improving service system outcomes

Increasing advocacy for, and promotion of, youth mental health and wellbeing in
their communities
In support of the various intended outcomes of the headspace model, a key component of activity focuses
on engaging with communities. The headspace program logic lists the following activities as part of this
work:
• promoting headspace services to local community and services, and promoting early help seeking for
young people aged 12 to 25;
• facilitating engagement and participation with young people and their families to better understand
community needs;
• engaging with GPs, schools and other local organisations to better understand community needs; and
• delivering community awareness activities, including psycho-education, mental health literacy and
stigma reduction activities.
In order to assess how well headspace advocates for and promotes youth mental health and wellbeing in
their communities, feedback on these activities was sought through a range of data collection activities.
These include interviews with Youth Reference Groups, interviews and focus groups with young people,
surveys of service and lead agency staff, and discussions with staff, GPs and other stakeholders at a
number of services during deep dive fieldwork (see Appendix D.5 for further details).
Stakeholders reported that headspace services are active in advocacy and promotion, and highly visible in
their local communities. Work to promote mental health literacy and help seeking with schools,
universities and community organisations more broadly received positive feedback. As discussed in
Section 3.1, however, the extent to which these activities are occurring is not measured through the hMDS
or other means. Feedback from services also indicates that many stakeholders suggested this activity is
under-resourced.
Advocacy and promotion activities are a key component of the work of headspace services, and feedback
gathered through this evaluation indicates that this is recognised as an effective aspect of their activities.
The evidence indicates that headspace is effective in supporting youth mental health through
advocacy and promotion activities.
The key enabling elements of the headspace model which support advocacy and promotion of youth mental
health and wellbeing in communities are:
• community awareness and engagement; and
• multi-disciplinary workforce.

Reducing stigma associated with mental illness and help seeking for young people,
their families and friends, and the community
Stigma in the context of the headspace model is the fear or embarrassment of seeking help for mental
health and wellbeing, and the negative judgement of, and lack of empathy for, those who do. In the
headspace program logic, stigma is identified as a blocker, preventing young people from being able to
identify when they need help and from seeking that help early.
National research into stigma indicates that most people in Australia with mental illness report
experiencing stigma, however the severity, nature, and experience of stigma vary depending on factors
such as mental illness type, age, gender, and cultural background . Approximately 29 per cent of people
116F

with mental illness reported discrimination or unfair treatment in the past year, as opposed to about 16 per
cent of those without mental illness. People with severe mental illnesses are likely to face high levels of
stigma, according to the 2011 National Survey of Mental Health Literacy and Stigma, although the nature
of stigma differs among illnesses. The impact of stigma may include preventing people who suffer from
mental illness from being able to engage socially or feel included. This stigma can lead to discrimination,
social exclusion and a reluctance to seek care .
117F

In order to examine whether headspace has been associated with a reduction in mental health related
stigma, this evaluation sought the views of headspace service and lead agency staff through both survey
and fieldwork methods, as well as reflections from school and university counsellors and young people
who do not use headspace, to gauge their views on how effective headspace has been in this domain (see
Appendix D.6 for further details).
Overall, the evidence collected suggests that stigma reduction activities are a continued focus of headspace
services, as they are for other services and organisations across the mental health sector. Discussions also
indicated that, for some families and segments of the community, stigma around mental health help
seeking continues to be strong, and services are continuing to focus efforts, including outreach, recruitment
and other engagement strategies, to reduce stigma and encourage support of mental health help seeking. A
number of cultural groups were discussed in these fieldwork conversations, along with the particular
challenges for young people from some culturally and linguistically diverse backgrounds where mental
illness is not easily accepted or understood.
Young people discussed how schools and the media are also working to improve rates of stigma, and that
the work headspace does is one of many things happening to help in its reduction. On balance, views are
positive that stigma around mental health and mental illness is reducing, and those close to the model
believe headspace has made a positive contribution.
This qualitative evidence indicates that the work of headspace service providers in community engagement
and mental health promotion and advocacy is considered by stakeholders to be an effective contribution to
stigma reduction around mental illness and help seeking. Further detail of qualitative data collected is at
Appendix B.
Overall, the evidence collected suggests that stigma reduction activities undertaken as part of the
headspace model are effective.
The key enabling elements of the headspace model which support stigma reduction associated with mental
illness and help seeking for young people, their families and friends, and the community are:
• community awareness and engagement; and
• multi-disciplinary workforce.

Improving pathways to care for young people

In the context of the headspace model, activities to promote service integration and coordination are
designed to improve pathways to care for young people.
Integration refers to individuals and organisations in different areas and sectors working together and
aligning their practices and policies to deliver high quality mental healthcare and achieve good outcomes . 118F

In the headspace model, service integration refers to bringing services together to function as one,
providing a seamless service experience for a young person, particularly if they require care involving
multiple service providers and supports .
119F

In the context of mental health services, there are two ways services can typically be integrated – vertically
and horizontally. Vertical integration refers to how services at different levels of healthcare, for example
primary, secondary and tertiary, work together to deliver services to an individual as the severity of their
condition changes over time. Horizontal integration refers to how services from different sectors or sub-
sectors work together, such as physical and sexual health and mental health services, to support the various
needs of an individual across multiple aspects of their health or wellbeing. Vertical or horizontal
integration may also occur between mental health and other service systems, such as housing or
employment .120F

Ensuring people, including young people, have access to services and supports they need where and when
they need them is critical to a well-functioning mental health service system. However, the Productivity
Commission has identified that, nationally, there are challenges with current pathways between care and
service integration across the entire mental health service system. These challenges include:
• the complex and disjointed nature of the mental health service system;
• a lack of information sharing and coordination between services, impacting on outcomes; and
• inconsistent services providing overlap in some areas and for some cohorts of people, with no services
for other groups .
121F

PHNs have a lead role to play in building linkages across the local service system and, as part of the
national network component of the headspace model, headspace services must work with PHNs in the role
of local commissioner of their services to link with other services. The establishment and maintenance of
effective community consortia is another key task for services to promote improved pathways to care for
young people. At the same time, the community awareness and engagement element of the model also
requires headspace services to work with their local community to build relationships and referral
pathways for young people in their care.
Evidence from young people, headspace service providers and other external stakeholders was examined,
as detailed in Appendix D.7, and was used to evaluate headspace’s effectiveness in improving pathways to
care through service integration and coordination.
Qualitative data show that young people and their families, and other external stakeholders, highly value
service integration and care coordination, to ensure young people are connected to other required services
when they need them.
Most young people accessing headspace indicate they received appropriate referrals to other services. At
the same time, a small number of young people who used alternative service providers such as GPs to
support their care pathway, or who did not feel they received the appropriate referral they needed from
their headspace service, reported mixed experience.
Under the service integration component of the headspace model, care providers are brought together,
often under one roof, to provide seamless care for a young person requiring multiple services and supports.
Evidence from headspace services and other external stakeholders indicated that headspace services
undertake a range of activities to support pathways to care through integration and care coordination.
These contributions were consistently recognised across stakeholder groups.
However, headspace services face a range of barriers to their ability to improve service integration and
care coordination. A key element of the model raised here was difficulty in accessing a multidisciplinary
workforce, which is designed to combine clinical and non-clinical staff to work together to holistically
meet the needs of young people.
Currently, there are capacity constraints within many health services, with integration difficult where a
service cannot take a referral, or is unable to work with headspace services to improve care coordination.
There are also instances where alternative services are not available, particularly in regional and remote
communities.
Similarly, headspace services encounter difficulties engaging in these activities at points in time based on
demand for services, and the need to balance clinical workloads with these additional activities and
managing referrals with existing wait times. For these reasons, there was mixed evidence from other
providers in the sector as to the effectiveness of headspace in supporting pathways to care through
integration and coordination.
Evidence from young people, headspace service providers and other external stakeholders indicates
that the headspace model is effective in improving pathways to care, however there are challenges
which impact this work, many of which are outside the control of headspace and rely on effective
functioning of the broader service system.
The key enabling elements of the headspace model which support improving pathways to care for young
people are:
• community awareness and engagement;
• service integration;
• national network (in particular the roles of PHNs and community consortia); and
• multi-disciplinary workforce.

Providing a localised service offering

In order to successfully meet the mental health needs of young people, headspace services need to be
tailored to the local service system, working closely with other providers as well as with schools and other
community organisations to build relationships with young people and their families.
PHNs play a central role in ensuring services are localised, responsive to the needs of the local community
and well-integrated. In their commissioning role, PHNs work with local headspace service providers to set
priorities and target activities to respond to local need.
Evidence from headspace service providers, consortium members, PHNs and school and university
counsellors was reviewed to explore the extent to which the headspace model is successful in providing a
localised service offering (details are collated at Appendix D.8).
Evidence from these sources indicates that headspace services work effectively with local communities
and providers to build relationships and understand what local needs services should target. The
consortium model, commissioning process and community engagement activities support headspace
services to localise their offerings, strengthening referral pathways and relationships to assist them to
support young people to access local services that align to their needs, where available.
PHNs reported that many services are well-integrated into their local communities, and provide services in
demand in their local community. headspace service providers indicated that community engagement
activities assist them to identify how best to respond to local need, and some lead agencies have a specific
focus on supporting these activities by also applying for additional grant funding from alternative sources
(such as state government grants) to support this work.
However, the evidence also indicates that the extent to which services are localised varies between
services. This variation is due to a range of factors, including:
• Capacity within the headspace service for community engagement and exploration of local needs, which
is constrained by high demand, staffing pressures and funding limitations. Some headspace services do
not have dedicated community engagement positions, and community engagement is often
deprioritised due to clinical service loads within services.
• headspace services’ ability to recruit specific workers or professions to deliver on specific support needs
in communities, particularly in regional and remote areas, for example Aboriginal wellbeing workers.
• Some PHNs indicated challenges as the local commissioning agency for headspace services in tailoring
services to the needs of the local community, while ensuring services still meet the requirements of the
hMIF. These PHNs also indicated that there is a lack of flexibility to use funding provided for a
headspace service to design localised services which directly address the specific needs of the
community. While some tailoring is afforded through headspace services, this does not allow the PHN
to commission a tailored service targeted at local need.
Evidence from headspace service providers and other external stakeholders indicates that the
headspace model enables localised services to a good extent, however this is inconsistent across
services and the local commissioning role of PHNs may be under-utilised.
The key enabling elements of the headspace model which support providing a localised service offering
are:
• community awareness and engagement;
• national network (in particular the roles of PHNs and community consortia); and
• multi-disciplinary workforce.

Other contributions to the local community

In addition to direct clinical, centre-based and other services provided to young people, a range of other
contributions and activities are also provided by headspace services, often outside of the headspace
service.
As described further in Appendix D.9, evidence from young people, headspace service providers and other
external stakeholders indicated that the headspace model also provides a range of additional contributions
to local communities that are highly valued by those communities. headspace is effective in using these
contributions to support other program objectives; however these are, at times, impacted by capacity
constraints within services.
Qualitative data demonstrates strong recognition of the contributions that headspace services make to their
local communities, through schools and other education institutions, community events and engagement,
and availability of resources and information.
All stakeholder groups had positive views as to the impact of these contributions, and these were tied to
other outcomes discussed above, such as improvements to mental health literacy, early help seeking, and
access for young people.
Furthermore, young people reported that the headspace model provides them with valuable development
opportunities. Being part of the governance and planning of services provides young people with
experience and improved capability, and has the potential to increase their confidence and self-esteem.
Evidence from young people, headspace service providers and other external stakeholders indicates
that the headspace model provides a range of additional contributions to local communities that are
highly valued by those communities.
The key enabling elements of the headspace model which support other contributions to the local
community are:
• community awareness and engagement; and
• youth participation.

Providing a ‘no wrong door’ approach

The ‘no wrong door’ approach is embedded within the enhanced access component of the headspace
model, where the service ensures all young people who contact or present are screened, and then are either
supported or re-directed to more appropriate local services and supports . This element of the model is
122F

designed to support mental health literacy, early help seeking and access to services. It is also intended to
support young people to get help when they need it, regardless of the severity of their mental health
problem.
Evidence from young people, service and lead agency staff, PHN representatives and community
consortium members was analysed for evidence of the effect of the ‘no wrong door' approach on the
headspace model and how well it achieves its outcomes (Appendix D.10).
Overall, there was significant support for headspace’s ‘no wrong door’ approach as a key enabler of
supporting young people. Evidence shows the approach supports young people by:
• ensuring they are able to engage with mental health supports in a way they feel comfortable;
• providing a free entry point into the mental health service system;
• providing a soft entry point into the mental health service system, with referrals to other services
available to support service integration for young people; and
• providing young people with access to initial services to support broader objectives, such as improved
mental health literacy and early help seeking, even where they may be referred to a more appropriate
service.
However, in discussing the ‘no wrong door’ approach, service and lead agency stakeholders consistently
raised that they report young people’s mental health needs are becoming increasingly severe and more
complex, with many cases being outside of the headspace model’s mild to moderate criteria. Common
presenting concerns were reportedly developmental disorders, personality disorders, eating disorders,
complex trauma and grief, and self-harm and suicidal ideation, including in ages under 12 years . 123F

headspace service staff interviewed commonly described a “missing middle” of clients who are too
complex to be seen under the headspace model’s mild to moderate remit, but who are not unwell enough to
be transitioned to overwhelmed TMHSs. These headspace providers indicated that the ‘no wrong door’
approach, coupled with other challenges in the service system such as referral services with limited or no
capacity for new referrals, significantly impacts headspace’s core business of supporting young people
with mild to moderate, high-prevalence mental health conditions and other contributions to communities
through outreach and engagement.
Evidence from young people, headspace service providers and other external stakeholders indicates
that the ‘no wrong door’ approach is an important and valued feature of the model, supporting
improved mental health literacy, early help seeking and access to required services. At the same
time, however, the level of demand for mental health support, and the volume of young people who
use headspace as the entry point into support leads to increased wait times for young people,
particularly those in the ‘mild to moderate’ group who are the headspace model’s primary target
cohort of young people.
The key enabling elements of the headspace model which support successful provision of the ‘no wrong
door’ approach are:
• enhanced access;
• service integration;
• national network (in particular the roles of PHNs and community consortia); and
• multi-disciplinary workforce.

Securing support for headspace of other primary care and mental health providers
In order to operate successfully, the headspace model requires services to work collaboratively and build
positive relationships with other services throughout their local service system and referral pathways.
The extent to which the headspace model is supported by other primary care and mental health providers
was explored through this evaluation, through interviews and focus groups with PHNs, school and
university counsellors, community consortium members and a small number of GPs (Appendix D.11).
Evidence demonstrates that there are a range of factors that impact on the level of support primary care
and mental health providers have for headspace, and in particular their likelihood to make referrals to
headspace services. These factors include concern about wait times within headspace services, challenges
in engaging in coordinated care with headspace services, and in building relationships with headspace
services when there is staff turnover.
The overall level of support for the headspace model is high, and headspace is viewed as a vital
community service. At the same time, day-to-day operational challenges associated with supporting young
people to find appropriate care were frequently raised by stakeholders when asked about their level of
support for the headspace model. Some of these challenges could be ameliorated, for example, there may
be opportunity to address staff turnover through adjustments to the funding model, ensuring adequate
reimbursement, to ensure competitive arrangements within the context of other services. Additionally, the
sharing of information between providers could be prioritised, so that the service pathways through which
a young person transitions are documented and information about outcomes is shared. While this issue is
not confined to the headspace model, as the highest profile and extensive form of support available for
young people in Australia in mental health service delivery, the headspace model could be a powerful tool
in improving care coordination across local service sectors.
The headspace model benefits from generally high levels of support from other primary care and
mental health providers, although operational pressures affect individuals’ referring decisions and,
at times, create frustrations.
The key enabling elements of the headspace model which assist in securing the support of other primary
care and mental health providers are:
• community awareness and engagement;
• service integration; and
• multi-disciplinary workforce.

Improving service system outcomes – in conclusion

The headspace model supports outcomes in improving the mental health service system, however there is
mixed evidence for how well it achieves these outcomes in certain areas.
The headspace model is effective in supporting youth mental health through advocacy and promotion
activities, and stigma reduction activities undertaken as part of the headspace model are effective. It is also
recognised as providing a range of additional contributions to local communities that are highly valued by
those communities.
The areas where the headspace model is less effective are related to the broader service system in which it
operates. The extent to which the model is effective in improving pathways to care, as well as the extent to
which it provides a localised service offering, are both constrained by the capacity of other services in the
headspace services’ referral pathways. At the same time, pressures associated with a ‘no wrong door’
approach are exacerbated when headspace services are unable to access referral pathways to more
intensive supports for high needs young people, in turn increasing wait times at headspace services as
young people with higher presenting needs are prioritised. Similarly, while overall the headspace model
benefits from generally high levels of support from other primary care and mental health providers, these
operational pressures affect individuals’ referring decisions and attitudes towards the headspace model.
The local commissioning role of PHNs may be under-utilised, here, and roles and responsibilities across
the national network are further discussed in section 5.4.3.
While various elements of the headspace model support improved service system outcomes, the most
relevant of these are:
• community awareness and engagement;
• service integration;
• multi-disciplinary workforce; and
• national network (in particular the roles of PHNs and community consortia).
These elements are the focus of potential recommendations to support better service system outcomes,
detailed in section 5.4.

4.2.3 Improving user experience outcomes

Providing an appropriate service approach for young people with mild to moderate,
high-prevalence mental health conditions
In the headspace model, appropriate care is defined as “the provision of evidence-based interventions for
each individual young person by matching the type, intensity, frequency, duration, location and mode of
treatment to their presenting need. This includes identification and consideration of factors such as: risk
and protective factors, stage of illness, psychosocial complexity, and developmental and sociocultural
factors” .
124F

As discussed in Appendix D.12, many elements of the current headspace model are closely aligned to the
needs of young people with mild to moderate, high-prevalence mental health conditions. For headspace
users, mild to moderate psychological distress is defined as a value of between 20 to 29 out of 50 on the
K10 questionnaire .125F

High-prevalence mental health conditions, such as depression and anxiety, are widely considered to be
able to be effectively treated and to respond well to early intervention . The design of the headspace model
126F

has prioritised supporting young people in this category. It includes a psychosocial model of supports
provided by peers and ,in practice, many staff working in headspace are early career clinicians with whom
the young person is likely able to identify and build rapport . headspace providers described how the
127F

staffing profile is driven by a combination of the available funding envelope, which is most competitive
for early career psychologists, and by the brand of the model, which appeals to younger staff with an
interest in working with young people.
Stakeholders also argued that the focus in the model on early intervention and prevention of mental ill-
health for young people, including through supporting improved mental health literacy, also aligns well to
the support of mild to moderate conditions. Integration of other factors impacting on mental health, such as
physical health, alcohol and drug use and employment and education, helps to attract young people, giving
a ‘soft entry’ into mental health services, appropriate for those with mild to moderate mental health
conditions.
Evidence was reviewed from headspace service providers, young people and consortium partners
(Appendix D.12), which confirmed that there is a widespread view that the model is well designed for this
cohort of young people, with the provision of support groups, skills training and peer workers particularly
recognised as powerful in the potential to help young people to tap into protective factors and support their
wellbeing. Youth representation in the design and delivery of services was also called out as key to the
appropriateness of the model for this cohort.
Evidence suggests that the headspace model provides a highly appropriate mental health service
approach for young people with mild to moderate, high-prevalence mental health conditions.
The key enabling elements of the headspace model which support appropriate care for mild to moderate,
high-prevalence mental health conditions are:
• enhanced access;
• early intervention;
• appropriate care; and
• evidence-informed practice.

Providing culturally appropriate and inclusive services

The headspace model includes a focus on the experience of service for young people and their families
from a diverse range of backgrounds through providing an accessible, welcome, inclusive and
nonstigmatising service. For young people from diverse population groups, this includes providing
translated information, guidance materials and posters and flags and other cultural symbols to make the
young person and their family feel welcome and included. Given the broad age range supported by
headspace, the experience of service must also be tailored for the level of maturity of the young person,
with very different needs for those aged 12 years compared with those approaching the age of 25.
In order to examine the extent to which headspace is successfully providing culturally appropriate and
inclusive services, data is drawn from hMDS user satisfaction surveys, interviews with headspace users
and non-users as well as Youth Reference Group members and school and university counsellors
(Appendix D.13).
Data from a range of sources indicates that headspace is broadly effective in providing culturally
appropriate and inclusive services for the general population of young people, and for LGBTQIA+ young
people. User satisfaction overall is very high, however, user satisfaction is significantly lower for
culturally and linguistically diverse young people and for Aboriginal and Torres Strait Islander young
people. In contrast, measures of satisfaction undertaken for this evaluation, including that of how welcome
young people felt and how respectful services were of a young person’s culture, gender or faith identity,
were all positive and in line with results for the general population of young people using headspace.
At the same time, discussions with young people and other stakeholders further highlighted that employing
staff with particular cultural backgrounds is a key mechanism to providing culturally appropriate care for
young people from that culture.
There were also differences between the age when young people felt headspace was appropriate for them,
with younger people more likely to see it as a service where they feel included and non-users being unclear
on which age groups the service was intended to assist.
Overall, there are mixed results from the data and insights gathered through this evaluation about
how well the headspace model effectively provides a culturally appropriate and inclusive model for
young people and their families, with strong satisfaction from the general population and
LGBTQIA+ young people, but significantly lower satisfaction levels on relevant measures from
culturally and linguistically diverse young people and Aboriginal and Torres Strait Islander young
people.
The key enabling elements of the headspace model which support culturally appropriate and inclusive care
are:
• community awareness and engagement;
• service integration;
• multi-disciplinary workforce; and
• national network (in particular the roles of PHNs and community consortia).

Enabling young people and their families to access support where, when and how
they want
A key element of the headspace program logic is that the services provided are appropriate for young
people. Through providing a positive experience of service, by ensuring young people feel that their needs
and interests are reflected in the services on offer, and that the services adapt to the needs of young people,
the overall objectives of the model are supported. Enabling young people and their families to access
support where, when and how they want to is a key indicator.
This evaluation examined a range of data and evidence regarding the extent to which headspace is
successful in these domains. Feedback from young people using headspace collected through hAPI
surveys, as well as direct consultation with young people, Youth Reference Group members and staff and
other stakeholders, provide evidence of relevance to this evaluation question (Appendix D.14).
Evidence shows headspace provides appropriate, accessible and youth friendly supports, with strong
positive responses from young people in surveys and interviews for these domains. The more contact
young people had with their headspace service, the more likely they were to rate the experience highly,
which is a further positive reflection on the appropriateness, accessibility and youth friendliness of the
headspace model.
Qualitative insights indicate that young people value the rapport built with headspace staff, and the easily
accessed location of their local headspace service. At the same time, for those not accessing headspace,
fear of being stigmatised arose in relation to the central location of headspace service sites and being seen
by others when seeking mental health support, while the need to be close to public transport was again
highlighted.
Barriers to accessibility were raised by users and non-users, including waiting times and the opening hours
of the service. A lack of flexibility to change counsellors within headspace if they were not the right
‘match’ with the young person was also raised as an area where headspace could be more 'youth friendly'.
Cultural and gender characteristics of the staff member were again very important for a young person to
feel comfortable.
Other stakeholders had positive views of the youth friendly, appropriate and accessible nature of the
services, with drop-in sessions and outreach highlighted as key enablers.
Evaluation results suggest that headspace is effective in enabling young people to access support where,
when and how they want it, and that it is generally appropriate, youth friendly and accessible, with some
issues around opening hours and waiting times proving a challenge.
The key enabling elements of the headspace model which support young people to access support where,
when and how they want it are:
• enhanced access;
• youth participation;
• family and friends participation; and
• multi-disciplinary workforce (as related to wait lists and capacity constraints).

Enabling young people to participate in the design and delivery of headspace

Ensuring young people are actively engaged in the design and delivery of the services they receive is
another key element of the headspace program logic. Youth participation at a governance and service level
is built into the headspace model, primarily through the role of Youth Reference Groups. Each service is
required to establish and maintain a Youth Reference Group to ensure youth participation in strategic
planning, service development, delivery and evaluation. Young people are also involved in their own care
through opportunities to engage in decisions throughout their episode of care, including regarding their
service planning and transitions. Through providing a positive experience of service by ensuring young
people feel listened to and involved in decision making, the overall objectives of the model are supported.
To examine the extent to which young people are participating in the design and delivery of services, and
how this relates to their experience of headspace, user satisfaction data was analysed, along with
interviews with headspace users and Youth Reference Group members (Appendix D.15).
Young people recognise and value the extent to which they are invited to codesign their service
experience, and rated this highly in satisfaction surveys. The extent to which this translates to improved
experience of headspace for their families is unclear, as satisfaction of families attending family-focused
sessions is not measured.
Youth Reference Group members highlighted a range of areas where their contribution to the governance
of their local headspace service had been valued and had helped to improve the service experience for
young people seeking mental health support.
Evaluation evidence suggests that the headspace model effectively enables young people to
participate in the design and delivery of headspace, and this is associated with positive experiences
of headspace for young people.
The key elements of the headspace model which enable young people to participate in the design and
delivery of headspace services are:
• youth participation;
• family and friends participation; and
• national network (in particular Youth and Family and Friends Reference Groups).

Improving user experience outcomes – in conclusion

The headspace model supports user experience outcomes, with evidence to suggest that it provides a
highly appropriate mental health service approach for young people with mild to moderate, highprevalence
mental health conditions. The model also successfully supports the participation of young people in the
design and delivery of headspace services, which is associated with strong positive views as to user
experience.
At the same time, however, the model has less success in providing culturally appropriate and inclusive
supports for young people from culturally and linguistically diverse backgrounds, and Aboriginal and
Torres Strait Islander young people. This is an area of ongoing focus for many services nationally.
In terms of enabling young people to access support where, when and how they want it, the model is
reasonably effective, however opening hours and waiting times detract from this.
While various elements of the headspace model support improved user experience outcomes, the most
relevant of these are:
• community awareness and engagement;
• multi-disciplinary workforce (as related to wait lists and capacity constraints); and
• national network (in particular the roles of PHNs and community consortia).
These elements are the focus of potential recommendations to support better user experience outcomes,
detailed in section 5.4.

4.2.4 Improving psychosocial outcomes

Improving mental health and wellbeing outcomes, considering clinical outcomes for
young people
The clinical scores of young people attending headspace, and other items of administrative data from the
hMDS, enable detailed statistical analysis as to the effectiveness of headspace in improving mental health
and wellbeing outcomes. Detailed analysis has been conducted on this data, and is presented in Appendix
E.
Overall, data from the hMDS shows that young people benefit from more engagement and treatment
through headspace, which is associated with greater improvements in mental health and wellbeing. The
improvement in young persons accessing six or more headspace sessions is on par with that observed from
psychotherapy treatment for depression more broadly . The follow up survey suggests that outcomes
128F

achieved during a headspace episode are sustained over the following 90 days.
The largest proportion of young people accessing the headspace model only attend once (36 per cent of
episodes of care within the data period were a single OOS), and only 19 per cent of episodes of care were
for six or more OOS.
Analysis of variation suggests that the majority of headspace services deliver a positive and statistically
significant improvement in mental health and broader outcomes, as measured by the K10, SOFAS and
MLT outcome measures. A smaller proportion of episodes achieved reliable change (meaning the change
is greater than a difference that could have occurred randomly). Similarly, a smaller proportion of episodes
which met clinical threshold on entry, achieved clinically significant change. There is merit in further
consideration and potential strategies to enhance the efficacy of the interventions provided, further
discussed in Chapter 5.
The number of OOS, a young person’s initial level of mental distress, and the individual service itself are
key drivers of variation in outcomes. Young people who present with high levels of mental distress and
who go on to access at least six to eight OOS achieve the greatest improvement in outcomes. In contrast,
there were no clear factors associated with those headspace services that had higher than the average
improvement of outcomes across their client group, indicating that the stronger outcomes are not the result
of specific features of the service providing care. In contrast, average improvement (in the K10 and MLT
outcome measures) is lowest among young persons who entered headspace with low levels of initial
distress (as measured by the K10).
Contrary to expectations of headspace service providers (as provided in service and lead agency survey
responses), LGBTQIA+ young people experienced lower improvements (but still positive) across all
measures than young people who do not identify as LGBTQIA+ . By contrast, culturally and linguistically
diverse cohorts achieved statistically similar improvements as young people who do not identify as
culturally and linguistically diverse. Improvements in the SOFAS and MLT outcome measures were
statistically significantly lower among the Aboriginal and Torres Strait Islander young people than the
general population of young people accessing headspace. However, when using the K10 outcome measure,
outcomes among the Aboriginal and Torres Strait Islander cohort and the general population of young
people accessing headspace are statistically similar.
Area-level analysis was also conducted (as described in Appendix A), to support analysis of outcomes
from headspace services where no control group exists. Using the Difference-in-Difference (DID) quasi-
experimental methodology, the impact of headspace services at the area-level, rather than the individual
,,
level, was evaluated .
129F 130F 131F

This design made use of longitudinal data to estimate the effect of headspace services by comparing the
changes in outcomes over time between areas where headspace services are introduced to different PHNs
at different points in time. Specifically, at each point in time, the approach compares outcomes between
PHNs that had headspace services and PHNs with fewer or no services (before experiencing an increase in
the number of services)
To examine how variations in headspace exposure influence area-level outcomes over time, the number of
mental-health related hospitalisations, intentional self-harm hospitalisations, illicit drug and alcohol related
hospitalisations, deaths from intentional self-harm, and Medicare-subsidised mental health specific
services among 12 to 25 year olds were examined. This report hypothesises that increasing exposure to
headspace services include a reduction in the number of hospitalisations and deaths from intentional self-
harm. Furthermore, increasing exposure to headspace should destigmatise the use of mental health services
and increase subsequent uses of mental health services as recorded by the MBS.
There is some evidence that the number of headspace services had a positive effect on some outcomes
such as reducing substance abuse hospitalisations and the number of self-harm hospitalisations. However,
these impacts are not consistent when using alternative variables to measure the headspace treatment effect
such as the number of headspace clients per 1,000 12 to 25 year olds and the ratio of MBS funded
headspace mental health services to MBS funded mental health services external to headspace.
Young people benefit from more engagement and treatment through the headspace model, which is
associated with greater improvements in mental health and wellbeing. Young people who present
with high levels of mental distress and who go on to access at least six to eight sessions achieve the
greatest improvement in outcomes. Clinically significant improvement is achieved for a smaller
proportion of young people.
While the model is associated with positive outcomes for young people, these vary for LGBTQIA+
and Aboriginal and Torres Strait Islander young people.
Analysis to explore longer-term impacts using current data through an area-level analysis failed to
identify any reliable effects where access to services through the headspace model leads to
improvements in hospitalisation rates.

Improving psychosocial outcomes through providing alternative service delivery

models
As outlined in Section 2.3.4 above, there are a range of different headspace service types now present
within communities across Australia, and increasing emphasis is being placed on diversifying the
headspace model by the Commonwealth Government. Additional satellite services have been funded to
support young people in smaller communities surrounding headspace services to offer them faceto-face
services of mental health and counselling support.
As intended, the types of services delivered by these alternative models differs to those offered by
headspace centres. The predominant focus of supports is on mental health and counselling, with only two
of the three other core services required to be provided by a satellite service, either directly by staff, or
through linkages with local providers of those supports.
As explored in Appendix E.9, there are mixed views from across stakeholders involved in delivering, or
working with, headspace services as to the impact of satellite services. As outlined in this report, there is
significant positive regard for headspace services, and communities and stakeholders view any headspace
service as a positive addition to achieving core objectives.
Stakeholders consulted from within satellite services or parent centres were of the view that the work they
were undertaking through the headspace model made an important contribution to their communities.
However, these stakeholders also indicated that the level of need in their local community warranted a
headspace centre, and that being able to implement the full headspace model would make the most
difference for young people locally. Similarly, PHNs as commissioners of services, indicated a preference
for headspace centres to better meet the needs of local young people through the holistic headspace model.
In terms of barriers and enablers to achieving effective outcomes, survey responses from staff in satellite
and outreach services did not differ from headspace centres, nor did they vary in how well these services
feel able to support headspace’s objectives. Respondents from across service types indicated similar
challenges in recruiting appropriate staff, managing wait times for young people, and challenges with
perceived complexity of presenting need.
With respect to clinical outcomes, only a small number (less than five) of alternative models were able to
be analysed in line with criteria established for this analysis. Services were typically excluded from
analysis if they had not been open long enough to move past their establishment phase, that is they had not
yet been operating for at least 12 months.
When comparing results for headspace centres versus satellites, centres have statistically significantly
higher average MLT improvements than their satellite counterparts. MLT improvements were four points
higher among centres than satellites. However, there were limited differences between outcomes for young
people based on the K10 and SOFAS outcome measures. The more detailed analysis of these outcomes is
contained in Appendix E.6. This analysis is supplemented the hMDS data with longitudinal area-level data
from the AIHW and Services Australia. This is done to compare outcomes over time for PHNs with few or
no headspace services, to PHNs that have experienced a growth in headspace services. Analysis of longer-
term impacts failed to identify any conclusive and consistent effects where access to services through the
headspace model leads to improvements in hospitalisation rates. There is some evidence that headspace
has a positive effect on some outcomes such as reducing substance abuse hospitalisations and the number
of self-harm hospitalisations but the impacts are typically lagged and/or inconsistent over time and
different measures of headspace exposure. Further details of the area-level analysis are available under
appendix E.8.
However, it should be noted that the number and scale of satellite services that were able to be analysed
within the data period for this evaluation was limited. A number of satellite services have since opened,
but were unable to be evaluated at this time, due to their short time in operation. Further evaluation of any
differences in outcomes for young people accessing headspace satellites and other models should be
undertaken once more services have reached full establishment, that is at least 12 months after they have
commenced operations.
The expansion of headspace services into new communities assists headspace to meet its objectives
by supporting a greater number of young people. However, there is recognition amongst
stakeholders that headspace centres are preferred over satellite services. The small sample size
makes it difficult to either affirm or disconfirm if clinical outcomes for young people support this
position.

Improving psychosocial outcomes – in conclusion

Clinical outcomes associated with headspace are positive, and sustained in the short- to medium-term.
Detailed statistical analysis conducted on the clinical measures incorporated within the headspace model
provide evidence that, overall, young people benefit from more engagement and treatment through the
headspace model, which is associated with greater improvements in mental health and wellbeing. For
young people who access six OOS or more, headspace is associated with similar improvements in mental
health and wellbeing as comparable psychotherapy treatments.
Furthermore, young people who present with high levels of mental distress and who go on to access at
least 6 to 8 sessions achieve the greatest improvement in outcomes. While the model is associated with
positive outcomes for young people in general, these are not as strong for LGBTQIA+ and Aboriginal and
Torres Strait Islander young people.
There is a consistency of results across services, however it is difficult to see the impact of headspace on
longer-term outcomes in the data, including using area-level methods which are designed to reliably detect
changes in population level indicators. As discussed in Section 3.1 above, longer-term outcomes are not
measured through the model. Further analysis to explore longer-term impacts was conducted as part of this
evaluation. This analysis supplemented the hMDS data with area-level data from the AIHW and Services
Australia, and failed to identify any conclusive and consistent effects where access to services through the
headspace model leads to improvements in hospitalisation rates. The lack of a consistent impacts of
headspace on the rate of mental-health related hospitalisations, substance-abuse hospitalisations, self-harm
hospitalisations, deaths from self-harm, mental health emergency department presentations and Medicare-
subsidised mental health services are important given the broad geographic coverage and value of the
investment in the headspace model.
The expansion of headspace services into new communities assists headspace to meet its objectives by
supporting a greater number of young people. However, there is recognition amongst stakeholders that the
full headspace model is preferred over satellite services. The small sample size makes it difficult to either
affirm or disconfirm if clinical outcomes for young people support this position.
The key element of the model of relevance to measuring psychosocial outcomes is:
• Monitoring and evaluation.
This element is the focus of potential recommendations to support better understanding of psychosocial
outcomes, detailed in section 5.4.

4.3 Overall effectiveness of

the headspace model
Analysis of extensive qualitative and quantitative evidence demonstrates that the headspace model is
effective in achieving many of its intended outcomes. There is some inconsistency, however, in outcomes
for different groups, and across some aspects of the model program logic.
Table 14 provides a summary of the effectiveness of key elements of the headspace model program logic.
The most relevant element of the hMIF is also identified.
Table 14: Effectiveness findings in summary

Program logic element Effectiveness findings - in summary

Measuring outcomes of the Data is collected and disseminated across a broad range of activities.
headspace model
Relevant hMIF element:
Monitoring and evaluation.

Gaps in activity data prevent measurement of some elements of the

headspace model, including community engagement and services
integration.
Longer-term outcomes associated with the model are not measured.
Improving intermediate outcomes The headspace model is effective in supporting important intermediate
Most relevant hMIF elements: outcomes for the general population of young people. These include
Community awareness and mental health literacy, early help seeking and increased access to
engagement; required services, which in turn improve the likelihood that young
people will seek support with their mental health and achieve
Enhanced access (minimising
improved psychosocial outcomes in the longer-term.
barriers to seeking professional
help); and
Multi-disciplinary workforce.

The headspace model has mixed success in supporting these

intermediate outcomes for ‘hard to reach’ groups.
Improving service system The headspace model is effective in supporting youth mental health
outcomes through advocacy and promotion activities, and stigma reduction
Most relevant hMIF elements: activities undertaken as part of the headspace model are also effective.
The model is also recognised as providing a range of additional
• Community awareness and contributions to local communities that are highly valued by those
engagement; communities.
• Service integration;
• Multi-disciplinary
workforce; and
• National network (in
particular the roles of
PHNs and community
consortia).

The headspace model has mixed effectiveness in areas related to the

broader service system in which it operates. The implementation of
the model is often impacted by the broader service system in which it
operates, especially in relation to:
 • improving pathways to care;
• providing a localised service offering; and
• providing a ‘no wrong door’ approach.

These outcomes are constrained by the capacity of other services,

workforce shortages and, in rural and regional areas, difficulty in
attracting MBS billing staff. These challenges lead to increased wait
times for services and reduce the generally high levels of support the
model receives from other primary care and mental health providers.
Improving user experience The headspace model provides a highly appropriate mental health
outcomes service approach for young people with mild to moderate,
Most relevant hMIF elements: highprevalence mental health conditions.
• Community awareness and
engagement;
• Multi-disciplinary workforce
(as related to wait lists
and capacity constraints);
and
• National network (in
particular the roles of
PHNs and community
consortia).

The model successfully supports the participation of young people in

the design and delivery of headspace services, which is associated
with strong, positive views as to user experience.
The model has less success in providing culturally appropriate and
inclusive supports for young people from culturally and linguistically
diverse backgrounds, and Aboriginal and Torres Strait Islander young
people.
The model is reasonably effective in enabling young people to access
support where, when and how they want it, however opening hours
and waiting times detract from this.
Improving psychosocial Young people benefit from more engagement and treatment through
outcomes the headspace model, which is associated with greater improvements
Most relevant hMIF element: in mental health and wellbeing.

• Monitoring and evaluation.

For young people who access six OOS or more, headspace is

associated with similar improvements in mental health and wellbeing
as comparable psychotherapy treatments.
The largest proportion of young people accessing the headspace model
only attend once (36 per cent of episodes of care within the data
period were a single occasion of service), and only 19 per cent of
episodes of care are for six or more OOS.
While the model is associated with positive psychosocial outcomes for
young people, the majority do not see a clinically significant change to
their outcomes.
Young people who present with high levels of mental distress and who
go on to access at least six to eight sessions achieve the greatest
improvement in outcomes.
Clinical outcomes, although positive, are not as strong for LGBTQIA+
young people as they are for the general population of young people
accessing the headspace model.
Clinical outcomes for Aboriginal and Torres Strait Islander young
people are mixed. When using the K10 outcome measure, this cohort
achieved statistically similar outcomes as the general population of
young people accessing the headspace model. However, outcomes are
not as strong for Aboriginal and Torres Strait Islander young people
when using the SOFAS and MLT outcome measures.
Analysis of longer-term impacts failed to identify any conclusive and
consistent effects where access to services through the headspace
model leads to improvements in hospitalisation rates.
The full headspace model is preferred over satellite services, however
the small sample size makes it difficult to either affirm or disconfirm
if clinical outcomes for young people support this position.
Source: KPMG 2022
5 Cost-effectiveness
and value of
headspace
5.1 The full cost of
delivering headspace
5.1.1 Estimating the cost of delivering headspace
The funding provided to operate headspace includes the national headspace grant; any additional funding
that a PHN, state or federal government may provide to deliver specific activities; activity-based funding
of services through the MBS; in-kind contributions; private donations; and any out-of-pocket payments
made by young people or their carers.
No single source captures these ranges of costs of delivering headspace. The department records the
national headspace grant costs but does not require services to record the division of the grant between
service provision and indirect costs, such as rent and utilities, office expenses and community awareness
expenses. The hMDS identifies the funding source for each occasion of service provided by headspace but
does not capture the value of funding for that occasion of service. Any funding of indirect costs, such as
in-kind contributions for physical space, can only be provided by the service itself and may be prone to a
range of data quality issues (e.g., definition and quantification of in-kind support may vary).
For this evaluation, the sources and methodologies to estimate the full cost of delivering headspace are as
follows.
• Direct and indirect headspace costs funded by the national headspace grant itself – available from the
administrative funding data.
• Direct headspace service costs not funded by the national headspace grant – the hMDS identifies the
volume, but not the value, of direct service provision that is not funded by alternative sources such as
the MBS, other government funding or in-kind contributions. The value of these contributions is
estimated as a volume-weighted average of the equivalent MBS benefit fees (see Appendix J).
• Indirect headspace costs not funded by the headspace grant – a bespoke data collation exercise was
completed but did not deliver consistent or reliable data and, as a result, these costs are not considered
within this analysis.
The following sub-sections examine each of these areas in more detail. Note that the cost analysis here
considers only headspace services that opened before 30 June 2019, to ensure each service had reached full
implementation. Given this selection criteria , there are 112 headspace services to estimate the costs of
132F

,
delivering headspace, three of which are satellite services .
133F 134F

5.1.2 Direct and indirect costs funded by the national headspace

grant
The national headspace grant is the major funding source for headspace. The cost of the headspace grant
for 2019-20 was $101 million.
Under the current funding model, all headspace services receive relatively similar annual funding amounts
regardless of the volume of services they deliver, with an average of $914,300 per centre. Satellite services
received an average of $404,100.
Figure 20: National headspace grant funding by service by volume during 2019-20
Source: KPMG analysis of the total cost dataset
Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS and
102,550 episodes. The outlier service on the right is a service that opened in July 2007 and is located in a metropolitan area . The
outlier service on the bottom left opened in 2015 and is located in a major city of Australia. In practice, it operates as part of another
headspace centre located nearby, which accounts for the outlier funding amount.

5.1.3 Indirect headspace costs not funded via the national

headspace grant
The contribution of other PHN and other government funding, in-kind contributions and any private
donations to indirect costs of delivering headspace were investigated by the deep dive case studies and
anonymous survey of headspace services. Across these bespoke data collection exercises, only three
services reported funding for indirect costs from sources other than the national headspace grant. These
responses are summarised in Table 15. Two out of three responses indicated that the value of the indirect
contributions was small, accounting for approximately one per cent of the total cost.

Table 15: Summary of responses to questions on indirect service contributions

Survey source Amount Share of total cost Form of contribution

Deep dive $18,370 1.2% Donations and small
grants
headspace service survey $5,000 Unavailable data Building/property/rent
headspace service survey $200,000 15.4% Building/property/rent
$52,277 Office equipment and
other donated goods
Source: KPMG summary of deep dive surveys and headspace service survey

Given the low response rate to cost-specific questions of the service survey and the small magnitude of the
in-kind contribution amount in most responses, it was not possible to extrapolate these indirect
contributions for all headspace services. This remains a limitation of the cost analysis.

5.1.4 Direct headspace services not funded via the national

headspace grant
In 2019-20, 43 per cent of all OOS were funded via the national headspace grant, with the remaining 57
per cent funded through other sources, including the MBS, other PHN and government funding, and in-
kind contributions from young people and/or their carers.
Figure 21: Distribution of OOS funding source during 2019-20
Source: KPMG analysis of the total cost dataset
Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS and
102,550 episodes.

During 2019-20, 35 per cent of total OOS were funded via the MBS, at an estimated cost of $14 million . 135F

This was down from over $16 million in 2019, potentially as a result of COVID19 restrictions which saw
widespread reductions in healthcare service use , and difficulties attracting and retaining private
136F

practitioners to provide MBS-billed services.

The majority (68 per cent) of OOS funded via the MBS were psychologist services, followed by GP visits
(14 per cent) and allied health services (four per cent). Four per cent of MBS funded OOS had missing
information regarding the service provider type.
Figure 22: MBS funded OOS by provider type

Source: KPMG analysis of the total cost dataset

Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS and
102,550 episodes. ‘Other psychologists’ refer to all other psychologists not labelled as clinical psychologists according to the service
provider survey.

The share of MBS funding varied considerably by service as shown in Figure 23. Figure 23 describes the
distribution of the percentage of OOS funded by the MBS across each headspace service. Fourteen
headspace services had less than one per cent of their OOS funded via the MBS. These services are all
headspace centres, with 13 located in regional or remote areas. Only one of these 14 services is in a
metropolitan area. Twenty-three headspace services had over half of OOS funded via the MBS. Fifteen of
these 23 services are in metropolitan areas with seven services in inner regional areas and one in a remote
area. Seven of these 23 services are in Queensland, seven in Victoria, seven in NSW, two in Tasmania and
one each in Northern Territory (NT) and WA. Among the three satellites out of the 112 services, around 14
to 43 per cent of OOS were funded via the MBS.
Figure 23: Proportion of OOS funded by the MBS by headspace service
Source: KPMG analysis of the total cost dataset
Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS.

Figure 24 shows that there is a strong correlation between the total number of OOS delivered by a
headspace service and the use of MBS funding.
Figure 24: MBS funded OOS vs total OOS

Source: KPMG analysis of the total cost dataset

Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS.

In-kind contributions to the delivery of headspace

In 2019-20, four per cent of total OOS were funded via in-kind contributions at an estimated cost of $1.3
million.
The in-kind contribution varied considerably by service but was not strongly driven by volume, rurality or
service type. For 60 per cent of headspace services, in-kind funding accounted for less than five per cent of
total OOS; in six headspace services, however, it accounted for over 15 per cent. All six of the services are
located in a regional area.
Figure 25: Histogram of in-kind contribution
Source: KPMG analysis of the total cost dataset
Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS and
102,550 episodes.

Private contributions to the delivery of headspace

In 2019-20, 0.2 per cent of total OOS were funded via private contributions at an estimated cost of
$81,000. Fifty out of 112 services had at least one occasion of service funded via private contributions.
However, the number of OOS funded via private contributions were small, ranging between 1 to 220, an
insignificant number compared to the total OOS delivered during the year.

Additional PHN program funding

In 2019-20, six per cent of OOS were funded via additional PHN funding in addition to the headspace
funding provided through the PHNs. In addition to the headspace grant, PHNs provided funding across
programs, including the new access program, the low intensity mental health fund, psychological therapies
and other PHN funding. The estimated cost of these services was $2.1 million based on the weighted
average of the equivalent MBS unit costs (see Appendix J).
The contribution of PHN funding varied considerably by service. For 73 per cent of headspace services,
PHN program funding accounted for less than five per cent of total OOS; in one service however, it
accounted for over 30 per cent. There were no clear relationships between other PHN funding and rurality
or service type.
Figure 26: Histogram of PHN funding contribution

Source: KPMG analysis of the total cost dataset

Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS and
102,550 episodes.

Other federal and state government funding

In 2019-20, 1.8 per cent of OOS were delivered through other Commonwealth Government funding, at an
estimated cost of $619,500 . The hMDS also captured OOS funded via the Victorian Government’s
137F

Enhancing Mental Health Support in Schools (EMHSS) initiative. In 201920, 7.6 per cent of Victorian
OOS (and two per cent of total OOS) were funded through this program. The support provided via the
EMHSS included both student counselling and psychological support. The unit cost of the EMHSS
services were therefore estimated as a weighted average of the equivalent MBS costs (Appendix J). In
total, the EMHSS costs of delivering headspace were estimated at $768,000. This may under-estimate the
true costs of the EMHSS services. However, no detailed costing of this initiative was available at the time
of writing.
Out-of-pocket costs for young people
In 2019-20, out-of-pocket costs were charged on 2.1 per cent of OOS. Based on an out-of-pocket cost of
$29 per service, derived from a weighted average of out-of-pocket costs for different providers from
national MBS data, the costs to young people or their carers summed to $250,300 for 2019-20. Table 16
summarises the range of out-of-pocket costs for young people by service provided.
Table 16: Range of out-of-pocket costs by service provided during 2019-20

Service provided Out-of-pocket costs

General practitioner $3
Psychiatrist $68
Clinical psychologist $40
Other psychologist $36
Allied health $37
Other $18
Weighted average $29
Source: KPMG 2022
Notes: Out-of-pocket costs are calculated as averages from the national MBS data across the entire population. National MBS
expenditure data for young people specifically is not publicly available, and the hMDS does not capture the specific outofpocket
costs for young people from headspace services.

Most services charged out-of-pocket fees on less than two per cent of occasions (72 per cent); six services
charged out-of-pocket fees on more than 10 per cent of OOS.
Figure 27: Histogram of out-of-pocket share by service

Source: KPMG analysis of the total cost dataset

Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS.

5.1.5 Total cost of delivering headspace

The total cost of operating and providing treatment at headspace in 2019-20 was estimated to be $123.3
million (Table 17), noting that some indirect costs were undetermined. In addition, the cost to government
by administering funding through PHNs is slightly higher. The total cost estimated in table 16 below is the
amount allocated to headspace services but exclusive of PHN operational costs. On average, the total cost
per centre was $1.1 million, and $494,000 per satellite service.
Table 17: Total costs of delivering headspace

Direct Indirect Total

headspace grant $100,867,056

MBS $13,974,946 $0 $13,974,946

Other PHN $2,145,654 Undetermined $2,145,654
Other Commonwealth $619,475 Undetermined $619,475
programs
Other state government $768,011 Undetermined $768,011
programs
In kind $1,328,042 Undetermined $1,328,042
Private $80,998 Undetermined $80,998
Other funding sources $3,270,170 Undetermined $3,270,170
Out-of-pocket $250,292 Undetermined $250,292
Total $123,304,645
Source: KPMG analysis of the total cost dataset
Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS and 102,550
episodes.

In 2019-20, 112 headspace services included in the cost analysis delivered 401,325 OOS. The average cost
per OOS was approximately $307. The average direct cost per OOS was $230 under the assumption that
the direct service costs account for 75 per cent of the total cost . This amount is higher than the average
138F

cost of a mental health session with clinical psychologists ($154) and GPs ($100). Given a typical length
139F

of an OOS at headspace was 40 – 89 minutes in 2019-20, the average direct cost per OOS is lower than the
Australian Psychological Society’s recommended fee for a session with a similar duration of $320 . 140F

Figure 28 shows the distribution of the total cost per OOS across services. There are a number of services
with considerably higher cost per OOS. Ten services had the total cost per OOS over twice the average
(over $620), among which three services had the total cost per episode more than three times the average
(over $1,000).
Figure 28: Distribution of the total cost per OOS by headspace service during 2019-20

Source: KPMG analysis of the total cost dataset

Notes: See Appendix F for detailed exclusion criteria. The sample includes 112 services delivering 401,325 OOS.

The cost of operating and providing treatment at headspace

Commonwealth funding through the national headspace grant continues to make the largest contribution to
delivering headspace. Under the current funding model, all services receive relatively similar annual
funding amounts regardless of the volume of services they deliver.
MBS payments make the next most sizeable contribution, however this varies widely by service with
higher volume services more likely to use private providers billed through the MBS. This aligns to the
challenges reported by rural and regional services in attracting private practitioners and GPs into the
headspace model.
A range of alternative funding is also obtained from PHNs, state and Commonwealth governments, in-kind
and private donations, however these represent only a small fraction of the costs of delivering headspace.
Out-of-pocket costs were only paid in 2.1 per cent of OOS, in line with the intention that the model
provides accessible, low or no cost service to young people.
Overall, 112 headspace services included in the 2019-20 cost analysis delivered 401,325 OOS at a cost of
$123.3 million. The average cost per OOS was approximately $307. The average direct cost per OOS was
$230, which is twice as much as the MBS fee (and any out-of-pocket costs) for a typical mental health
session . However, the average direct cost per OOS is slightly lower than the Australian Psychological
141F

Society’s recommended fee for a 46 to 60 minute session of $260 . 142F

In the devolved governance model, the full costs of delivering headspace are not measured or collected
centrally. It proved particularly difficult to obtain reliable and comprehensive data on in-kind contributions
and other funding of indirect costs.
The national headspace grant makes the largest contribution to delivering headspace, followed by
MBS payments. Overall, the total cost for the 112 services included in the cost analysis was $123.3
million, resulting in an average direct cost per OOS of $230.

5.2 Economic evaluation of

services provided by
headspace
5.2.1 How cost-effective is headspace?
Economic evaluation aims
This economic evaluation aims to present a transparent, acceptable and robust assessment of costs and
outcomes of headspace activity compared to a plausibly defined situation in which headspace is absent
from the sector. The evaluation aims to reflect all pivotal considerations concerning headspace activity,
even if these considerations are necessarily implemented in a simplified way. Such simplifications are
inherent to any modelling exercise and dictated by limitations of data and evidence available to evaluators.
However, it is essential that main aspects and implications of headspace activity are accounted for. This
includes modelling the services provided, their effects, proportions of clients accessing different types of
services, and the consequences of not accessing such services. The evaluation also aims to present
outcomes in a format that will be most suitable to inform decision making.
A summary of the approach to economic evaluation is presented in Table 18 below, with further details of
the methodology provided in Appendix K. Key limitations of the economic evaluation have been
identified and summarised in Appendix K. Key points to note include:
Table 18: Overview of economic evaluation

Evaluation Value
parameter
Type Cost-effectiveness analysis.
Model population Young persons with mild to moderate mental health needs that fall within the
scope of services correspondingly provided by headspace (consistent with the
target population for headspace services).
Intervention headspace program.
Comparator The world without the headspace program.
Outcomes QALYs; Costs; Consequences of not receiving a mental health treatment.
Methods Decision tree;
K10 score to QALY conversion;
Expected values approach to costs and health outcomes; and
Extrapolation of gains in mental health outcomes.
Evaluation period 2019-20.
Time horizon 18 months.
Perspective Extended payer perspective (funder and patient out-of-pocket costs).
Discounting 5% 143F

Source: KPMG model framework design

Notes: K10 Kessler Psychological Distress Scale; QALY quality-adjusted life year

Scope of headspace activity for economic evaluation

The headspace program delivers a range of supports that include not only mental health services, but also
other individual and social benefits associated with its presence. These include the existence of safe
spaces, promoting mental health awareness and increasing mental health literacy, stigma reduction and
early help seeking . As these outputs may be difficult to define and measure, it is important to outline the
scope of benefits that the economic evaluation has the aim and capacity to account for. The benefits of
headspace activity reflected in this economic evaluation are:
1. the services provided by headspace to its target client population; and
2. the increased uptake of such services by the target population due to increasing accessibility of mental
health care. The latter can take different forms, including increased affordability (e.g., due to the lack
of out-of-pocket costs for the clients), increased availability (e.g., physical presence in otherwise
underserved areas) and increased acceptability (e.g., greater awareness and lower barriers to asking for
support).

Model population
Defining the specific population for the economic evaluation is important because the model population
may not fully align with the population being supported in practice by headspace services. For example,
headspace services are reporting many young people accessing headspace services currently have more
severe levels of distress or complexity of presentation than headspace’s target population of young people
with mild to moderate, high-prevalence mental health conditions.
The target population for the purposes of economic evaluation comprises cohorts of young people with
predominantly mild to moderate mental health needs that fall within the scope of services provided by
headspace.

Comparator
The comparator is broadly defined as the state of the world in which headspace is absent. Admittedly, this
definition allows for many different interpretations of what such a world could look like. Consequently, the
following steps were taken to operationalise this concept:
3. It was proposed that the comparator would not presume the existence of any major alternative policies,
programs or mechanisms substituting for headspace. While there are many possibilities of what could
be put in place instead of headspace, focusing on any particular solution would be highly speculative
as there are no specific plans for such a substitution that would be relevant to this evaluation. Instead,
the comparator should broadly reflect the state of the current system in places where headspace was
not implemented.
4. The analysis simplified the definition of the comparator to three key parameters:
 a. the number of young people that would and would not receive mental health treatment;
b. the number of young people that would end up accessing MAT; and
c. the effectiveness of corresponding services if provided outside of headspace. Data from the hMDS
are used to make informed assumptions regarding (b) and (c).
5. Regarding (2a), the proposed definition was presented to the headspace Evaluation Reference Group
and a consensus approach was taken to define what the parameter value should be. As a starting point
for the discussion, the evaluators put forward the range of zero to 35 per cent of young people, which
was informed by the proportion of headspace delivered services that were MBS-funded services. It is
presumed that, in the absence of headspace, these services could plausibly still be provided through
other practices if sufficient capacity was available. The Evaluation Reference Group suggested that
the comparator value would fall on the lower end of the spectrum and the range of zero to 20 per cent
was agreed to inform the evaluation as the proportion of headspace clients that would access treatment
in the absence of headspace. This lower range was due to a number of challenges young people have
accessing alternative services, including limited or no other local services, and the costs of accessing
private practitioners. There was no literature available to support determination of this assumption for
the comparator group.

Model framework
A visual representation of two states of the world, with or without headspace, is provided in Figure 29.
This provides the basis for an incremental analysis. An incremental analysis is only concerned with those
elements that are subject to change. For this analysis, this report identified three areas, while important
elements of mental health service provision, are not likely to change with or without headspace.
6. Data available to the analysis (see section 4.1.4) suggest that the number and type of MBSfunded mental
health service provision outside of headspace is not affected by the presence of headspace.
7. It is assumed that there would be no change in the number and type of services, if any, funded and
provided entirely in the private system, that is funded out-of-pocket or by private health insurance. It
seems reasonable to think that young people who access such services may not be headspace clients
and, conversely, few of the current headspace clients would end up using privately funded mental
health services considering their socio-economic profiles and the cost of such services.
8. There is a core group of young people who do not access mental health services in either scenario
despite having mental health needs. This group is by definition the same in both states of the world.
Figure 29: Mental health service use by target population with or without the headspace program

Source: KPMG economic evaluation framework.

Notes: MH mental health; MBS Medicare Benefits Schedule

Figure 29 highlights the two elements that are subject to change in the world with headspace or the world
without headspace comparison. headspace clients who are currently accessing mental health services
(subsidised by MBS or otherwise) and those that access acute services due to, or as a consequence of,
having mental health needs. The latter includes hospital admissions and ED presentations.
This evaluation assumes that in the world without headspace fewer people would end up accessing mental
health services. It is critical for the evaluation to capture the consequences of mental health needs
remaining unaddressed. These consequences can be found in both (poorer) mental health outcomes and in
(higher) downstream demand for acute health services resulting from the unaddressed need.
The model structure for the comparison of the world with headspace against a world without headspace is
presented in Figure 30. In the ‘world with headspace’ scenario, all headspace clients access either one, two
or three or more OOS.
According to the results of the effectiveness analysis (see Appendix E.5 Appendix), a single OOS does not
deliver any significant improvements in mental health outcomes. Episodes with only two OOS do not meet
the criteria of MAT (defined in this economic evaluation as three or more OOS) and the observed
improvement in mental health outcomes is relatively small. Therefore, in the base case, this analysis
assumed that episodes with only two OOS do not produce an improvement in mental health outcomes.
Finally, the analysis assumed that episodes with three or more OOS meet the MAT requirement and
produce a significant improvement in the young person’s mental health outcome. While the literature
suggests that at least four OOS are required for an effect to be reliably obtained, analysed data from the
hMDS indicate that this effect is also present for closed episodes with at least three OOS (see Figure 60 in
Appendix E.5) . 144F

Given the discrepancy between the literature and the data, this analysis gives preference to the data that are
specific to headspace. This interpretation favours the headspace program.
Figure 30: The model structure for comparing the world with or without headspace
Source: KPMG developed model structure.
Note: QALY - quality adjusted-life year; OOS – occasion of service

Base case results

Table 19 below summarises the difference in costs and QALYs between the world with headspace and
without headspace for a base case analysis.
The base case is developed under the following key assumptions:
For headspace:
• Proportion of funding provided to headspace services attributable to service provision is 75 per cent.
Calculation of cost per episodes is presented in Appendix K.
• Outcome is QALYs gained per episode, which is calculated based on the transformation of K10 measures
into AQoL-8D . The benefit includes QALYs gained during the episode of treatment, sustained effects
145F

reported up to three months post treatment from the follow up survey and an extrapolation of sustained
effects up to 12 months after the last observed data point. Details for calculation of outcomes is
presented in Appendix K.
• Outcomes are adjusted for RTM effects.
• Treatment effects from episodes with one OOS or two OOS are assumed to be zero as these episodes do
not meet the MAT criteria.
For comparator:
• Ten per cent of headspace closed episodes are assumed to get alternative treatments in the world without
headspace. The rest of 90 per cent closed episodes are assumed to receive no treatment, hence, the
treatment effects after RTM adjustment are zero. In addition, these episodes would have higher
probability of hospitalisation as a consequence. Details on consequences are discussed in Appendix K.
• Two elements are used to define the comparator costs: 1) the scheduled fees for the observed mix of
initial appointments; and 2) the Australian Psychological Society national schedule of recommended
fees and item numbers for psychological services for the treatment . More details are presented in
146F

Appendix K.
• Outcomes of treatment from providers in the world without headspace are assumed to be the same as
treatments from headspace.
Compared to the ‘world without headspace’ scenario, the headspace program results in an average
incremental cost of $755 and an incremental QALY gain of 0.02. Combined, this generates an ICER of
$44,722, on par with established benchmarks on cost-effectiveness ratios.
Table 19: Results of incremental cost-effectiveness analysis

Scenario Costs QALYs ICER

World with headspace $842 0.019
World without headspace $87 0.002
Difference between the scenarios $755 0.017 $44,722
Source: KPMG analysis of hMDS.
Note: QALYs - quality-adjusted life year; ICER - incremental cost-effectiveness ratio

Sensitivity analyses
Table 19 presented results of a base case scenario. The scenario’s results were estimated with a model
where its input parameters were estimated with the most plausible assumptions. However, these estimates
are subject to uncertainty. The implications of this uncertainty are explored in several sensitivity analyses
presented in Table 20. These analyses examine how the point estimates are affected based on changes in
parameters as described above in Table 89. Each sensitivity analysis is described in detail in Appendix K.
The sensitivity analysis suggests that, despite it being a central unknown parameter in the economic
evaluation, varying the proportion of people receiving mental health treatment in the ‘world without
headspace’ scenario within the assumed range has little effect on the ICER.
More impactful is the value of the cost per OOS, which is determined by an assumption of the proportion
of funding provided to headspace services attributable to direct mental health treatment provision. Varying
this proportion between 60 and 90 per cent of the total services’ budget results in a considerable ICER
spread between $34,751 and $54,693, respectively.
The analysis also identifies that the assumption of three or more OOS representing MAT is an impactful
assumption. The base case assumption is based on similar outcomes observed in patients receiving three or
more OOS and four or more OOS. This, however, favours headspace, in light of the literature which
suggests only treatment comprising four or more sessions is effective and considered adequate. By
assuming that MAT is achieved with four or more OOS, the analysis results in an ICER of $56,894,
considerably above its base case value.
Extrapolating the benefit for up to five years, instead of 12 months within the base case analysis, has the
largest impact on the incremental gains, producing an ICER of $20,205.
The base case analysis adjusted mental health outcomes to account for regression to the mean. If the raw
outcome measures are used instead, the ICER changes to a lower value of $32,567.
The base case analysis assumes that closed episodes with only two OOS would not produce any changes in
the young person’s mental health outcomes. By allowing the young person to receive a partial treatment
benefit, the ICER becomes $35,713.
The analysis varied the average fees charged per OOS delivered outside of the headspace program and
found that it had negligible impacts on the ICER.
The analysis also found negligible changes to the ICER if it excluded the out-of-pocket costs from the cost
calculations.
The sensitivity analysis also varied the relative effectiveness of similar services provided outside of the
headspace program. These services can be 20 per cent less effective or 20 per cent more effective than the
headspace services. Both of these options had negligible impacts on the ICER.
The base case assumed that not receiving treatment increases a person’s probability of seeking care in the
acute system. In the scenario analysis, it was assumed that receiving treatment that does not meet the MAT
requirement would also lead to an increased probability of hospitalisation. This had negligible impacts on
the ICER.
Table 20: One-way sensitivity analysis of selected evaluation parameters

Scenario Change in Change in ICER

cost QALY
Base case (BC) $754.94 0.017 $44,722
Proportion receiving treatment in the 'no headspace' scenario (BC: 10%)
0% $837.05 0.019 $44,628
20% $672.82 0.015 $44,840
Proportion of headspace services’ budgets attributable to service provision (BC: 75%)
60% $586.62 0.017 $34,751
90% $923.25 0.017 $54,693
Relative effectiveness of similar services provided outside of headspace (BC: 100%)
80% $754.94 0.017 $43,909
120% $754.94 0.017 $45,739
Fees charged outside of headspace (BC: $260 per OOS)

$198 $773.19 0.017 $45,803

$320 $737.27 0.017 $43,676
Extrapolation of benefit to 5 years (BC: 12 months) $754.94 0.037 $20,205
MAT achieved in 4+ OOS (BC: 3+ OOS) $754.94 0.013 $56,894
No RTM for health outcomes (BC: apply RTM) $754.94 0.023 $32,567
2 OOS give partial benefit (BC: no gain from 2 OOS) $754.94 0.021 $35,713
Exclude out-of-pocket costs (BC: include out-of-pocket) $802.87 0.017 $47,561
Any treatment that is not MAT increases probability of hospital $753.15 0.017 $44,616
admission (BC: no treatment only)
Note: QALYs - quality-adjusted life year; ICER - incremental cost-effectiveness ratio; BC - base case; OOS - occasion of service;
MAT - minimum adequate treatment; RTM - regression to the mean

5.2.2 How does the cost-effectiveness of headspace vary across

services
The variation in costs and effects across services are shown on a cost-effectiveness plane in Figure 31
below. While headspace can be considered cost-effective, on average, the cost-effectiveness varies largely
across headspace services. Applying the $50,000/QALY threshold, 62 out of 112 (55.4 per cent) services,
,
would be considered cost-effective considering their individual ICERs . 147F 148F

Figure 31: headspace service-specific incremental costs and effects

Source: KPMG analysis of the cost-effectiveness dataset.
Notes: QALY Quality-adjusted life year.

Table 21 presents a one-way exploration of headspace service types surrounding the base case. The sub-
group analysis suggests that there is considerable variation in the ICER by the services’ state or territory,
regionality, maturity and size in terms of episodes treated during 2019-20. The spread in the ICER is most
noticeable when splitting headspace services by their regionality, maturity and size.
The ICER is the lowest among services located within major cities of Australia ($40,881) and is the
highest among services within remote or very remote areas of Australia ($460,052). This implies the cost
per QALY gained in remote or very remote areas is 11 times the respective cost in major cities. This is
likely driven by the fact that most of the headspace services (59) and episodes (30,737) are located and
treated, respectively, within major cities. By contrast, only four services are located in remote or very
remote areas and they collectively treated 497 episodes, or 5,858 OOS.
Table 21 also shows there is a positive correlation between the services’ maturity and their respective
ICERs. Services opened less than four years have an average ICER of $74,143. This cost per QALY
gained is approximately 1.7 times that of services which opened more than four years ago. This indicates
that maturity allows a service to become more established in the local community, ensuring appropriate
staffing meets local needs and using experience to deliver services efficiently.
Lastly, Table 21 highlights that larger services (in terms of numbers of episodes closed) are generally more
cost-effective than smaller services. Services that treated at least 600 episodes, the largest category, have
the lowest ICER of $34,267, whereas the smallest service category has an ICER at around $138,586.
Among the four categories explored an inverse relationship is observed between the service size and
ICER.
Table 21: Sub-group analysis of headspace service ICERs

Scenario Change in total Change in total

ICER
costs QALYs
State or territory
ACT $350,373.81 5.842 $59,979
Victoria $8,888,910.18 155.679 $57,098
Northern Territory $592,309.99 10.481 $56,512
Tasmania $750,280.77 13.755 $54,546
Western Australia $4,745,608.39 91.580 $51,819
New South Wales $11,216,121.63 258.017 $43,471
South Australia $2,801,958.57 64.483 $43,452
Queensland $8,496,737.86 238.015 $35,698
Regionality
Remote or very remote Australia $1,265,065.43 2.750 $460,052
Outer regional Australia $4,866,644.43 88.325 $55,099
Inner regional Australia $9,990,150.37 215.466 $46,365
Major cities of Australia $21,720,440.96 531.310 $40,881
Service maturity
Opened less than four years ago $4,989,204.79 67.292 $74,143
Opened between four and six years ago $10,549,958.46 239.906 $43,975
Opened between six and eight years ago $9,254,878.52 220.543 $41,964
Opened more than eight years ago $13,048,259.43 310.110 $42,076
Service size (in terms of number of episodes)
Delivered less than 200 episodes $3,125,863.47 22.555 $138,586
Delivered between 200 and 400 episodes $12,296,890.37 240.265 $51,181
Delivered between 400 and 600 episodes $13,060,634.18 301.916 $43,259
Delivered more than 600 episodes $9,358,913.18 273.116 $34,267
Note: QALYs - quality-adjusted life year; ICER - incremental cost-effectiveness ratio; BC - base case; OOS - occasion of service;
MAT - minimum adequate treatment; RTM - regression to the mean

5.2.3 Cost-effectiveness of headspace – in conclusion

The results of the cost-effectiveness analysis indicate that headspace is broadly cost-effective, based on the
assumptions used within the base case cost-effectiveness analysis. The ICER determined for headspace
services was $44,722 having assumed that benefits associated with access headspace services last for 18
months, and having adjusted for RTM within outcomes. This is lower than the threshold used to assess
similar healthcare services, of $50,000 per QALY.
There is a large variation in cost-effectiveness across services. This stems both from the variation in cost
per episode of care and the variation in outcomes achieved. As discussed in the cost analysis section, under
the current funding model, all services receive relatively similar annual funding amounts, regardless of the
volume of services they deliver. Even assuming outcomes are similar across services, this alone can lead to
a large variation in average cost per OOS (larger services would be more cost efficient than smaller
services). The effectiveness and the cost-effectiveness analysis further show there is also considerable
variation in outcomes and QALYs gained across services. This may be due to the extrapolation of benefits
beyond the last observed outcome at the follow up time, which amplified QALY gains in services with
better treatment outcomes and exacerbated the variation in cost-effectiveness across services.
The key unknowns modelled within the cost-effectiveness analysis, including the proportion of young
people who would receive alternative treatment if there were no headspace, the effectiveness of treatments
provided outside of headspace, and the cost of these alternative services, were not key drivers of the results
of the analysis. Sensitivity analysis shows that these assumptions had only minor impacts on the overall
ICER for headspace services.

6 Factors affecting
the future
implementation,
sustainability and
enhancement of
headspace
6.1 Barriers and enablers to
headspace meeting its
objectives
This evaluation found that effective outcomes are being achieved overall and for headspace’s objectives.
This includes objectives enabling mental health literacy and stigma reduction, positive service experience
(including youth friendly and inclusive services) and connecting young people into specialist services,
through consortium partnerships and pathway mapping across local service systems. These elements of the
headspace model all operate to a high standard, however there is variation in the effectiveness of the model
in each domain across different cohorts of young people.
Evidence reviewed suggests that the 16 elements of the headspace model, core and enabling, work
together to support these objectives, and that they combine to provide unique and complementary
contributions to outcomes of young people.
Qualitative data analysis conducted as part of this evaluation indicated that seven components of the
headspace model prove to be an ongoing challenge for services to deliver. These are:
• community awareness and engagement;
• four core streams;
• service integration;
• national network;
• multi-disciplinary workforce;
• blended funding; and
• monitoring and evaluation.
The barriers and enablers associated with these elements of the model are discussed in the following
sections.

6.1.1 Community awareness and engagement

Community awareness building and engagement is undertaken by headspace services to increase mental
health literacy, reduce stigma, encourage early help seeking and promote access, while building strong
relationships with young people, their family and friends, other local services and the broader community.
As discussed in Chapter 3, the importance of community awareness and engagement activities was
highlighted by stakeholders from all groups consulted for this evaluation, identifying this as a key enabler
of the objectives of the headspace model, including mental health literacy, early help seeking and stigma
reduction. At the same time, it consistently came up as a barrier to services achieving their objectives, with
insufficient resourcing able to be allocated to community awareness and engagement, largely due to
funding shortages and funding being prioritised for clinical services.
There appears to be tension between the desire to enhance awareness raising and promotion activities with
the need to deliver direct services to young people. This may constrain headspace services’ capacity to
meet the objectives of increasing mental health literacy, improving early help seeking and combatting
stigma about mental illness and help seeking behaviour. While these objectives are being effectively met
across headspace for the general population of young people attending, there are significant differences in
the effectiveness of headspace in meeting these objectives for young people from culturally and
linguistically diverse backgrounds, LGBTQIA+ young people and Aboriginal and Torres Strait Islander
young people. It is possible that, with increased focus and funding for community awareness and
engagement activities, these outcomes could be improved.

6.1.2 Four core streams

The headspace model includes the provision of an enhanced primary care platform with four core service
streams – mental health, physical and sexual health, alcohol and other drugs, and vocational and
educational support – to holistically address the main mental health and wellbeing needs of young
people within the local community.
As discussed in Chapter 2, headspace services are primarily focused on the provision of mental health
support and intake and assessment activities, with the three other core streams significantly less of a focus.
Eighty-one per cent of services provided by headspace centres, and 91 per cent of those provided by
satellite services are focused on the two main activities of mental health support and intake and
assessment. Theoretically, the consortium component of the headspace model should be an enabler to
delivering all four core streams, with opportunities for the lead agency to engage consortium members
who can provide expertise in one or more of the core streams. In practice, when a young person requires
physical or sexual health services, alcohol or other drug services or vocational and educational support,
they are more likely to be referred elsewhere than to receive support through headspace.
Stakeholders identified workforce attraction, retention and funding limitations as the key barriers to
providing the four core streams. While anecdotally, the provision of the four core streams in a ‘one stop
shop’ is valued by young people and other stakeholders of headspace, the strong results for the model
(despite services predominately providing only one of the four streams) indicates that the other three are
less critical to the overall effectiveness of the model in supporting mental health and wellbeing outcomes
for young people.

6.1.3 Service integration

The headspace model is designed to bring services together to function as one, providing a seamless
service experience for a young person, particularly if they require care involving multiple service providers
and supports.
As discussed in Section 3, headspace services are consistently recognised by external stakeholders for their
contributions to support pathways to care through integration and care coordination. However, there are
challenges around service constraints with many health services experiencing high demand. This is
particularly the case in regional and remote communities where there is a lack of alternative services
available.
At the same time, headspace service staff consulted reported that young people’s mental health needs are
becoming increasingly severe and complex, with most presentations being outside the headspace model’s
mild to moderate criteria. headspace’s service capacity to refer young people with more severe conditions
to more appropriate services is constrained by a combination of capacity issues in local services, and high
levels of demand within headspace services which reduce the capacity for staff to balance clinical
workloads with the additional activities required for successful service integration.
headspace service providers are impacted by a range of factors associated with the effective functioning of
the broader service system. The consequence of this is that many headspace services have long wait lists of
young people to receive clinical support. In 2018, an investment of $152 million over seven years was
made by the Commonwealth Government to fund the hDMEP aimed at addressing these challenges by
increasing access and reducing wait times. The effectiveness of this program of work is outside the scope
of the current evaluation, however the issue of wait times continues to be raised by stakeholders when
discussing headspace services.

6.1.4 National network

The network of headspace services across Australia collaborates to share learning, innovation and best
practice and, in turn, facilitates continuous improvement of services to enhance youth mental health and
wellbeing outcomes. It is composed of all headspace centres, satellites and other services, headspace
National, PHNs, lead agencies, Consortia, and Youth and Family and Friends Reference Groups.
The national network of headspace services is an important enabling component of the headspace model.
Facilitated by headspace National, the network supports evidence-informed practice and the continuous
improvement of services. It also provides a key interface between headspace National, with its role
focused on the headspace model, and PHNs, who focus on local commissioning of headspace services on
behalf of the Commonwealth Government, the primary funder of headspace services in Australia.
Stakeholders interviewed from across the headspace landscape identified tension between the delivery of a
nationally consistent model in a regional context and subsequent impacts on how effectively the headspace
model meets its objectives. For example, headspace service providers are required to satisfy both
headspace National and PHN requirements in their operations, which is an area of challenge for all parties
as they negotiate model fidelity and localised service delivery.
While evidence gathered during the evaluation suggests that the headspace model is generally achieving
good outcomes, the consequence of this tension between primary stakeholders for headspace services is
increased administrative burden, and possibly also reduces the capacity of services to meet the needs of
young people in a timely and tailored way.

6.1.5 Multi-disciplinary workforce

headspace's multi-disciplinary workforce includes clinical and non-clinical workers required from a range
of disciplines and backgrounds – with the right knowledge, skills and expertise – who work together to
holistically meet the mental health and wellbeing needs of young people, and their families and friends,
within the local community.
Attracting and retaining a multi-disciplinary workforce is a critical enabler of the headspace model, and a
consistent area of challenge for headspace service providers. Difficulties in competing with income offered
through private psychology clinics or CYMHS mean that headspace services struggle to attract and retain
clinicians. Often, headspace clinicians are early in their career, and pass-through headspace services on to
better pay and longer tenure after a few years. Similar difficulties are seen with recruitment and retention
of GPs.
While the headspace model is designed to support young people with mild to moderate, highprevalence
mental health conditions, a consistent and substantial proportion of young people attending headspace
have multiple risk factors alongside mental illness. With limited referral pathways to specialist or tertiary
mental health services, headspace services are challenged to meet the needs of these young people in-
house. The higher prevalence of early career clinicians employed within headspace can lead to reduced
expertise and capacity to respond to these young people with higher needs. This also has subsequent
impacts on workflow, such as increased waiting times for young people to see a clinician at headspace,
leading to reduced capacity to undertake community engagement and engage with young people with
milder presentations. With a less experienced workforce, the continuous investment of resources into
robust clinical governance is also an ongoing area of effort required for headspace services and lead
agencies.

6.1.6 Blended funding

The headspace model uses multiple funding streams and in-kind contributions to increase income
diversity, flexibility and sustainability of the service in accordance with the needs of the headspace service,
young people and their community to ensure access to no or low-cost services.
The use of blended funding to support the headspace model means that private practitioners who charge to
the MBS are working alongside salaried staff who are funded through the headspace core grant. This
funding model has the advantage of supporting a variety of multi-disciplinary workers who are able to
charge to the MBS, including GPs, psychiatrists and psychologists, and incentivises a high volume of
service delivery.
At the same time, for many services in non-metropolitan areas, local workforce shortages mean the MBS
model is not viable. Rural and regional services describe using their core headspace grant to fund salaried
clinicians, and then struggle to resource other key roles, such as community engagement officers and case
managers. This is borne out by findings in Chapter 4 which found that while, in general, services deliver a
similar level of average improvement per episode, the cost of delivering that improvement can vary
widely. In addition, the MBS billing model results in headspace services not being able to bill for core
parts of service delivery, including case management, work with a young person’s family, and missed
appointments where young people do not attend.
The pressures on headspace services to utilise grant funding for clinical services impacts their capacity to
prioritise other components of the model, particularly case coordination, referral, community awareness
and engagement activities and support for young people presenting with more complex conditions. This
results in a reduced impact on mental health literacy, early help seeking and stigma reduction and
increased wait times.

6.1.7 Monitoring and evaluation

Monitoring and evaluation includes the continual collection and review of comprehensive information to
facilitate service planning, delivery, evaluation and continuous improvement for headspace services,
PHNs and the national network.
As discussed in Chapter 3, monitoring and evaluation activities undertaken by headspace National and
PHNs are extensive and contribute to continuous improvement of programs and services, as well as
contributing to the evidence base of youth mental health care more broadly. headspace services, lead
agencies, PHNs and headspace National all see a portion of the data collected through the hMDS, based on
their specific reporting requirements, but none of these organisations have a full picture of the service.
This fragmented view reflects the complex governance and stakeholder landscape for the model, and
anecdotal evidence suggests it may exacerbate differences in viewpoint and reduce collaboration.
Data collection, measurement and monitoring activities undertaken by headspace National through the
hMDS capture short-term changes in mental health outcomes for young people, however measurement of
the longer term impacts of headspace could be improved. Separate from the monitoring and evaluation
activities undertaken by headspace National, the lack of centralised economic data to identify and improve
the cost-effectiveness of headspace was a further limitation noted by a range of stakeholders.
The lack of data linkage and potential for this to be used to better understand headspace’s impact on young
people, as well as the lack of centrally collected, reliable and complete costs data, are barriers to the
monitoring and evaluation of the headspace model. These barriers prevent the development of a cohesive
picture of the costs of service provision and the value of headspace to young people, their families and the
community more broadly.

6.1.8 In summary
Enablers and barriers for the headspace model
Enablers and barriers for the headspace model are closely related to services’ activities around
community awareness and engagement; providing four core streams of services; service integration;
the national network; attracting and retaining a multi-disciplinary workforce; the blended funding
model; and monitoring and evaluation. Each of these areas is challenging in the broader mental
health system in Australia, and they interact to increase wait times and reduce access of ‘hard to
reach’ groups.

6.2 External factors that have

impacted or will impact
headspace objectives
being delivered
The key barriers and enablers of headspace described in Section 5.1 above provide insight into a number
of external factors that have an impact on headspace objectives being delivered.

6.2.1 Broader mental health system challenges

A common theme when examining the barriers and enablers of headspace is its place within the broader
mental health system in Australia. As described in Chapter 1, there have been a number of major inquiries
and reports into the system in which headspace operates in Australia.

Limited referral pathways

One of the implications of the challenges to the broader system is that it can be difficult for young people
to find and access appropriate support, with limited services available within the health sector, as well as in
other social and vocational supports, and particularly in nonmetropolitan areas. This is a key barrier for
headspace, where successful service integration requires the presence and capacity of local services so that
young people can experience seamless service and supports involving multiple service providers.

Stigma and discrimination

Another external factor which impacts headspace is the level of stigma and discrimination in the broader
Australian community against people with mental illness. The reduction of this stigma is part of the
community awareness and engagement component of the headspace model, but headspace is only one of
many services and systems attempting to improve attitudes towards mental health in Australia. In some
communities, stigma and shame around mental illness are strong barriers to young people seeking help,
and to headspace achieving its objectives.

Workforce shortages
Workforce shortages in key roles and locations is another key barrier to headspace meeting its objectives,
which is largely external to the model itself. Particularly in rural and remote areas, attracting and
maintaining a multi-disciplinary and culturally diverse workforce is challenging, and is made more
difficult by the structure of the headspace core grant, which works most effectively in combination with
MBS billing to bring private practitioners, including psychologists, GPs and psychiatrists, into the
headspace service. In rural and remote settings, these professionals are few in number and tend to focus on
private practice.

Demand for services and complexity of presenting need

Under-investment in prevention and early intervention across the mental health service system is likely to
result in further increased demand from young people requiring mental health support. The success of
headspace’s brand leads to high demand for headspace services, and its ‘no wrong door’ approach means
all young people are seen and supported, even those who are subsequently referred on to more appropriate
care. This, combined with the external factors associated with increased need in mental health support for
young people in Australia, act as constraints on headspace service providers’ ability to meet the model’s
objectives.

6.2.2 In summary
External factors impacting headspace
A range of external factors provide challenges for the implementation of the headspace model, in
particular the limited referral pathways available in many areas, stigma and discrimination against
those with mental illness, workforce shortages and high demand for services and complexity of
presenting need. headspace service providers work diligently within the headspace model to
compensate and adjust for these external factors and ensure the objectives of headspace are met.

6.3 Changes required to the

design of headspace to
enable it to meet its
objectives
6.3.1 Is the current design of headspace sustainable in the short
and medium-term to deliver outcomes for young people?
Despite the challenges faced in implementing the headspace model, as discussed in Chapter 3, the model is
working well across all domains. The 16 components of the headspace model, as set out in the hMIF, all
operate well and are each important to the overall program logic of the headspace model. Through
investment and activities in mental health literacy and stigma reduction, early help seeking and increased
access to support is encouraged. Through headspace, young people are supported into pathways of care
which seamlessly integrate with other services to meet the young person’s needs, without the young person
being required to know how to navigate the service system themselves. Engaging in youth friendly,
welcoming and inclusive supports across a range of psychosocial domains helps young people improve
their mental health and wellbeing outcomes.
While challenges in the broader service system can make this model difficult to deliver in certain locations
or periods of time, the model itself is sound and has strong logic supported by robust evidence.

6.3.2 Changes required to the design of headspace

Changes to the design of headspace
On balance, this evaluation has not found any evidence to suggest that changes are required to the
design of the headspace model in order to enable it to meet its objectives more effectively. Despite
challenges in meeting the needs of some cohorts, and constraints and limitations brought about by
broader mental health system issues, headspace is achieving its intended outcomes with its current
design.

6.4 Changes required to the

implementation of
headspace to enable it to
meet its objectives
Given the challenges, enablers and barriers faced by the headspace model, there are a number of areas
where implementation of headspace services could be enhanced to enable it to meet its objectives even
more effectively. As discussed above, while it is effective overall, there is variation in outcomes and the
model has mixed success in supporting ‘hard to reach’ cohorts of young people compared to young people
in the general population attending headspace. Wait times are an area of criticism for the model and the
complex governance arrangements are burdensome. To help address these challenges, a number of
recommendations are put forward below, for implementation in the short- to medium-term.

6.4.1 ‘Hard to reach’ groups

A number of key findings from this evaluation indicate that more needs to be done to support ‘hard to
reach’ groups to engage with headspace services and to improve outcomes achieved for those who do
engage. Three groups of ‘hard to reach’ young people have engagement and outcomes which differ from
those of the broader population of young people accessing headspace: young people who identify as
LGBTQIA+, young people from culturally and linguistically diverse backgrounds and Aboriginal and
Torres Strait Islander young people. Relevant findings and reflections for each group are summarised
below.
In particular, one strategy commonly endorsed to support engagement and ongoing support for ‘hard to
reach’ groups involves enhancing representation of those groups within the workforce. Collaboration and
service integration of community services designed specifically for ‘hard to reach’ groups, such as
Aboriginal Community Controlled Health Services (ACCHSs), is also a common strategy to building
better referral pathways and engagement with priority groups .149F

When individuals from these priority groups engage with services, many strategies can be employed to
maintain engagement and achieve positive outcomes. Stakeholders from headspace services were able to
describe various ongoing activities undertaken to promote cultural safety and inclusion, including through
training, policies and procedures. The provision of culturally safe and appropriate services is key to
successful engagement with priority groups, for example through employment of identified workers,
ensuring an inclusive and respectful physical environment, provision of flexible support models including
outreach, peer and group opportunities, and engagement with their community (such as family, friends and
Elders).

Young people who identify as LGBTQIA+

As described in Chapter 3, analysis of clinical data indicates that LGBTQIA+ young people experienced
less improvement across all measures than young people who do not identify as LGBTQIA+ . At the same
time, data indicates that LGBTQIA+ young people are significantly more likely to be older than 21 years
of age when attending a headspace service, a trend which is consistent with LGBTQIA+ young people’s
help seeking behaviours more broadly. There are many factors which impact mental health help seeking
behaviours of LGBTQIA+ young people, including the age at which young people generally negotiate
sexual orientation and gender norms, perceived judgements by health care workers and presence of
existing natural support through family and friends .
150F

While LGBTQIA+ young people tend to access headspace at a later age, they are significantly more likely
to present with more risk factors, but are as equally likely as young people from the general population to
be presenting in the early stages of a mental health condition. For LGBTQIA+ young people, access rates
of headspace supports are high and have remained stable over the data period.
One interpretation of this data is that for LGBTQIA+ young people, late help seeking may be constraining
the clinical effectiveness of support they receive at headspace, although this cohort reports high user
satisfaction levels and feels headspace is a safe and welcoming place for them.
Another interpretation of this result may be that the headspace clinical model is unsuitable for LGBTQIA+
young people, and that this group has unique or particular needs which the model does not support. Given
the high user satisfaction from young people who identify as LGBTQIA+, this does not seem to be a likely
explanation.
Whilst delayed help seeking is common amongst LGBTQIA+ young people, this may be a key factor
leading to relatively poor clinical outcomes for this group. headspace is uniquely placed to strengthen the
role of community awareness and engagement activities for LGBTQIA+ young people and focus on
encouraging early help seeking, including outreach to schools and participation in community events with
a view to normalising mental health and wellbeing help seeking for LGBTQIA+ young people,
particularly for those aged under 21 years.

Young people from culturally and linguistically diverse backgrounds

By contrast, culturally and linguistically diverse cohorts achieved statistically similar improvements on
clinical outcome measures as young people who do not identify as culturally and linguistically diverse.
Culturally and linguistically diverse young people are also significantly more likely to be older than
21 years of age when attending a headspace service, but are as equally likely as young people from the
general population to be presenting with low mental health risk factors or early stages of a disorder.
For young people in these groups, access rates may be constrained by stigma, as described in focus groups,
interviews and survey responses from young people who identify as culturally and linguistically diverse.
These results indicate that for culturally and linguistically diverse young people, community awareness
and engagement activities could be undertaken, targeting stigma and mental health literacy. This may
improve early help seeking and access of the headspace model, and increase the volume of young people
from these groups who would benefit from support through headspace.

Aboriginal and Torres Strait Islander young people

Improvements in SOFAS and MLT outcome measures were statistically significantly lower than the
average analytical sample among the Aboriginal and Torres Strait Islander cohort. However, when using
the K10 outcome measure, outcomes among the Aboriginal and Torres Strait Islander cohort and the
general young person population accessing headspace are statistically similar. It should be noted that while
the K10 is widely used in population health surveys and as a clinical measure of distress, there are
questions as to its cultural appropriateness for Aboriginal and Torres Strait Islander young people . There
151F

would be value in exploring more reliable measures of mental health and wellbeing in Aboriginal and
Torres Strait Islander young people, for use within headspace. For example, the modified Kessler scale
MK-K5 may be a useful alternative assessment tool to support a reliable understanding of the
psychological distress levels of Aboriginal and Torres Strait Islander young people presenting at headspace
,
services .
152F 153F

Aboriginal and Torres Strait Islander young people attending headspace fare well on two of the three
indicators of early help seeking examined in this evaluation. They are significantly more likely to be under
the age of 21 compared to the general population of young people attending headspace and are as likely to
be presenting with low mental health risk as young people from the general population. At the same time,
however, Aboriginal and Torres Strait Islander young people are more likely to be presenting in later
stages of a disorder than the general population of young people attending headspace.
For Aboriginal and Torres Strait Islander young people, accessing headspace services occurs earlier, but
the associated gain in psychosocial outcomes associated with mental health and wellbeing is lower than for
young people from the broader population attending headspace. For this group, more needs to be done to
enhance the capability of headspace services to work with Aboriginal and Torres Strait islander young
people to enhance service take up and retention and achieve improved outcomes.

Opportunities to enhance outcomes for ‘hard to reach’ groups

In light of findings that the headspace model has mixed success in reaching and supporting young people
from ‘hard to reach’ groups, consideration should be given to emphasising key components and roles
within the model which are designed to support these outcomes. Prioritising the use of data to drive
engagement and workforce tailoring for priority cohorts should also be considered.
One strategy commonly endorsed to support engagement and ongoing support for ‘hard to reach’ groups
involves enhancing representation of those groups within the workforce. Building the headspace
workforce to reflect the local population should be a priority. This may include local strategies to targeting
identified staffing needs, such as providing dedicated internships and traineeships for workers from
priority populations, opportunities for upskilling and development the existing workforce from these
cohorts, or support for formal qualifications (e.g., Certificate IV in Youth Work). When individuals from
priority groups engage with services, many strategies can be employed to maintain engagement and
achieve positive outcomes. Key to this is the provision of culturally safe and appropriate services. This can
be achieved through employment of identified workers, ensuring an inclusive and respectful physical
environment, provision of flexible support models including outreach, peer and group opportunities, and
engagement with their community (such as family, friends and Elders). Collaboration and service
integration of community services designed specifically for ‘hard to reach’ groups, such as ACCHSs , is 154F

also an important strategy to building better referral pathways and engagement with priority groups.
Each of these strategies lends itself to an increased focus on community awareness and engagement
activities to improve outcomes associated with headspace services. Throughout the evaluation,
stakeholders consistently raised the importance of this work in building early help seeking and mental
health literacy but indicated that it was an area of the model which is often underresourced and time
consuming. Increased emphasis, planning and resourcing for community awareness and engagement
activities could have a material impact on the extent to which young people from ‘hard to reach’ groups
seek help from headspace, and on the extent to which this is associated with clinical improvements.
Table 22: ‘Hard to reach’ groups recommendations

Recommendations
1. The headspace model has mixed success in reaching and supporting young people from ‘hard to reach’
groups. Enhancing representation of these groups within the workforce may support engagement and
ongoing support for young people who identify as part of ‘hard to reach’ cohorts.
Lead agencies and headspace services should draw on PHN needs analyses to prioritise their
workforce needs, and implement strategies to diversify the headspace workforce to be representative
of the local community and to enhance engagement with relevant ‘hard to reach’ groups.
Source: KPMG 2022

6.4.2 Service integration

Analysis demonstrates that the more OOS a young person receives (with reliable improvement commonly
achieved after six or more sessions), the better mental health outcomes they will achieve. Around 19 per
cent of young people who accessed headspace during 2019-20 received six or more OOS by episode
closure. Around 36 per cent of young people accessing headspace are receiving just one occasion of
service.
Data limitations prevented further insights regarding the trend towards single OOS. A more detailed
investigation would be beneficial to better understand the causation and benefits of single sessions therapy,
including reviewing existing research. In particular, there is opportunity to explore:
• why young people are accessing only one or a low number of sessions;
• whether these young people are being referred onwards after the session and to where;
• what the research shows about single session therapy benefits and risks; and
• effective approaches to measure effectiveness of single session therapy.
While some of this trend toward single OOS in an episode of care may be explained by out of scope cases
being referred onward, there was limited evaluation data to either support or reject this hypothesis. Instead,
there are two key factors that became clear throughout the evaluation, contributing to this trend and which
require a service integration response – pathways to care and service demand. These are discussed below.
Pathways to Care
There are many factors constraining headspace’s ability to deliver seamless pathways to care and
integrated services. While there is ample support and evidence for the ‘no wrong door’ approach, there is
limited capacity for headspace services to refer young people out to more appropriate or specialised
services as required. As a consequence, headspace services end up supporting young people who may not
be suitable for the model, often resulting in a bottleneck and leading to long wait times. Add to this high
clinical workloads, and headspace services’ capacity to prioritise service integration and collaboration
activities is diminished.
Feedback from young people also suggested that many young people’s perception of headspace was
skewed, with many non-users suggesting they felt they were either too complex or not complex enough to
access the service. Many reported a lack of awareness of youth mental health services and how to
determine which services will best meet their individual needs. Many young people, carers and workers
reported an experience of young people being bounced between services based on unclear service criteria
or reduced capacity.

Service Demand
Demand for youth mental health services, including headspace services, has overtaken capacity. Sector-
wide workforce pressures also continue to constrain the volume of young people able to access support
and further contribute to long wait times and reduced service offerings. Challenges associated with
supporting young people to find appropriate care were frequently raised by stakeholders.
The combination of headspace’s ‘no wrong door’ approach, along with these service systems challenges,
significantly impacts headspace services’ ability to carry out the core business of supporting young people
with mild to moderate, highprevalence mental health conditions and delivering early intervention and
community engagement. Whilst some of these challenges are not exclusive to headspace, the service is in a
unique position to address challenges resulting from reduced care pathways, increased service demand and
workforce issues, through prioritising solutions to service integration.
Of relevance here is the National Mental Health and Suicide Prevention Agreement (National Agreement),
which sets out the shared intention of the Commonwealth, state and territory governments to work in
partnership to improve the mental health of all Australians . This agreement will seek to improve service
155F

integration within the mental health sector. Additionally, the National Initial Assessment and Referral
(IAR) for Mental Healthcare Project is another initiative from the Commonwealth which supports PHNs
and their contractors to establish effective systems for initial assessment and referral for individuals. The
IAR State of Play Report indicates that whilst resource intensive, collaboration and co-design with local
stakeholders is a key enabler of the strategy and has led to exciting observations .The report also notes
156F

digital decision support tools and smart referral forms, and education and training as other key enablers of
the IAR.

Opportunities to enhance service integration

This evaluation has found that better mental health outcomes for young people are associated with more
sustained engagement with headspace services. However, sustained engagement with headspace for young
people is constrained by two key challenges, finding the most appropriate care to meet their needs, and
wait times driven by high service demand.
Efforts to enhance local service integration are underway at the federal and state levels. These include
PHN commissioned work, headspace National service integration initiatives, the IAR, the National
Partnership Agreement and associated bilaterals activity with elements aimed to enhance integration
between headspace services, community services and tertiary mental health services for young people.
While the details of these activities are outside the scope of this evaluation, findings related to service
integration and the benefit to young people of sustained engagement with headspace indicate their
importance.
Through a focus within these activities on driving a shared commitment and approach to integrated
assessment and care across services, young people’s experience of engagement pathways could be
improved and their ongoing engagement with appropriate care could be better supported. In the context of
headspace, this may include projects to co-design and implement shared tools and procedures across
services that address regional needs and capability, drawing on the IAR and the PHN regional
commissioning role. The work already underway by headspace National through the Demand
Management and Enhancement Program and service integration exploration is also designed to support
these outcomes. Successful service integration also requires system-level support through government
investment and leadership across the Commonwealth, and state and territory jurisdictions. This work has
begun under the National Agreement, and associated state/territory bilateral agreements, and will continue
to be an ongoing area of focus.
With considered planning and active evaluation, these various pieces of work have the potential to
collectively enhance local partnerships, capability and resourcing, leading to improved early experiences
of young people with the mental health service system, particularly headspace services.
Table 23: Service integration recommendations

Recommendations
2. There is a need to further enhance integration with headspace services and local mental health and
other service providers. This should build on the current service integration piloting and evaluation
activity underway through the IAR and the PHN regional commissioning role. It should also consider
the National Agreement, and bilateral agreements developed with each state and territory in relation
to specific strategies to support service integration.

This would support access to more appropriate initial connections to services for young people and
provide greater clarity for referrers locally. It would also support regional service connections and
providers’ understanding of services and supports available during and following a young person’s
 EOC with headspace.

Source: KPMG 2022

6.4.3 Governance and commissioning

National network
The national network, linking a national headspace model with a local commissioning approach, is an area
of tension and should be explored for improved collaboration and flexibility. Each lens brings a vital factor
driving the success of the headspace model, however roles and relationships across the national network
are complex. Anecdotal evidence indicates that, at times, these conflicting priorities create inefficiency and
frustration.
Benefit may be gained by undertaking a refresh of roles and responsibilities across the network, clearly
identifying areas of operations which are the remit of each stakeholder. Increased clarity and span of
control for stakeholders within the national network may help reduce duplication and inflexibility and lead
to greater impact on the needs of young people.
As key stakeholders within their local community, PHNs are uniquely positioned to undertake needs
analyses of the young people in their area. Yet, PHNs perceive limited scope for regional variation within
the headspace model and reduced opportunity to utilise core headspace grant funding to commission
services that match the needs analysis undertaken in the local community. Theoretically, the hMIF
provides a level of agility to meet localised needs, however this is not understood or experienced by all
stakeholders, including PHNs and lead agencies. There is an opportunity for greater sharing and
communication regarding the adaptability of the hMIF and its ability to be agile and respond to local
needs.
An implementation refinement project could be undertaken to explore how the PHN local lens could be
better used to commission a model consistent with the hMIF that responds to identified regional need.
Whilst the hMIF allows for local responses in theory, this project would specifically take into account the
PHNs’ needs analyses, as well as the local service landscape and any gaps or areas of particular priority, in
the design and commissioning phase. The outcome of this project may be a new commissioning approach
to headspace services, where the hMIF provides a minimum standard across the 16 components of the
headspace model, but PHNs would select from these in order to build a localised and bespoke model to
meet the needs of their community, for example by including varied proportions of service stream effort in
the headspace service funding agreement. The headspace model and brand are valuable in reducing stigma,
enhancing community recognition and securing funding. This proposed approach would retain fidelity to
important components of the hMIF model, whilst also demonstrating the flexibility of the model to adapt
and respond to local needs.
Whilst there was overall improvement in mental health outcomes for young people accessing headspace
services, the reliable improvement and clinically significant change results warrant consideration. The
hMIF provides a solid framework for commissioners and lead agencies to guide the delivery of the
headspace services; however, quality control of the delivery of evidence-based interventions is generally
indirect, being led by the professional ethics of individual staff and clinicians employed within lead
agencies. With a smaller proportion of episodes of care achieving reliable change (that is, the change is
greater than a difference that could have occurred randomly), a focus on fidelity to the therapeutic model
and measuring subsequent outcomes may be warranted. It is possible, for example, that improved
outcomes may be achieved through enhancements to the quality components of clinical governance and a
focus on efficacy and fidelity to evidence informed treatments. Enhancement in this area could also have a
positive impact on attendance and retention rates.
The governance and commissioning roles within the national network of the headspace model are critical
enablers of strong implementation and positive outcomes for young people. Findings from this evaluation
highlight that service planning and clinical governance are important areas of focus in governance and
commissioning activities. Table 24 sets out three recommended areas of activity to improve the operation
of these elements of the model.

Table 24: Governance and commissioning recommendations

Recommendations
3. This evaluation has identified tension between different stakeholders regarding the agility of the model
to address local needs, and constraints on the capacity to tailor headspace services locally.

Government should work with PHNs and headspace National to undertake a refresh of roles and
responsibilities across the network. This should focus on clarifying the scope of roles in planning,
commissioning, delivering and tailoring headspace services.
4. There is a high degree of consistency of service mix across headspace services, with AOD, physical
and sexual health and vocational support representing a very low proportion of services provided.
Stakeholder feedback has suggested this may not always reflect local or regional need, and that
headspace service planning inconsistently draws on PHN needs analyses to inform and update the
local headspace service mix of the four core streams. It would be expected, for example, that a region
with significant substance misuse issues for young people may need a greater mix of AOD support
services at the local headspace service, or similarly where there are areas with higher rates of chronic
health issues in younger populations, physical and sexual health services should be appropriately
prioritised.

Government should consider investing in an implementation refinement project to explore how the
PHN local lens could be better used to commission a model consistent with the hMIF that responds to
identified regional need. This could allow greater capacity to reflect the PHNs’ local needs analysis
 and the local service landscape, including areas of high need. The project should consider the
potential risks of reducing the consistency of costs and outcomes across headspace services and
ensure mechanisms are in place to maintain a level of fidelity to core elements of the headspace
model.
5. Whilst there was overall improvement in mental health outcomes for young people accessing
headspace services, reliable improvement and clinically significant change results were lower than
expected. This suggests that clinical governance and the quality control of the delivery of evidence-
based interventions could be enhanced.

PHNs should take an active role in ensuring that headspace lead agencies prioritise clinical
governance which ensures quality service provision and adherence to evidence-based approaches.
With support and monitoring from PHNs, lead agencies should formalise processes to regularly
monitor efficacy, performance against outcome benchmarks and evidence-based approaches, where
these are not already in place. This could be achieved through mechanisms such as: ensuring
interventions meet recommended practice guidelines; setting and achieving clear benchmarks for
outcomes; regularly monitoring service outcomes data; and supporting staff to access focused
training and supervision.
Source: KPMG 2022

6.4.4 Monitoring and reporting

Monitoring and evaluation activities undertaken by headspace National and PHNs are extensive and
contribute to continuous improvement of programs and services, as well as contributing to the evidence
base of youth mental health care more broadly. However, the existing framework for monitoring and
evaluating headspace services, including work conducted by headspace National, lead agencies and PHNs,
does not allow for assessment of longterm outcomes and impact of services on individuals or the broader
community. There is strong support for data linkage activities across the service system, and further
investigation into these activities should be prioritised.
To support data linkage, data from headspace should be collected in a way that allows it to be linked to
other datasets so that outcomes of young people who access headspace can be tracked over time and
compared to those who do not access headspace. This data can then be used for evaluation and research
purposes to track long-term outcomes in a more meaningful way. Where possible, these should also be
supplemented by studies using experimental or quasi-experimental designs so that outcomes can be
rigorously measured and attributed to headspace. Similarly, data about outreach and engagement activities
should be collected for monitoring and evaluation purposes. Data collection activities should be conducted
in a consistent way through regular reporting with mechanisms in place to collate a common set of
outreach and engagement activity data for evaluation purposes. This could be done at regular intervals,
with care taken to avoid placing an undue burden on headspace services.
Noting that additional data capture is under development by headspace National and PHNs, a number of
other areas currently lack sufficient or reliable data to conduct in-depth analysis or, at times, any analysis.
Such areas include:
• data on outreach and engagement activities – activity type, duration, and number of young people
participating;
• post/follow up data for episodes involving single OOS;
• reason for closure data – to differentiate between unplanned exits and planned exits;
• data to record when someone was referred on to another service – service type referred to, stage in care at
point of referral (e.g., intake, mid treatment, exit), if referral was taken up;
• data on service users who identify as having a disability, and who identify as neurodiverse;
• secondary consultation data to demonstrate and quantify service integration activities and investment;
and
• mechanisms to understand and report on funding and FTE resourcing.
These areas could be enhanced through ongoing or point-in-time data and evaluation activities, however
this needs to be balanced against the creation of unnecessary additional administrative burden for
headspace service providers and young people accessing headspace services.
Table 25: Monitoring and reporting recommendations

Recommendations
6. Despite extensive reporting undertaken across activities within the headspace model, a number of gaps
in data collection were identified through the evaluation. Filling these gaps could support better
monitoring and evaluation of outcomes associated with the headspace model.

The following data should be collected by headspace National to inform future evaluation and
continuous improvement processes:
• outreach and engagement activity data – including activity type, duration, and number of young
people participating;
• outcomes data beyond 90 days post EOC – with a particular focus on episodes involving a single
OOS;
• reason for closure data – to differentiate between unplanned exits and planned exits;
• referral data – service type referred from and to, stage in care at point of referral (e.g., intake, mid-
treatment, exit), whether referral onwards was taken up;
• demographic data – enabling service users to identify as having disability, and to identify as
neurodiverse;
 • funding data – capturing ongoing, in-kind support and specific MBS items claimed through
headspace services in hAPI; and
• workforce data – capturing more detailed workforce information including full-time equivalent
workforce available and their characteristics.
The extent to which the needs of young people are being met at an area-level, as estimated through
PHN local needs analysis, should be considered a priority monitoring activity by PHNs.
7. While data is collected extensively across activities within the headspace model, the longer-term
impacts of headspace are not measured.

Data from headspace should be collected in a way that allows it to be linked to other datasets, so that
outcomes over time of young people who access headspace can be better understood when compared
to those who do not access headspace. Ethical considerations should also be prioritised, for example
to ensure that individuals cannot be identified in the data. The administrative burden of additional
data collection activities for providers and young people accessing headspace should be balanced
against the benefits provided through enhanced reporting.
• Linked data sets might include:
• self-harm hospitalisations;
• substance abuse hospitalisations;
• suicide deaths;
• MBS mental health services accessed;
• PBS usage;
• mental health related emergency department presentations;
• education and employment outcomes; and
• income support use.
Data linkage should be supported by government, and should be complementary to data linkage
being conducted under the National Agreement.
8. A number of areas across the headspace program logic could benefit from further evidence to
understand the best implementation approach to support improved outcomes for young people.

Data linkage should be supplemented by studies using experimental or quasi-experimental designs so

that outcomes can be rigorously measured and attributed to headspace. Where this is not achievable
through control or comparison group analysis using linked data, government should allocate funding
for one-off experimental studies. Priority examples include:
• exploring differences between centre and satellite headspace services;
• research into single session interventions, given that approximately 36 per cent of episodes of care
have a single OOS, and wait times lead to disengagement of young people before treatment;
• examining how AOD, physical and sexual health and/or vocational assistance support mental health
and wellbeing, both in the short and medium-to-long-term;
• exploring the most appropriate intake and assessment approaches when engaging with Aboriginal
and Torres Strait Islander young people;
• exploring the most reliable measures of mental health and wellbeing in Aboriginal and Torres Strait
Islander young people, for use within headspace;
• examining the extent to which young people and families experience more streamlined and less
fragmented pathways of care in the medium-term.
Detailed logic documents should be developed to support the collection of appropriate data.
Source: KPMG 2022

6.5 Changes required to the

funding arrangements of
headspace to enable it to
meet its objectives
6.5.1 Are the current funding arrangements sustainable?
As outlined in Section 4.1 above, headspace services are currently funded through a blended funding
model, including core grants received from the department, through PHNs as the commissioning body, and
use of MBS billing by practitioners providing supports through headspace services.
Currently, there is no specific funding model used to determine the grant contributions made by the
department to headspace services. A model was previously used, however this has been moved away from
in recent years, and all headspace services now receive similar volumes of grant funding. One-size-fits-all
approaches to providing funding to headspace services are not cost-effective, and this is demonstrated by
the significant variability in cost-effectiveness between individual headspace services discussed in Chapter
4.
In addition, headspace services have significantly varied success making use of the blended funding
model. Some services provide significantly more OOS than otherwise possible through grant funding, as a
result of MBS-billed services from private practitioners. In other services, particularly in nonmetropolitan
locations, a model that relies heavily on MBS billing is not viable or sustainable as there are local
workforce shortages, which impact the ability for these headspace services to deliver MBSbased clinical
services.

6.5.2 Changes required to current funding arrangements

Based on the challenges described above with the current funding model, and conclusions regarding the
current sustainability of funding arrangements, a new funding model should be developed to guide funding
for all headspace services moving forward. The funding model should be flexible, and take into account
the individual characteristics of each headspace service.
Table 26: Funding arrangements recommendations

Recommendation
9. headspace services do not currently collect or report the full costs of operation, with in-kind
contributions and indirect costs not captured under funding agreement requirements. Without
accurate data regarding the full costs of operating a headspace service, the cost-effectiveness of the
headspace model can only be estimated, as has been done through this evaluation.

Government should prioritise the collection of full and accurate data to inform a more detailed review
of current cost information across all headspace services. This could be done through individual
engagement with headspace services, or compulsory survey of all headspace services. This would
confirm current costs of delivering the headspace model, including in-kind contributions provided to
services and other indirect costs. This would identify differences in costs for different headspace
services based on location, and other service-specific factors. The official count of headspace
services should also be revisited to improve clarity of funding arrangements, e.g., the count of
headspace services could be updated to reflect the number with a Trade Mark Licence Deed.

10. While the headspace model is broadly effective in achieving its intended outcomes, a number of areas
related to funding are challenging for services providing headspace. Difficulty in attracting and
retaining a multi-disciplinary workforce varies across regions, as does the need to undertake
extensive community engagement activities with ‘hard to reach’ groups. At the same time, across the
headspace services included in this evaluation, the number of OOS funded each year varies widely,
while funding levels within the core headspace grant are relatively consistent across services. This
variation in demand and service provision leads to considerable differences in the estimated
economic efficiency across headspace services.

Government should develop a variable funding model based on demand and regional need which
accounts for differences in location, population and service delivery modes and volumes. This should
consider core funding components, such as administrative costs and management costs, as well as
more variable cost components which may include:
• location of the headspace service, including regionality and areas of workforce shortages, with
increased allowance for salaried staff where access to MBS-based staff is challenging;
• the size of the population to be supported by the headspace service, including the number of young
people within the headspace service catchment and geographically proximate communities to be
supported by the service, and associated required service FTE; and
• the headspace service type to be implemented, including whether the service is a headspace centre,
satellite service or outreach service.
A separate funding model, or specific element, should be considered for establishment costs required
for a new headspace service.

Government should consider how a revised funding model may apply to existing headspace services,
in addition to new services established in future.
Source: KPMG 2022

6.6 Broader system changes

that would support
headspace to meet its
objectives
6.6.1 Broader system changes
This evaluation has found a number of areas where headspace services can be improved over time.
However, there are a range of identified barriers and enablers which are also impacted by the broader
mental health service system and are not within the remit of individual headspace services, or the
headspace program overall, to control. As such, there are also a range of broader system-level changes that
would support headspace to meet its objectives going forward.

Increased prevention and early intervention services

As noted by the Productivity Commission and discussed in Section 2.4.3 above, there is currently
underinvestment in prevention-based supports for mental health across Australia. Increased focus and
provision of prevention- and early intervention-based services, to reduce the ultimate need for treatment in
the longer term, would support headspace objectives, by decreasing overall demand for treatment services,
and reducing current pressure on headspace wait times.

Improved service integration and pathways

As found in this evaluation, there are significant challenges for headspace services in managing the mental
health needs of young people who are presenting with more severe need than headspace is designed to
manage, and with more complex needs overall. This is especially the case where there are limited or no
local services to refer these young people to, either due to a lack of services or existing services with their
own wait time pressures.
Strengthening service integration across the mental health service system and ensuring all Australians,
including young people, have access to required services through clear pathways would support headspace
to better meet its objectives. This includes reducing pressure on headspace service wait times, reducing the
level of risk headspace services are experiencing from young people who were not intended to be
supported by the headspace model, and enabling more effective referral pathways with these services.
While focus on prevention and early intervention is more sustainable in the longer term, there is currently
significant unmet need for tertiary and other specialist supports, and this demand must be met for these
individuals in parallel with an increased focus on prevention.

Development of national mental health workforce

The workforce shortages experienced by headspace services, particularly for GPs, psychologists and
psychiatrists and in regional, rural and remote areas, are also experienced across the broader mental health
service system. Development of system-wide initiatives to support growth in the mental health workforce,
including in regional and remote areas, should also support headspace services to access those professions
which currently present a challenge.

6.6.2 Existing service system changes underway

As discussed in Section 1.1.3 above, there continues to be broad sector reform across the mental health
sector, and it is important to acknowledge broader service system changes that are already underway, or
that have been recommended that will also support headspace services into the future.

Recommendations from the Productivity Commission Inquiry into Mental Health

The Productivity Commission’s Mental Health Inquiry Report made a number of detailed
recommendations that will go some way to address the broader system changes described above. These
recommendations include:
• creating a person-centred mental health system;
• focusing on children’s wellbeing across the education and health systems;
• supporting the mental health of tertiary students;
• increasing informed access to mental healthcare services;
• linking consumers with the services they need;
• increasing the efficacy of Australia’s mental health workforce;
• developing best practice governance to guide a whole-of-government approach; and
• funding arrangements to support efficient and equitable service provision.
The specific actions and reforms described under these recommendations that are relevant to the service
system challenges impacting headspace services include:
• Filling gaps and addressing barriers in the services available to people who need support due to mental
ill-health, and their families and carers.
• Removing barriers to collaboration within and between different parts of the mental health system, by
actively encouraging information sharing and coordination between health service providers; creating
systems and processes that bring together the range of treatments and supports that people may
choose; and by reforming funding, to incentivise better cooperation and collaboration across mental
health services.
• Improving coordination and integration between health and other services to better promote recovery.
• Improving the efficacy of supports delivered through schools and workplaces, to promote better mental
health and early intervention.
• Establishing an evaluation and monitoring system that focuses on outcomes, and ensures that mental
health services are effective in supporting recovery.
• Supporting the development of single care plans for consumers with moderate to severe mental illness
 who are receiving services from multiple providers.
• Regional commissioning bodies developing and maintaining on-line navigation portals that include
detailed clinical and non-clinical referral pathways, which can be accessed by clinical and nonclinical
service providers.
• Aligning the skills, costs, cultural capability, availability and location of mental health practitioners with
the needs of consumers through the forthcoming National Mental Health Workforce Strategy.
• Developing a national plan to increase the number of psychiatrists in clinical practice –particularly those
practising outside major cities and in sub-specialities with significant shortages.
• Strengthening the peer workforce by providing one-off, seed funding to create a professional association
for peer workers, and in collaboration with State and Territory Governments, develop a program to
educate health professionals about the role and value of peer workers in improving outcomes.
• Developing a new whole-of-government National Mental Health Strategy that aligns the collective
efforts of health and non-health sectors.
• Strengthening cooperation between PHNs and Local Hospital Networks by requiring comprehensive
joint regional planning and formalised consumer and carer involvement.
The Commonwealth Government has indicated support either in full, or in part for all 21 recommendations
made within the Productivity Commission’s report.

Development of the National Agreement on Mental Health and Suicide Prevention

Significant work has recently been completed to develop the National Agreement on Mental Health and
Suicide Prevention, a joint agreement between the Commonwealth and states and territories, negotiated
through the Health National Cabinet Reform Committee. This agreement is intended to deliver a
comprehensive, coordinated, consumer-focused and compassionate mental health and suicide prevention
system, for all Australians, and address a number of the Productivity Commission’s specific
recommendations. In April 2022, the Agreement and bilateral agreements with State and Territory
Governments were finalised, and implementation is currently being considered.

6.7 Evaluation conclusion

This evaluation has examined the headspace model across a number of criteria. A range of data and
evidence has been analysed to assess the model's alignment to need and the fidelity of the model in
practice, including in terms of take-up and reach of service provision. The effectiveness of the headspace
model has been assessed against intermediate outcomes, service system outcomes, user experience
outcomes and psychosocial outcomes achieved through the model. The economic value of the headspace
model has also been assessed, alongside the sustainability of the model going forward.
Through the range of methods and analyses applied, this evaluation concludes that the headspace model
provides a comprehensive and complete set of components to address the mental health needs of young
people. The model incorporates components which are designed to prevent mental illness, through mental
health literacy, early help seeking and stigma reduction, and to treat mental illness whatever the presenting
need. While the model is intended to support young people with mild to moderate, high-prevalence mental
health conditions, through the 'no wrong door' approach and as a result of capacity pressures across the
mental health service sector which constrain referral pathways, every young person presenting at a
headspace service receives support of some kind.
When outcomes are examined, the model achieves mixed results for young people from 'hard to reach'
groups, across outcome areas. The model achieves its intended outcomes for the general population of
young people across domains, and the cost-effectiveness of direct services provided through the headspace
model is on par with established benchmarks on cost-effectiveness ratios. When longer-term benefits are
included in the analysis, the headspace model appears to be even more cost-effective, but more data is
required to substantiate this.
While the model is associated with positive psychosocial outcomes for young people, the majority do not
see a clinically significant change to their outcomes. In general, associated psychosocial outcomes only
become comparable to other psychotherapies once six or more sessions have been accessed.
There are opportunities to improve the efficiency and effectiveness of the model, through targeting the key
areas of 'hard to reach’ groups, service integration, governance and commissioning and monitoring and
evaluation. Pressures and reforms in the broader mental health services sector currently, and will continue
to, effect the headspace model. In its role as a national program to support the mental health and wellbeing
of young people, there is an opportunity to greater leverage the headspace platform for broader reform in
the sector.

Appendix A :
Evaluation Scope and
Method
A.1 The evaluation of the national headspace program
A.1.1 Overview
KPMG and its research partners, the Social Policy Research Centre at the University of New South Wales,
and batyr, were commissioned by the Department to evaluate the national headspace program, as delivered
through headspace services. This evaluation builds on previous evaluations of headspace and focuses on
the time period since the completion of the most recent evaluation conducted for the Commonwealth in
2015.
As headspace delivers a range of services and supports for young people and their families, this evaluation
sought to understand its overall contribution to the objective of improving mental health and wellbeing for
young people.
This evaluation represented an important opportunity to take stock of what is being delivered at individual
services and across the headspace network, and how this aligns with the core intent and expectations of the
headspace program.
The scope of this evaluation was focused on headspace services, as provided in individual services around
Australia, looking at the period from July 2015 to end of June 2020. Several aspects of the broader
program were explicitly out of scope, including the operations and performance of headspace National and
eheadspace. Other programs were also excluded from this evaluation, including the Individual Placement
Support trial, funded by DSS, and the EPYS Program provided at selected headspace services. These
initiatives have been the subject of separate evaluations.
It should also be noted that, while the evaluation primarily considered the period from July 2015 to June
2020, there were challenges associated with ensuring all stakeholders relate their views only to this period.
Stakeholders, who are described in more detail in Appendix B, were engaged following ethics approval for
the evaluation being granted in May 2021 through to December 2021. There may be differences between
these views and the data captured through headspace services between July 2015 and June 2020.

A.1.2 Current environment impacting the evaluation

It is also important to consider the context in which this evaluation was completed. The period between
2020 and 2022 was impacted by a range of factors. Firstly, the black summer bushfires created increased
demand for mental health services. Thereafter, the COVID-19 pandemic has had significant impacts on
service delivery, and an increased focus on mental health. The evaluation scope was also directly impacted
by COVID-19, with the last four months of the evaluation period from March 2020 to June 2020 part of
Australia’s first pandemic wave, with lockdowns and restrictions in place. During this period, services,
including headspace services, were required to shift service modalities to provide telehealth and virtual
services. In addition, the number of OOS delivered by headspace may also have been impacted, with fewer
young people able to access services.
Since the opening of the first headspace service in 2007, there has also been broader sector reform. Some
of the significant recent changes include:
• the establishment of the National Mental Health Commission (2012) and its review of mental health
services in 2015;
• the work of Australia’s National Suicide Prevention Adviser;
• the endorsement of the Fifth National Mental Health Plan in 2017, committing all Australian
Governments to eight priority areas;157F

• the Victorian Royal Commission into Victoria’s Mental Health System (final report delivered in February
2021);
• the Productivity Commission’s review into Mental Health (with the final report publicly released in
November 2020);
• the National Mental Health and Suicide Prevention Plan (released in May 2021); and
• the continuing negotiation of the National Agreement on Mental Health and Suicide Prevention.
These developments in Australia’s mental health landscape are important factors to take into account for
this evaluation of headspace. At the same time, this evaluation will help inform policy and investment
decisions about the future direction of headspace. The evaluation outputs will also feed into the reform
agenda shaping mental health service delivery in Australia for the next decade and beyond.

A.1.3 Evaluation Questions

This evaluation is targeting four domains of inquiry:
1. understanding headspace;
2. the effectiveness of headspace in achieving program outcomes;
3. the cost-effectiveness and value of headspace; and
4. factors affecting the future implementation, sustainability, and enhancement of headspace.
For each of these four domains of inquiry, a range of evaluation questions were specified and have been
answered through this evaluation. It should be noted that evaluation question 3.1 – what is the full cost of
headspace, was not able to be answered in full in this report. Data limitations impacting the collection of
complete cost information are discussed in Section A.4.3 below. These questions fall broadly into three
categories of evaluation, combining to provide important insights to inform policy and funding decisions
for decision-makers. In line with guidance for evaluating complex health programs , these categories
158F

include a process evaluation, an economic evaluation and an outcome evaluation using statistical methods
rather than an experimental design, which is unfeasible within project timeframes in the absence of pre-
existing data linkage arrangements.
The key Evaluation Questions are outlined in Table 27 below:
Table 27: Evaluation Questions

Domain of Evaluation Question Evaluation sub-questions

Inquiry
Understanding 1.1. What is being provided at a) What is currently available?
headspace headspace? b) How has this changed over the last five years?
c) How does this align with stakeholder
expectations and the objectives of headspace?
1.2. What does ‘success’ look like a) What would be a ‘positive outcome’ for a
for headspace? young person utilising headspace services?
b) What measures would show success?
c) How does this align with current measurement
and reporting of performance of headspace?
Effectiveness of 2.1. How effective is headspace in a) How well does headspace advocate for and
headspace in increasing mental health literacy, promote youth mental health and wellbeing in
achieving program early help seeking and access to their communities?
outcomes required services (generally and for b) To what extent has headspace reduced stigma
‘hard to reach’ groups, including associated with mental illness and help seeking
those who are at greater risk and for young people, their families and friends, and
less likely to seek help)? the community?
2.2. How effective is headspace in a) To what extent does headspace provide an
improving pathways to care for appropriate service approach for young people
young people through service with mild-moderate high-prevalence conditions?
integration and coordination? b) To what extent is headspace providing a
localised service offering, and what are the
barriers and enablers to this?
c) What other contributions does headspace make
to local communities?
d) To what extent does a ‘no wrong door’
approach assist headspace to meet its objectives?
e) What is the level of support for headspace
from other primary care and mental health
service providers?
f) To what extent does headspace assist young
people who do not meet the criteria for headspace
services to access alternative pathways of care?
2.3. How effective is headspace in a) To what extent does headspace provide a
ensuring young people can access culturally appropriate and inclusive service for
the help they need in an young people and their friends and families,
appropriate, accessible and youth including for vulnerable and diverse population
friendly way? groups and different age groups?
b) To what extent does headspace enable young
people and their families to access support where,
when, and how they want it, and what are the
barriers and enablers to this?
c) How is the establishment of alternative service
delivery models (e.g., satellites, outposts)
assisting headspace to meet its program
outcomes?
d) To what extent do young people participate in
the design and delivery of headspace, and how
does this influence young people and their
families’ experience of headspace?
2.4. How effective is headspace in a) To what extent do young people accessing
improving mental health and headspace achieve improvements in mental
wellbeing outcomes, including health, wellbeing, social and economic
physical health, social and participation, and life satisfaction?
economic participation (i.e., b) To what extent are outcomes sustained over
education or employment) and time?
quality of life? c) What factors are associated with positive
outcomes, and how do they vary across
population groups, presenting issues and amount
and type of services received?
d) To what extent does a positive regard for
headspace relate to improved outcomes for young
people?
Cost-effectiveness 3.1. What is the full cost of a) What financial and in-kind contributions
and value of headspace? support headspace and how does this vary
headspace between services (services types and locations)?
b) What are the out-of-pocket costs for young
people and how does this vary between services
(service types and locations)?
3.2 What is the overall cost- a) How does this vary between services (service
effectiveness of headspace? types and locations)?
Factors affecting 4.1. Are there any changes required a) What are the barriers and enablers to
the future to the design, implementation, and headspace meeting its objectives?
implementation, funding arrangements of headspace b) Are there any broader system changes that
sustainability, and to enable it to meet its objectives? would support headspace to meet its objectives?
enhancement of c) Is the current design of headspace sustainable
headspace in the short and medium-term to deliver
outcomes for young people? What changes, if
any, are required to support the sustainability of
headspace?
d) Are the current funding arrangements
sustainable?
e) Are there external factors that have impacted
or will impact headspace objectives being
delivered?
Source: KPMG 2022

A.2 Evaluation governance

The evaluation of the national headspace program directly involved a number groups and organisations,
who provided oversight to evaluation methods and activities, guidance on interpretation of findings and
data, and other input as required. These roles are described in more detail below.

A.2.1 Commonwealth Department of Health

As the ultimate sponsor for the evaluation, the department has provided ongoing oversight for the entire
evaluation, and input at key decision points. The department’s Youth Mental Health Section within the
Mental Health Services Branch were primary contacts, with frequent evaluation progress and status
meetings held with this section. In addition to these meetings, the Section also:
• developed initial evaluation questions for the evaluation, with input from headspace National;
• provided input into and endorsed the evaluation framework developed;
• provided input into the approach to ethics approval; and
• assisted with key insights to support interpretation of findings during the evaluation.
In addition to the oversight of the Youth Mental Health Section, the department’s Health Economics and
Modelling Branch within the Health Economics and Research Division were also consulted at key stages
of the evaluation, to test key evaluation methods and activities and provide endorsement for the Evaluation
Framework developed.

A.2.2 Evaluation Reference Group

An Evaluation Reference Group (ERG) was formed to provide advice and oversight for the evaluation, to
help ensure that the evaluation met its agreed objectives. The ERG provided guidance, advice and
feedback in relation to the evaluation methodology, its implementation in line with the evaluation
framework developed described above, and evaluation findings as these progressively emerged.
The members of the ERG collectively held deep knowledge and expertise in youth mental health and
epidemiology, mental health service delivery, and complex quantitative and qualitative evaluation
techniques. The group also included youth and lived experience representatives to ensure these
perspectives were closely incorporated at every stage of the evaluation. Membership of the ERG included
the following expertise:
• Youth mental health policy
• Economic evaluation methodology
• Program evaluation methodology
• Mental health commissioning
• Consumer experience
• Youth service delivery
The ERG met regularly throughout the evaluation, aligned to key milestones for the evaluation. Meetings
focused on:
• feedback and endorsement of the evaluation framework;
• feedback on the methodological approach to the evaluation, including cost-effectiveness analysis;
• updates on evaluation progress;
• discussion of findings as the evaluation continued; and
• discussion of recommendations for the headspace program.

A.2.3 Ethics Approval

This evaluation also received formal ethics approval, as evaluation activities met the National Health and
Medical Research Council definition of human research.
Ethics approval is an important safeguard for evaluation projects of this kind. The process of external
review ensures research methods, design and implementation approaches have been thoughtfully selected
to protect the autonomy and wellbeing of the people participating in research. This was a particularly
important consideration for the evaluation of the national headspace program, given that its key service
cohort is made up of vulnerable young people who are experiencing mental ill health or other wellbeing
challenges.
In determining the appropriate process for securing ethics approval, the evaluation team gave close
consideration to the department’s areas of focus for this evaluation and the range of young people intended
to participate in primary data collection. One important area of focus for this evaluation was the mental
health outcomes of Aboriginal and Torres Strait Islander young people. Producing insights regarding these
outcomes required the collection, analysis and reporting of data relating to Aboriginal and Torres Strait
Islander young people. This fell under the Australian Institute of Aboriginal and Torres Strait Islander
Studies (AIATSIS) definition of Aboriginal and Torres Strait Islander research in the AIATSIS Code of
Ethics.
Ethics approval for the evaluation was sought from the AIATSIS Ethics Committee, and took multiple
rounds of iteration to be completed. The approval looked at all human research activities to be completed
for the evaluation, including the components relating to research conducted with Aboriginal and Torres
Strait Islander young people. Creating a robust research approach that met ethical research requirements
was an intensive process, and overall project timelines for the evaluation were delayed to ensure this could
be completed.
A range of activities were undertaken to support ethics approval for the evaluation, and mechanisms were
established to support the safety of human research conducted, including:
• engagement with a range of local organisations within each of the locations proposed for deep dive
consultation, including local ACCHS or Aboriginal Land Councils, PHNs and headspace lead
agencies;
• inclusion of Aboriginal and Torres Strait Islander researchers to undertake consultation with Aboriginal
and Torres Strait Islander young people to ensure cultural safety;
• oversight of evaluation method development from a senior Aboriginal and Torres Strait Islander
researcher;
• endorsement from Gayaa Dhuwi, the new Aboriginal and Torres Strait Islander social and emotional
wellbeing, mental health and suicide prevention national leadership body, with the opportunity to
continue engagement with the organisation should culturally sensitive topics be raised by young
people;
• development of deliberate protocols in the event culturally sensitive information was raised during
consultation; and
• further engagement with local ACCHSs as part of stakeholder engagement at deep dive locations.
The original evaluation design proposed engagement with eight locations in detail, however support from
local Aboriginal organisations to support culturally safe engagement was not able to be secured in two
proposed locations, and the evaluation proceeded with six locations as a result. A more detailed
description of this evaluation activity is described in the section below.

A.3 Evaluation activities and data sources

The program evaluation activities undertaken to support the analysis are outlined below.

A.3.1 Review of program documentation

A desktop review of headspace program documentation was undertaken. The review analysed documents
prepared by both the department and headspace National. Additionally, a desktop review of headspace
services was undertaken to provide a preliminary understanding of what is being delivered at headspace
services. Documents reviewed include:
• headspace annual reports;
• headspace program logic and hMIF;
• previous external headspace evaluations;
• previous evaluation and research work undertaken by headspace National;
• publicly available information such as PHN demographics; and
• media releases, social media feeds, headspace service websites.

A.3.2 Consultation with policy owners and mental health sector

A targeted series of consultations with key mental health and headspace policy owners within the
department and broader mental health sector stakeholders were undertaken to explore how the intent and
objectives of the program were defined within the evaluation timeframe, and how these may have evolved
over time. These interviews focused on identifying the degree of alignment between the documented
objectives of headspace and the priorities or objectives of key policy owners – particularly in the context
of the network’s recent rapid expansion. They also addressed the role of headspace within the broader
Australian mental health service delivery landscape.
Representatives of the following organisations were consulted during the evaluation:
• Mental Health Division through the Mental Health Services Branch, Commonwealth Department of
Health;
• Population Health, Indigenous Health and Primary Care Divisions, Commonwealth Department of
 Health;
• headspace National;
• corresponding mental health policy areas from each state and territory;
• primary care and community mental health peak and research bodies, including:
o Orygen;
o Community Mental Health Australia;
o National Mental Health Consumer & Carer Forum; and
o Mental Health Australia;
• all PHNs across Australia; and
• headspace services in two metropolitan areas, Craigieburn and Bankstown, in addition to the deep dive
site consultations described below.

A.3.3 Deep dive site consultations

Information about activities undertaken by headspace services is collected through administrative or
service delivery data sets by headspace National. However, services, such as mental health advocacy work,
community outreach and efforts by staff to connect young people who do not meet headspace service
criteria with other local supports, are not captured in these data sets.
To develop a more comprehensive understanding of the headspace services offered and delivered within
these categories, the evaluation team undertook deep dive analysis into services and supports offered by a
selection of headspace services. The headspace services were selected based on the following criteria, and
were reviewed by the ERG, department and headspace National.
• Geographic location – ensuring the inclusion of a mix of metropolitan, regional, and remote locations.
• Socio-economic profile – ensuring the inclusion of sites serving communities across the income
spectrum, as indicated by the average weekly income for the Australian Bureau of Statistics (ABS)
Statistical Area 3 region in which each service is located.
• Range of headspace services – ensuring the inclusion of headspace centres as well as satellite services
and other models.
• Diversity of client cohort groups – ensuring the inclusion of services supporting communities with high
concentrations of Aboriginal and Torres Strait Islander young people and young people from culturally
and linguistically diverse backgrounds.
• Lead agency organisation types – ensuring the inclusion of services run by a mix of lead agencies,
including major national healthcare providers, local NGOs and Aboriginal Controlled Health
Organisations.
• Length of operation of headspace service – ensuring the inclusion of services which have been in
operation across the spectrum of the headspace program’s lifespan, selected from sites that have been
operational for more than 10 years, more than five years, more than two years and more than one year.
In conducting deep dive qualitative research, the evaluation team:
• conducted interviews with representatives of the commissioning PHN, lead agency, Youth Advisory
Group, senior clinical and administrative staff, local area GPs, TMHSs and Indigenous health or
community organisations;
• explored the demographic profile of the service’s users and their experiences through discussions with
staff; and
• explored the clinical, non-clinical and advocacy and community activities undertaken at the service and
the associated resourcing.
A standard questionnaire and data-collection matrix was developed for use in undertaking each of the deep
dive case studies. This ensured that consistent information was sought and analysed across all site
locations. The deep dive sites for the evaluation and their specific characteristics include:
Table 28: Geographic regions of focus for the evaluation

Location Service Characteristics Demographics (based on 2016 Census)

Duration of Aboriginal Culturally and

Average
operation In previous and Torres linguistically
Regionality Type household
(as at July evaluation Strait Islander diverse young
income
2020) young person person

Hub for Eden,

Outer
Bega >2 years Narooma, and N 4.4% 15.4% $961
Regional
Cooma spokes
Satellite
Inner
Gympie >2 year (Maroochydore N 3.6% 12.9% $938
Regional
is parent centre)
Joondalup >6 years Metro EPYS Hub Y (2015) 0.6% 46.6% 159F $1,957
Katherine >1 year Remote Centre N 48.9% 8.9% $1,485
Mt Isa >5 years Remote Centre N 24.5% 14.7% $1,833
Centre; parent for
Murray Inner Victor Harbor & Y (2015,
>10 years 3.7% 15.3% $965
Bridge Regional Mt Barker 2009)
satellite sites
Source: KPMG 2022

Given that deep dive sites were not selected to be a representative sample of the headspace model across
Australia, qualitative data from deep dive research activities were used to augment other data and
information collected from across the evaluation activities. To protect the confidentiality of service users
and providers in each site, the detailed deep dive case studies prepared during the evaluation fieldwork are
not available for publication.

A.3.4 Survey of young people

A survey was administered to young people in two distinct groups: (i) current and former headspace users,
and (ii) non-headspace users. The survey covered the following areas, with some variation to account for
whether young people have previously accessed headspace services:
• demographic information, including age, gender, country of origin, Aboriginal and Torres Strait Islander
status, languages spoken other than English, education and training, employment, and income sources;
• mental health literacy and help seeking;
• mental health service use;
• specific headspace service questions; and
• outcomes instruments, such as psychological distress and quality of life through K10 measures.
Young people who have used headspace services were recruited through headspace channels. Current users
of headspace services were invited to participate in the survey while completing required headspace
surveys to support hAPI data collection. Former headspace users who had accessed headspace services in
the 18 months to June 2021, and agreed to be contacted about future research, were invited to participate.
Young people who had not previously used headspace services were a community sample, aged 12 to 25,
and were recruited through promotion on batyr’s social media channels, and through engagement with
schools and universities nationally.
A total of 3,683 young people responded to the survey, with 1,234 responses from young people who
indicated they had never used a headspace service, and 2,449 responses from young people who indicated
they had used a headspace service.

A.3.5 Focus groups and interviews

A series of structured interviews and focus groups were undertaken with key headspace target populations
and stakeholders. These included:
• Forty-seven interviews with young people who have used headspace services – to obtain selfreported
information on outcomes, as well as to explore in depth young people’s experiences with headspace.
Young people who have used headspace services were recruited through headspace National, with all
current and former headspace clients who completed the survey described above given the option to
volunteer for focus groups, interviews, or both.
• Four focus groups with 10 young people from priority population groups who have accessed headspace
services, including Aboriginal and Torres Strait Islander young people, young people from culturally
and linguistically diverse backgrounds, young people with disability, LGBTQIA+ young people, and
young people in rural or remote areas – to understand how these groups experience headspace and
how the program may be enhanced to better meet their needs. These young people were recruited in
the same way as other headspace users described in the point above, with additional screening
undertaken to understand with which groups they may identify.
• Fifteen focus groups with 76 young people who do not access headspace – exploring awareness of, and
views about, headspace services and potential barriers to access, attitudes towards help seeking and
mental health, and how headspace advocates for and promotes youth mental health and wellbeing in
communities, and other contributions to local communities. Young people were recruited by batyr
through the organisation’s social media channels, as well as volunteers who had completed the young
people’s survey described above.
• Eight focus groups with 35 school and university counsellors – exploring accessibility and availability of
headspace services and current barriers to access, the appropriateness of headspace services in meeting
identified needs of young people, and observations about young people’s attitudes towards help
seeking and mental health, particularly any changes over time. School and university representatives
were recruited through batyr’s school and university networks.
• Five interviews with five GPs – exploring the referral pathways and process in more detail from this
stakeholder group’s perspective. GPs were firstly recruited through headspace services in each of the
six deep dive sites, however engagement levels were unexpectedly low from GPs as a stakeholder
group. Following low take up from this method, PHNs consulted were also asked to contact local GP
networks to find interested GPs. Following both methods of recruitment, only five GPs were able to
participate in an interview. Of the five GPs interviewed, three were based in Queensland (Gympie and
Gold Coast), one in NSW (Bega) and one in the NT (Katherine). The ERG also had one GP
representative member to provide additional input throughout the evaluation. It should be noted that
this GP works within a headspace service.

A.3.6 Survey of headspace services and their lead agencies

Beyond those that were established in 2019-2020, a survey was sent to all headspace services and their
lead agencies across Australia. The survey was designed to test key evaluation questions, as well as
whether barriers, enablers and other factors raised by stakeholders within the deep dive site consultations
described above were also experienced by other services. This survey covered the following areas:
• characteristics of the headspace service and the services it delivers;
• how services are meeting headspace objectives, including for young people who are hard to reach;
• key barriers and enablers for services in meeting objectives of the headspace program; and
• high level funding questions (headspace services only).
A total of 89 responses were received to the survey, however it should be noted that two responses
received only completed the requested financial information, as services were given the option to
complete these questions separately at a later time. In addition, 13 responses only completed the
location and type of headspace service they were representing, with no other questions answered.
Of the 76 surveys with answers other than demographic information, 34 per cent were from headspace
services in metropolitan locations, 47 per cent were in regional locations, and 18 per cent were in rural
or remote locations. Compared to the spread of headspace services nationally, metropolitan locations
were under-represented, and rural and remote services over-represented within the respondent group.
Approximately five per cent of responses came from satellite or other service models, which is a
similar proportion to the headspace network more broadly.

A.3.7 Analysis of the headspace Minimum Dataset

The hMDS comprises data on services delivered through headspace services. hMDS data is collected from
both young people and service providers. Data is collected on:
• young person profile;
• OOS; and
• EOC.
Service providers and young people enter required information for specified OOS received from a
headspace service, based on a standard set of questions that are used across all services. This data forms
the basis of the hMDS. Over time, the data collected for the hMDS has changed, with additional fields of
collection added, or some fields no longer collected. Wherever this has impacted the data presented
throughout this report, this is stated directly.
The hMDS has been analysed for the following information:
• headspace clients and their demographics;
• number of EOC, and the total number of OOS that make up each EOC;
• the young person’s primary issue and main services provided during each OOS;
• service provider characteristics for each OOS;
• client outcomes based on the K10, MyLifeTracker and Social and Occupational Functioning Assessment
Scale (SOFAS) scores; and
• funding sources for each OOS, such as out-of-pocket costs for young people and MBS funding.
The hMDS data analysed was for the period from 2015-16 to 2019-20. This included data on closed EOCs
that were created within each financial year. For 2019-20, this included EOCs that were closed by
December 2020, when data was extracted from the hMDS to support the evaluation.
This information has been used by the evaluation team to develop findings on services provided by
headspace and how they have changed over time, the characteristics of headspace clients, the services they
receive, outcomes achieved and how these contribute to the efficiency and value of the program.

A.3.8 Area-level effectiveness analysis

To support analysis of outcomes from headspace services where no control group exists, another form of
quasi-experimental methodology known as Difference-in-Differences (DID) was applied to further
,,
evaluate the impact of headspace services at the area-level, rather than the individual level .
160F 161F 162F

This DID design made use of longitudinal data to estimate the effect of headspace services by comparing
the changes in outcomes over time between areas. Specifically, the approach compared outcomes over
time for PHNs with few or no headspace services, to PHNs that have experienced a growth in headspace
services. The hypotheses were that PHNs that have seen an increase in headspace services will have:
• a reduction in the number of mental-health, self-harm and substance-abuse related hospitalisations and
the number of suicides; and
• an increase in the number of Medicare-subsidised mental health specific services as increasing exposure
to headspace should de-stigmatise the need to seek mental health care, especially outside the
headspace program.
To examine how variations in headspace exposure influence area-level outcomes over time, outcome
measures, aggregated by PHNs, were obtained from the Australian Institute of Health and Welfare
(AIHW) and Services Australia. These data included the population of 12 to 25 year olds from 2008-09 to
2018-19 and the number of mental-health related hospitalisations; intentional self-harm hospitalisations
related hospitalisations; illicit drug and alcohol related hospitalisations; deaths from intentional self-harm;
and Medicare-subsidised mental health specific services among 12 to 25 year olds.

A.3.9 Cost-effectiveness analysis

To determine the cost-effectiveness of headspace services, the costs of delivering headspace services were
brought together with the effectiveness of headspace services to allow quantification of the costs required
to deliver a unit improvement in QALYs. The cost analysis included the funding sources of each cost (e.g.,
MBS or headspace grant funding), deep dive locations as well as use of the headspace service survey, to
assist in assessing the proportion of costs that should be considered in-scope for the estimation of ICER.
The estimation of the ICERs also requires the definition of a comparator group which would be young
people with mental health needs in a world where headspace (or an equivalent program) was never
implemented.
ICERs were calculated for headspace as a whole and as individual services. To support this cost-
effectiveness analysis, the following analysis was also undertaken:
• Sensitivity analysis: to test the sensitivity of the results to assumptions surrounding the parameter inputs,
such as the proportion of headspace clients that would continue to seek treatment in a world without
headspace, the proportion of headspace services’ budgets attributable to treatment service provision,
relative effectiveness and fees charged per OOS provided outside of the headspace program and
definition of what constitutes as MAT.
More detailed explanations of the methods used for this analysis are contained in Section 4.2 and the
corresponding appendices, outlined in Section 4.2.

A.4 Data limitations and considerations

When reading this report, there are a number of key data limitations and considerations which should be
considered. In addition to those outlined here, some more specific data limitations are also discussed,
where relevant, throughout the report and appendices.

A.4.1 headspace minimum dataset

Overall compliance with required data collection

During analysis of the hMDS data provided by headspace National, it was observed that there is
considerable variation in data compliance across difference headspace services, and this was confirmed in
discussion by the headspace National data team. Some headspace services, despite being well-established
in metropolitan suburban areas, report low levels of service delivery. Other services appear to have low
levels of compliance with surveys, by both young people and staff providing services.
Compliance by young people has fluctuated more and been heavily impacted during the COVID-19
pandemic. Prior to COVID-19, young person compliance sat at between 80-85 per cent and dropped to
below 60 per cent during COVID-19 as young people did not want to use iPads in services and did not
always complete surveys they received via email or text prior to their appointments. Young person data
compliance is currently sitting at about 65 per cent. However, at the national level over the past 18-months,
Service Provider data compliance has remained above 90 per cent. 163F

There are also varying levels of compliance for specific items within the hMDS. For example, a significant
number of OOS did not have the main type of service provided to the young person recorded, which meant
these OOS were not able to be included in some analysis for the evaluation.

Young person treatment pathways and outcomes

The current hMDS dataset captures the treatment course of young people accessing headspace. This
includes listing the primary issue of a young person attending a headspace service, the main services they
received, and the evolution of their mental health outcomes via the K10, SOFAS and MLT outcome
measures.
However, information about the pathway before and after accessing headspace is less complete. In
particular, survey questions about the main reason for accessing headspace services, and whether a young
person has seen a mental health professional prior to attending headspace. had variable compliance.
The long-term therapeutic or clinical benefits of accessing headspace on employment, education or
training also requires a follow up response from the young person post-episode closure. The current young
person follow up survey provides some indication of the wider and longer-term impacts of headspace on
schooling and employment outcomes; however survey response rates are low.

headspace staff and workers

The hMDS includes a service provider dataset that contains information about the staff member or ‘service
provider’ who delivered the OOS, such as their age, gender and profession. This data indicates the type of
service providers who provided at least OOS at a headspace service but does not provide details on the full
staffing mix at the site. There is currently no systematic data collection of the headspace service workforce
aside from what is captured in this service provider survey, and a higher level annual workforce survey
which has been recently implemented.
There is also currently no clear picture of the number of FTE employed across headspace services. This is
made more complicated by the use of medical contractors who are not identifiable within the hMDS.
Service provider profession was recently added to the hMDS during July 2019. Previously, the hMDS
collected information on the service provider’s role rather than their profession. This has been
inconsistently completed as service providers may take on multiple roles within a headspace service.

A.4.2 Data linkage

The evaluation considered data linkage as a preferred evaluation method, in order to compare outcomes for
young people using headspace services to that of young people who have not used headspace services.
However, whether personal data collected from young people can be used to support data linkage within
current consent processes has not been investigated. In addition, to undertake data linkage for this
evaluation, data would have had to be collected from individual headspace services for linkage. It was
estimated by the data linkage authority that this type of data linkage would take approximately 18 months
to compete, which was not feasible for the evaluation period. The area-level effectiveness analysis was
undertaken in place of direct data linkage with other key datasets.

A.4.3 Cost data and cost-effectiveness

The hMDS dataset collates information about the funding source of each occasion of service, such as the
headspace grant, specific PHN funding agreements, in-kind contribution, or the MBS. However, it does
not capture the amount of funding provided for that OOS. The MBS item for MBS-funded OOS is not
captured, for example. These were required to be estimated based on the type of professional providing the
service in order to quantify the approximate costs of delivering headspace services.
Indirect costs and funding are also not captured within any current data collection activities. While the
evaluation sought more detailed input on these indirect and in-kind costs from headspace services through
both deep dive engagement and the headspace service and lead agency survey, very few headspace
services were able to provide this information.

A.4.4 Qualitative data collection

Engagement with GPs during the evaluation was limited (with five GPs consulted), with very few
volunteering to participate in a discussion despite multiple recruitment methods being used, including
through headspace services and PHN GP networks.

Appendix B :
Consultation
Details of stakeholders consulted and a summary of the themes they raised are contained below.

B.1 Stakeholders consulted

B.1.1 headspace services
Table 29: Stakeholders consulted from headspace services

Organisation Location Stakeholders consulted

Murray Mallee General Murray Bridge, South • Centre manager
Practice Network Australia
• Business Manager at Murray Mallee GP
Network
• CEO, Murray Mallee GP Network
• Two mental health clinicians
• Social worker and team leader
• Community engagement team leader across
three centres and Kangaroo Island
• Aboriginal youth and engagement manager

Gidgee Healing Mount Isa, Queensland • Centre Manager

• CEO, Gidgee Healing
• Senior Research Fellow, Mount Isa Centre for
Rural and Remote Health
• Acting Regional Manager, Deadly Choices
• Family Wellbeing (FW) Regional Manager,
Gidgee Healing
• North West Remote Health (NWRH)
psychologist
• Director, Mount Isa Centre for Rural and
Remote Health
• Court Link Officer, Magistrates Court
• Family Wellbeing Worker, Gidgee
• IPS manager
• Practice Manager
• Community Engagement Officer

Black Swan Health Joondalup, Western • COO, Black Swan Health

Australia
• Clinical governance and compliance manager
• Acting headspace Joondalup centre manager
• Psychologist, acting clinical lead and
predominantly allied mental health
commission role
• Triage coordinator
Anglicare NT Katherine, Northern • Executive Manager Mental Health – Anglicare
Territory NT
• Operations Manager – Primary Mental Health
Services – headspace Darwin and Katherine
• Centre Manager – headspace Katherine
• headspace Senior Clinician
• headspace Senior Clinician Primary
• Youth Mental Health Clinician

Grand Pacific Health Bega, New South Wales • CEO of Grand Pacific Health
• Clinical Lead
• Intake manager
• Family clinician and mental health clinician
• IPS manager
• Youth care coordinator
• Mental health clinician
• Peer worker
• Acting service manager
• Bushfire recovery and community engagement
• Mental health clinician (Cooma)
• Senior administrator (Cooma)
• Executive manager of primary care portfolio at
Grand Pacific Health

Flourish Australia Bankstown, New South • Acting Clinical and Operations Manager at
Wales headspace
• Acting Team Leader
• Acting Senior Clinical Manager headspace
• Member of Youth Reference Group

Orygen Craigieburn, Victoria • Program Manager, Primary Clinical Services at

Orygen
• Director – National and Local Clinical Service
Innovation
• Manager, Clinical Services
• Senior Access Clinician

Youturn Gympie • CEO of Youturn

Source: KPMG 2022

B.1.2 Primary Health Networks

Table 30: Stakeholders consulted from Primary Health Networks

Organisation Location Attendee Name

Country SA Primary South Australia • Manager, Mental Health and AOD, Youth Portfolio
Health Network
Western Queensland PHN Queensland • Coordinator, Primary Mental Health Care
Commissioning

WA Primary Health Western Australia • Current and former contract manager for Joondalup
Alliance (WAPHA) 164F

headspace
• Metro Operations Manager, WAPHA

NT Primary Health Northern Territory • Procurement Coordinator NT PHN

Network
• Health Commissioning Manager NT PHN
• Health Commission Lead NT PHN

South Eastern NSW PHN New South Wales • Contract manager for headspace Bega
• Manager in charge of population data, planning and
reporting

South Western Sydney New South Wales • Mental Health Program Advisor
PHN
• Research and Evaluation Coordinator
• Mental Health Coordinator
North Western Melbourne Victoria • Executive Director, Service Development &
PHN Reform at North Western Melbourne Primary
Health Network
• Director MH and Wellbeing
• Manager CYMH

Central Queensland, Wide Queensland • Coordinator/ Contract Manager for Central

Bay, Sunshine Coast PHN Queensland
• MHAOD Coordinator/ Commissioner for Gympie
headspace
• Primary HealthCare Officer
• Business Support Officer/ intake and referrals
All other PHNs National • Various representatives from all other PHNs in
workshops
Source: KPMG 2022

B.1.3 Indigenous organisations

Table 31: Stakeholders consulted from Indigenous organisations

Organisation Location Attendee Name

Moorundi Aboriginal Murray Bridge, South • Manager of Social and Emotional Wellbeing
Controlled Health Australia team
Organisation
Arche Health / Wangen Joondalup, Western • Executive Manager Health Services
Murduin Aboriginal Australia
Health Service • Manager
• RN on the Aboriginal Health team and at
Joondalup Health Centre

Bega Local Aboriginal Bega, New South Wales • Aboriginal Community Liaison
Land Council
Katherine West Health Katherine, Northern • Manager Population Health, Katherine West
Board Territory Health Board
Source: KPMG 2022

B.1.4 Tertiary Mental Health Services

Table 32: Stakeholders consulted from tertiary mental health services

Organisation Location Attendee Name

Bega Child and Bega, New South • Acting Manager, Community Mental Health
Adolescent Mental Health Wales
Services • Adult Clinical Leader
• CAMHS Clinical Leader

Victorian Child and Across all of Victoria • Clinical Director of the Alfred Child and Youth
Adolescent Mental Health Mental Health Service and headspace
Services
• Executive Director - Orygen
• Associate Program Director CYMHS – Eastern
Health
• Divisional Manager – Child and Youth Mental
Health Service at Austin Health
• Manager for Albury-Wodonga Health (regional)
• Manager - Monash Health
• Clinical director of the CYMHS program –
Goulburn Valley, Shepparton
• Manager of Goulburn Valley CAMHS
• Manager at Ballarat CAMHS
• Clinical psychologist, Monash Health
• Clinical psychiatrist, Bendigo Health
Source: KPMG 2022

B.1.5 General Practitioners

Table 33: General Practitioners consulted

Location
Bega, New South Wales
Mount Isa, Queensland
Gold Coast, Queensland
Katherine, NT
Gympie, Queensland
Source: KPMG 2022

B.1.6 headspace National

Table 34: Stakeholders consulted from headspace National

Organisation Location Attendee Name

headspace National Melbourne, Victoria • Executive Director, Strategy, Impact and Policy
Executive
• Chief Operating Officer
• Head of Centre Services
• Head of Clinical Leadership
• National Centre Services Manager
• Chief Scientific Advisor
• Executive Director Clinical Practice
• Evaluation Manager
• Evaluation Team Lead – Centre-based services

headspace National Board Melbourne, Victoria • Board Chair

• Three Board Members
• Youth Advisor to the Board
Source: KPMG 2022

B.1.7 Commonwealth Government

Table 35: Stakeholders consulted from Commonwealth Government

Organisation Location Attendee Name

Department of Health ACT • First Assistant Secretary - Mental Health Division
• First Assistant Secretary - Population Health
Division
• First Assistant Secretary - Indigenous Health
Division
• Assistant Secretary - Mental Health Services and
Evidence Branch
• Assistant Secretary – Primary Health Networks
Branch
Source: KPMG 2022

B.1.8 State and Territory Governments

Table 36: Stakeholders consulted from State and Territory Governments

Organisation Location Attendee Name

ACT Office of Mental ACT • Suicide Prevention Officer
Health and Wellbeing
Mental Health Commission NSW • Deputy Commissioner
of NSW
Health NSW NSW • Executive Director, Mental Health Branch
• Senior Clinical Advisor, Child and Youth Mental
Health/Senior Child and Adolescent
• Senior Manager, Mental Health – Children and
Young People

NT • Senior Director
Mental Health Alcohol and
Other Drugs Branch, NT • Suicide Prevention Coordinator
Health

Queensland Mental Health QLD • Commissioner

Commission
Queensland Health’s QLD • Executive Director
Mental Health, Alcohol and
Other Drugs Branch
SA Mental Health SA • Commissioner
Commission
Department of Health and SA • Director, Policy Planning and Safety, Office of the
Wellbeing SA Chief Psychiatrist
• Three departmental employees

Mental Health, Alcohol and VIC • Executive Director, Mental Health and AOD
Other Drugs Section, VIC System Operations and Commissioning
Department of Health and
Human Services • Manager, 0-25 System Redesign, Programs and
Performance - Mental Health Services
• Chief Adviser -Transformation, Mental Health &
Wellbeing Division

WA Mental Health WA • Four employees of WA Mental Health Commission

Commission
National Mental Health National • CEO
Commission
Source: KPMG 2022

B.1.9 Peak bodies

Table 37: Stakeholders consulted from peak bodies

Organisation Location Attendee Name

Orygen Vic • Executive Director
• Director, Strategy and Policy

Community Mental Health National • CEO, Community Mental Health Australia

Australia
• CEO Mental Health Coalition of South Australia
• Policy Officer Northern Territory Mental Health
Commission
• Policy Officer Mental Health Council of Tasmania
• Senior Policy Officer Western Australian
Association for Mental Health
• Policy Officer Western Australian Association for
Mental Health

National Mental Health National • SA Consumer Representative

Consumer & Carer Forum
• WA Consumer Representative
• Deputy Carer Co-Chair - Representative of Mental
Health Carers Australia
• Consumer Executive Member – Aboriginal and
Torres Strait Islander Consumer Representative

Mental Health Australia National • Acting CEO

• Senior Policy Advisor
• Senior Policy and Projects Officer
Source: KPMG 2022

B.2 Stakeholder engagement themes

The following tables present reflections common across each stakeholder group engaged in the course of
the evaluation, identified by data source.

B.2.1 Young people who use headspace

Evaluation activities and data sources

Qualitative themes are sourced from a mix of data described below.
• Survey - administered to young people who currently use headspace services were invited to participate
in the survey while completing required headspace surveys to support hAPI data collection. You
people who formerly had accessed headspace services in the 18 months to June 2021, and agreed to be
contacted about future research, were invited to participate.
• Focus groups and interviews - 47 interviews. All current and former headspace clients who completed
the survey described above were given the option to volunteer for focus groups, interviews, or both.
• hMDS - comprises data collected from both young people and service providers on services delivered
through headspace services.
Table 38 Stakeholder engagement themes from young people who use headspace

Evaluation Question Themes

How effective is headspace in • Young people felt that they knew more about mental health
increasing mental health problems in general because of attending headspace.
literacy?
• Young people from culturally and linguistically diverse
backgrounds, from Aboriginal or Torres Strait Islander
backgrounds, and for young people who speak a language other
than English at home thought that headspace had increased their
mental health literacy to a similar degree.
• The extent to which young people indicated that headspace had
helped them to improve their mental health literacy steadily
increased with the number of OOS they had accessed.
• Most young people interviewed who currently use headspace
services reported their mental health literacy had improved due
to their participation in therapeutic encounters with headspace
counsellors and clinical psychologists.
• Respondents articulated they had learned about mental health,
specific concepts, obtained a diagnosis in many cases, and had
gained more insight into their own conditions, discussing these
using concepts and language derived from written material and
their therapists. Some attributed new knowledge and positive
outcomes wholly to headspace.
• A minority of respondents felt headspace had not helped them
much if at all.
• A key aspect of building mental health literacy with young people,
emphasised in interview responses from headspace users, is the
level of rapport and engagement established between the young
person and the headspace worker with whom they are
connected.
• Apart from learning more about mental health through headspace,
young people also cited that their mental health literacy
increased by accessing information from their private
psychologists and by researching and reading information
online.
• After coming to headspace, using the services, and learning coping
strategies to manage their mental health, many of the young
people in reference groups interviewed reported that they had
gained more confidence to speak about their experiences and
had then actively promoted the service to friends and more
broadly.

How effective is headspace in • 77-78 per cent of young people presenting were aged under
increasing early help seeking? 20 years (2015-16 to 2019-20).
• hMDS data in the period indicates that just under half of the young
people presenting (46.1 per cent) were in the early help seeking
category.
• headspace users relayed that young people are increasingly aware
of mental health issues, and that stigma has reduced over time.
headspace visibility and outreach meant that young people were
sometimes already aware of headspace, or were referred to
headspace early via school or through their GP.
• Waitlists were raised by some as an inconvenience and others as a
severe challenge.
• Reference group participants commented that headspace’s
promotional activities were effective, and services appeared
accessible and used ‘soft entry points’ as well as referral
pathways from GPs.
• Youth reference group participants commented that headspace’s
promotional activities in the community and outreach in schools
meant that young people could be linked to help early through
GPs and school counsellors.

How effective is headspace in • Young people using headspace were generally referred by GPs, via
increasing access to required schools, or on parental suggestion.
services?
• While most accessed services face-to-face, due to the pandemic, or
because of distance, some users preferred flexible appointments
using telephone or online platforms, such as Zoom.
• headspace services were reported as conveniently located, with
some element of discretion preferred, to avoid stigma and the
risk of people observing them accessing the service.
• Once accessing the service, users reported it to be friendly and
welcoming.
• One downside reported by a small number of young people who
had accessed headspace was a long waiting time (about a
month) between intake and assignment to a counsellor or
 psychologist.
• Young people who had accessed headspace services reported
appropriate referrals, for example to dieticians or other
specialists, however referrals to psychiatrists were difficult.
Young people noted this is not a reflection of headspace but that
psychiatrists are hard to come by in general and waiting lists for
appointments are lengthy.
• A minority of young people were not happy with their encounter
with headspace, citing other services or clinicians who helped
them more.
• Most found the help they received from headspace to be beneficial,
or that headspace led to a referral that helped more.
• Young people reported that headspace worked well with other
external service providers such as dietitians, specialists to meet
the needs of users, using a client-centred approach.
• Staff provided support in a respectful and non-judgmental manner
and worked diligently to ensure that users accessed the required
services from headspace or from external organisations.
• Confidentiality was raised as an important issue when the young
people were referred to other services from headspace.
• Young people also commented that headspace staff would
diligently try to meet the needs of users first rather than just
‘redirecting them’. However, due to the increasing complexity
of young people presenting and the bounds of the headspace
model, it was not uncommon for young people to require more
support than headspace service can provide. In these situations,
headspace staff provide continuity of support until other more
appropriate services can be put in place.

How effective is headspace in • The hMDS user satisfaction data indicates that there is no
supporting ‘hard to reach’ significant difference between the improvements in mental
groups, including those who are health literacy reported by young people who access headspace
at greater risk and less likely to from different culturally and linguistically diverse backgrounds
seek help? In increasing access or by young people who identify as LGBTQIA+, however
for hard-to-reach groups? satisfaction was significantly lower for Aboriginal and Torres
Strait Islander young people compared to the general population
of young people attending headspace.
• Youth reference group members noted that headspace actively
worked to increase mental health literacy across all groups of
young people, including those who are hard to reach. They
noted that unless young people were willing to accept help,
these groups would remain difficult to reach.
• headspace users from ‘hard-to-reach’ cohorts interviewed reported
that it took them time to decide to seek help and pointed to other
young people who were ‘hard- to-reach’ and resistant to seeking
help.
• Young people cited the importance of outreach in public spaces
and schools as one way of engaging with harder to reach people,
as well as ensuring that people are made aware it is a free
service.
• Family attitudes that downplayed distress due to mental health
issues were cited as preventing young people from seeking help,
therefore young people from culturally and linguistically
diverse backgrounds in particular thought headspace could
educate families to reduce stigma.
• Young people in the reference groups reported that headspace
successfully engaged with the LGBTQIA+ community. The
specific groups run by headspace meant that they could meet
and connect with other young people in a space where they felt
comfortable and were treated with respect.
• Young people noted that waiting times could deter hard-to-reach
clients from accessing help, especially if they had taken the
difficult step to ask for help.
• Aboriginal and Torres Strait Islander young people spoke of their
challenges around depression, drugs, and abuse and that they
sometimes did not access services due to stigma.
• Young people from culturally and linguistically diverse
backgrounds reported they would like to see more cultural
diversity among headspace staff, especially so their family
backgrounds and religious considerations could be better
understood.
• Aboriginal and Torres Strait Islander young people who had
accessed headspace services had a range of views, including
 that headspace could be more culturally competent (and include
more First Nations staff), liaise with ACCHSs, and also detailed
culturally positive practices.
• LGBTQIA+ young people had mainly positive encounters with
headspace, with some exceptions, complaints centred on the
quality of clinical support, rather than issues related to sexuality.
• Young people with disability were generally positive about
headspace, but there are limits to what headspace can do in
relation to some conditions, including Autism Spectrum
Disorder which requires specific diagnostic tools and specialist
support.
• Remote and rural residents referred to the small-town effect where
‘everyone knows everyone’s business’ and some cited stigma in
relation to seeking help, but no young people users from this
subcohort who had accessed headspace services reported any
specific issues with accessing headspace or the quality of
service.
• There were hurdles to overcome in outreach and bringing First
Nations young people into the service.
• It was difficult for the young people in the reference group to
assess whether headspace was effective in increasing access for
hard-to-reach groups with the exception of LGBTQIA+ young
people. In one area, headspace had organised a festival for
LGBTQIA+ young people. The festival aims to celebrate and
raise awareness of the LGBTQIA+ young people. A headspace
youth group for LGBTQIA+ young people in the area provided
a supportive environment for young people to meet and access
information.

How well does headspace • Most youth reference group members interviewed endorsed the
advocate for and promote youth way headspace staff actively promoted the service on social
mental health and wellbeing in media and through outreach in schools and stalls in the
their communities? community.
• Having a regular presence on social media and promotional
activities, such as leaflets and groups in schools and booths in
shopping centres, was seen by this group as increasing
awareness of headspace and mental health issues for young
people, thereby increasing mental health literacy.

To what extent has headspace • N/A

reduced stigma associated with
mental illness and help seeking
for young people, their families
and friends, and the
community?
How effective is headspace in • Young people who had accessed headspace services reported
improving pathways to care for receiving appropriate referrals to dieticians and other
young people through service professionals as well as assistance with practical matters
integration and coordination? relating to housing, income, and employment, both in-house and
via referral from headspace services.
• Referrals and links to psychiatrists were reported by some
headspace users to be more problematic, indicating they were
not able to be linked with a psychiatrist when needed from the
headspace service.
• While many started their mental health journey with headspace,
some ended up being referred to, or choosing to, seek help from
a private practice psychologist and/or psychiatrist.
• The survey showed that 85 per cent of young people indicated that
it was extremely important that headspace would connect them
with other services if they needed them.
• Similarly, 95 per cent of young people’s parents agreed it was
extremely important that headspace connect their child to other
services as required.
• Cost was a major barrier in referrals for some headspace users. For
example, one user was referred elsewhere for an expensive test.
Cost barriers were also identified as limiting access to
psychiatrists.
• In some cases, there was some frustration from headspace users
that headspace could not support them with these services, and
they were referred to a more expensive service as part of a
coordinated care model.
• headspace users also indicated that they used alternative services,
such as their GP, to provide integrated care, referrals, and care
coordination, rather than relying on headspace.
 • A small minority of headspace users interviewed also felt that
headspace did not understand what their problem was, so
referred them to the wrong type of professional.
• Young people representing Youth Reference Groups from deep
dive locations reported that staff from headspace were
‘constantly connecting with other services. They noted that
headspace and the mental health sector were trying to improve
integration and coordination between different mental health
services to facilitate the pathway for young people through the
service system, as well as with broader social supports such as
those available through Centrelink and Medicare.
• The Youth Reference Group members also reported that processes
were in place at headspace to support young people through
referrals to other services, such as private psychologists, to
support effective care coordination.
• Youth Reference Group members spoke about young people with
self-harm or suicidal thoughts and how headspace staff
organised and supported them through the referral and transition
process, for example taking them to the hospital emergency
department or contacting CAMHSs.

To what extent is headspace • N/A

providing a localised service
offering, and what are the
barriers and enablers of this?
What other contributions does • headspace users indicated they have experienced a range of
headspace make to local contributions to their local communities from their local
communities? headspace service. The most common of these were
contributions through outreach activities in communities – in
schools, at public events, and near public transport hubs.
• However, when asked what headspace could do differently,
headspace users suggested increasing their profile through
social media and more in-school presence.
• headspace also provides young people with opportunities to
contribute to governance via the Youth Reference Groups and as
mental health ambassadors.
• In one area, Youth Reference Group participants spoke about the
contribution of headspace to the community through supporting
them to organise festivals around issues of importance to them,
such as young people who identify as LGBTQIA+,
homelessness and social justice. Members of the Youth
Reference Group worked on the organising committees for
these events with support from headspace staff and other key
services in the area.

To what extent does a ‘no- • Young people consistently recognised the benefits of the nowrong-
wrong-door’ approach assist door approach and had strong positive regard for it as part of the
headspace to meet its headspace service model.
objectives?
What is the level of support for • N/A
headspace from other primary
care and mental health service
providers?
To what extent does headspace • Responses from the young person satisfaction matrix indicate that
provide culturally appropriate headspace is an appropriate and inclusive service for the general
and inclusive service for young population of young people (responses range from neutral to
people and their friends and strongly agree) and for a number of indicators. This was
families, including for particularly the case for LGBTQIA+ young people as well, with
vulnerable and diverse scores significantly higher than the general population on six
population groups and different indicators.
age groups?
• Aboriginal and Torres Strait Islander young people were
statistically less satisfied than the general population of young
people accessing headspace.
• Young people completing the survey as part of this evaluation were
asked to reflect on the service they had received over the
previous 12 months and to rate on a five-point scale, from
‘always’ to ‘never’ how they felt about five statements. Results
indicate that young people responding to this survey had
positive experiences with headspace, with a large majority
indicating ‘always’ in response to the indicator statements.
• When analysed for any differences between Aboriginal or Torres
Strait Islander young people, LGBTQIA+ young people or as
speaking a language other than English at home, survey results
were similarly high, with no significant difference between
groups.
 • In interviews and focus groups, headspace users indicated that
cultural diversity of staff was important to them (this was
mentioned most often by culturally and linguistically diverse
young people).
• Sometimes there was a gender preference, also based on cultural
considerations (for example, for a female young person to see a
female counsellor).
• Although Aboriginal and Torres Strait Islander young people
continue to access headspace services, some young people
noted services would benefit from hiring more First Nations
staff.
• Members of Youth Reference Groups noted that headspace offered
a range of supports for diverse groups. They commented that
headspace provided inclusive services, particularly for
LGBTQIA+ young people.
• There were some concerns from Youth Reference Group
participants that young people who fell outside the age ranges of
12 to 25 fell through service gaps.

To what extent does headspace • Young people were asked in the evaluation survey about their
enable young people and their experiences with headspace services over the previous 12
families to access support months. Sixtysix per cent of headspace users responding to the
where, when and how they want survey indicated that headspace services ‘always’ tried to see
it, and what are the barriers and them when they wanted.
enablers to this?
• The fewer the number of OOS the young person had, the more
likely they were to indicate an answer other than ‘always'.
• Young people who had accessed headspace services described in
focus groups and interviews that they found headspace staff
easy to talk to, nonjudgmental and relatable, and appreciate that
the people who work at headspace can be quite young but still
qualified and experienced.
• headspace users interviewed described that they accessed support
either face-to-face by going to a service, which were well-
located and near public transport, or online (mainly due to the
pandemic, or distance).
• Some users talked about wanting access to online resources while
waiting for their first appointment (or between appointments).
• Barriers included opening hours (as users aged in their 20s were
more likely to be at work during the day), being able to move to
another counsellor if they were not the right ‘match’ with the
headspace staff member, the cultural or gender characteristics of
the staff member being too different so that they could not
relate; however, the actual logistics of appointments were not a
problem for many headspace users.
• A number of interviewees had been to headspace and then later to a
clinical psychologist, and the majority much preferred the
therapeutic relationship with the psychologist in private
practice, while also acknowledging that headspace had been
useful at the time or pointed them in the right direction. A
minority felt headspace had been of very little use to them and
they were glad they had ‘moved on’.
• In interviews with Youth Reference Groups, young people noted
that accessible locations, the high recognition of the headspace
brand for example on social media, outreach activities in
schools and the community and the youth friendly approach to
providing help and advocacy contributed to enabling young
people to access services.
• Youth Reference Group participants identified several barriers to
accessing support: waiting lists, staffing shortages, and
resourcing.

To what extent do young people • The views of young people captured in the hMDS young person
participate in the design and satisfaction matrix indicate that most are very satisfied with
delivery of headspace, and how their experience of being involved in the design and delivery of
does this influence young headspace, with the majority selecting ‘strongly agree’ or
people and their families ‘agree’ for the statements.
experience of headspace?
• Reference group members in one area participated in a review of
the forms young people filled out when they first presented to
headspace. They suggested changes to the forms to ‘make it as
easy and straightforward to fill out as possible’. Making the
process simple was especially important for young people who
attended headspace alone.
• Members of another reference group helped to facilitate groups of
 likeminded people around issues they felt strongly about, to the
support young people on the headspace waiting list in their area.
The young people hoped to start a ‘climate’ group.
• While the hMDS collects satisfaction data directly from young
people, it does not survey family members participating in
family and friend focussed OOS. This makes the extent to
which including young people in design and decision making is
associated with improved service experience for families.

How is the establishment of • N/A

alternative service delivery
models assisting headspace to
meet its program outcomes?
Source: KPMG 2022

B.2.2 Young people who do not use headspace

Evaluation activities and data sources

Qualitative themes are sourced from a mix of data described below.
• Focus groups and consultation with non-headspace users to understand their perceptions of headspace,
and experiences in supporting their mental health and wellbeing.
• Survey of 1,432 young people who do not use headspace
Table 39 Stakeholder engagement themes from young people who do not use headspace

Evaluation Question Themes

How effective is headspace in • N/A
increasing mental health
literacy?
How effective is headspace in • When asked why they do not seek support from headspace, they
increasing early help seeking? responded they do not feel their need is severe enough to
warrant taking the time or resources away from those in need.

How effective is headspace in • N/A

increasing access to required
services?
How effective is headspace in • N/A
supporting ‘hard to reach’
groups, including those who are
at greater risk and less likely to
seek help? In increasing access
for hard-to-reach groups?
How well does headspace • In discussions about headspace and its role in advocating and
advocate for and promote youth promoting mental health in local communities surrounding
mental health and wellbeing in headspace services, young people were able to readily identify
their communities? occasions where they had observed a headspace presence at
community events, for example at schools and university
oweeks, as well as on social media.

To what extent has headspace • Discussions with young people from culturally and linguistically
reduced stigma associated with diverse backgrounds indicated that they felt there was limited
mental illness and help seeking understanding of the cultural sensitivities around mental health,
for young people, their families and that this was true of the headspace model as well as of
and friends, and the mainstream services more generally.
community?
How effective is headspace in • Young people who had not accessed headspace services were asked
improving pathways to care for if they had sought support from their GP for mental health. Of
young people through service the 1,432 young people who had not used headspace services,
integration and coordination? and who answered this question, 537 indicated they had sought
support from their GP.
• These young people were asked a follow up question about other
services their GP had referred them to for additional support.
Twelve per cent of these young people reported receiving a
referral from their GP to both headspace services as well as
other mental health services. Four per cent of young people
reported receiving a referral to headspace services only, while
81 per cent of young people indicated that their GP had referred
them to other services but not a headspace service.
• It is not known why these 16 per cent of young people who were
referred to headspace chose not to access the services.

To what extent is headspace • N/A

providing localised service
offering, and what are the
barriers and enablers of this?
What other contributions does • Many young people who do not use headspace described hearing
headspace make to local from headspace services through their schools.
communities?
• Non-headspace users at university indicated that, while they
recalled headspace services visiting their school, these types of
community engagement activities were more limited through
their university.
• A small minority of non-headspace users indicated that the impact
of these community engagement activities on them depended on
who was running school-based sessions.
• Where the representative was a young person and easier to identify
with for students, non-headspace users described this as being
more effective in promoting headspace services and mental
health wellbeing, than with other headspace service staff with
whom young people did not identify.
• A small number of non-headspace users also identified youth
ambassadors for headspace services as an element of their
community engagement and outreach activities, and that this
supported great awareness of services and other outcomes, such
as early help seeking and reducing stigma around seeking
support.

To what extent does a ‘no- • N/A

wrong-door’ approach assist
headspace to meet its
objectives?
What is the level of support for • As outlined above in ‘How effective is headspace in improving
headspace from other primary pathways to care for young people through service integration
care and mental health service and coordination’, young people who completed the young
providers? people’s survey were asked about other services they may have
accessed to support their mental health. Twelve per cent of
young people who had not visited a headspace service, but
sought help elsewhere, reported receiving a referral from their
GP to both headspace services as well as other mental health
services. Four per cent of these young people reported receiving
a referral to headspace services only, while 81 per cent of young
people indicated that their GP had referred them to other
services but not a headspace service.
• Similar to the experience of headspace clients, the level of support
for headspace services from other parts of the service system
varies between individual services.

To what extent does headspace • Interviews and focus groups found there is recognition amongst
provide culturally appropriate nonheadspace users that headspace services appear to cater well
and inclusive service for young to the LGBTQIA+ young people in the community and have
people and their friends and knowledge of issues affecting these young people.
families, including for
vulnerable and diverse • There was also some indication from young people who had not
population groups and different accessed headspace services from culturally and linguistically
age groups? diverse backgrounds that they would consider using headspace
services, as they are able to assess the service without parental
consent, especially where they encounter cultural stigma related
to mental health support.
• A key caveat was the importance of appropriate staff members, for
example that young Muslim women need a female worker, and
that the mix of headspace service staff may not always provide
the right support.
• Aboriginal and Torres Strait Islander young people who do not use
headspace indicated that they thought there was some variation
in the appropriateness of services between locations.
• Neurodivergent young people who have not used headspace
indicated in interviews and focus groups that they did not
necessarily identify with the service. The neurodiverse flag is
not present, and their interactions with headspace staff did
improve their level of trust in the service.
• There was also some indication from non-users that they identified
with the brand more when they were younger (high school age),
with this dropping off as they got older.
• Amongst non-headspace users, there was also very inconsistent
understanding of what age groups were eligible for support
from headspace services.

To what extent does headspace • Feedback from non-headspace users indicated that opening hours
enable young people and their predominantly in business hours did not support young people
families to access support with full time study and workloads to access services.
where, when and how they want
it, and what are the barriers and • There was positive feedback, through interviews and focus groups,
enablers to this? from non-headspace users who have accessed website resources
from headspace.
• Some non-headspace users recognised that headspace also has
online and telephone counselling services through eheadspace
for those who cannot attend a service in person. They saw these
examples of telehealth services as important for those who
cannot attend a physical service.
• Non-users of headspace also discussed the location and
accessibility of the physical headspace centres near them. Many
non-headspace users knew where their local service was located
but highlighted that this was sometimes not accessible from
local communities due to travel durations and lack of public
transport. In these discussions, young people identified greater
flexibility for outreach services as being potentially beneficial.
• There was some hesitancy from the group around using the service
from the public setting of a service, as they did not want to be
seen walking through the door. These young people thought the
presentation of the building would draw unwanted attention,
and in small communities, young people were concerned about
their privacy.
• In contrast, other non-users spoke positively about the bright and
vibrant brand of the headspace service and thought this looked
welcoming and inviting.
• When discussing accessibility of headspace, non-users also
highlighted that they thought that providing services without
cost was an important benefit of the headspace model.

To what extent do young people • N/A

participate in the design and
delivery of headspace, and how
does this influence young
people and their families
experience of headspace?
How is the establishment of • N/A
alternative service delivery
models assisting headspace to
meet its program outcomes?
Source: KPMG 2022

B.2.3 headspace service providers

Evaluation activities and data sources

Qualitative themes are sourced from a mix of data described below. Fifty-two headspace providers
contributed.
• Consultations with ‘deep dive’ locations – detailed consultation with a cross section of six headspace
services nationally to explore what services headspace offers, and the contributions of services to their
local community.
• Survey of headspace service and lead agency representatives – testing key evaluation questions, as well
as whether barriers, enablers and other factors raised by stakeholders within the deep dive site
consultations described above, were also experienced by other services.
Table 40 Stakeholder engagement themes from headspace service providers

Evaluation Question Themes

How effective is headspace in • Ninety-three per cent of surveyed service and lead agency staff
increasing mental health working within the headspace model have generally high levels
literacy? of confidence that the services they provide lead to increases in
mental health literacy for young people.
• Staff indicated that broader community engagement by the
headspace service, including through social media campaigns,
education and awareness activities with local schools, and
through establishing partnerships with local councils,
universities and colleges improved mental health literacy.
• Online and printed resources are provided by services to support
mental health literacy. These are seen to be frequently accessed
by young people and their families and include material on
various services available in the local area, how to make
informed decisions about referral pathways, and how young
people can support their own mental health.
• A case management component of work undertaken with young
people and families, upskilling them on mental health support
options, and capacity building strategies improved mental
health literacy.
• headspace services, such as psychoeducation, are provided as part
 of clinical services and group work services that focus on
mental health literacy, capacity building, and accessing support
for young people.
• Safety planning and information on available supports are
provided at the intake and assessment stage with young people
accessing headspace services.
• Staff identified access issues as a key barrier to improving mental
health literacy for young people. This was raised in terms of
strong demand pressures with young people waiting to access
the service, as well as in terms of access issues caused for the
service by COVID19, which introduced barriers to engagement
between services and schools and other community
organisations.

How effective is headspace in • Eighty-seven per cent of service and lead agency respondents
increasing early help seeking? indicated that staff working within the headspace model have
generally high levels of confidence that the services they
provide lead to increases in early help seeking behaviour.
• Staff identified that strong brand recognition and social media
presence of headspace services and promotion and advocacy
work of services, including the community engagement roles
like school events, contributed to increased early help seeking.
• In terms of barriers to increasing early help seeking, these are
similar to those identified for improving mental health literacy,
such as the impact of waiting times constraining the extent to
which services can provide early support and early referrals to
other services.
• Staff also saw community engagement as a key mechanism
through which services promote early help seeking, which is
limited due to staffing challenges and funding constraints.
• The potential to improve the service’s contribution to early help
seeking through additional intake engagement workers,
supporting early intervention and low-level needs early was
highlighted.
• Perceptions that headspace services are supporting high-risk or
highneeds young people, discouraging others from seeking
support for mild to moderate needs was seen as a barrier.
• Staff noted pressure on services from supporting higher needs
young people as reducing the capacity to provide early
intervention support to those who seek help early.
• In line with this, school and university counsellors interviewed
frequently raised the issue of waiting times, and anticipated
delays in receiving support as a reason young people do not
seek support from headspace, constraining the extent to which
headspace can provide an early intervention service for young
people.

How effective is headspace in • A large majority of staff from services and lead agencies indicated
increasing access to required by survey that waiting lists (83 per cent) and workforce
services? attraction and retention (76 per cent) are the key barriers to
supporting increased access to their headspace service.
• Many staff indicated that the funding model such as insufficient
funding for salaried staff, again including community
engagement staff, acted as barriers in enabling access to
required services for young people.
• Others noted the difficulties in being able to afford an accessible
site.
• The impact of the pandemic was also noted as impacting the
ability for young people to access their services.
• Representatives interviewed across all deep dive sites shared
anecdotally that most young people accessing headspace
selfreferred into their services, and the ability to do this ensured
headspace provided a ‘soft entry’ into mental health support,
without the need for formal referral through GPs or other
avenues.
• The credibility and power of the headspace brand was noted as a
key strength across services in encouraging young people to
proactively access services.
• For some services, the use of satellite sites has allowed young
people to conveniently access headspace in their local area, and
services offering after-hours access was highlighted as
important to enable young people to access services outside
school and work.
 • Stakeholders also discussed the topics of ease of location and
being close to transport as key aspects of the model which
support access to headspace services.
• Again, waiting times for support through services was frequently
raised in interviews with stakeholders in deep dive sites as a
barrier to access for young people.

How effective is headspace in • Responses from the survey of service and lead agency staff
supporting ‘hard to reach’ indicate that most staff surveyed see the headspace model as
groups, including those who are less effective in meeting these objectives for young people from
at greater risk and less likely to ‘hard to reach’ groups.
seek help?
In increasing access for hard-to- • Centre and lead agency staff who responded ‘yes’ regarding seeing
reach groups? differences in outcomes for young people from these groups
compared with the general population of young people were
then given the option to rate the difference between groups on a
sliding scale. Averaged results from staff indicate that
Aboriginal and Torres Strait Islander young people, culturally
and linguistically diverse young people and young people with
disability all fare below the general population of young people
attending headspace in terms of the service's impact on their
mental health literacy.
• Results from staff also indicate that engagement with LGBTQIA+
young people result in better mental health literacy than for
other groups of young people.
• Relevant to meeting the needs of Aboriginal and Torres Strait
Islander young people, the staff survey also highlighted that, in
regional areas with high Aboriginal and Torres Strait Islander
populations, specific Aboriginal Social Emotional Wellbeing
Workers are important. Staff in these roles support adaptation
of presentations and other resources for Aboriginal and Torres
Strait Islander young people. While these roles were
highlighted as contributing strongly to improved mental health
literacy for young people accessing headspace, respondents to
the service and lead agency survey also noted difficulty
recruiting staff for these roles in small communities.
• The need to be able to provide services for community in
community was also emphasised as a challenge for the
headspace model.
• For young people with disability, service providers noted that they
had limited referral pathways with disability services, and that
disability service providers do not refer young people into
headspace unless it is funded on their NDIS plan, which is rare.
• The overall perception is that young people with disability access
other services rather than headspace, and some providers
indicated this is more appropriate due to headspace clinicians
not having experience working with dual diagnoses.
• For young people from culturally and linguistically diverse
backgrounds, headspace service providers described the
barriers in having access to culturally and linguistically diverse
staff with the capability to work with those arriving with
significant trauma, with multicultural mental health issues, and
with different language and cultural skills. They also spoke
about limitations in the capacity to undertake outreach to
culturally and linguistically diverse communities to promote
service access.
• School and university counsellors agreed with the importance of
having local workers who identify as Aboriginal or Torres
Strait Islander to support outcomes for Aboriginal and Torres
Strait young people. Communities with large Aboriginal and
Torres Strait Islander populations have benefited from outreach
and work designing the services with the community. This has
built trust that, in turn, supports engagement with headspace by
young people and improved mental health literacy.
• Focus groups with counsellors also identified challenges for
culturally and linguistically diverse communities, including
international students who have not had the same education
around mental health throughout earlier schooling as other
young people from the general population.
• Responses from staff at headspace services and lead agencies
indicated that the majority thought the headspace model was
less effective in encouraging early help seeking for Aboriginal
and Torres Strait Islander young people, culturally and
linguistically diverse young people and young people with
disability.
• As with mental health literacy, respondents felt that the outcomes
were stronger for LGBTQIA+ young people than for those
from the general population of young people attending
headspace.
• Respondents had similar views about increased access for the
‘hard to reach’ cohorts as they had for increased mental health
literacy and improved early help seeking. Young people from
Aboriginal and Torres Strait Islander backgrounds, culturally
and linguistically diverse cohorts and young people with
disability were seen to have worse access rates to headspace
services compared with those from the general population of
young people attending headspace. LGBTQIA+ young people
were perceived to have better rates of access than all other
groups, including the general population of young people
attending headspace.
• In response to a prompt in the service and leady agency survey to
describe the barriers and enablers to support these cohorts, a
common theme was related to challenges for rural and remote
services. Issues with the other parts of the service system were
raised for regional areas with limited capacity of tertiary
services, bulk billing services and affordable psychiatry. The
ability to attract specialist psychologists, AOD workers,
vocational workers and GPs were all identified as difficult in
remote areas. Turnover and a limited overall pool of workers
across providers and PHN roles was also highlighted.
• One respondent also described challenges they face in a regional
area with the headspace service funding model, where outreach
activities to take services to remote communities are not funded
but are expected by stakeholders across their local area.
• Other responses again highlighted a concern about insufficient
funding for salaried staff, including community engagement of
staff. Challenges around finding staff with the right skillset or
cultural background were particularly salient for regional and
remote staff.
• Deep dive consultations and discussions with Aboriginal and
Torres Strait Islander community organisations illustrated key
themes required to engage and assist Aboriginal and Torres
Strait Islander young people. Stakeholders emphasised the need
for young people to be able to see people like themselves in the
staff at their local headspace, and for it to feel like a safe and
culturally appropriate place for them to seek help.
• In consultations with metropolitan services, stakeholders reported
the importance of having members of staff from a wide range
of cultural backgrounds, and of the important role they play in
reducing stigma and building mental health literacy for
different communities.
• The headspace model promotes centre-based support and
emphasises the importance of making each service look
culturally appropriate and welcoming to members of the
Aboriginal and Torres Strait Islander community. Services
display Aboriginal and Torres Strait Islander flags, and draw on
local culture, art and language to show visible signs of welcome
to the local Indigenous community.
• Services in areas of high Aboriginal and Torres Strait Islander
population also prioritise having staff from the local Indigenous
community, and engagement with Elders and well-known local
Aboriginal and Torres Strait Islander people on the consortium
in an advisory capacity, so they can be seen to be endorsing the
use of the headspace services for their people.
• Stakeholders in regional and remote areas described the
centrebased model as a barrier to Aboriginal and Torres Strait
Islander young people seeking support, due to high levels of
selfconsciousness and stigma associated with mental illness.
• Indigenous models of care, centring the person within their family,
community and culture, were also described as more effective
in assisting Aboriginal and Torres Strait Islander young people
than a more individual-centric model provided in mainstream
clinical practice.
• Where outreach is conducted to Aboriginal and Torres Strait
Islander communities, by trusted service providers without the
need for appointments or to be seen to be seeking help, barriers
may be reduced, and positive outcomes supported.
• For young people who identify as sexuality or gender diverse,
headspace has become a brand which provides a safe space for
them to seek support, connect with peers and manage their
 wellbeing. Stakeholders consistently recognised this as a
strength of headspace.
• Consultations indicated that headspace had achieved this success
in improving access of this group through its brand recognition,
social media presence and through peer-to-peer networking.

How well does headspace • Service and lead agency survey responses indicated strong levels
advocate for and promote youth of confidence from staff that their service is successful in
mental health and wellbeing in increasing mental health literacy. When asked to describe key
their communities? enablers of this, responses identified broader community
engagement by the headspace service as a key aspect of their
observed success in this area. Examples included activities such
as social media campaigns, education and awareness activities
with local schools, and the establishment of partnerships with
local councils, universities, and colleges.
• Community Development Officers were highlighted as
particularly critical to this work, however some services
identified only having funding for 0.6 FTE for this role, which
they consider to be insufficient.

To what extent has headspace • In response to the survey, 93 per cent of service and lead agency
reduced stigma associated with respondents consider their headspace service to be reducing
mental illness and help seeking stigma.
for young people, their families
and friends, and the community? • When considering the extent to which headspace has been
successful in reducing stigma for family, friends and the
community, it is less clear from the data. Qualitative evidence
from interviews and discussions at a range of sites indicated
that, while some success is being made in reducing stigma in
young people, this is due to a range of factors including the
work of schools and the media more broadly in highlighting
and normalising mental health help seeking.

How effective is headspace in • There were consistent views from stakeholders across deep dive
improving pathways to care for locations that headspace services undertake a range of activities
young people through service to support integration with other services and coordination of
integration and coordination? care for young people. These include case coordination for
young people, establishment of relationships with other local
services, such as NDIS access workers, cultural healing
services, and other family-based supports, and direct referrals
to other services.
• These stakeholders also indicated how this work was an ongoing
and important aspect of ensuring access to services for young
people.
• Case coordination work was consistently raised by deep dive
representatives as critical to the success of the headspace model
in supporting service integration and better outcomes for young
people.
• Services invest time in building relationships with other local
services, including local mental health services, and other
support services that contribute to aspects of a young person’s
wellbeing.
• The level of investment in these relationships differs between
services and depends on the capacity of other services to
engage, loss of relationships when other organisations lose
time-limited grant funding, and the focus of the management of
individual headspace services on this relationship building
versus other elements of service delivery.
• Relationships and resulting service integration with psychosocial
supports, including cultural healing, NDIS access, and family
supports, enables headspace to facilitate cross-referrals.
• There were differences reported by deep dive stakeholders
between metropolitan and regional and remote services with
respect to service integration. The availability of other services,
and their capacity, particularly in non-metropolitan locations,
has impacted the ability of some headspace services to support
integration.
• Where services do not have capacity to take on new clients, this
impacted referrals made by headspace services, and
opportunities for care coordination and service integration for
young people.
• The most common barriers identified were waitlists and lack of
capacity in local referral services, followed by limited local
services for specific conditions or treatment needs, and lack of
local services to meet more acute needs.
• With respect to case coordination in particular, deep dive
 representatives also described challenges in documenting and
demonstrating the volume of time spent on coordination
activities and balancing these activities with direct clinical
services for workers within headspace services, especially
where the headspace services rely on MBS billing to support
services.
• Case coordination is also more challenging for young people with
more severe distress levels and complex mental health support
needs.

To what extent is headspace • There are a multitude of examples of how services have been
providing localised service tailored to the needs of the local community. Representatives
offering, and what are the from deep dive locations demonstrated a strong level of
barriers and enablers of this? community engagement and awareness enabled by the
consortium arrangements and a local workforce with local
networks to support this.
• Services tailored to their communities include: introduction of a
bushfire recovery role to tackle climate-related anxiety with
young people, increased focus on outreach services where there
is increased need, for example in remote Aboriginal
communities, or neighbouring communities impacted by
bushfires and the introduction of new consortium partnerships
with additional local services, responding to particular stressors
for young people in the local community, such as domestic and
family violence and family wellbeing services.
• Many services are well-integrated into their local communities and
provide services in demand with local community. Community
engagement activities assist headspace services to identify how
best to respond to local need, and some lead agencies have a
specific focus on supporting these activities by also applying
for additional grant funding from alternative sources to support
this work.
• Services and lead agencies through deep dive discussions and the
service and leady agency survey consistently identified that
there is limited capacity for outreach and community
engagement activities within services, to identify local needs
and tailor services, and reach those in local communities who
may not use the centre model.
• Community engagement positions are sometimes part-time roles
based on available funding. Some headspace services do not
have dedicated community engagement positions, and
community engagement is often de-prioritised due to clinical
service loads within services.
• Services and lead agencies also indicated they often have trouble
recruiting specific workers to meet the needs of the local
community. These may be for specific professional positions or
positions related to a specific cohort of young people, such as
Aboriginal wellbeing workers or workers with culturally and
linguistically diverse backgrounds.
• headspace service staff and lead agencies indicated there are some
challenges in localising services where there is increasing
complexity and severity in the presenting needs of young
people. These young people are not the focus of the headspace
model, and tailoring services to meet their needs is difficult.
• There is increased pressure on service capacity from young people
with more intensive needs which impacts on capacity to focus
on tailored offerings.

What other contributions does • Service and lead agency representatives were asked to indicate
headspace make to local what types of services their service provides to young people
communities? and the community more broadly. Of the 69 respondents who
answered this question, 58 (or 84 per cent) indicated that their
services work with local schools and community groups, while
44 (or 64 per cent) indicated they provide outreach services to
local communities.
• Deep dive site representatives, as well as survey respondents,
indicated that community engagement such as this is a critical
and successful part of the headspace model, however, it is an
onerous obligation, and is often not able to be adequately
resourced within current funding for headspace services.
• Engagement with schools and universities includes a range of
activities, including presentations to schools on supporting their
mental health and wellbeing, where young people can find
resources to support their mental health, information regarding
services available, and participation in open days and fair days
 in universities.
• Outreach services provided also differed significantly between
headspace services, often linked to preferences and needs of the
local community.

To what extent does a ‘no- • There was significant support for headspace’s ‘no-wrong-door’
wrong-door’ approach assist approach to supporting young people. The approach supports
headspace to meet its young people by: ensuring they are able to engage with mental
objectives? health supports in a way they feel comfortable, providing a free
entry point into the mental health service system, providing a
soft entry into the mental health service system, with referrals
to other services available to support service integration for
young people and providing them with access to initial services
to support broader objectives, such as improved mental health
literacy and early help seeking, even where they may be
referred to a more appropriate service.
• Service and lead agency stakeholders across deep dive services, as
well as those responding to the survey of headspace services,
consistently indicated that, anecdotally, young people’s mental
health needs are becoming increasingly severe and more
complex, with many cases being outside of the headspace
model’s mild to moderate criteria.
• headspace service staff interviewed commonly described a
“missing middle” of clients who are too complex to be seen
under the headspace model’s mild to moderate remit, but are
not unwell enough to be transitioned to overwhelmed TMHSs.
• There were also consistent views from services and lead agencies
that there is significant demand placed on services by the ‘no
wrong door’ approach. While this is largely regarded as
essential to ensure young people presenting with high risk,
distress, need, or acuity are not turned away without assistance,
the value of this element of the model is particularly high where
tertiary mental health services are unable to meet demand for
higher needs young people. Rural and remote areas highlighted
this as an issue.
• Commonly in smaller regional and remote areas, where there are
limited private practices and TMHSs, local services will
redirect a young person back to headspace services to
counteract their own wait times. This has resulted in headspace
services in these circumstances taking on these young people to
ensure they receive some form of support and needing to
provide intensive case management and crisis support services.
• Another reported effect of the ‘no wrong door’ approach, coupled
with the high visibility and brand recognition of headspace, is
that services spend a proportion of time fielding general
enquiries from and about the local service sector.
• Stakeholders also described that the combined impact of these
flowon effects of the ‘no wrong door’ approach are to increase
the waiting times for young people with mild to moderate
conditions with lower risk profiles to access services. Wait
times have reportedly increased over time for some headspace
services. However, data capture for wait times has only recently
commenced, and longer-term trends in wait times are not able
to be determined.
• The no-door-wrong approach supports headspace to reach young
people and support mental health literacy, early help seeking
and access to services. It also supports young people to get help
when they need it, regardless of the severity of their mental
health problem.
• The no-wrong-door approach, coupled with other challenges in the
service system, such as referral services with limited or no
capacity for new referrals, significantly impacts headspace’s
core business of supporting young people with mild to
moderate, high-prevalence mental health conditions and other
contributions to communities through outreach and
engagement.

What is the level of support for • N/A

headspace from other primary
care and mental health service
providers?
To what extent does headspace • N/A
provide culturally appropriate
and inclusive service for young
people and their friends and
families, including for
vulnerable and diverse
population groups and different
age groups?
To what extent does headspace • Survey responses from 54 per cent of service and lead agency staff
enable young people and their indicate that most people working within headspace services
families to access support where, believe that their service provides services that are youth
when and how they want it, and friendly, appropriate, and accessible.
what are the barriers and
enablers to this? • When asked to describe enablers and barriers to their service
providing youth friendly, appropriate, and accessible services,
respondents provided a range of responses. Some identified the
youthful, friendly and welcoming service design as a key
enabler, others that the physical site is important, needing to be
accessible for young people, and large enough to support
engaging private providers.
• The flexible service model with service-based and some outreach
services, as well as having multiple referral pathways, strong
staff knowledge and relationships with the local service system
and a ‘no wrong door’ approach that aim to meet the needs of
young people, were also highlighted.
• The role of Youth Reference Groups in service design was also
identified as a key enabler, with services designed by young
people for young people and including youth friendly
approaches such as ‘walk and talk’ sessions, sessions held
outdoors and experiential learning approaches. Similarly,
services noted that they try to employ younger staff to help
make the service more ‘youth friendly’.
• Many of the barriers service providers described in response to
this question are related to the key enablers, highlighting the
ongoing challenges they face with limited referral pathways in
some communities, waitlists for tertiary mental health services
where young people have more complex or acute needs, and
headspace service waitlists impacting accessibility as well. In
regional areas, the distance between towns and the lack of
public transport were also raised as barriers to access.
• Providers felt that their headspace service was less able to support
access rates of young people with disability, young people from
culturally and linguistically diverse backgrounds and
Aboriginal and Torres Strait Islander young people.

To what extent do young people • N/A

participate in the design and
delivery of headspace, and how
does this influence young people
and their families experience of
headspace?

How is the establishment of • Of the six responses received to the headspace service and lead
alternative service delivery agency survey from satellite or outreach service respondents,
models assisting headspace to there were no discernible differences in responses received to
meet its program outcomes? enablers and barriers identified, or how well these services are
able to support headspace’s objectives.
• These respondents indicated similar challenges in recruiting
appropriate staff, managing wait times for young people, and
challenges with perceived complexity of presenting need.
• One satellite service respondent indicated that the small funding
amount received by headspace satellites meant they were only
able to employ a single clinician, and for this service, this
contributed to wait times.
• Deep dive stakeholders linked to satellite services either directly
or as a parent centre recognised the value of the work they were
undertaking and the contribution headspace, in any form,
makes to communities. However, these stakeholders also
indicated that the level of need in their local community
warranted a headspace centre, and that being able to implement
the full headspace model would make the most difference for
young people locally.
Source: KPMG 2022

B.2.4 Primary Health Networks

Evaluation activities and data sources

Qualitative themes are sourced from a mix of data described below. Twenty PHNs contributed.
• Consultations with ‘deep dive’ locations – detailed consultation with a cross section of PHNs nationally.
• Workshops with PHN representatives – testing key evaluation questions, as well as whether barriers,
enablers and other factors raised by stakeholders within the deep dive site consultations were also
experienced by other services.
Table 41 Stakeholder engagement themes from Primary Health Networks

Evaluation Question Themes

How effective is headspace in • N/A
increasing mental health
literacy?
How effective is headspace in • N/A
increasing early help seeking?
How effective is headspace in • N/A
increasing access to required
services?
How effective is headspace in • N/A
supporting ‘hard to reach’
groups, including those who are
at greater risk and less likely to
seek help? In increasing access
for hard-to-reach groups?
How well does headspace • N/A
advocate for and promote youth
mental health and wellbeing in
their communities?
To what extent has headspace • N/A
reduced stigma associated with
mental illness and help seeking
for young people, their families
and friends, and the community?
How effective is headspace in • PHN representatives attending an evaluation data collection
improving pathways to care for workshop were asked to rate how well-established headspace
young people through service service pathways are with primary care and mental health
integration and coordination? services, on a five-point scale from ‘not established’ to ‘well
established’. Pathways with GPs were rated in the middle
between not established and well established (3 out of 5).
Pathways with state and territory mental health programs, such
as CAMHSs and CYMHSs, were rated closer to not established
than well established (2.8 out of 5). Pathways with other mental
health services were rated closer to well established than not
established (3.3 out of 5).
• PHNs outlined two key enablers for headspace services in support
of service integration and care coordination – formal
agreements with services and relationship building. Most PHNs
indicated headspace effectively supports where there are strong
Memoranda of Understanding or Service Level Agreements
with external organisations, to make clear agreed protocols and
roles and responsibilities in place between services. Similarly,
relationships were a key enabler identified by most PHNs to
support pathways to care and service integration, especially in
regional and remote locations where there are fewer services
available to support young people.
• PHNs also identified a range of challenges which impact on the
ability of headspace services to support integration and care
coordination in improving pathways to care, and which are
often outside the control of headspace services. These include:
relationships between headspace services and tertiary mental
health services, such as CAMHSs, are impacted by limited
capacity within tertiary services to engage in these activities
with significant clinical work and wait lists. The capacity of
other services impacts service integration, even where strong
relationships exist, as young people may not be able to access
the service at all, preventing integration and care coordination
from occurring. Limited infrastructure to support shared
records between services reduce the level of care coordination a
young person may receive. Inconsistent eligibility criteria
across other services and significant gaps in where eligibility
criteria for tiers of the service system end, especially with
tertiary mental health services, impact when referrals can be
made from different headspace services.
• PHNs also highlighted the challenge for headspace services in
managing care coordination and service integration activities,
within existing funding limits, and with workforce challenges
within services. These activities meant staff are taken away
from clinical supports. Particular challenges were described in
engaging with local GP services. Wait times have impacted on
the relationships held with GPs at some headspace services, and
 difficulties working with some GPs impact the level of
horizontal integration with physical and sexual health services
over and above the small volume of physical health services
provided within services.

To what extent is headspace • PHNs and deep dive representatives identified the consortium
providing localised service model and use of Youth Reference Groups were key to
offering, and what are the localising service offerings. Consortium members operating in
barriers and enablers of this? local communities have deep insight into challenges faced by
young people, and what services may be required to support
these.
• Some PHNs indicated that the commissioning process for services
allows consideration of local need to be built into lead agency
selection, with specific local considerations part of the selection
process. This view was not shared by all PHNs.
• A small number of PHNs indicated that issues recruiting specific
workers has resulted in some services focusing on employing
any available workers, with less focus on the types of staff
required to meet local need. Competition with other providers
for workforce reduces local collaboration.
• Some PHNs also indicated challenges as the local commissioning
agency for headspace services in tailoring services to the needs
of the local community, while ensuring services still meet the
requirements of the headspace model integrity framework.
• These PHNs also indicated that there is no flexibility to use
funding provided for a headspace service to design localised
services which directly address the specific needs of the
community. While some tailoring is afforded through
headspace services, this does not allow the PHN to commission
a tailored service targeted at local need.

What other contributions does • N/A

headspace make to local
communities?
To what extent does a ‘no- • There were consistent views from PHNs that there is significant
wrong-door’ approach assist demand placed on services by the ‘no wrong door’ approach.
headspace to meet its While this is largely regarded as essential to ensure young
objectives? people presenting with high risk, distress, need or acuity are not
turned away without assistance, the value of this element of the
model is particularly high where tertiary mental health services
are unable to meet demand for higher needs young people.

What is the level of support for • Some PHNs acknowledged challenges for local headspace
headspace from other primary services to engage with, and receive support from, local GPs.
care and mental health service
providers?
To what extent does headspace • N/A
provide culturally appropriate
and inclusive service for young
people and their friends and
families, including for
vulnerable and diverse
population groups and different
age groups?
To what extent does headspace
enable young people and their
families to access support where, • N/A
when and how they want it, and
what are the barriers and
enablers to this?
To what extent do young people • N/A
participate in the design and
delivery of headspace, and how
does this influence young people
and their families experience of
headspace?
How is the establishment of • PHNs reported that there are mixed views from across
alternative service delivery stakeholders involved in delivering or working with headspace
models assisting headspace to services as to the impact of satellite services. There is
meet its program outcomes? significant positive regard for headspace services, and
communities and stakeholders view any headspace services as a
positive addition to achieving core objectives.
• PHNs as commissioners of services indicated a preference for
headspace centres to better meet the needs of local young
people through the holistic headspace model.
Source: KPMG 2022

B.2.5 Other Service Providers

Evaluation activities and data sources

Qualitative themes are sourced from a mix of data described below.
• Consultations with ‘deep dive’ locations – detailed consultation with:
o Indigenous Organisations (six contributed)
o Tertiary Mental Health Services (14 contributed)
o General Practitioners (five contributed)
o Secondary School and University Counsellors.

Table 42 Stakeholder engagement themes from other service providers

Evaluation Question Themes

How effective is headspace in • There was general recognition from service providers that mental
increasing mental health health literacy has improved over time and that the stigma
literacy? associated with mental health has reduced. They also agreed
that encouragement to seek help early has also increased. They
noted that this change was unlikely to be attributed solely to
headspace, but a product of ongoing work in schools, on social
media, and by other organisations as well.
• Counsellors agreed that headspace resources were effective in
increasing mental health literacy as well as encouraging young
people on how to seek help.
• Service providers identified that challenges persist for culturally
and linguistically diverse communities – including international
students who may not have received education on mental health
through their earlier schooling.
• Service providers from communities with large Aboriginal and
Torres Strait Islanders populations reported outreach activities
and work designing the services with the community has built
trust that in turn supports mental health literacy.

How effective is headspace in • Providers noted that school pastoral care teams have had an impact
increasing early help seeking? on early help seeking behaviour, for example by encouraging
younger students to get help.

How effective is headspace in • N/A

increasing access to required
services?
How effective is headspace in • N/A
supporting ‘hard to reach’
groups, including those who are
at greater risk and less likely to
seek help? In increasing access
for hard-to-reach groups?
How well does headspace • Consultations with GPs and consortium members from
advocate for and promote youth surrounding community services as part of the fieldwork for
mental health and wellbeing in this evaluation elicited broadly positive views about the work
their communities? headspace staff undertake to engage with schools and to drive
and participate in community events and mental health
awareness raising activities.
• Fieldwork teams observed services delivering on a detailed
calendar of events and activities regarding mental health and
wellbeing, actively participating in pre-existing events as well
as driving the planning and implementation of specific events of
their own, for example around headspace week.
• Services also described targeted outreach to different segments and
cohorts in their local communities, for example engaging with
church youth groups and with Police Community Youth Centre
(PCYC) programs for young people.

To what extent has headspace • Interviews with school and university counsellors indicated a
reduced stigma associated with general recognition that mental health literacy has improved
mental illness and help seeking over time for young people in Australia, that stigma about
for young people, their families mental illness has been reduced and help seeking is widely
and friends, and the encouraged, with a tendency to talk more openly about mental
community? health today.
• There was a view from participating counsellors that headspace
resources contribute to increasing mental health literacy,
 including a general improvement in young people’s knowledge
of how to seek help for their mental health and wellbeing.
Providers acknowledged that these observed changes could not
be attributed to headspace alone, but also to broader work
happening in schools, social media, and other organisations as
well.
• School and university counsellors also identified challenges for
culturally and linguistically diverse communities related to
stigma. Discussions noted that, within some cultural groups,
stigma has an ongoing impact on menta health help seeking
behaviour.
• Service providers indicated that, for some families and in some
segments of the community, stigma around mental health help
seeking continues to be strong, and services are continuing to
focus efforts, including outreach, recruitment, and other
engagement strategies, to reduce stigma and encourage support
of mental health help seeking.
• Several cultural groups were discussed in fieldwork conversations,
along with the challenges for young people from some
culturally and linguistically diverse backgrounds where mental
illness is not easily accepted or understood.

How effective is headspace in • Schools and university counsellors from across Australia indicated
improving pathways to care for that relationships and referral pathways between their services
young people through service and other external services within the community were critical
integration and coordination? to support effective outcomes, and that headspace played a role
in this.
• Ninety-eight per cent of school principals and wellbeing
coordinators indicated that being able to connect students to
other services if they need them was an important part of the
headspace model in previous research undertaken by Colmar
Brunton for headspace National.
• When asked whether headspace has improved service integration,
a lower proportion of principals and wellbeing coordinators
indicated their support. Sixty-nine per cent agreed that
headspace services strengthened relationships between service
providers and schools, and 67 per cent agreed that headspace
services improved the coordination of local services.
• There was mixed feedback from counsellors as part of focus
groups completed specifically for this evaluation, in particular
university counsellors, regarding the referral process for
headspace services. Following a referral, some counsellors
described there being limited communication regarding what
support the young person was receiving, especially while on a
wait list for headspace services, and whether the young person
would benefit from ongoing support from the school or
university while waiting for headspace support.
• Service providers indicated the referral process was ‘smooth and
easy to use’, especially where the counsellor was engaging with
headspace directly to support the young person’s access to the
service.
• Service providers also identified challenges with service
integration and care coordination for young people in the
‘missing middle’.
• Counsellors were uncertain about how to support young people
who did not have a severe enough mental health problem for
local CAMHSs or CYMHSs, but who were not within the mild-
moderate target group of headspace services.
• A small minority of counsellors indicated there was limited
communication regarding where else a young person might be
referred if the headspace service indicated it could not support
the young person.
• Some counsellors also discussed the challenge of current wait
times within headspace services as a deterrent to referrals,
especially where there was limited information provided back
to the school or university about what other support was
available to the young person during their wait for headspace
services.

To what extent is headspace • N/A

providing localised service
offering, and what are the
barriers and enablers of this?
What other contributions does • School and university counsellors consistently described the types
headspace make to local of community engagement and outreach activities of headspace
communities? services. Some identified where headspace had visited their
local school or university to provide information and resources
for young people. However, the exact nature and frequency of
these activities varied between local communities. Some
headspace services have delivered more of these activities to
their local communities than others, and the reach of these
activities also varied. For example, in regional areas, the focus
was stronger on community engagement and outreach within
the immediate area around the service, with other surrounding
communities less of a focus from the perspective of counsellors.
• Some school and university counsellors also indicated they use
online headspace resources to support their own work,
including as part of their practice, or to refer young people to, to
support their mental health and wellbeing. This was recognised
as a key strength of headspace, and a key contributor to
communities.
• headspace engages with schools and universities by giving
presentations, sharing resources, running sessions in schools,
participating in open days, and having a presence at fair days.

To what extent does a ‘no- • N/A

wrong-door’ approach assist
headspace to meet its
objectives?
What is the level of support for • In consultation with a small sample of GPs, there was good
headspace from other primary understanding of what headspace services delivered at a high
care and mental health service level, and acknowledgement of the work headspace does to
providers? support early intervention and young people with mild to
moderate conditions. However, this sample of GPs consistently
described a range of challenges, specifically in regional areas,
that impact on their support for headspace through referrals.
• Wait times for some headspace services have, at times, deterred
GPs from making referrals to their local service, out of concern
for the young person in the intervening period before being able
to access recommended treatment options.
• While there was good understanding of the broad offerings of the
headspace model, GPs also described challenges in
understanding what specific staff and specialist service areas a
headspace service might have, such as AOD workers,
occupational therapists, dietitians, or specialist psychological
services.
• Challenges were also described with operating a shared care model
with headspace services. These GPs described reluctance of
headspace services to take a GP’s diagnosis at the time of
referral or intake, and limited opportunity to discuss ongoing
progress and any other onwards referrals with GPs to support
effective care coordination.
• These challenges have also, at times, prevented GPs from
supporting headspace services through referrals.
• Specific to smaller regional locations, GPs also discussed
challenges with competing for the same staff. Where there is
only one local worker who provides a specific type of support,
referrals are often made to that person, regardless of which
service they work for.
• Staff have been lost from headspace services to another local
organisation in some instances, and support often shifts with the
person.
• School and university counsellors across the country had a strong
understanding of the professional and clinical services provided
by their local service, especially mental health and GP services.
• There was more limited recognition of other services provided,
including vocational, AOD and occupational therapy services.
Support for headspace services was also impacted by continuity
of relationships.
• A small number of counsellors described instances where a
headspace service had changed its management, and this
impacted the level of engagement they were able to have with
the service based on the approach of the manager. This, in turn,
impacted perceptions of the quality of the service.
• Staff turnover in other key roles, such as community engagement
coordinators, also impacted relationships, and where there was
a stable staffing group, opportunities for engagement were more
common. These views were consistent from counsellors across
different locations, including metropolitan and regional
 services.
• Similar to the experiences of GPs, some counsellors also described
challenges in engaging with headspace services for care
coordination. It was common that these counsellors did not
receive information regarding what happened with their referral
for a young person after it was made, unless the young person
returned to the counsellor and filled them in. This meant
counsellors were unsure of what additional support might be
required for a young person over and above headspace services
received.
• School and university counsellors in consultations also supported
referral to headspace services in most circumstances, however
some challenges were reported in supporting headspace.
• Some counsellors indicated they have stopped referring young
people to headspace services due to current wait times at their
local service, and they would prefer an alternative service that
might see a young person more quickly.
• A small minority of university counsellors also indicated that there
was limited benefit in referring a young person to headspace
services, as they were not able to provide an additional service
in addition to what their university support team could provide.
This varied depending on the resources available at institutions
and mental health supports offered.

To what extent does headspace • School and university counsellors identified that, in some
provide culturally appropriate communities, informal community outreach to remote
and inclusive service for young communities was beneficial. The extent to which this happened
people and their friends and varied between services.
families, including for
vulnerable and diverse
population groups and different
age groups?
To what extent does headspace • There was consistent feedback from school and university
enable young people and their counsellors that, often, young people prefer face-to-face
families to access support supports when they are seeking the type of counselling and
where, when and how they want psychology headspace services provide.
it, and what are the barriers and
enablers to this? • School and university counsellors also identified alternative
service formats, for example drop-in centres and sessions,
outreach into schools where a young person can attend a session
with a headspace clinician at school, and social groups, as
important services, particularly for hard-to-reach groups.
• School and university counsellors identified that service location
was important to access, with some indicating they did not refer
to headspace as they knew the closest service was not
accessible for high school students who cannot drive.
• Where there is a distance to travel to a service, access requires
parental support, which is not always what the young person
wants, or parents may be unsupportive.
• Deep dives resulted in a consistent theme of lengthy waitlists and
access to a multi-disciplinary workforce as strong barriers to
accessing support.

To what extent do young people • N/A

participate in the design and
delivery of headspace, and how
does this influence young
people and their families
experience of headspace?
How is the establishment of • There are mixed views from across stakeholders involved in
alternative service delivery delivering or working with headspace services as to the impact
models assisting headspace to of satellite services. There is significant positive regard for
meet its program outcomes? headspace services, and communities and stakeholders view
any headspace services as a positive addition to achieving core
objectives.
Source: KPMG 2022

B.2.6 Commonwealth government, state and territory governments and peak

bodies

Evaluation activities and data sources

Qualitative themes are sourced from a mix of data described below. Thirty-seven stakeholders contributed.
• Consultations with the department.
• Stakeholder consultations with State and Territory Governments
• Consultation with representatives from three national peak bodies and Orygen.
Table 43 Stakeholder engagement themes from Commonwealth Government, state and territory governments and peak
bodies

Evaluation Question Themes

How effective is headspace in • N/A
increasing mental health
literacy?
How effective is headspace in • N/A
increasing early help seeking?
How effective is headspace in • Although a considerable amount of money has been spent on
increasing access to required strategies to reduce waiting times, people are still getting ‘stuck’
services? on waiting lists, and therefore more serious cases are not being
provided with the care they need.
• The system appears overwhelmed and unable, at times, to meet the
needs of every young person who presents at a service.
• Some respondents highlighted the need for different models in
metropolitan versus remote regions.
• Adjusting opening hours was suggested to improve access
regarding offering appointments before and after hours and on
weekends.

How effective is headspace in • Only in recent times has there been more engagement with
supporting ‘hard to reach’ organisations that support Aboriginal and Torres Strait Islander
groups, including those who are young people. Some headspace services are engaging with this
at greater risk and less likely to cohort better than others.
seek help? In increasing access
for hard-to-reach groups? • States that had larger Indigenous populations felt the model was
not culturally adaptive. Improvements could be made with a
focus on employing Indigenous staff or more proactive outreach
as not all young people feel comfortable presenting at a service.

How well does headspace • headspace's branding is strong, services are accessible and
advocate for and promote youth welcoming.
mental health and wellbeing in
their communities? • One respondent thought there could be more proactive outreach.

To what extent has headspace • N/A

reduced stigma associated with
mental illness and help seeking
for young people, their families
and friends, and the
community?
How effective is headspace in • Integration needs trust between state-based services and other
improving pathways to care for services through PHNs, but often cases are not connected to
young people through service other services due to poor communication, leaving young
integration and coordination? people without the supports they need.
• Vertical integration between headspace and primary health care
providers is paramount.
• Integration could be improved with shared triage systems which
would improve efficiency for clinicians and prevent re-telling of
stories.
• Partnerships in consortia, where used, worked well.
• Some stakeholders noted differences in integration with other
services across services.

To what extent is headspace • Some areas provide various headspace services with different
providing localised service offerings and cater for a variety of cohorts of headspace users.
offering, and what are the
barriers and enablers of this? • One of the biggest barriers is still long waiting lists for care, and
clinical governance.
• Workforce shortage and clinical experience limits what level of
care some services can provide. Attracting staff is difficult.

What other contributions does • headspace provides support to schools and community in situations
headspace make to local around suicide and general engagement around suicide
communities? prevention.
• It was queried whether there could be better partnerships with other
entities, such as education departments, to get information to
schools for example.

To what extent does a ‘no- • N/A

wrong-door’ approach assist
headspace to meet its
objectives?
What is the level of support for • Responses varied from some services and jurisdictions having
headspace from other primary flexible support from primary care and mental health service
care and mental health service providers and others not having consistent support.
providers?
• A few respondents noted that recent lockdowns were difficult for
some services with surrounding private services and family
services being closed.
• It can be difficult to get appointments in public health services and
the default seemed to be headspace.

To what extent does headspace • Only in recent times has there been more engagement with
provide culturally appropriate organisations that support Aboriginal and Torres Strait Islander
and inclusive service for young young people. Some headspace locations are doing better than
people and their friends and others.
families, including for
vulnerable and diverse • It was noted that Aboriginal organisations often do not engage with
population groups and different headspace and that a better understanding of healing in
age groups? Indigenous youth was required.
• Greater recruitment required of a workforce with more live
experience and peer workers.

To what extent does headspace • Some respondents noted the strengths of the headspace model
enable young people and their allows for offering a soft entry, easy access for youth to mental
families to access support health care with no referral required.
where, when and how they want
it, and what are the barriers and • The branding is visible and has a recognisable name which
enablers to this? represented a good starting point for mental health care for
youth and their families.

To what extent do young people • N/A

participate in the design and
delivery of headspace, and how
does this influence young
people and their families
experience of headspace?
How is the establishment of • One respondent did not support satellite services, although these
alternative service delivery services leveraged the headspace brand; they thought this would
models assisting headspace to put headspace in a compromising position regarding meeting its
meet its program outcomes? program outcomes.
Source: KPMG 2022

KPMG | 15

©2022 KPMG, an Australian partnership and a member firm of the KPMG global organisation of independent member firms affiliated with KPMG International Limited, a
private English company limited by guarantee. All rights reserved. The KPMG name and logo are trademarks used under license by the independent member firms of the KPMG
global organisation. Liability limited by a scheme approved under Professional Standards Legislation.
Evaluation of the National headspace Program – Final Report
June 2022

Appendix C :
headspace services as
at 30 June 2020
Table 44: headspace services open at 30 June 2020

Date
Service State PHN Service type Analysis inclusion
opened
August
Adelaide SA Adelaide headspace centre All analysis
2015
October
Albany WA Country WA headspace centre All analysis
2007
Albury- December
VIC Murray headspace centre All analysis
Wodonga 2014
November
Alice Springs NT Northern Territory headspace centre All analysis
2008
Armadale WA Perth South headspace centre June 2015 All analysis
March
Ashfield NSW Central & Eastern Sydney headspace centre 2015 All analysis

March
Bairnsdale VIC Gippsland headspace centre All analysis
2017
Ballarat VIC Western Victoria headspace centre July 2013 All analysis
March
Bankstown NSW South Western Sydney headspace centre All analysis
2015
Bathurst NSW Western NSW headspace centre July 2008 All analysis
December
Bega NSW South Eastern NSW headspace centre All analysis
2018
Bendigo VIC Murray headspace centre July 2012 All analysis
Bentleigh VIC South Eastern Melbourne headspace centre July 2015 All analysis
September
Berri SA Country SA headspace centre All analysis
2008
Bondi Junction NSW Central & Eastern Sydney headspace centre May 2016 All analysis
Broken Hill NSW Western NSW headspace centre July 2017 All analysis
December
Brookvale NSW Northern Sydney headspace centre All analysis
2014
August
Broome WA Country WA headspace centre All analysis
2008
January
Bunbury WA Country WA headspace centre All analysis
2013
Central QLD, Wide Bay & March
Bundaberg QLD headspace centre All analysis
Sunshine Coast 2017
March
Caboolture QLD Brisbane North headspace centre All analysis
2016
Cairns QLD Northern QLD headspace centre April 2012 All analysis
November
Campbelltown NSW South Western Sydney headspace centre All analysis
2007
August
Camperdown NSW Central & Eastern Sydney headspace centre All analysis
2008
September
Canberra ACT ACT headspace centre All analysis
2008
March
Capalaba QLD Brisbane South headspace centre All analysis
2016
Castle Hill NSW Western Sydney headspace centre June 2016 All analysis
Chatswood NSW Northern Sydney headspace centre May 2013 All analysis
March
Coffs Harbour NSW North Coast headspace centre All analysis
2008
January
Collingwood VIC North Western Melbourne headspace centre All analysis
2012
Craigieburn VIC North Western Melbourne headspace centre April 2014 All analysis
Dandenong VIC South Eastern Melbourne headspace centre April 2013 All analysis
Darwin NT Northern Territory headspace centre May 2007 All analysis
Devonport TAS Tasmania headspace centre June 2013 All analysis
December
Dubbo NSW Western NSW headspace centre All analysis
2014
Edinburgh North SA Adelaide headspace centre May 2007 All analysis
March
Elsternwick* VIC South Eastern Melbourne headspace centre All analysis
2008
Frankston VIC South Eastern Melbourne headspace centre June 2008 All analysis
Fremantle WA Perth South headspace centre July 2008 All analysis
Geelong VIC Western Victoria headspace centre July 2007 All analysis
March
Geraldton WA Country WA headspace centre All analysis
2016
Central QLD, Wide Bay &
Gladstone QLD headspace centre April 2016 All analysis
Sunshine Coast
December
Glenroy VIC North Western Melbourne headspace centre All analysis
2008
Hunter New England &
Gosford NSW headspace centre June 2007 All analysis
Central Coast
February
Goulburn NSW South Eastern NSW headspace centre 2017 All analysis

December
Grafton NSW North Coast headspace centre 2017 All analysis

Greensborough VIC Eastern Melbourne headspace centre April 2016 All analysis
March
Griffith NSW Murrumbidgee headspace centre All analysis
2016
Central QLD, Wide Bay & Satellite from
Gympie QLD July 2018 All analysis
Sunshine Coast Maroochydore
Service provision
Outpost from
Hastings* VIC South Eastern Melbourne April 2020 and outcomes
Dandenong
analysis only
January
Hawthorn VIC Eastern Melbourne headspace centre All analysis
2014
Central QLD, Wide Bay &
Hervey Bay QLD headspace centre June 2008 All analysis
Sunshine Coast
February
Hobart TAS Tasmania headspace centre All analysis
2012
February
Horsham VIC Western Victoria headspace centre All analysis
2017
Hurstville NSW Central & Eastern Sydney headspace centre July 2014 All analysis
March
Inala QLD Brisbane South headspace centre All analysis
2012
Darling Downs & West February
Ipswich QLD headspace centre All analysis
Moreton 2013
Joondalup WA Perth North headspace centre July 2014 All analysis
January
Kalgoorlie WA Country WA headspace centre All analysis
2015
Service provision
September
Katherine NT Northern Territory headspace centre and outcomes
2019
analysis only
January
Knox VIC Eastern Melbourne headspace centre All analysis
2013
Hunter New England & Satellite from
Lake Haven* NSW May 2015 All analysis
Central Coast Gosford
January
Launceston TAS Tasmania headspace centre All analysis
2009
January
Lismore NSW North Coast headspace centre All analysis
2014
Service provision
Satellite from
Lithgow NSW Nepean Blue Mountains June 2019 and outcomes
Bathurst
analysis only
March
Liverpool NSW South Western Sydney headspace centre All analysis
2014
February
Mackay QLD Northern QLD headspace centre All analysis
2013
Hunter New England & September
Maitland NSW headspace centre All analysis
Central Coast 2008
Mandurah WA Perth South headspace centre July 2018 All analysis
Central QLD, Wide Bay & January
Maroochydore QLD headspace centre All analysis
Sunshine Coast 2013
December
Meadowbrook QLD Brisbane South headspace centre All analysis
2014
September
Melton VIC North Western Melbourne headspace centre All analysis
2018
January
Midland WA Perth North headspace centre All analysis
2013
March
Mildura VIC Murray headspace centre All analysis
2015
March
Miranda NSW Central & Eastern Sydney headspace centre All analysis
2014
Morwell VIC Gippsland headspace centre July 2008 All analysis
Service provision
Satellite from
Mount Barker SA Country SA June 2020 and outcomes
Murray Bridge
analysis only
Mount Druitt NSW Western Sydney headspace centre August All analysis
2008
March
Mount Gambier SA Country SA headspace centre 2016 All analysis

December
Mount Isa QLD Western QLD headspace centre All analysis
2014
Murray Bridge SA Country SA headspace centre June 2008 All analysis
December
Narre Warren VIC South Eastern Melbourne headspace centre All analysis
2014
Hunter New England &
Newcastle NSW headspace centre April 2013 All analysis
Central Coast
Nowra NSW South Eastern NSW headspace centre April 2012 All analysis
January
Nundah QLD Brisbane North headspace centre All analysis
2012
Onkaparinga SA Adelaide headspace centre May 2012 All analysis
January
Orange NSW Western NSW headspace centre All analysis
2016
February
Osborne Park WA Perth North headspace centre All analysis
2012
January
Parramatta NSW Western Sydney headspace centre All analysis
2012
Penrith NSW Nepean Blue Mountains headspace centre May 2013 All analysis
Pilbara Regional Outreach/Regional
WA Country WA May 2018 All analysis
Trial* Trial
Port Adelaide SA Adelaide headspace centre April 2015 All analysis
Port Augusta SA Country SA headspace centre April 2013 All analysis
January
Port Macquarie NSW North Coast headspace centre All analysis
2013
Satellite from
Portland VIC Western Victoria July 2018 All analysis
Warrnambool
Queanbeyan NSW South Eastern NSW headspace centre April 2015 All analysis
January
Redcliffe QLD Brisbane North headspace centre All analysis
2014
Central QLD, Wide Bay & January
Rockhampton QLD headspace centre All analysis
Sunshine Coast 2014
February
Rockingham WA Perth South headspace centre All analysis
2014
Service provision
Satellite from February
Rosebud VIC South Eastern Melbourne and outcomes
Frankston 2020
analysis only
Shepparton VIC Murray headspace centre April 2013 All analysis
Southport QLD Gold Coast headspace centre May 2008 All analysis
November
Sunshine VIC North Western Melbourne headspace centre All analysis
2007
January
Swan Hill VIC Murray headspace centre All analysis
2016
Hunter New England & January
Tamworth NSW headspace centre All analysis
Central Coast 2013
February
Taringa QLD Brisbane North headspace centre All analysis
2015
Darling Downs & West
Toowoomba QLD headspace centre July 2015 All analysis
Moreton
Townsville QLD Northern QLD headspace centre June 2008 All analysis
March
Tweed Heads NSW North Coast headspace centre All analysis
2015
Service provision
Satellite from December
Victor Harbor SA Country SA and outcomes
Murray Bridge 2019
analysis only
Wagga Wagga NSW Murrumbidgee headspace centre July 2008 All analysis
Warrnambool VIC Western Victoria headspace centre June 2008 All analysis
Darling Downs & West
Warwick QLD headspace centre July 2008 All analysis
Moreton
February
Werribee VIC North Western Melbourne headspace centre 2014 All analysis

Whyalla SA Country SA headspace centre April 2018 All analysis

February
Wollongong NSW South Eastern NSW headspace centre 2008 All analysis

December
Wonthaggi VIC Gippsland headspace centre All analysis
2018
February
Woolloongabba QLD Brisbane South headspace centre All analysis
2014
Source: KPMG 2022
Note” Services marked with an asterisk * represent those services not recorded in the Commonwealth Government’s official count of
headspace services. The Elsternwick service is counted as one service with Bentleigh

KPMG | 15

©2022 KPMG, an Australian partnership and a member firm of the KPMG global organisation of independent member firms affiliated with KPMG International Limited, a
private English company limited by guarantee. All rights reserved. The KPMG name and logo are trademarks used under license by the independent member firms of the KPMG
global organisation. Liability limited by a scheme approved under Professional Standards Legislation.
Evaluation of the National headspace Program – Final Report
June 2022

Appendix D :
Effectiveness in
achieving
intermediate
outcomes
D.1 How effective is headspace in increasing mental health
literacy?
D.1.1 Mental health literacy
Table 45 Overview of mental health literacy objectives of headspace

Objective Short term-impacts Medium-term impacts

Increasing mental • Young people accessing • Young people are better able to
health literacy - headspace services improve manage their mental health in
knowledge about their mental health literacy the medium- to long-term,
mental health, how to (knowledge about mental including identifying when
seek help and how to health, how to seek help, and they need to seek help and
manage mental health how to manage mental health) support

The headspace program logic sets out the above objectives and impacts associated with increasing mental
health literacy. In this context, mental health literacy is defined as knowledge about mental health, how to
seek help and how to manage mental health. Through improving mental health literacy, the headspace
model supports the medium-term impact that young people are better able to manage their mental health in
the medium- to long-term, including identifying when they need to seek help and support. Ultimately,
improved mental health literacy contributes to long-term impacts of improved health outcomes for young
people and increased social and economic participation outcomes for young people over their life course.
As the world experiences unprecedented challenges in the face of COVID-19, good mental health literacy
in young people and their key support people may lead to better outcomes for those with mental illness,
either by assisting early help seeking by young people themselves, or by their support people identifying
early signs of mental disorders and seeking help on their behalf . In the headspace context, mental health
165F 166F

literacy refers to knowledge about mental health, how to manage mental health and how to go about
accessing support with mental health concerns.
Previous evaluation work undertaken by headspace National indicates that the headspace model is
effective in building mental health literacy for young people. For example, Colmar Brunton conducted a
review which found consistent feedback across stakeholder groups that headspace supports better
understanding of mental health, ill health and seeking help . 167F

In order for the current evaluation to examine the extent to which the headspace model is effective in
increasing mental health literacy, a range of data and evidence was reviewed from across the fieldwork
activities conducted for this project. These are described below, and include analysis of the hMDS,
interviews with service users, interviews with Youth Reference Group members, interviews with
university and school counsellors and survey responses from service and lead agency staff.

Evidence of the contribution of headspace to increased mental health literacy

Perspectives of young people who use headspace
A key indicator of the extent to which the headspace model has a positive effect on the mental health
literacy of young people attending its services is measured through the hMDS young person satisfaction
matrix. Young people attending headspace are given the option to complete a satisfaction survey on their
second OOS, and subsequently at every fourth visit during that EOC. The survey asks them to rate 14
statements on a five point scale of ‘strongly agree’ to ‘strongly disagree’. Statement 11 is an indicator of
self-reported change in mental health literacy:

“I feel that I know more about mental health problems in general because of
attending headspace”
Figure 32: Distribution of responses to “I feel that I know more about mental health problems in general because of
attending headspace” from 2015-16 to 2019-20

Source: KPMG master dataset covering completed and ongoing episodes of care from created during 2015-16 to 2019-20
Notes: See Appendix F for a description of how the master dataset is derived. Sample includes 379,130 episodes of care.

Within the period of analysis, 125,209 out of 379,130 episodes observed between 2015-16 to 201920 had
responses given by young people to this statement, for 33 per cent of the total number of episodes of care.
Overall, around 66 per cent of respondents either agreed or strongly agreed (20 per cent) with the
statement, indicating that they attribute an increase in their mental health literacy to their interactions with
headspace.
Responses reflected a similar pattern across genders, as well as for young people from culturally and
linguistically diverse backgrounds, Aboriginal and Torres Strait Islander young people, and for young
people who speak a language other than English at home.
The extent to which young people indicated that headspace had helped them to improve their mental health
literacy steadily increased with the number of OOS they had accessed. Where the young person had
received one to two OOS, 53 per cent of responses agreed or strongly agreed with the question statement,
compared with 79 per cent agreement where they had received 20 or more OOS.
Figure 33: Proportion of episodes of care for young people who agreed their mental health literacy had improved after
using headspace services, based on the number of OOS accessed during their episode of care
Source: KPMG analysis of hMDS covering completed and ongoing episodes of care created from 2015-16 to 2019-20
Notes: Sample includes 379,130 episodes of care. Analysis considers last observed response to the question “more about mental
health problems in general because of attending headspace”.

The vast majority of headspace users interviewed reported their mental health literacy had improved due to
their participation in therapeutic encounters with headspace counsellors and clinical psychologists. They
articulated they had learned about mental health, specific concepts, obtained a diagnosis in many cases,
and had gained more insight into their own conditions, discussing these using concepts and language
derived from written material and their therapists. Some attributed new knowledge and positive outcomes
wholly to headspace, for example:

“I've learned a lot. It’s been good since then…like the strategies they give you at
headspace. I just want to thank headspace for changing my life and I still continue to
grow. I still take what I learnt from headspace every day.”

and

“I think it's solely because of headspace that I got better. I would've never gotten
better if I hadn't gone.”
At the same time, a minority felt headspace had not helped them much or at all:

“I might be an outlier. I think there’s a lot of people that headspace probably have
helped but maybe it’s just not for me or maybe I haven’t found the right person.”

A key aspect of building mental health literacy with young people, emphasised in interview responses
from headspace users, is the level of rapport and engagement established between the young person and
the headspace worker with whom they are connected.
Apart from learning more about mental health through headspace, users also cited other sources of
knowledge, including from private psychologists and by researching and reading information online.
Overall, Youth Reference Group participants interviewed reported that their mental health literacy had
improved through engagement with headspace:

“I think headspace has really helped, one: understanding how you’re feeling and
what’s going on, and then, two: after understanding it, learning how to cope with it,
how to deal with it, different strategies to help you through.”

After coming to headspace, using the services, and learning coping strategies to manage their mental
health, many of the young people in reference groups interviewed reported that they had gained more
confidence to speak about their experiences and had then actively promoted the service to friends and
more broadly:

“I have two friends who I said, you know, you’ve got to go to headspace, you’ve got
to do something about how you’re feeling, and both of them have now continued to
come to headspace and they now both promote it and they talk about it and, you
know, they’re on track to seeking help and it’s something that is really positive.”

Perspectives of headspace service providers

The survey of staff from services and lead agencies included two questions to elicit staff views of the
efficacy of the headspace model on improving the mental health literacy of young people. The first
question asked them to use a five point rating scale of ‘very well’ to ‘not at all well’ in response to the
prompt:

Based on your observation of young people at your headspace service, how well
does the service increase mental health literacy? For example, building
understanding of where to seek support, understanding of mental ill health and
treatments, and reduction of stigma to support help seeking.
Figure 34: Responses from lead agency and headspace survey representatives on how effective headspace services are
in increasing mental health literacy
Source: KPMG Analysis of the Survey of headspace services and their lead agencies
Note: 60 staff at either services or lead agencies responded to this question in the survey.

A total of 93 per cent of service and lead agency respondents selected ‘very well’ or ‘well’ in response to
this question, indicating that staff working within the headspace model have generally high levels of
confidence that the services they provide lead to increases in mental health literacy for young people.
The survey then prompts a further, free text response to the question:

Why have you chosen this response? What are the barriers and enablers to this
service achieving this objective?

Responses identified a number of ways in which headspace services contribute to improved mental health
literacy, including the following.
• Broader community engagement by the headspace service, including through social media campaigns,
education and awareness activities with local schools, and through establishing partnerships with local
councils, universities and colleges.
• Online and printed resources provided by services to support mental health literacy. These are seen to be
frequently accessed by young people and their families, and include material on various services
available in the local area, how to make informed decisions about referral pathways, and how young
people can support their own mental health.
• A case management component of work undertaken with young people and families, upskilling them on
mental health support options, and capacity building strategies.
• Psychoeducation provided as part of clinical services.
• Group work services that focus on mental health literacy, capacity building, and accessing support for
young people.
• Safety planning and information on available supports provided at the intake and assessment stage with
young people accessing headspace services.
Analysis of comments received in response to this survey question identified access issues as a key barrier
to improving mental health literacy for young people. This was raised in terms of strong demand pressures
with young people waiting to access the service, as well as in terms of access issues caused for the service
by COVID-19, which introduced barriers to engagement between services and schools and other
community organisations.

Effectiveness of headspace in increasing mental health literacy

Qualitative data show that young people using headspace and staff working within the headspace model
have strong, positive views about the effect headspace has on increasing mental health literacy. Young
people highlighted the useful strategies headspace had given them in identifying and managing their
mental health issues, while they also indicated that finding a staff member at headspace with whom they
could build a rapport was an essential enabler of improving the mental health literacy of young people.
The staff surveyed presented a range of elements from across the hMIF which they see as key to enabling
their service to contribute to increasing the mental health literacy of the young people they support. These
range from clinical, community and information related activities, indicating that improving mental health
literacy is embedded across the headspace model. The key barrier identified was limitations to young
people’s ability to access the service due to wait times, and limited capacity across the local service system
to engage during COVID-19 restrictions.
It is important to note that there are some limitations around the data, with limited evidence from family,
friends and community stakeholders, and data being self-reported by young people and services about their
own performance or improvement on this measure. At the same time, this evidence supports analysis
commissioned by headspace, with similar themes from the report into stakeholder views conducted by
Colmar Brunton, giving some confidence that findings are reliable.
Evidence from young people using headspace and headspace service staff indicates that the headspace
model is effective in increasing the mental health literacy of young people engaging with its services.

D.2 How effective is headspace in increasing early help seeking?

D.2.1 Early help seeking
Table 46 Overview of early help seeking objectives of headspace

Objective Short-term impacts Medium-term impacts

Increasing early help • Young people and families • Young people, their families and
seeking – at an earlier accessing headspace services communities are better able to
age (e.g., under 21 have increased knowledge identify when someone needs
years); at relatively low about, and willingness to, seek help, and support appropriate,
mental health risk help early help seeking
status; assessed as at
less than threshold • Young people, their families and • Earlier identification and
stage of illness communities (living near treatment of emerging mental
headspace centres and health problems for young
satellites) have improved people
attitudes towards mental health
• Young people increase help
and mental illness (stigma
seeking behaviour for mental
reduction)
health and wellbeing issues

In the headspace program logic, increasing early help seeking is key to improving short-term impacts for
young people and families in increasing their knowledge about, and willingness to seek help with, mental
health issues. It is also associated with having improved attitudes towards mental health and mental illness.
These, in turn, relate to a range of medium-term impacts around help seeking, early identification of
emerging mental health problems and increased help seeking behaviour. In the headspace context, again,
these are all identified as contributors to the long-term impacts headspace is seeking for improved
outcomes over the life course.
The headspace model includes ‘early intervention’ as a core service component, defined as “the
identification and provision of intervention and support services as early as possible in the development of
mental health difficulties to prevent or delay the onset of mental ill-health or reduce the impact associated
with mental ill-health and improve outcomes” . 168F

A young person's mental health is an investment into the future. Management and improvements as they
move into adulthood benefits not only the young person but, in the long-term, the economy. Mental illness
for young people usually manifests before the age of 21, indicating the importance of treatment and
assistance provided early in life, early in illness and early in an incident . Early intervention programs
169F 170F

assist a young person by identifying risk factors early or providing timely treatment for problems that can
alleviate the potential harm from mental illness. Treating risk factors and symptoms early is seen as not
only improving the social and emotional wellbeing of young people, but also as a cost-effective approach
to improving lifelong outcomes for them . 171F

A key evaluation question for this project examines the extent to which the headspace model is associated
with increased levels of early help seeking from young people. For the purposes of this evaluation, early
help seeking is defined as a young person engaging with headspace when they are:
• under 21 years of age;
• at relatively low mental health risk status; and
• assessed as at less than the threshold stage of illness.
To examine the extent to which the headspace model is succeeding in contributing to increased early help
seeking behaviour, relevant data and evidence was reviewed from across the fieldwork activities conducted
for this project. These are described below, and include analysis of the hMDS and survey responses from
service and lead agency staff.
Administrative data from the hMDS 172F

The hMDS collects a range of measures relevant to the definition of early help seeking used for this
project. In terms of the age at which young people are presenting at headspace for information and
support, this has remained relatively stable over the data period, with around three-quarters of young
people presenting aged under 20 years.
Figure 35: Distribution of age by young person from 2015-16 to 2019-20
Source: KPMG analysis of the hMDS
Notes: See Appendix F for a description of how the master dataset is derived. Sample 90,110 young people for 2019-20; 98,270
young people for 2018-19; 87,510 young people for 2017-18; 79,322 young people for 2016-17; 70,940 young people for 2015-16.
Data labels are not included for categories with less than five per cent for clarity purposes.

The hMDS also contains data regarding each young person’s mental health risk status and clinical stage of
illness and diagnosis (where relevant). These measures are collected by the clinical service provider as part
of the intake and assessment process, and then are reassessed on every occasion of service.
The mental health risk status measure considers the presence of risk and protective factors, such as
unstable or unsafe living conditions, relationship problems and bullying. It also considers the presence of
symptoms of mental disorder, such as anxiety or depression. A young person would be considered to be
undertaking early help seeking if presenting with either ‘no risk factors or symptoms of mental health
problems’ or ‘risk factors present’, indicating the presence of one or more situational factors making them
vulnerable to developing a mental health problem. hMDS data in the period indicates that just under half
of the young people presenting (46.1 per cent) were in this early help seeking category.
Figure 36: Mental risk status on initial OOS for all episodes of care during 201920
Source: KPMG analysis of the hMDS
Notes: See Appendix F for a description of how the master dataset is derived. The initial OOS recorded in the main extract during
2019-20 is examined. The sample consists of 73,712 OOS.

Stage of illness data collected by service providers indicates the extent of progression of a disorder at a
particular point in time, showing where the young person sits on a continuum of the course of an illness.
To make this assessment, the service provider considers the severity, persistence and recurrence of
symptoms, as well as biological and social impacts associated with the disorder. Similar to the mental
health risk status measure, where a young person presents with ‘no symptoms of mental health problems
or disorder’ or with ‘mild to moderate general symptoms of mental health problems and/or high risk
psychosocial stressors’ (e.g., bullying or relationship problems), they would meet the definition of ‘early
help seeking’.
As can be seen in Figure 37, over time, the proportion of young people in these categories during their
initial OOS has remained relatively stable year on year, at just under half (between 41 and 48 per cent).
Figure 37: Stage of illness during initial OOS for all episodes of care between 201516 and 2019-20
Source: KPMG master dataset
Notes: See Appendix F for a description of how the master dataset is derived. The initial OOS recorded in the main extract during
2019-20 is examined. The sample consists of 73,712 OOS for 2019-20; 89,789 occasions for 2018-19; 79,603 for 201718; 72,479 for
2016-17; 65,612 for 2015-16. Data labels are not included for categories with less than 0.5 per cent for clarity purposes.

Perspectives of young people who use headspace

headspace users relayed that young people are increasingly aware of mental health issues, and that stigma
has reduced over time. headspace visibility and outreach meant that young people were sometimes already
aware of it, or were referred to it early via school or through their GP. However, waitlists were raised by
some as an inconvenience and others as a severe challenge:

“I guess the doctor was quite helpful but I find that the waiting period can be quite
long, and it makes it quite difficult… It just feels so long away at the time, I guess,
especially when you’re depressed.”

Reference group participants commented that headspace’s promotional activities were effective and
services appeared accessible and used ‘soft entry points’ as well as referral pathways from GPs.
Youth Reference Group participants commented that headspace’s promotional activities in the community
and outreach in schools meant that young people could be linked to help early through GPs, and school
counsellors:

“Their signage is really good. Everyone knows where it is. People notice it and
people ask what it is. I think that’s really good but in terms of stigma.”
and

“It shows ‘Here we are, if you need our help. Just come inside. It’s all fine’.”

Groups that are run out of headspace, for example a yoga group in one area, serve as a soft entry point for
seeking help:

“People who’ve never been to headspace before come to the yoga and then go
‘Actually I might use this service’.”

Perspectives of staff within headspace services

The survey of staff from services and lead agencies included two questions to elicit staff views of the
efficacy of the headspace model on increasing early help seeking behaviour. The first question asked them
to use a five point rating scale of ‘very well’ to ‘not at all well’ in response to the prompt:

Based on your observation of young people at your headspace service, how well
does your service increase early help seeking behaviour?
Figure 38: Responses from lead agency and headspace survey representatives on how effective headspace services are
in increasing early help seeking

Source: KPMG analysis of the survey of headspace services and their lead agencies
Notes: Sixty staff at either services or lead agencies responded to this question in the survey.

A total of 87 per cent of service and lead agency respondents selected ‘very well’ or ‘well’ in response to
this question, indicating that staff working within the headspace model have generally high levels of
confidence that the services they provide lead to increases in early help seeking behaviour.
The survey then prompts a further, free text response to the question:

Why have you chosen this response? What are the barriers and enablers to this
service achieving this objective?
Responses identified a number of ways in which headspace services contribute to increased early help
seeking. These include:
• strong brand recognition and social media presence of headspace services; and
• promotion and advocacy work of services, including the community engagement roles, for example
school events conducted.
In terms of barriers to increasing early help seeking, these are similar to those identified for improving
mental health literacy:
• the impact of waiting times constraining the extent to which services can provide early support and early
referrals to other services;
• staff saw community engagement as a key mechanism through which services promote early help
seeking, which is limited due to staffing challenges and funding constraints;
• the potential to improve the service’s contribution to early help seeking through additional intake
engagement workers, supporting early intervention and low level needs early;
• perceptions that headspace services are supporting high-risk or high-needs young people, discouraging
others from seeking support for mild to moderate needs; and
• pressure on services from supporting higher needs young people reducing the capacity to provide early
intervention support to those who seek help early.
In line with this, school and university counsellors interviewed frequently raised the issue of waiting times
and anticipated delays in receiving support, as a reason young people do not seek support from headspace,
constraining the extent to which headspace can provide an early intervention service for young people.
Perspectives of young people who do not use headspace
A theme emerging from interviews with young people who do not use headspace services was that, when
asked why they do not seek support from headspace, they responded they do not feel their need is severe
enough to warrant taking the time or resources away from those in need. This not only indicates that the
focus and purpose of headspace has been misunderstood but that there may be an unmet need in the
community, and that young people could benefit from seeking support before their mental health problems
become more severe.

Effectiveness of headspace in improving early help seeking

Evidence reviewed from a range of sources indicates that early help seeking is an area of continued focus
for the headspace model, with good results despite barriers.
Administrative data in the hMDS regarding the age, mental health risk status and stage of illness of young
people presenting in the period for an occasion of service indicate that around three-quarters of young
people presenting were aged under 20 years. Just under half of those presenting (46.1 per cent) in the
period had either ‘no risk factors or symptoms of mental health problems’ or ‘risk factors present’,
indicating the presence of one or more situational factors making them vulnerable to developing a mental
health problem. Furthermore, just under half (41-48 per cent) of young people presenting at headspace in
the period had ‘no symptoms of mental health problems or disorder’ or ‘mild to moderate general
symptoms of mental health problems and/or high risk psychosocial stressors (e.g., bullying or relationship
problems). These data provide a broad indicator of ‘early help seeking’, where the young person is
presenting at a young age, at an early stage of illness or with low risk factors. They show that a substantial
proportion of young people attending headspace meet a broad definition of ‘early help seeking’.
Staff at headspace are confident that their service provides increases in early help seeking behaviour, with
87 per cent of service and lead agency respondents selecting ‘very well’ or ‘well’ in response to this
question. Enablers of this were identified as community engagement activities and strong brand
recognition, while wait times, workforce limitations and misconceptions of headspace as being for high-
needs young people were all also raised by school and university counsellors and young people who do not
use headspace.
This data provides a range of lenses through which to examine the question as to whether headspace is
improving early help seeking in young people. The evidence suggests that headspace is effective in
improving early help seeking behaviour in young people with increased access by those from younger age
groups, although waiting times at services was identified as a key barrier to early help seeking behaviour.
While mental health risk factors and stage of illness for young people attending headspace have remained
relatively stable, data indicates that almost half of those attending headspace are seeking help for mild
mental health conditions or are engaging in ‘early help seeking’.

D.3 How effective is headspace in increasing access to required

services?
D.3.1 Access to services through headspace
Table 47 Overview of access to service objectives of headspace

Objective Short-term impacts Medium-term impacts

Increasing access to • Young people from a diverse • Young people receive appropriate,
required services – the range of backgrounds access evidence-based treatment early
number of young and engage with headspace
people accessing • Young people increase help
services
headspace seeking behaviour for mental
• Young people and families can health and wellbeing issues
access headspace services in a
timely manner, and at low or
no cost

Young people accessing required services through headspace is central to the headspace program logic.
Young people and families being able to access services in a timely manner at low or no cost leads to
young people receiving appropriate, evidence-based treatment early and increased help seeking behaviour
into the medium-term. Again, this leads to long-term impacts in the headspace program logic for improved
outcomes over the life course.
As evidenced by the Royal Commission's work into Victoria's mental health system, there are a number of
barriers many young people come up against when seeking care. Demand has overtaken capacity,
community-based services are under-supplied, unsuitable or driven by crisis, services are poorly integrated
and families, carers and supporters are left out .173F

Given the increased level of funding the headspace model has received in recent years, and the expansion
in the number of services around the country, a key measure of its effectiveness is to also examine the
extent to which the increased number of services and service funding is associated with an increase in the
overall number of young people accessing headspace.

Administrative data from the hMDS 174F

headspace National hMDS data demonstrate that, over time, the number of young people accessing
support through headspace has increased with the increase in number of services. The slight drop in 2019-
20 may be due to the COVID-19 pandemic reducing access rates, in line with comments from providers
interviewed across the evaluation. The data indicate that the average number of young people accessing
headspace per service is increasing, despite expansion of new services into regional and rural areas which
have smaller populations, potentially leading to lower average numbers of young people accessing these
services. This corresponds to an observed increase in need for the age group served by headspace, with
headspace National research indicating that over onethird of young people in Australia reported high to
very high levels of distress in 2018, compared with only nine per cent doing so in 2007 . While this
175F

headspace National research indicates that mental health needs are highest in 18 to 21 year olds, hMDS
data indicates that the average age of a young person attending a service over the past five years has
consistently been just over 17 years of age.
Table 48: Average number of young people accessing headspace per year

Financial Year Number of young people Number of services Average young people
accessing headspace in operation accessing headspace per
service per year
2015-16 70,940 98 724
2016-17 79,322 102 778
2017-18 87,510 106 826
2018-19 98,270 113 870
2019-20 90,110 118 764
Source: KPMG analysis of the hMDS master dataset

Perspectives of young people who use headspace

Young people using headspace talked about their entry points to accessing the service, and were generally
referred by GPs, via schools, or on parental suggestion:

“So I was referred to headspace from my GP. I had experienced two panic attacks in
February of that year, 2020, in a three-week period, and on that DSM-5, it said that I
met the criteria to go see somebody.”

and

“Mum just found them, she did everything for me, obviously my current mindset back
then, I didn't want to do anything.”
and

“It was pretty much an open day festival and there was a bunch of stalls opened up
and one of the stalls was headspace. So I saw the headspace stall and I was just
wondering what they do and all that, because I know that headspace had something
to do with mental health and helping with you know, like letting people know that.”

While most accessed services face-to-face, due to the pandemic, or because of distance, some users
preferred flexible appointments using telephone or online platforms such as Zoom.
headspace services were reported as conveniently located, with some element of discretion preferred, to
avoid stigma and the risk of people observing them accessing the service. For example:

“I would say in terms of a discrete location definitely something that is still kind of
easy to find and not just be, like, "Where is it? They say it's here but I can't - I don't
see anything". But not necessarily being so bold with being, like, "This is
headspace".”
Once accessing the service, users reported it to be friendly and welcoming:

“I felt really, really welcome because I'm assuming most other offices have the same
thing, but they've got all supportive things everywhere about how they welcome every
type of person. They were just really kind to me as well because I mean I think I was
13 or 12. So I was obviously quite scared but they were really, really nice to me
which was good. It was very colourful, bright.”

One downside reported by a small number of users was a long waiting time (about a month) between
intake and assignment to a counsellor or psychologist.
headspace users reported appropriate referrals, for example to dieticians or other specialists, with one
reported lack – referrals to psychiatrists were difficult:

“So they referred me to a dietician that was there once a week and then I saw them
and got some advice on my meal plans and what to cook and things like that. So, that
was really helpful because it was at the same place. I didn't have to go anywhere.”

and

“I'd like for them to sort of provide better access to psychiatrists whether that is
having one on hand at these centres or knowing where the most readily available
ones are in the local area.”

While a minority of young people were not happy with their encounter with headspace, citing other
services or clinicians that helped them more, most found the help they received from headspace to be
beneficial, or that headspace led to a referral that helped more. Therefore, headspace largely fulfilled its
role as a ‘onestop-shop’ service.
Young people in the reference groups reported that headspace worked well with other services to meet the
needs of users, using a client-centred approach. Staff provided support in a respectful and non-judgmental
manner and worked diligently to ensure that young people accessed the required services from headspace
or from external organisations. Confidentiality was raised as an important issue when the young people
were referred to other services from headspace:

“It's a daunting thing to have to do, especially if you have anxieties or stuff like that.
 But I think that headspace works really well with the exterior services that other
places provide and they really communicate very well and they do provide a really
good level of confidentiality.”
Young people also commented that headspace staff would diligently try to meet the needs of young people
first rather than just ‘redirecting them’. However, they noted that sometimes it reached a point where a
young person needed more complex support and headspace staff would continue to support them until
other services were in place.

Perspectives of headspace service providers

The survey of staff from services and lead agencies includes a question to draw out staff views of the
efficacy of the headspace model in increasing access to required services. The question asked respondents
to select from a range of options based on qualitative data obtained through other streams of activity across
the evaluation. It also offered a free text option for them to describe other issues they feel are barriers to
access for young people:

What factors are barriers for your service in enabling access to required
services for young people? Choose all that apply.
Figure 39: Barriers to access for young people

Source: KPMG analysis of the survey of headspace services and their lead agencies
Notes: Fifty-nine staff at either services or lead agencies responded to this question in the survey.

A large majority of respondents indicated waiting lists and workforce attraction and retention as the key
barriers to supporting increased access to their headspace service. In response to the free text ‘other’
option, a number of themes emerged. Here, many responses were related to the funding model, for
example some cited insufficient funding for salaried staff, again including community engagement staff,
while others noted the difficulties in being able to afford an accessible site. The impact of the COVID-19
pandemic was also noted as impacting the ability for young people to access their services.
Representatives interviewed across all deep dive sites shared anecdotally that the majority of young people
accessing headspace self-referred into their services, and the ability to do this ensured headspace provided
a ‘soft entry’ into mental health support, without the need for formal referral through GPs or other avenues.
The credibility and power of the headspace brand was noted as a key strength across services in
encouraging young people to proactively access services.
For some services, the use of satellite sites have allowed young people to conveniently access headspace in
their local area, removing the need for additional travel into regional centres. Services offering after-hours
access was highlighted as important to enable young people to access services outside school and work.
Stakeholders also discussed the topics of ease of location, and being close to transport as key aspects of the
model which support access to headspace services.
Again, waiting times for support through services was frequently raised in interviews with stakeholders in
deep dive sites as a barrier to access for young people.

The extent to which headspace is effective in increasing access to required services

Data from a range of sources indicates that headspace is broadly effective in increasing access to required
services, but that workforce and demand pressures continue to constrain the volume of young people able
to access support.
At the same time, hMDS data demonstrate that, over time, the number of young people accessing support
through headspace has increased steadily with the increase in number of services. Young people and
headspace staff value the ‘soft entry’ approach to accessing headspace through GPs or schools, mostly
face-to-face but with flexible options during COVID-19. Flexibility in opening hours was also valued, and
features of the physical site were identified as improving access, for example a central location close to
public transport.
Data indicates that headspace is effective in increasing access to required services. Barriers to increased
access raised by various stakeholders were long waiting times between intake and assignment to a
counsellor or psychologist. Insufficient funding for salaried staff was also raised as a barrier, including
community engagement staff, and the costs of an accessible site.
D.4 How effective is headspace in supporting ‘hard to reach’
groups, including those who are at greater risk and less
likely to seek help?
D.4.1 Supporting ‘hard to reach’ groups
Table 49 Overview of objectives of headspace for ‘hard to reach’ groups

Objective Short-term impacts Medium term-impacts

Increasing mental • Young people accessing • Young people are better able to
health literacy - headspace services improve manage their mental health in
knowledge about their mental health literacy the medium- to long-term,
mental health, how to (knowledge about mental including identifying when
seek help and how to health, how to seek help, and they need to seek help and
manage mental health how to manage mental health) support

Increasing early help • Young people and families • Young people, their families and
seeking - at an earlier accessing headspace services communities are better able to
age (e.g., under 21 have increased knowledge identify when someone needs
years); at relatively low about, and willingness to, seek help, and support appropriate,
mental health risk help early help seeking
status; or when assessed
as at less than threshold • Earlier identification and
stage of illness treatment of emerging mental
health problems for young
people
• Young people increase help
seeking behaviour for mental
health and wellbeing issues

Reducing stigma • Young people, their families and •

associated with communities (living near
mental health and headspace centres and
mental illness - the satellites) have improved
fear or attitudes towards mental health
embarrassment of and mental illness (stigma
seeking help for reduction)
mental health and
wellbeing, and the
negative judgment
of and lack of
empathy for those
that do

Increasing access to • Young people from a diverse • Young people receive appropriate,
required services - the range of backgrounds access evidence-based treatment early
number of young and engage with headspace
people accessing services
headspace
• Young people and families can
access headspace services in a
timely manner, and at low or
no cost

A key component of this evaluation was to consider the four objectives listed in the table above and to
examine how well the headspace model meets the needs of ‘hard to reach’ groups with respect to these
objectives.

How headspace supports ‘hard to reach’ groups

Engaging groups considered to be marginalised from mainstream health services can be difficult due to
ongoing perceptions and experiences of stigma and discrimination. Groups already at high risk of stigma
include Aboriginal and Torres Strait Islander young people, young people who identify as LGBTQIA+,
young people with disability and young people from some culturally and linguistically diverse
communities, where there may be significant stigma and taboo associated with mental illness. The Mission
Australia Youth Survey 2021 highlighted that, among young people participating in the study, 51.5 per
cent cited mental health as their top obstacle to success, compared with 83.2 per cent of young people of
gender diverse backgrounds who reported poorer mental health on numerous measures. In this study,
although the majority of Aboriginal and Torres Strait Islander young people engage in education, value
their family and friends, and feel positive about the future, they also report greater challenges than their
peers who do not identify as Aboriginal or Torres Strait Islander, including being less likely to feel happy
or very happy with their lives .176F

Other research highlights enduring issues with systemic barriers and unconscious bias within the health
system, where young people from diverse backgrounds are less likely to have their needs met, due to
,,,
factors such as lack of cultural competence and misdiagnoses . 177F 178F 179F 180F

Based on feedback from stakeholders across the headspace landscape, as well as on broader academic and
grey literature regarding stigma and service access, for the purposes of this evaluation, ‘hard to reach’
groups are defined as:
• Aboriginal and Torres Strait Islander young people;
• young people from culturally and linguistically diverse backgrounds;
• young people who identify as LGBTQIA+; and
• young people with disability.
It is important to note that young people within these groups are diverse and have a variety of experiences
and perspectives on issues associated with mental health. At the same time, exploring the evidence for how
well the headspace model supports members of these groups allows for consideration of potential systemic
factors which may reduce its efficacy for young people across the spectrum of potential life experiences.
The evidence for how well the headspace model supports young people from ‘hard to reach’ groups across
key objectives of the model is reviewed below. Overall, while there is a high degree of similarity in
feedback from these groups across the key objectives, there is also some evidence to suggest that meeting
the needs of some stakeholder groups is a greater challenge than for others. In particular, workforce
shortages of key staff reduce the ability for services to make young people from ‘hard to reach’
backgrounds feel welcome.
Perspectives of headspace service providers
Responses from the survey of service and lead agency staff were sought as to whether they see any
differences in effectiveness of the headspace service in improving the mental health literacy, early help
seeking and access of young people who identify as being part of these groups. The responses indicate that
a majority of staff surveyed see the headspace model as less effective in meeting these objectives for
young people from ‘hard to reach’ groups.
Figure 40: Survey responses about whether the headspace model is less effective for particular cohorts compared with
the general population of young people

Source: KPMG analysis of the headspace service and lead agency survey
Notes: A total of 60 responses were received for Aboriginal and Torres Strait Islander young people and culturally and linguistically
diverse young people, 59 responses were received from LGBTQIA+ young people and 58 responses were received from young
people with disability.

How effective is headspace in increasing mental health literacy for ‘hard to reach’
groups, including those who are at greater risk and less likely to seek help?
Perspectives of young people who use headspace
The hMDS user satisfaction data was examined for any variation between young people from different
backgrounds, or ‘hard to reach’ groups. Data during the period indicates that there is no significant
difference between the improvements in mental health literacy reported by young people who access
headspace from different culturally and linguistically diverse backgrounds or by young people who
identify as LGBTQIA+, however satisfaction was significantly lower for Aboriginal and Torres Strait
Islander young people compared to the general population of young people attending headspace.
Youth Reference Group members noted that headspace actively worked to increase mental health literacy
across all groups of young people including those hard-to-reach. They noted that unless young people were
willing to accept help, these groups would remain difficult to reach:
“I personally think that headspace is doing very well at trying to, engaging and actively going
out of their way to seek those [hard to reach] people but as long as those people are not willing
to engage back, their treatment is going to be hard for them.”

Perspectives of headspace service providers

headspace service and lead agency staff who responded ‘yes’ regarding seeing differences in outcomes for
young people from ‘hard to reach’ groups compared with the general population of young people were
then given the option to rate the difference between groups on a sliding scale. Averaged results indicate
that staff believe Aboriginal and Torres Strait Islander young people, culturally and linguistically diverse
young people and young people with disability all fare below the general population of young people
attending headspace in terms of the service's impact on their mental health literacy. Results from staff also
indicate that engagement with LGBTQIA+ young people results in better mental health literacy than for
other groups of young people.
Figure 41: Responses from service and lead agency survey: how well does your centre provide services that support
mental health literacy for young people from priority cohorts?

Source: KPMG analysis of headspace service and lead agency survey

Notes: A total of 60 responses were received for this question. The survey response is a sliding scale ranging from ”worse” to “much
better”. The figure reports the average responses.

Relevant to meeting the needs of Aboriginal and Torres Strait Islander young people, the headspace
service and lead agency survey also highlighted that, in regional areas with high Aboriginal and Torres
Strait Islander populations, specific Aboriginal Social Emotional Wellbeing Workers are important. Staff
in these roles support adaptation of presentations and other resources for Aboriginal and Torres Strait
Islander young people. While these roles were highlighted as contributing strongly to improved mental
health literacy for young people accessing headspace, respondents to the service and lead agency survey
also noted difficulty recruiting staff for these roles in small communities. The need to be able to provide
services for community in community was also emphasised as a challenge for the headspace model.
For young people with disability, service providers noted that they had limited referral pathways with
disability services, and that disability service providers do not refer young people into headspace unless it
is funded on their NDIS plan, which is rare. The overall perception is that young people with disability
access other services instead of headspace, and some providers indicated they believe this is more
appropriate due to headspace clinicians not having experience working with dual diagnoses.
For young people from culturally and linguistically diverse backgrounds, headspace service providers
described the barriers in having access to culturally and linguistically diverse staff with the capability to
work with those arriving with significant trauma, with multicultural mental health issues, and with
different language and cultural skills. They also spoke about limitations in the capacity to undertake
outreach to culturally and linguistically diverse communities to promote service access.
School and university counsellors agreed with the importance of having local Aboriginal workers to
support outcomes for Aboriginal and Torres Strait Islander young people. In particular, communities with
large Aboriginal and Torres Strait Islander populations have benefited from outreach and work designing
the services with the community. This has built trust that, in turn, supports engagement with headspace by
young people and improved mental health literacy. Focus groups with counsellors also identified
challenges for culturally and linguistically diverse communities, including international students who have
not had the same education around mental health throughout earlier schooling as other young people from
the general population.
Effectiveness of headspace in increasing mental health literacy in ‘hard to reach’ groups
Overall, the evidence suggests there is some variation in how effective headspace is in supporting mental
health literacy in ‘hard to reach’ groups. Young person satisfaction data during the period indicates that
there is no significant difference between the improvements in mental health literacy reported by young
people who access headspace from different cultural backgrounds, with high levels of satisfaction across
relevant measures for all groups.
However, staff views of the success of the model in this domain showed concerns that Aboriginal and
Torres Strait Islander young people, culturally and linguistically diverse young people and young people
with disability all fare below the general population in terms of the service's impact on their mental health
literacy. At the same time, staff suggested that engagement with LGBTQIA+ young people results in better
mental health literacy than the general population of young people attending headspace. The importance of
having culturally appropriate staff in building trust and engagement to support mental health literacy were
highlighted.

How effective is headspace in increasing early help seeking for ‘hard to reach’
groups, including those who are at greater risk and less likely to seek help?
As described in section 5.1.2, early help seeking is defined in this evaluationas a young person engaging
with headspace when they are:
• under 21 years of age;
• at relatively low mental health risk status; and
• assessed as at less than the threshold stage of illness.
To examine the extent to which the headspace model is succeeding in contributing to increased early help
seeking behaviour in ‘hard to reach’ groups, relevant data and evidence was reviewed from across the
fieldwork activities conducted for this evaluation. These are described below, and include analysis of the
hMDS and survey responses from service and lead agency staff.
Administrative data from the hMDS
The hMDS collects demographic data of young people attending headspace services, asking them whether
they identify as Aboriginal or Torres Strait Islander, culturally and linguistically diverse or LGBTQIA+.
Considering the differences in early help seeking indicators for young people who identify as part of one
(or more) of these groups provides a useful lens to understand the extent to which the headspace model
supports young people from ‘hard to reach’ groups in increasing early help seeking behaviour.
While there is a substantial proportion of missing data against these categories, when looking at the data
related to the age of the young person attending headspace (at the first OOS in each EOC in the data
period), there are significant differences between groups. For those young people who identify as
Aboriginal or Torres Strait Islander, 84 per cent of young people attending headspace are aged under 21,
compared with 77 per cent of those who do not identify as within this group. For culturally and
linguistically diverse young people, 76 per cent are under 21 years of age, compared with 78 per cent of
those who do not identify as from a culturally and linguistically diverse background. For LGBTQIA+
young people, 74 per cent are under the age of 21, compared with 78 per cent of those who do not identify
as LGBTQIA+ young people. On this measure, the differences between the ‘harder to reach’ young people
and the general population are statistically significant, with Aboriginal and Torres Strait Islander young
people significantly more likely to be under the age of 21, and culturally and linguistically diverse and
LGBTQIA+ young people significantly more likely to be older than 21 years of age when attending a
headspace service.
Figure 42: Distribution of age by young person during 2019-20
Source: KPMG master dataset
Notes: See Appendix F for a description of how the master dataset is derived. Sample includes 90,110 young people with ongoing
episodes of care during 2019-20. Data labels are not included for categories with less than 0.5 per cent for clarity purposes.

When considering the hMDS data regarding each young person’s mental health risk status collected by the
clinical service provider as part of the intake and assessment process, young people who identify as
LGBTQIA+ are significantly less likely to present with low levels of risk factors. Relevant data for young
people from other ‘hard to reach’ groups (Aboriginal and Torres Strait Islander young people and
culturally and linguistically diverse young people) are not significantly different from those who do not
identify as members of these groups. As discussed in section 5.1.2 above, the mental health risk status
measure considers the presence of risk and protective factors, such as unstable or unsafe living conditions,
relationship problems and bullying, alongside the presence of symptoms of mental disorder such as
anxiety or depression. A young person presenting with either ‘no risk factors or symptoms of mental health
problems’ or ‘risk factors present’, indicating the presence of one or more situational factors making them
vulnerable to developing a mental health problem, would be considered to be undertaking early help
seeking.
Figure 43: Mental health risk status on initial OOS for all episodes of care during 201920
Source: KPMG master dataset
Notes: See Appendix F for a description of how the master dataset was derived. The initial occasion of service recorded was
examined in the main extract during 2019-20. The sample consists of 73,712 OOS.

Using stage of illness data collected by service providers to observe the extent of progression of a disorder
at a particular point in time, overall, there is little variation between ‘hard to reach’ groups and those who
are from the general population. Aboriginal and Torres Strait Islander young people are, however,
significantly less likely to be presenting in early stages of a disorder than young people who do not
identify as Aboriginal or Torres Strait Islander with ‘mild to moderate general symptoms of mental health
problems and/or high risk psychosocial stressors’ (e.g., bullying or relationship problems), meeting the
definition of ‘early help seeking’.
Figure 44: Stage of illness during initial OOS for all episodes of care during 2019-20
Source: KPMG master dataset
Notes: See Appendix F for a description of how the master dataset was derived. The initial occasion of service recorded in the main
extract during 2019-20 was examined. The sample consists of 73,712 OOS for 2019-20. Data labels are not included for categories
with less than 0.5 per cent for clarity purposes.

Perspectives of young people who use headspace

Some of the interviewed headspace users were in ‘hard to reach’ groups (culturally and linguistically
diverse young people, young people who identify as LGBTQIA+ , young people with disability,
Aboriginal and Torres Strait Islander young people). Observations from those from regional and remote
areas, who also experience service access barriers, have also been included. headspace users from these
groups discussed that it may have taken them a while to decide to seek help, and pointed to other young
people who were ‘hard to reach’ and resistant to seeking help:

“I feel like people that, I guess don't have enough courage or don't have family
members to help them reach out. They're not going to really come across it and then,
yeah, they're going to struggle. So I feel like they definitely would be a couple of
people that are missing out on the services and what it has to offer as well.”

Young people cited the importance of outreach in public spaces and schools as one way of engaging with
harder to reach people, as well as making sure that people know it is a free service:

“because I know a lot of people think, "all this counselling and all that, must cost a
fortune." It didn’t cost me a cent.”
One young person commented that outreach programs, where headspace staff visited young people in their
homes, helped to support early help seeking from ‘hard to reach’ groups:

“And that’s where their kind of outreach comes in as well so that they’re going on to
those young people as kind of a safe space that will come to their homes instead of
bringing an Indigenous young person who’s lived on a community their entire life,
they’re not going to enjoy being in four walls in a sterile environment.”
Family attitudes that downplayed distress due to mental health issues were cited as preventing young
people from seeking help, therefore young people from culturally and linguistically diverse backgrounds in
particular thought headspace could educate families to reduce stigma.
Young people in the reference groups reported that headspace successfully engaged with the young people
in the LGBTQIA+ community. The specific groups run by headspace meant that they could meet with and
connect other young people in a space where they felt comfortable and treated with respect:

“The first time that I went to the LGBTQ group, first up I was asked what my
pronouns were which was something that was really unique and different that I really
appreciated. So off the bat you have appropriately addressed someone and I think
that was very important.”

The young people noted that waiting times could deter ‘hard to reach’ clients from accessing help,
especially if they had taken the difficult step to ask for help:

“Sometimes you can get yourself to a point you need help right now, but if you've got
to wait three months you're just not going to be motivated enough and you're
probably just going to go downhill even more.”

Perspectives of headspace service providers

Responses from staff at services and lead agencies indicated that the majority thought the headspace model
was less effective in encouraging early help seeking for Aboriginal and Torres Strait Islander young
people, culturally and linguistically diverse young people and young people with disability. As with mental
health literacy, respondents felt that the outcomes were stronger for LGBTQIA+ young people than for
those from the general population of young people attending headspace. This contrasts with administrative
data from the hMDS, however, it indicates that LGBTQIA+ young people are actually less likely to
present with low levels of mental health risk factors than the general population of young people attending
headspace.
Figure 45: Responses from lead agency and headspace services survey to ‘how well does your centre provide services
that support early help seeking for young people from priority cohorts?’

Source: KPMG analysis of headspace service and lead agency survey

Notes: A total of 60 responses were received for this question.

Effectiveness of headspace in increasing early help seeking for ‘hard to reach’ groups
The data regarding early help seeking for different cohorts of young people indicates that there are some
variations between groups on different indicators. Aboriginal and Torres Strait Islander young people are
significantly more likely to be under the age of 21, and culturally and linguistically diverse young people
and LGBTQIA+ young people are significantly more likely to be older than 21 years of age when
attending a headspace service, compared with the general population of young people attending headspace.
At the same time, mental health risk data is broadly the same for young people across all backgrounds and
cultural groups, except for those who identify as young LGBTQIA+ people, who are significantly less
likely to present with low levels of risk factors. On measures of stage of illness, Aboriginal and Torres
Strait Islander young people are significantly less likely to be presenting in early stages of a disorder than
young people who do not identify as Aboriginal or Torres Strait Islander.
As with responses regarding mental health literacy and young people from ‘hard to reach’ groups, staff at
services and lead agencies felt that the headspace model was less effective in encouraging early help
seeking for Aboriginal and Torres Strait Islander young people, culturally and linguistically diverse young
people and young people with disability but that LGBTQIA+ young people were more likely to engage in
early help seeking than young people from the general population attending headspace. This is in contrast
with the data, which indicates this group is either the same as the general population or ‘slower’ in seeking
help, depending on the indicator. LGBTQIA+ young people are more likely to be over the age of 21 and
less likely to present with low levels of mental health risk factors than the general population of young
people presenting at headspace.
headspace users from these ‘hard to reach’ groups discussed that it may have taken them ‘a while’ to
decide to seek help, with family attitudes reducing young people’s help seeking behaviour. These young
people thought that outreach in public spaces and schools, as well as highlighting that headspace is a free
service, are useful ways to counteract this barrier.
How effective is headspace in increasing access for ‘hard to reach’ groups?
There is strong evidence across the literature that Aboriginal and Torres Strait Islander people,
LGBTQIA+ people, refugees, asylum seekers, people from culturally diverse backgrounds, and people
living with disability have barriers to accessing mental health support. These may take the form of stigma,
discrimination, racism, persistent socioeconomic disadvantage, lack of knowledge, loneliness, and
,,
trauma . A lack of information due to language barriers and culturally capable services can also be a
181F 182F 183F

deterrent to access care. Living in rural and remote areas carries a set of unique risk factors for mental
illness, including isolation and environmental events such as droughts and bushfires. As discovered by the
Royal Flying Doctor service survey in 2018, those living in regional and remote areas may have access to
a very small number of services, if any. This can result in long wait times, unsuitability of treatments, a
lack of services in a community, or people being required to travel significant distances to receive mental
health services, incurring additional expenses and time . 184F

Understanding the extent to which young people from ‘hard to reach’ populations are accessing headspace
services is important for this evaluation in establishing an assessment of the effectiveness of the headspace
model in meeting its key objectives. hAPI data, service and lead agency survey data and discussions with
stakeholders during deep dive fieldwork highlighted key barriers, enablers and achievements in this area.
Perspectives of young people who use headspace
As discussed above, culturally and linguistically diverse young people reported familial opprobrium as a
hurdle to seeking help due to stigma associated with admitting mental health issues. In addition, young
people from culturally and linguistically diverse backgrounds reported they would like to see more cultural
diversity among headspace staff, especially so their family backgrounds and religious considerations could
be better understood. Aboriginal and Torres Strait Islander young people spoke of their challenges around
depression, drugs, and abuse and that they sometimes did not access services due to stigma.
Aboriginal and Torres Strait Islander young people had a range of views, including that headspace could
be more culturally competent (and include more First Nations staff), liase with ACCHSs, and also detailed
culturally positive practices:

“I think in terms of Aboriginal and Torres Strait Islander stuff you’d need to sort of
probably have strong connections with ACCHOs, the Community Controlled Health
Organisations, because that’s where Aboriginal and Torres Strait Islander people
go.”

and

“They [headspace] allow different support people into those sessions wanting to
bring a mum or a dad, any support person, they go on to community and speak to the
Elders about what would be best moving forward for their young people, but also in
collaboration with the young people, so ultimately at the end of the day those young
people needing support get the best outcome and it’s also done in the safest way that
means a community can respond if something happens.”

and

“I would say if we had more community engagement, yes it would be amazing. As of

right now community engagement in community seems to be pretty good.”
LGBTQIA+ young people had mainly positive encounters with headspace, with some exceptions, but
complaints were about a lack of what they perceived as good enough clinical support, rather than issues
related to sexuality. People with disability were generally positive about headspace, but there are limits to
what headspace can do in relation to some conditions, including Autism Spectrum Disorder which requires
specific diagnostic tools and specialist support.
Remote and rural residents referred to the small town effect where ‘everyone knows everyone’s business’
and some cited stigma in relation to seeking help, but no users from this subcohort reported any specific
issues with accessing headspace or quality of service.
There were hurdles to overcome in outreach and bringing First Nations young people into the service:

“I don't know where to start because most Aboriginals here in Murray Bridge are
isolated and too scared to go to places.”
Bringing cultural competence to outreach was discussed:

“That’s where their kind of outreach comes in as well so that they’re going on to
those young people as kind of a safe space that will come to their homes instead of
bringing an Indigenous young person who’s lived on a community their entire life,
they’re not going to enjoy being in four walls in a sterile environment.”

Many of the users who were interviewed or in focus groups were from ‘hard to reach’ groups. While they
had accessed the service, they agreed that it was sometimes despite attitudes of peers or family in
particular.Cultural and gender diversity in staffing was identified as important:

“Some of the stuff that may like restrict kids from coming, I think… like how I wanted
someone that was a female. Sometimes, it can be a bit hard because of who's like on
the branch, who works there. I think that's a bit of a barrier.”
Individual willingness to seek help was also identified as a barrier – even with outreach and referral,
ultimately, the young person needs to agree to seek assistance:

“I think it’s more just people aren’t willing to accept help if they think it’s too hard or
they have all these different ideas and different expectations that are negative, and
it’s not like that at all. So I think just people’s own perceptions of getting help, I
 guess, is probably the biggest disadvantage they have.”
It was difficult for the young people in the reference group to assess whether headspace was effective in
increasing access for ‘hard to reach’ groups with the exception of LGBTQIA+ young people. In one area,
headspace had organised a festival for LGBTQIA+ young people. The festival aims to celebrate and raise
awareness of the LGBTQIA+ young people. A headspace youth group for LGBTQIA+ young people in
the area provided a supportive environment for young people to meet and access information:

“headspace put on a festival for LGBTQIA+ people to basically, I don’t know,

celebrate their existence in [the area] because there is, you know, a lot of stigma and
quite a lot of issues around our representation….and headspace also has a group a
youth group for LGBTQIA+ people to get together and either just relax or get
educated about stuff. And yeah, headspace is catering to a lot of LGBTQIA+ folks, at
least the young people in the [area].”

Perspectives of headspace service providers

Respondents had similar views about increased access for the ‘hard to reach’ cohorts as they had for
increased mental health literacy and improved early help seeking. Young people from Aboriginal and
Torres Strait Islander backgrounds, culturally and linguistically diverse cohorts and young people with
disability were seen to have worse access rates to headspace services compared with those from the
general population of young people attending headspace. LGBTQIA+ young people were perceived to
have better rates of access than all other groups, including the general population of young people
attending headspace.
Figure 46: Responses from the lead agency and headspace service survey to ‘how well does your centre provide
services that support access for young people from priority cohorts?’

Source: KPMG analysis of headspace service and lead agency survey

Notes: A total of 60 responses were received for this question.

Regional and remote challenges

In response to a prompt in the service and lead agency survey to describe the barriers and enablers to
support these cohorts, a common theme was related to challenges for rural and remote services. Issues with
the other parts of the service system were raised for regional areas with limited capacity of tertiary
services, bulk billing services and affordable psychiatry. The ability to attract specialist psychologists,
AOD workers, vocational workers and GPs were all identified as difficult in remote areas. Turnover and a
limited overall pool of workers across providers and PHN roles was also highlighted.
One respondent also described challenges they face in a regional area with the headspace service funding
model, where outreach activities to take services to remote communities are not funded, but are expected
by stakeholders across their local area.
Other responses again highlighted a concern about insufficient funding for salaried staff, including
community engagement of staff. Challenges around finding staff with the right skillset or cultural
background were particularly salient for regional and remote staff.
Perspectives from deep dive fieldwork
Deep dive consultations and discussions with Aboriginal and Torres Strait Islander community
organisations within deep dive locations illustrated key themes required to engage and assist Aboriginal
and Torres Strait Islander young people. Stakeholders emphasised the need for young people to be able to
see people like themselves in the staff at their local headspace, and for it to feel like a safe and culturally
appropriate place for them to seek help. In consultations with metropolitan services, stakeholders reported
the importance of having members of staff from a wide range of cultural backgrounds, and of the
important role they play in reducing stigma and building mental health literacy for different communities.
The headspace model promotes centre-based support, and emphasises the importance of making each
service look culturally appropriate and welcoming to members of the Aboriginal and Torres Strait Islander
community. Services display Aboriginal and Torres Strait Islander flags, and draw on local culture, art and
language to show visible signs of welcome to the local Indigenous community.
Services in areas with larger Aboriginal and Torres Strait Islander populations also prioritise having staff
from the local Indigenous community, and engagement with Elders and well-known local Aboriginal and
Torres Strait Islander people on the consortium in an advisory capacity, so they can be seen to be endorsing
the use of the headspace services for their people.
Stakeholders in regional and remote areas described the centre-based model as a barrier to Aboriginal and
Torres Strait Islander young people seeking support, due to high levels of self-consciousness and stigma
associated with mental illness. Indigenous models of care, centring the person within their family,
community and culture were also described as more effective in assisting Aboriginal and Torres Strait
Islander young people than a more individual-centric model privileged in mainstream clinical practice.
Where outreach is conducted to Aboriginal and Torres Strait Islander communities, by trusted service
providers without the need for appointments or to be seen to be seeking help, barriers may be reduced and
positive outcomes supported.
For young people who identify as sexuality or gender diverse, headspace has become a brand which
provides a safe space for them to seek support, connect with peers and manage their wellbeing.
Stakeholders consistently recognised this as a strength of headspace. Consultations indicated that
headspace had achieved this success in improving access of this group through its brand recognition, social
media presence and through peer-to-peer networking.
Administrative data from the hMDS
While services described ongoing work to continue to build and maintain engagement with the local
Aboriginal and Torres Strait Islander community, hMDS data shows a slight decrease in the last year in the
proportion of Aboriginal and Torres Strait Islander young people accessing headspace services. Overall,
data indicate that headspace is reaching culturally diverse young people, as culturally and linguistically
diverse representation has increased (to 10 per cent of all clients), and Aboriginal and Torres Strait Islander
headspace clients are a higher percentage (seven per cent) than this age group population as a whole (six
per cent).The proportion of Aboriginal and Torres Strait Islander young people has remained steady, with a
higher proportion accessing headspace services than their share of the total population.
Table 50: Share of young people accessing headspace who are Aboriginal and Torres Strait Islander

Financial Year Percentage of young people who are Percentage of young people who are
Aboriginal and Torres Strait Islander - Aboriginal and Torres Strait Islander -
hMDS ABS , 185F 186F

2015-16 8% 5%
2016-17 8% 5%
2017-18 8% 5%
2018-19 8% 6%
2019-20 7% 6%
Source: KPMG master dataset. See Figure 70 in Appendix F; Australian Bureau of Statistics (ABS) estimated resident population
statistics

Over recent years, the proportion of young people attending headspace who are from culturally and
linguistically diverse backgrounds has steadily increased, as shown in Table 51 below. This suggests
increased help seeking from these young people, which will require headspace to provide increasing
numbers of young people with culturally appropriate support.
Table 51: Share of young people accessing headspace with culturally and linguistically diverse backgrounds

Financial Year Percentage of young people with culturally and

linguistically diverse backgrounds
2015-16 8%
2016-17 8%
2017-18 8%
2018-19 9%
2019-20 10%
Source: KPMG master dataset. See Figure 70 in Appendix F.

There has been a relatively high proportion of headspace clients who identify as LGBTQIA+ accessing
headspace services over time, with the proportion steadily growing over recent years.
Table 52 :Share of young people who identify as LGBTQIA+

Financial Year Percentage of young people who identify as

LGBTQIA+
2015-16 18%
2016-17 19%
2017-18 19%
2018-19 20%
2019-20 25%
Source: KPMG master dataset. See Figure 70 in Appendix F.

Effectiveness of headspace in supporting access to services for ‘hard to reach’ groups

A range of data and evidence indicates that headspace is effective in supporting access to services for ‘hard
to reach’ groups, although this continues to be an ongoing challenge for the headspace model.
Administrative data from the hMDS, shows that, over time, access rates have slightly improved for ‘hard
to reach’ groups, however those working within headspace suggested the service is less effective in
supporting the access rates of Aboriginal and Torres Strait Islander young people, culturally and
linguistically diverse young people and young people with disability. LGBTQIA+ young people were
again perceived to be better supported, with higher perceived rates of access than all other groups, which is
upheld by administrative data on young people attending headspace.
Feedback from young people highlighted the continued importance of having staff from the young
person’s cultural group, and the ongoing need to work to reduce stigma and build trust in order to support
access from ‘hard to reach’ groups.
The headspace model does not achieve the same results for ‘hard to reach’ groups compared to the general
population of young people.

D.5 How well does headspace advocate for and promote youth
mental health and wellbeing in their communities?
D.5.1 Advocacy and promotion activities
In support of the various intended outcomes of the headspace model, a key component of activity focuses
on engaging with communities. The headspace program logic lists the following activities as part of this
work:
• promoting headspace services to local community and services, and promote early help seeking for
young people aged 12 to 25;
• facilitating engagement and participation with young people and their families to better understand
community needs;
• engaging with GPs, schools and other local organisations to better understand community needs; and
• delivering community awareness activities including psycho-education, mental health literacy and stigma
reduction activities.
In order to assess how well headspace advocates for and promotes youth mental health and wellbeing in
their communities, feedback on these activities was sought through a range of data collection activities.
These include interviews with Youth Reference Groups, interviews and focus groups with young people,
surveys of service and lead agency staff, and discussions with staff, GPs and other stakeholders at a
number of services during deep dive fieldwork.

Evidence of the contribution of headspace to advocacy and promotion of youth

mental health
Perspectives of young people who use headspace
Most Youth Reference Group members interviewed endorsed the way headspace staff actively promoted
the service on social media and through outreach in schools and stalls in the community:

“They’re [the local headspace site] very active on their social medias, Instagram - I
know they’ve got regular content there.“

Having a regular presence on social media and promotional activities, such as leaflets and groups in
schools and booths in shopping centres, was seen by this group as increasing awareness of headspace and
mental health issues for young people, thereby increasing mental health literacy. These activities were seen
to contribute to young people gaining an understanding of how they felt and why, and where they could
access support.
Perspectives of young people who do not use headspace
In discussions about headspace and its role in advocating and promoting mental health in local
communities surrounding centres, young people were able to readily identify occasions where they had
observed a headspace presence at community events, for example at schools and university oweeks, as
well as on social media.
Perspectives of headspace service providers
As described above, service and lead agency survey responses indicated strong levels of confidence from
staff that their service is successful in increasing mental health literacy. When asked to describe key
enablers of this, responses identified broader community engagement by the headspace service as a key
aspect of their observed success in this area. Examples included activities such as social media campaigns,
education and awareness activities with local schools, and the establishment of partnerships with local
councils, universities and colleges. Community Development Officers were highlighted as particularly
critical to this work, however some services identified only having funding for 0.6 FTE for this role, which
they consider to be insufficient.
Perspectives from deep dive fieldwork
Consultations with GPs and consortium members from surrounding community services as part of the
fieldwork for this evaluation elicited broadly positive views about the work headspace staff undertake to
engage with schools and to drive and participate in community events and mental health awareness raising
activities. Fieldwork teams observed services delivering on a detailed calendar of events and activities
regarding mental health and wellbeing, actively participating in pre-existing events as well as driving the
planning and implementation of specific events of their own, for example around headspace week.
Services also described targeted outreach to different segments and cohorts in their local communities, for
example engaging with church youth groups and with Police PCYC programs for young people.

Effectiveness of headspace in supporting youth mental health through advocacy

and promotion activities
There is limited data beyond anecdotal reports of the value of these activities, and the extent to which they
are occurring is not measured through the hMDS or other means. However, stakeholders report that
headspace services are active in advocacy and promotion, and highly visible in their local communities.
Work to promote mental health literacy and help seeking with schools, universities and community
organisations more broadly received positive feedback.
The evidence indicates that headspace is effective in supporting youth mental health through advocacy and
promotion activities. As noted above, however, this stream of work is not captured in data collection
across the hMIF or hAPI systems, and services suggested it is underresourced.

D.6 To what extent has headspace reduced stigma associated with

mental illness and help seeking for young people, their
families and friends, and the community?
D.6.1 Stigma reduction
Table 53 Overview of objectives of headspace for stigma reduction

Objective Short-term impacts Medium-term impacts

Reducing stigma • Young people, their families and • Young people, their families and
associated with mental communities (living near communities are better able to
health and mental headspace centres and identify when someone needs
illness - the fear or satellites) have improved help, and support appropriate,
embarrassment of attitudes towards mental health early help seeking
seeking help for mental and mental illness (stigma
health and wellbeing, • Earlier identification and
reduction)
and the negative treatment of emerging mental
judgment of and lack of health problems for young
empathy for those that people
do • Young people increase help
seeking behaviour for mental
health and wellbeing issues

As can be seen from the headspace objectives and impacts in the table above, stigma in this context is the
fear or embarrassment of seeking help for mental health and wellbeing, and the negative judgment of, and
lack of empathy for, those who do. It acts as a blocker in the headspace program logic, preventing young
people from being able to identify when they need help and seeking that help early.
National research into stigma indicates that most people in Australia with mental illness report
experiencing stigma, however the severity, nature, and experience of stigma vary depending on factors
such as mental illness type, age, gender, and cultural background . About 29 per cent of people with mental
187F

illness reported discrimination or unfair treatment in the past year, as opposed to about 16 per cent of those
without mental illness. People with severe mental illnesses are likely to face high levels of stigma,
according to the 2011 National Survey of Mental Health Literacy and Stigma, although the nature of
stigma differs among illnesses. The impact of stigma may include preventing people who suffer from
mental illness from being able to engage socially or feel included. This stigma can lead to discrimination,
social exclusion and a reluctance to seek care . 188F

In order to examine whether headspace has been associated with a reduction in mental health related
stigma, this evaluation sought the views of headspace service and lead agency staff through both survey
and fieldwork methods, as well as reflections from school and university counsellors and young people
who do not use headspace, to gauge their views on how effective headspace has been in this domain.

Evidence of the contribution of headspace to reduced stigma associated with mental

illness and help seeking
Perspectives of young people who use headspace
Young people in the reference groups from ‘hard to reach’ groups reported that barriers to seeking help
included stigma, fear that service providers would not listen to them or break confidentiality and tell their
parents about their mental health issues:

“And yeah, there’s just this fear that the person you get won’t listen to you or they’ll
tell – one of the biggest fears is telling parents stuff. I’ve seen a lot of people very
scared of their parents being told things that they don’t want their parents to be told,
like the confidentiality being broken”

and

“Like headspace or many other health services. They're too nervous with family
issues or have a feeling that they're going to get judged by their friends and all that”

Perspectives of young people who do not use headspace

Discussions with young people from culturally and linguistically diverse backgrounds indicated that they
felt there was limited understanding of the cultural sensitivities around mental health, and that this was
true of the headspace model as well as of mainstream services more generally.
Perspectives of headspace service providers
The strong levels of confidence described above from service and lead agency survey responses regarding
increasing mental health literacy are also an indication of the extent to which respondents consider their
headspace service to be reducing stigma. In response to the prompt:

Based on your observation of young people at your headspace service, how well
does the service increase mental health literacy? For example, building
understanding of where to seek support, understanding of mental ill health and
treatments, and reduction of stigma to support help seeking.
A total of 93 per cent of service and lead agency respondents selected ‘very well’ or ‘well’.
Interviews with school and university counsellors indicated a general recognition that mental health
literacy has improved over time for young people in Australia, that stigma about mental illness has been
reduced and help seeking is widely encouraged, with a tendency to talk more openly about mental health
today. There was a view from participating counsellors that headspace resources contribute to increasing
mental health literacy and reducing stigma, including a general improvement in young people’s knowledge
of how to seek help for their mental health and wellbeing. Discussions acknowledged that these observed
changes could not be attributed to headspace alone, but also to broader work happening in schools, social
media and other organisations as well.
School and university counsellors also identified challenges for culturally and linguistically diverse
communities related to stigma. Discussions noted that, within some cultural groups, stigma has an ongoing
impact on menta health help seeking behaviour.
When considering the extent to which headspace has been successful in reducing stigma for family, friends
and the community, the data is less conclusive. Qualitative evidence from interviews and discussions at a
range of sites indicated that, while some progress is being made in reducing stigma in young people, this is
due to a range of factors including the work of schools and the media more broadly in highlighting and
normalising mental health help seeking.
Discussions also indicated that, for some families and in some segments of the community, stigma around
mental health help seeking continues to be strong, and services are continuing to focus efforts, including
outreach, recruitment and other engagement strategies, to reduce stigma and encourage support of mental
health help seeking. A number of cultural groups were discussed in these fieldwork conversations, along
with the particular challenges for young people from some culturally and linguistically diverse
backgrounds where mental illness is not easily accepted or understood.

Effectiveness of headspace in reducing stigma associated with mental illness and

help seeking
Overall, the evidence collected suggests that stigma reduction activities undertaken as part of the
headspace model are effective and are a continued focus of headspace services, as they are for services and
organisations across the mental health sector. Despite this, qualitative data indicates that stigma continues
to be an issue, particularly for young people from culturally and linguistically diverse or Aboriginal and
Torres Strait Islander young people. Young people discussed how schools and the media are also working
to combat stigma, and that the work headspace does is one of many things happening to help in its
reduction. On balance, views are positive that stigma around mental health and mental illness is reducing,
and those close to the model believe headspace has made a positive contribution.

D.7 How effective is headspace in improving pathways to care for

young people through service integration and
coordination?
D.7.1 Service Integration and Coordination
Table 54 Overview of objectives of headspace for service integration and coordination

Objective Short term-impacts Medium term-impacts

Improving the pathway • headspace services deliver • Young people and families
to care through service services across and beyond experience more streamlined
integration and four core streams (mental and less fragmented pathways
coordination - bringing health, physical health, of care
services together to alcohol and drug use,
function as one, • The local service system for youth
vocational programs)
providing a seamless mental health is better
service experience for a • headspace services deliver integrated and coordinated
young person integrated/coordinated care

The headspace program logic sets out two core objectives related to service integration and coordination,
as outlined in the table above. Medium-term impacts for young people and families and the local service
system are also expected.
Integration refers to individuals and organisations in different areas and sectors working together and
aligning their practices and policies to deliver high quality mental healthcare and achieve good outcomes . 189F

In the headspace model, service integration refers to bringing services together to function as one,
providing a seamless service experience for a young person, particularly if they require care involving
multiple service providers and supports . 190F

In the context of mental health services, there are two ways services can typically be integrated – vertically
and horizontally. Vertical integration refers to how services at different levels of healthcare, for example
primary, secondary and tertiary, work together to deliver on this. Horizontal integration refers to how
services from different sectors or sub-sectors work together, such as physical and sexual health and mental
health services. Vertical or horizontal integration may also occur between mental health and other service
systems, such as housing, or employment . 191F

Ensuring people, including young people, have access to services and supports they need where and when
they need them is critical to a well-functioning mental health service system. However, the Productivity
Commission has identified that nationally, there are challenges with current pathways between care and
service integration across the entire mental health service system. These challenges include:
• the complex and disjointed nature of the mental health service system;
• a lack of information sharing and coordination between services, impacting on outcomes; and
• some services providing overlap in some areas and for some cohorts of people, with no services for other
groups .192F

This section explores headspace’s effectiveness in improving pathways to care through service integration
and coordination.

Evidence of headspace’s contribution to improving pathways to care though service

integration and coordination
Perspectives of young people who use headspace
headspace users reported receiving appropriate referrals to dieticians and other professionals as well as
assistance with practical matters relating to housing, income, and employment, both in-house and via
referral from headspace services. Referrals and links to psychiatrists were reported by some headspace
users to be more problematic, indicating they were not able to be linked with a psychiatrist where needed
from the headspace service. While many started their mental health journey with headspace, some ended
up being referred to, or choosing to, seek help from a private practice psychologist and/or psychiatrist. The
experience of most headspace users receiving appropriate referrals where required is consistent with the
Colmar Brunton survey undertaken for headspace National, where 85 per cent of young people indicated
that it was extremely important that headspace would connect them with other services if they needed
them . Similarly, 95 per cent of young people’s parents agreed it was extremely important that headspace
193F

connect their child to other services as required .

194F

headspace’s affordability is a key asset. Cost was a major barrier in referrals for some headspace users. For
example, one user was referred elsewhere for an expensive test (e.g., for Autism Spectrum Disorder). Cost
barriers were also identified as limiting access to psychiatrists. In some cases, there was some frustration
from headspace users that headspace could not support them with these services and they were referred to
a more expensive service as part of a coordinated care model. This is consistent with the headspace model
as it is focused on young people with mild to moderate, high-prevalence mental health conditions,
however the ‘no wrong door’ approach has created a level of expectation for some young people.
headspace users also indicated that they used alternative services, such as their GP, to provide integrated
care, referrals and care coordination, rather than relying on headspace for this:

“I spoke with the GP again and we kind of decided that he could have referred me
back to headspace for more sessions. But we kind of decided that, that probably
wasn't quite what I needed. And so, instead he referred me to a different psychologist
out at [suburb].”
A small minority of headspace users interviewed also felt that headspace did not understand what their
problem was, so referred them to the wrong type of professional:

“It didn't do much simply because it wasn't - they didn't really know what the root of
the issue was.”

Young people representing Youth Reference Groups from deep dive locations reported that staff from
headspace were ‘constantly connecting with other services’. They noted that headspace and the mental
health sector were trying to improve integration and coordination between different mental health services
to facilitate the pathway for young people through the service system, as well as with broader social
supports such as those available through Centrelink and Medicare:

“I know from just connecting with other services there’s a lady who works for
Medicare sometimes comes to headspace and has appointments, like she’ll do one
day a month or something and then the counsellors if they think they have a young
person who needs to see the Medicare lady they can book that for them and they can
have a face-to-face with someone from Medicare to sort some stuff out. So I think
that’s really good because she comes to the headspace building so they don’t even
have to be sent somewhere else, they can just come back to headspace to see her.”

The Youth Reference Group members also reported that processes were in place at headspace to support
young people through referrals to other services, such as private psychologists, to support effective care
coordination:

“Having to approach people because it's daunting. It's a daunting thing to have to
do, especially if you have anxieties or stuff like that. But I think that headspace works
really well with the exterior services that other places provide and they really
communicate very well and they do provide a really good level of confidentiality.”
From the experiences of Youth Reference Group members, although headspace ‘may not be equipped to
handle extreme cases of need or support, they are there to help with the due process of getting you that
kind of support’. Youth Reference Group members spoke about young people with self-harm or suicidal
thoughts and how headspace staff organised and supported them through the referral and transition
process, for example taking them to the hospital emergency department or contacting CAMHSs:

“I can say from peoples’ experience that I know, have gone from either having really
mild anxiety and depression to having those kinds of thoughts and they did have the
due process of going from headspace to CAMHS, but the person from headspace,
their psychologist or whoever they were with, was with them the entire time.”

Perspectives of young people who do not use headspace

Young people who completed the young people’s survey were asked about other services they may have
accessed to support their mental health. Young people who had not accessed headspace services were
asked if they had sought support from their GP for mental health. Of the 1,432 young people who had not
used headspace services, and who answered this question, 537 indicated they had sought support from
their GP.
These young people were asked a follow up question about other services their GP had referred them to for
additional support. Twelve per cent of these young people reported receiving a referral from their GP to
both headspace services, as well as other mental health services. Four per cent of young people reported
receiving a referral to headspace services only, while 81 per cent of young people indicated that their GP
had referred them to other services but not a headspace service. Figure 47 below demonstrates this split
between responses. It is not known why these 16 per cent of young people who were referred to headspace
chose not to access the services.
Figure 47: Young people responses to other services their GP referred them to (young peoples’ survey)

Source: Evaluation survey of young people who have and have not accessed headspace services

Non-headspace users interviewed also identified referral pathways to headspace they have experienced in
the past. Most commonly, non-headspace users reported their GP referring them to headspace services,
especially as a fee alternative to private psychology services. School counsellors were also a source of
referrals for non-headspace users. However, a small minority of non-headspace users reported not using
these referrals or headspace services, as it would mean to repeat their story to another support person, with
limited care coordination available to prevent this. These non-users reported a desire for strengthened care
coordination that would prevent them being required to repeat their circumstances between services.
Perspectives of headspace service providers
There were consistent views from stakeholders across deep dive locations that headspace services
undertake a range of activities to support integration with other services and coordination of care for young
people. These activities include case coordination for young people; establishment of relationships with
other local services such as NDIS access workers, cultural healing services, and other family-based
supports; and direct referrals to other services. These stakeholders also indicated how this work was an
ongoing and important aspect of ensuring access to services for young people. This is supported by other
evaluation work undertaken by headspace National, including the Colmar Brunton survey. Of the 47 lead
agencies that completed the Colmar Brunton survey, 96 per cent agreed that headspace encourages broader
service collaboration, and 85 per cent agreed headspace improves coordination of local services . 195F

Similarly, independent chairs and consortium members working with headspace services indicated that the
services have a positive impact on other mental health services , including that:
196F

• 83 per cent agreed headspace encourages broader service collaboration;

• 82 per cent agreed headspace services support warm referrals;
• 84 per cent agreed headspace services have a positive impact by supporting continuity of care; and
• 83 per cent agreed pathways to care have improved for young people experiencing mental health
problems since a headspace service was introduced in their community.
Case coordination work was consistently raised by deep dive representatives as critical to the success of
the headspace model in supporting service integration and better outcomes for young people. Case
coordination supports access for young people, not only to headspace but also to other services to which
they may be referred.
Services invest time in building relationships with other local services, including local mental health
services, and other support services that contribute to aspects of a young person’s wellbeing. The level of
investment in these relationships differs between services, and depends on the capacity of other services to
engage, loss of relationships when other organisations lose time-limited grant funding, and the focus of the
management of individual headspace services on this relationship building versus other elements of
service delivery. Relationships and resulting service integration with psychosocial supports, including
cultural healing, NDIS access, and family supports, enables headspace to facilitate cross-referrals.
These examples illustrate headspace’s effectiveness in supporting improved pathways to care and service
integration. However, there were differences reported by deep dive stakeholders between metropolitan and
regional and remote services with respect to service integration. The availability of other services, and
their capacity, particularly in non-metropolitan locations, has impacted the ability of some headspace
services to support integration. Where services do not have capacity to take on new clients, this impacted
referrals made by headspace services, and opportunities for care coordination and service integration for
young people.
Service integration and coordination to support pathways to care is also impacted by a number of barriers.
The survey of representatives from services and lead agencies included a specific question relating to
barriers to care pathways. The question asked them to indicate whether or not a list of predeveloped factors
were barriers to supporting care pathways for young people. These factors were included based on barriers
identified by stakeholders from deep dive services engaged. Figure 48 provides a breakdown of the
number of survey respondents who indicated each factor was a barrier they experience. The most common
barriers identified were waitlists and lack of capacity in local referral services, followed by limited local
services for specific conditions or treatment needs, and lack of local services to meet more acute needs.
With respect to case coordination in particular, deep dive representatives also described challenges in
documenting and demonstrating the volume of time spent on coordination activities, and balancing these
activities with direct clinical services for workers within headspace services, especially where the
headspace services rely on MBS billing to support services. Case coordination is also more challenging for
young people with more severe distress levels and complex mental health support needs
Figure 48: Barriers to supporting pathways to care identified by service and lead agency representatives

Source: KPMG analysis of the headspace service and lead agency survey

Perspectives of other external stakeholders including school and university counsellors,

GPs and PHNs
Schools and university counsellors from across Australia indicated that relationships and referral pathways
between their services and other external services within the community were critical to support effective
outcomes, and that headspace played a role in this. Ninety-eight per cent of school principals and
wellbeing coordinators indicated that being able to connect students to other services if they need them
was an important part of the headspace model in previous research undertaken by Colmar Brunton for
headspace National.
However, when asked whether headspace has improved service integration, a lower proportion of all
principals and wellbeing coordinators indicated their support. Sixty-nine per cent agreed that headspace
services strengthened relationships between service providers and schools, and 67 per cent agreed that
headspace services improved the coordination of local services. Government schools were less likely to
agree that headspace services have contributed to these improvements (66 per cent for strengthened
relationships and 64 per cent for improved coordination) . 197F

There was mixed feedback from counsellors as part of focus groups completed specifically for this
evaluation, in particular university counsellors, regarding the referral process for headspace services.
Following a referral, some counsellors described there being limited communication regarding what
support the young person was receiving, especially while on a wait list for headspace services, and
whether the young person would benefit from ongoing support from the school or university while waiting
for headspace support. Other counsellors indicated the referral process was ‘smooth and easy to use’,
especially where the counsellor was engaging with headspace directly to support the young person’s access
to the service.
Some counsellors also identified challenges with service integration and care coordination for young
people in the ‘missing middle’. Counsellors were uncertain about how to support young people who did
not have a severe enough mental health problem for local CAMHSs or CYMHSs, but who were not within
the mild-moderate target group of headspace services. A small minority of counsellors indicated there was
limited communication regarding where else a young person might be referred if the headspace service
indicated it could not support the young person. Some counsellors also discussed the challenge of current
wait times within headspace services as a deterrent to referrals, especially where there was limited
information provided back to the school or university about what other support was available to the young
person during their wait for clinical services.
PHN representatives attending an evaluation data collection workshop were asked to rate how well
established headspace service pathways are with particular primary care and mental health services, on a
five point scale from ‘not established’ to ‘well established’. Pathways were rated as the following:
• pathways with GPs were rated in the middle between not established and well established (3 out of 5);
• pathways with state and territory mental health programs, such as CAMHSs and CYMHSs were rated
closer to not established than well established (2.8 out of 5); and
• pathways with other mental health services were rated closer to well established than not established (3.3
out of 5).
PHNs outlined two key enablers for headspace services in support of service integration and care
coordination – formal agreements with services and relationship building. Most PHNs indicated headspace
effectively supports where there are strong Memoranda of Understanding or Service Level Agreements
with external organisations, to make clear agreed protocols and roles and responsibilities in place between
services. Similarly, relationships were a key enabler identified by most PHNs to support pathways to care
and service integration, especially in regional and remote locations where there are fewer services
available to support young people.
PHNs also identified a range of challenges which impact on the ability of headspace services to support
integration and care coordination in improving pathways to care, and which are often outside the control of
headspace services. These include:
• Relationships between headspace services and tertiary mental health services, such as CAMHSs, is
impacted by limited capacity within tertiary services to engage in these activities with significant
clinical work and wait lists, and changing eligibility criteria to access TMHSs.
• The capacity of other services impacts service integration, even where strong relationships exist, as
young people may not be able to access the service at all, preventing integration and care coordination
from occurring.
• Limited infrastructure to support shared records between services, reducing the level of care coordination
a young person may receive.
• Inconsistent eligibility criteria across other services and significant gaps in where eligibility criteria for
tiers of the service system end, especially with tertiary mental health services, impacting when
referrals can be made from different headspace services.
PHNs also highlighted the challenge for headspace services in managing care coordination and service
integration activities, within existing funding limits, and with workforce challenges within services. These
activities meant staff are taken away from clinical supports. Particular challenges were described in
engaging with local GP services. Wait times have impacted on the relationships held with GPs at some
headspace services, and difficulties working with some GPs impacts the level of horizontal integration
with physical and sexual health services over and above the small volume of physical health services
provided within services.

Effectiveness of headspace in improving pathways to care through service

integration and care coordination
Qualitative data show that young people and their families, and other external stakeholders, value service
integration and care coordination highly, to ensure young people are connected to other required services
when they need them.
Most young people accessing headspace indicate they received appropriate referrals to other services, with
mixed experience for a small number of young people who used alternative service providers such as GPs
to support their care pathway, or did not receive the appropriate referral they needed from their headspace
service.
headspace services and other external stakeholders indicated that headspace services undertake a range of
activities to support pathways to care through integration and care coordination. These contributions were
consistently recognised across these stakeholder groups.
However, several challenges impact on the ability of headspace services to improve service integration and
care coordination. There are capacity constraints within many health services currently, with integration
difficult where a service cannot take a referral, or work with headspace services to improve care
coordination. There are also instances where there are not alternative services available, particularly in
regional and remote communities. Similarly, headspace services encounter difficulties engaging in these
activities at points in time based on demand for services, and the need to balance clinical workloads with
these additional activities and managing referrals with existing wait times. For these reasons, there was
mixed evidence from other providers in the sector as to the effectiveness of headspace in supporting
pathways to care through integration and coordination.
Evidence from young people, headspace service providers and other external stakeholders indicates that
the headspace model is effective in improving pathways to care, however there are challenges which
impact this work, many of which are outside the control of headspace and rely on effective functioning of
the broader service system. It is important to note that there are some limitations around the data, with data
being self-reported by young people, and headspace service staff about their own performance in this area.
However, there was consistency in recognition of headspace’s role and the types of challenges identified
by different stakeholder groups, suggesting findings are reliable.

D.8 To what extent is headspace providing a localised service

offering, and what are the barriers and enablers to this?
D.8.1 How headspace services provide a localised service
As described above in Section 2.1, the headspace model is made up of a number of core and enabling
components. Two of these components have specific links to localised services or offerings, including:
• Community Awareness and Engagement – activities are intended to identify local needs and high-
prevalence issues through community consultation and local data analysis.
• headspace Consortium – provides lead agencies with strategic direction and resources to enhance the
service’s capacity to meet local community needs.
In the distributed governance model underpinning headspace, PHNs play a central role in ensuring
services are localised, responsive to the needs of the local community and well-integrated. In their
commissioning role, PHNs work with local headspace service providers to set priorities and target
activities to respond to local need.
There are a multitude of examples of how services have been tailored to the needs of the local community.
Representatives from deep dive locations demonstrated a strong level of community engagement and
awareness enabled by the consortium arrangements and a local workforce with local networks to support
this. Examples of how services are tailored to their communities include:
• introduction of bushfire recovery role to tackle climate-related anxiety with young people;
• increased focus on outreach services where there is increased need, for example in remote Aboriginal and
Torres Strait Islander communities, or neighbouring communities impacted by bushfires; and
• introduction of new consortium partnerships with additional local services, responding to particular
stressors for young people in the local community, such as domestic and family violence and family
wellbeing services.
The examples highlighted indicate headspace’s approaches to building relationships with young people
and local services to identify high-prevalence issues, and in increasing access to required services specific
to local need. The extent to which individual headspace services are tailored to their local communities is
impacted by a range of enablers and barriers, outlined below.

Enablers to localised services

PHNs and deep dive representatives identified the consortium model and use of Youth Reference Groups
were key to localising services offerings. Consortium members operating in local communities have deep
insight into particular challenges faced by young people, and what services may be required to support
these. Similarly, Youth Reference Group members use their knowledge of their peers to support localised
service offerings, and have been responsible for initiating local offerings in some services.
Some PHNs also indicated that the commissioning process for services allows consideration of local need
to be built into leady agency selection, with specific local considerations part of the selection process. This
view was not shared by all PHNs, as outlined further under barriers below.
PHNs reported that many services are well-integrated into their local communities, and provide services in
demand with local community. headspace service providers indicated that community engagement
activities assist them to identify how best to respond to local need, and some lead agencies have a specific
focus on supporting these activities by also applying for additional grant funding from alternative sources
(such as local and state government grant funding rounds) to support this work.

Barriers to localised services

Centres and lead agencies through deep dive discussions, and the service and lead agency survey
consistently identified that there is limited capacity for outreach and community engagement activities
within services, to identify local needs and tailor services, and reach those in local communities who may
not use the centre model. Community engagement positions are sometimes part time roles based on
available funding. Some headspace services do not have dedicated community engagement positions, and
community engagement is often de-prioritised due to clinical service loads within services.
Centres and lead agencies also indicated they often have trouble recruiting specific workers to meet the
needs of the local community. These may be for specific professional positions or positions related to a
specific cohort of young people, such as Aboriginal wellbeing workers or workers with culturally and
linguistically diverse backgrounds. A small number of PHNs indicated that this has resulted in some
services focusing on employing any available workers, with less focus on the types of staff required to
meet local need.
Some PHNs also indicated challenges as the local commissioning agency for headspace services in
tailoring services to the needs of the local community, while ensuring services still meet the requirements
of the headspace model integrity framework. These PHNs also indicated that there is no flexibility to use
funding provided for a headspace service to design localised offerings which directly address the specific
needs of the community. While some tailoring is afforded through headspace services, the extent to which
PHNs can commission a tailored service targeted at local need is limited.
Finally, headspace service staff and lead agencies indicated there are some challenges in localising services
where there is increasing complexity and severity in the presenting need of young people. These young
people are not the focus of the headspace model, and tailoring services to meeting their needs is difficult.
In addition, increased pressure on service capacity from young people with more intensive needs impacts
on capacity to focus on tailored offerings.

Extent to which headspace provides localised service offerings

The qualitative data helps illustrate the number of ways in which headspace services are localised for their
communities. headspace services work with local communities and providers to build relationships and
understand what local needs services should target. The consortium model, commissioning process and
community engagement activities support services to localise offerings.
However, qualitative data also indicates that the extent to which services are localised varies significantly.
This variation is due to a range of factors, including:
• capacity for community engagement and exploration of local needs;
• ability to recruit specific workers or professions to deliver on specific support needs in communities
particularly in regional and remote areas, for example Aboriginal wellbeing workers;
• challenges with the flexibility of the headspace model and how funding can be used to target local needs
specifically; and
• increasing complexity of presenting need, which is not within headspace services’ usual target cohort to
provider tailoring.
Overall, evidence from headspace service providers and other external stakeholders indicates that the
headspace model enables localised services, however this is inconsistent across services, and the link
between local needs analysis work undertaken by PHNs and implementation of headspace services
could be strengthened.
It should be noted that this qualitative information is largely self-reported by individual stakeholders with
limited corroboration from other stakeholders. While there is less consistency in qualitative evidence,
this is to be expected based on the roles and involvement of different groups within headspace
services.

D.9 What other contributions does headspace make to local

communities?
D.9.1 Contributions to local communities
In addition to direct clinical, centre-based and other services provided to young people, a range of other
contributions and activities are also provided by headspace services, often outside of the service. A number
of headspace stakeholders commented on these contributions.

Evidence of contributions to local communities in which headspace works

Perspectives of young people who use headspace
headspace users indicated they have experienced a range of contributions to their local communities from
their local headspace service. The most common of these were contributions through outreach activities in
communities – in schools, at public events, and near public transport hubs:

“Near the [name of suburb] train station, there's a little path thing. They used to
have, I guess, like a little fun day where they would have little free games to play and
then packages to hand out and stuff. And more often than not, when [suburb] Youth
Centre, or the [region] City Council did a youth event, they would partner with
headspace to, again, bring more of that information out, to try to get it to young
people.”
However, when asked what headspace could do differently, headspace users suggested increasing their
profile through social media and more in-school presence:

“I think definitely high school education. If headspace went to schools it would make
a big difference because we were just told in school, they were like these are
symptoms of depression. “

headspace also provides young people with opportunities to contribute to governance via the Youth
Reference Groups and as mental health ambassadors. This adds to its reach and is potentially most
effective as peers can inspire other young people to seek help, it also provides development opportunities
for the young people involved:

“Well, it started off in school settings in 2020, when I first joined up as an

ambassador. And then one door opens up to another and I've been able to give
speeches on different panels, universities, corporate.”

In one area, Youth Reference Group participants spoke about the contribution of headspace to the
community through supporting them to organise festivals around issues of importance to them such as
supporting LGBTQIA+ young people, homelessness and social justice. Members of the Youth Reference
Group worked on the organising committees for these events with support from headspace staff and other
key services in the area. These events aimed to raise awareness and reduce stigma:

“There’s some community services that have planned an event that has of course
been postponed but it is an event to raise awareness for homelessness and funds for
social justice advocates [name of location] that’ll go straight back into a few homes
and things in the community to help tackle that issue. [Name of service], they have
been really active with promoting the event and we’ve had [name] from headspace
and some Youth Reference Group members as well working on that committee for
that event.”
headspace services’ involvement in these sorts of community events were not consistently described
across services, with outreach and community engagement activities differing between communities.
While many of these activities contribute directly to other outcomes, for example in reducing stigma
around mental illness, increasing mental health literacy and early help seeking, they also provide valuable
development opportunities for young people. Being part of the organising committees for events,
participating in service design, working groups and project teams through Youth Reference Group
participation provides young people with experience and improved capability, and has the potential to
increase their confidence and self-esteem.
Perspectives of young people who do not use headspace
Many young people who do not use headspace described hearing from headspace services through their
schools.
Non-headspace users at university indicated that while they recalled headspace services visiting their
school, that these sorts of community engagement activities were more limited through their university. A
small minority of non-headspace users indicated that the impact of these community engagement activities
on them depended on who was running school-based sessions. Where the representative was a young
person and easier to identify with for students, non-headspace users described this as being more effective
in promoting headspace services and mental health wellbeing, than with other headspace service staff with
whom who young people did not identify.
A small number of non-headspace users also identified youth ambassadors for headspace services as an
element of their community engagement and outreach activities, and that this supported great awareness of
services and other outcomes such as early help seeking and reducing stigma around seeking support.
Perspectives of headspace service providers
headspace service and lead agency representatives were asked to indicate what types of services their
service provides to young people and the community more broadly. Of the 69 respondents who answered
this question, 58 (or 84 per cent) indicated that their services work with local schools and community
groups, while 44 (or 64 per cent) indicated they provide outreach services to local communities. Deep dive
site representatives, as well as survey respondents indicated that community engagement such as this is a
critical and successful part of the headspace model, however is an onerous obligation, and is often not able
to be adequately resourced within current funding for headspace services.
Engagement with schools and universities includes presentations to school students on supporting their
mental health and wellbeing, where young people can find resources to support their mental health,
information regarding services available, and participation in open days and fair days in universities.
Outreach services provided also differed significantly between headspace services, often linked to
preferences and needs of the local community.
Perspectives of other external stakeholders including school and university counsellors,
GPs and PHNs
School and university counsellors also described the types community engagement and outreach activities
of headspace services. Some identified where headspace had visited their local school or university to
provide information and resources for young people. However, the exact nature and frequency of these
activities varied between local communities. Some headspace services have delivered more of these
activities to their local communities than others, and the reach of these activities also varied. For example,
in regional areas, the focus was stronger on community engagement and outreach within the immediate
area around the service, with other surrounding communities less of a focus from the perspective of
counsellors.
Some school and university counsellors also indicated they use online headspace resources to support their
own work, including as part of their practice, or to refer young people to, to support their mental health
and wellbeing. This was recognised as a key strength of headspace, and a key contributor to communities.

Extent to which headspace makes other contributions to local communities

Qualitative data demonstrates strong recognition of the contributions that headspace services make to their
local communities, through schools and other education institutions, community events and engagement,
and availability of resources and information.
All stakeholder groups had positive views as to the impact of these contributions, and these were tied to
other outcomes discussed above, such as improvements to mental health literacy, early help seeking, and
access for young people.
Furthermore, the headspace model provides valuable development opportunities for young people. Being
part of the governance and planning of services provides young people with experience and improved
capability, and has the potential to increase their confidence and self-esteem.
Evidence from young people, headspace service providers and other external stakeholders indicates that
the headspace model provides a range of additional contributions to local communities that are highly
valued by those communities.
It is important to note that there are some limitations with the data, being self-reported by young people,
and headspace service staff and external providers about their own performance in this area. However,
there was consistent recognition of the contributions made, and the challenges services face in delivering
these contributions, suggesting findings are reliable.

D.10 To what extent does a ‘no wrong door’ approach assist

headspace to meet its objectives?
There was significant support for headspace’s ‘no wrong door’ approach to supporting young people. The
approach supports young people by:
• ensuring they are able to engage with mental health supports in a way they feel comfortable;
• providing a free entry point into the mental health service system;
• providing a soft entry into the mental health service system, with referrals to other services available to
support service integration for young people; and
• providing them with access to initial services to support broader objectives such as improved mental
health literacy and early help seeking, even where they may be referred to a more appropriate service.
headspace service and lead agency stakeholders across deep dive services, as well as those responding to
the survey of headspace services consistently indicated that anecdotally, young people’s mental health
needs are becoming increasingly severe and more complex, with many cases being outside of the
headspace model’s mild to moderate criteria. Common presenting concerns were reportedly developmental
disorders, personality disorders, eating disorders, complex trauma and grief, and selfharm and suicidal
ideation, including in ages under 12 years old.
headspace service staff interviewed commonly described a “missing middle” of clients who are too
complex to be seen under the headspace model’s mild to moderate remit, but not unwell enough to be
transitioned to overwhelmed TMHSs.

D.10.1 The effect of a ‘no wrong door’ approach

There were also consistent views from headspace services and lead agencies, PHNs, and other deep dive
stakeholders that there is significant demand placed on services by the ‘no wrong door’ approach in the
hMIF. While this is largely regarded as essential in order to ensure young people presenting with high risk,
distress, need or acuity are not turned away without assistance, this element of the model is challenging
where tertiary mental health services are unable to meet demand for higher needs young people.
Commonly in smaller regional and remote areas, where there are limited private practices and TMHSs,
local services will redirect a young person back to headspace services to counteract their own wait times.
This has resulted in headspace services in these circumstances taking on these young people to ensure they
receive some form of support, and needing to provide intensive case management and crisis support
services. This has in turn led to workforce attraction and retention challenges, as remuneration and job
security with short-term funding cycles are not commensurate with the level of clinical risk associated
with supporting young people with more complex needs.
Another reported effect of the ‘no wrong door’ approach, coupled with the high visibility and brand
recognition of headspace, is that services spend a proportion of time fielding general enquiries from and
about the local service sector. As discussed in Section 3.1.2 above, this contribution of time and expertise
to the local community is not accounted for in data collection and reporting, but some services described it
as drawing on a large proportion of their time.
Stakeholders also described that the combined impact of these flow-on effects of the ‘no wrong door’
approach are to increase the waiting times for young people with mild to moderate conditions with lower
risk profiles to access services. Wait times have reportedly increased over time for some headspace
services based on anecdotal comments from headspace service stakeholders. However, data capture for
wait times has only recently commenced, and longer-term trends in wait times are not able to be
determined.
These examples illustrate the potential tensions for headspace services between maintaining scope in terms
of age and acuity, while also providing a ‘no wrong door’ approach and meeting goals and objectives to
increase access to required services and a pathway to appropriate care which may not realistically be
available.

D.10.2 Extent to which a ‘no-wrong door’ approach supports headspace program

objectives
Young people, as well as headspace service providers and external stakeholders consistently recognised the
benefits of the ‘no wrong door’ approach and had strong positive regard for it as part of the headspace
model. The ‘no wrong door approach supports headspace to reach young people and support mental health
literacy, early help seeking and access to services. It also supports young people to get help when they
need it, regardless of the severity of their mental health problem.
However, the ‘no wrong door’ approach, coupled with other challenges in the service system such as
referral services with limited or no capacity for new referrals significantly impacts headspace’s core
business of supporting young people with mild to moderate, high-prevalence mental health conditions and
other contributions to communities through outreach and engagement.
Evidence from young people, headspace service providers and other external stakeholders indicates that
the ‘no wrong door’ approach is an important and valued feature of the model, supporting improved
mental health literacy, early help seeking and access to required services. At the same time, however, the
level of demand for mental health support, and the volume of young people who use headspace as the
entry point into support, leads to increased wait times for young people, particularly those in the ‘mild to
moderate’ group who are the headspace model’s primary target cohort of young people.
It should be noted that this information is qualitative in nature, and self-reported by stakeholder groups,
however there is consistency between stakeholders as to the benefits and impacts of the ‘no wrong door’
on headspace achieving its objectives, suggesting the information is reliable.

D.11 What is the level of support for headspace from other

primary care and mental health service providers?
D.11.1 Evidence of the level of support for headspace from other primary care and
mental health service providers

Perspectives of young people who use headspace

While headspace clients weren’t able to provide much comment on their understanding of the level of
support for headspace from other mental health and primary care services, many young people who had
accessed headspace indicated they were referred to headspace by other services, including their GP, school
counsellor or parents. Referrals to headspace indicates that for some headspace services, there is good
support from other service providers. However, as discussed further below, this is not consistent across
services.
Perspectives of young people who do not use headspace
As outlined in Section D.7 above, young people who completed the young people’s survey were asked
about other services they may have accessed to support their mental health. Twelve per cent of young
people who had not visited a headspace service, but sought help elsewhere reported receiving a referral
from the GP to both headspace services, as well as other mental health services. Four per cent of these
young people reported receiving a referral to headspace services only, while 81 per cent of young people
indicated that their GP had referred them to other services but not a headspace service. Figure 49 below
demonstrates this split between responses. Similar to the experience of headspace clients, the level of
support for headspace services from other parts of the service system varies between individual services.
Figure 49: Young people responses to other services their GP referred them to (young peoples’ survey)

Source: Evaluation survey of young people who have and have not accessed headspace services

Perspectives of other external stakeholders including school and university

counsellors, GPs and PHNs
In consultation with a small sample of GPs, there was good understanding of what headspace services
delivered at a high level, and acknowledgement of the work headspace does to support early intervention
and young people with mild to moderate conditions. However, this sample of GPs described a range of
challenges, specifically in regional areas, that impact on their support for headspace through referrals. For
example, wait times for some headspace services have at times deterred GPs from making referrals to their
local service, out of concern for the young person in the intervening period before being able to access
recommended treatment options.
While there was good understanding of the broad offerings of the headspace model, GPs also described
challenges understanding what specific staff and specialist service areas a headspace service might have,
such as AOD workers, occupational therapists, dietitians, or specialist psychological services.
Challenges were also described with operating a shared care model with headspace services. These GPs
described reluctance of headspace services to take a GP’s diagnosis at the time of referrals or intake, and
limited opportunity to discuss ongoing progress and any other onwards referrals with GPs to support
effective care coordination. These challenges have also at times prevented GPs from supporting headspace
services through referrals.
Some PHNs also acknowledged challenges for local headspace services to engage with and receive
support from local GPs.
Specific to smaller regional locations, GPs also discussed challenges with competition for the same staff.
Where there is only one local worker who providers a specific type of support, referrals are often made to
that person, regardless of which service they work for. Staff have been lost from headspace services to
another local organisation in some instances, and support often shifts with the person.
School and university counsellors across the country had a strong understanding of the professional and
clinical services provided by their local service, especially mental health and GP services. There was more
limited recognition of other services provided, including vocational, alcohol and other drug, occupational
therapy services etc.
The majority of school and university counsellors consulted nationally indicated a strong level of support
for headspace services. School principals and wellbeing coordinators completing the Colmar Brunton also
indicated strong support for headspace services, with 92 per cent agreeing that headspace is a vital
community service for young people and families . Eighty-four per cent of these respondents also indicated
198F

they would recommend headspace if a young person needed support for mental health issues.
School and university counsellors in consultations also supported referral to headspace services in most
circumstances, however some challenges were reported in supporting headspace. Some counsellors
indicated they have stopped referring young people to headspace services due to current wait times at their
local service, and they would prefer an alternative service that might see a young person more quickly. A
small minority of university counsellor also indicated that there was limited benefit referring a young
person to headspace services, as they were not able to provide additional services in addition to what their
university support team could provide. This varied depending on the resources available at particular
institutions and mental health supports offered.
Similar to the experiences of GPs described above, some counsellors also described challenges engaging
with headspace services for care coordination. It was common that these counsellors did not receive
information regarding what happened with their referral for a young person after it was made, unless the
young person returned to the counsellor and shared that information directly. This meant counsellors were
unsure what additional support might be required for a young person over and above headspace services
received.
Support for headspace services was also impacted by continuity of relationships. A small number of
counsellors described instances where a headspace service had changed its management, and this impacted
the level of engagement they were able to have with the service based on the approach of the manager. In
turn, this impacted perceptions of the quality of the service. Staff turnover in other key roles such as
community engagement coordinators also impacted relationships, and where there was a stable staffing
group, opportunities for engagement were more common. These views were consistent from counsellors
across different locations, including metropolitan and regional services.

D.11.2 Level of support for the headspace program from other primary care and
mental health service providers
Stakeholder groups have significant positive regard for headspace services, including school
representatives, consortia members and other community partners, with all of these groups indicating that
it is a vital community service for young people.
However, qualitative evidence demonstrates that there are a range of factors that impact on the level of
support these primary care and mental health providers have for headspace, and in particular their
likelihood to make referrals to headspace services. These factors include concern about current wait times
within headspace services, challenges engaging in coordinated care with headspace services, and in
building relationships with headspace services when there is staff turnover.
While evidence in this area was largely qualitative, there was consistency in both the level of support for
headspace as a vital community service, and challenges described in supporting headspace services
through referrals.
The headspace model benefits from generally high levels of support from other primary care and mental
health providers, although operational pressures affect individuals’ referring decisions and, at times, create
frustrations.

D.12 Providing an appropriate service approach for young people

with mild to moderate, high-prevalence mental health
conditions
D.12.1 Mild to moderate, high-prevalence mental health conditions and the
headspace model
In the headspace model, appropriate care is defined as “the provision of evidence-based interventions for
each individual young person by matching the type, intensity, frequency, duration, location and mode of
treatment to their presenting need. This includes identification and consideration of factors such as: risk
and protective factors, stage of illness, psychosocial complexity, and developmental and sociocultural
factors” .
199F

Many elements of the current headspace model are closely aligned to the needs of young people with mild
to moderate, high-prevalence mental health conditions. For headspace users, mild to moderate
psychological distress is defined as a value of between 20 to 29 out of 50 on the K10 questionnaire.

Evidence of how the headspace model provides an appropriate service approach for
young people with mild to moderate, high-prevalence mental health conditions
High-prevalence mental health conditions, such as depression and anxiety, are widely considered to be
able to be effectively treated and to respond well to early intervention . The design of the headspace model
200F.

has prioritised supporting young people in this category. It includes a psychosocial model of supports
provided by peers, and in practice, many staff working in headspace are early career clinicians with whom
the young person is likely able to identify and build rapport . headspace providers described how the
201F

staffing profile is driven by a combination of the funding envelope available, which is competitive for
early career psychologists, and by the brand of the model, which appeals to younger staff with an interest
in working with young people.
Stakeholders also argued that the model also supports mild to moderate conditions, with a focus on early
intervention and prevention of mental ill-health for young people, including improved mental health
literacy, and integration for other factors impacting on mental health such as physical health, alcohol and
drug use and employment and education.
Representatives across a range of stakeholder groups consistently confirmed the view that the model is
well designed for this cohort of young people, with the provision of support groups, skills training and
peer workers particularly recognised as powerful in the potential to help young people to tap into
protective factors and support their wellbeing. Youth representation in the design and delivery of services
was also called out as key to the appropriateness of the model for this cohort.
The Colmar Brunton survey conducted for headspace National , while not explicitly exploring the
202F

appropriateness of the model for its target cohort, found a number of responses that supported this. Lead
agencies were asked what the impact of a community having no headspace service would be. There were a
number of responses which indicated the focus on early intervention for young people means pressure is
taken off tertiary services, and reaching young people early is a strength of the model.

“I believe that early intervention approaches to mental health are critical –and this
is very much the strength of the headspace platform. If the remit of headspace as an
early intervention program were diluted, it’s almost certain there would be a drift to
 provision of acute care and supports for more complex needs at the expense of early
intervention. Early intervention works! The immediate outcomes might not be as easy
to report, but the long-term benefits are clear!”
The same survey also saw significant positive regard for the headspace model from young people
themselves:
• 80 per cent of young people agreed they have a better understanding of the mental health issues;
• 79 per cent agreed they were better able to manage their general health and wellbeing;
• 78 per cent agreed headspace reduced the impact of mental health issues on their day-to-day life; and
• 70 per cent agreed headspace supported them to stay at work or school.

Extent to which headspace provides an appropriate approach for young people with
mild to moderate, high-prevalence mental health conditions
There is significant research as to the appropriateness of early intervention models for high-prevalence
mental health conditions such as anxiety and depression. There was also consistent feedback from across
stakeholder groups that the model is targeted at this cohort, and particular elements of the headspace model
such as peer workers, group work, focus on skills and training, and young people’s involvement in
designing the service assist with the appropriateness of the service.
Evidence suggests that the headspace model provides a highly appropriate mental health service approach
for young people with mild to moderate, high-prevalence mental health conditions.

D.13 Providing a culturally appropriate and inclusive service for

young people and their friends and families, including for
vulnerable and diverse population groups and different age
groups
D.13.1 Culturally appropriate and inclusive service
Table 55 Overview of objectives of headspace for culturally appropriate and inclusive services

Objective Short term impacts Medium term impacts

Ensuring young people • Young people from a diverse • Local service system provides
can access the help they range of backgrounds access more youth-friendly, accessible
need in an appropriate, and engage with headspace and inclusive services as a
accessible and youth services result of learning through
friendly way - partnerships, shared
• Young people and families feel
providing an accessible, professional development etc
their needs and interests are
welcome, inclusive and
understood and reflected in
nonstigmatising service
their local headspace service
(participation outcomes)
• headspace services meet the
expectations of their friends
and family and Youth
Reference Group.
• Young people and families report
that headspace services are
accessible, welcoming,
inclusive and non-stigmatising

The headspace model includes a focus on the experience of service for young people and their families
from a diverse range of backgrounds through providing an accessible, welcome, inclusive and
nonstigmatising service. For young people from diverse population groups, this includes providing
translated information, guidance materials and posters and flags and other cultural symbols to make the
young person and their family feel welcome and included. Given the broad age range supported by
headspace, the experience of service must also be tailored for the level of maturity of the young person,
with very different needs for 12 years compared with those approaching age 25.
In order to examine the extent to which headspace is successfully providing culturally appropriate and
inclusive services, data is drawn from hMDS user satisfaction surveys, interviews with headspace user and
non-users as well as Youth Reference Group members and school and university counsellors.

Evidence of the effectiveness of headspace in providing young people with

culturally appropriate and inclusive services
Perspectives of young people who use headspace
The young person satisfaction matrix is based on survey responses of young people attending headspace
on their second occasion of service, and subsequently at every fourth visit during that episode of care. The
survey asks them to rate 14 statements on a five-point scale of ‘strongly agree’ (5) to ‘strongly disagree’
(1). Table 56 summarises the estimated probabilities of young persons responding ‘agree’ or ‘strongly
agree’ to each satisfaction domains.
Table 56: Probability young person responded ‘agree’ or ‘strongly agree’ to satisfaction domains across episodes
created from 2015-16 to 2019-20

Culturally and Aboriginal &

LGBTQIA+ Linguistically Torres Strait
Young person Overall young person Diverse young Islander young
satisfaction domains: person person

I felt comfortable at
88.0% 88.8%* 86.9%* 85.6%*
headspace
It was easy for me to get to
85.2% 84.0%* 83.1%* 82.4%*
my headspace centre
I could attend appointment
times that suited me (i.e.,
81.4% 81.3% 78.9%* 80.8%
didn’t interfere with study or
work)
I felt that headspace staff
91.3% 91.2% 90.4%* 89.2%
listened to me
I felt that headspace staff
involved me in making
83.1% 83.3%* 80.8%* 81.2%*
decisions about what would
happen next
I felt that my views and
88.5% 88.4% 86.9%* 85.3%*
worries were taken seriously
I felt that I was able to raise
85.2% 84.4%* 83.5%* 81.9%*
any concerns that I had
I feel that my mental health
improved because of my 63.5% 62.3%* 63.5% 62.5%
contact with headspace
I feel that other aspects of my
life improved because of my 61.6% 59.9%* 60.8% 61.8%
contact with headspace
I feel that I can deal more
effectively with my problems
63.5% 62.3%* 63.3% 62.5%*
because of attending
headspace
I feel that I know more about
mental health problems in
66.5% 66.2% 67.0% 64.8%*
general because of attending
headspace
I was generally satisfied with
86.6% 87.6%* 84.9%* 83.5%*
headspace
I got help for the things I
75.6% 75.2% 73.1%* 73.4%*
wanted to get help with
If a friend needed this sort of
help, I would suggest 88.7% 89.9%* 87.5%* 86.9%*
headspace
Source: KPMG analysis of hMDS of episodes created within 2015-16 to 2019-20. Sample includes 379,130 episodes.
Notes: Statistics are based on the episode’s last observed response.
*: Significantly different from baseline at 5 per cent.

Responses for each statement indicate that headspace is an appropriate and inclusive service for the
general population of young people (responses range from neutral to strongly agree), and for a number of
indicators this was particularly the case for LGBTQIA+ young people, Culturally and Linguistically
Diverse cohorts and Aboriginal and Torres Strait Islander young people as well, with predictive
probabilities of scoring ‘agree’ or ‘strongly agree’ being similar to the general population of young people
across all indicators.
To explore this domain further, young people completing the survey as part of this evaluation were asked
to reflect on the service they had received over the previous 12 months and rate on a five-point scale, from
‘always’ to ‘never’ how they felt about five statements:
Figure 50: Summary statistics on young people’s ratings of service at headspace
Source: KPMG analysis of Young person survey. Total number of finished surveys: 3,004. Response rate to the analysed questions:
36 per cent.

As can be seen in Figure 50, results indicate that young people responding to this survey had positive
experiences with headspace, with the large majority indicating ‘always’ in response to the indicator
statements. When analysed for any differences between young people from Aboriginal or Torres Strait
Islander backgrounds, young people identifying as LGBTQIA+ or as speaking a language other than
English at home, results were similarly high, with no significant difference between groups.
In interviews and focus groups, headspace users indicated that cultural diversity of staff was important to
them (this was mentioned most often by culturally and linguistically diverse young people):

“So I think it's a lack of cultural diversity in headspace and especially because I'm
ethnic I'd specifically asked for an ethnic person and then they said that there wasn't
any. At least in my area. Just the advice that they give is very tailored to white
Australians kind of thing. There's no kind of perspective when it comes to ethnic
clients and what their home life might be like.”
and

“And when I think about appropriate cultural competency, I think of someone or a

counsellor understanding the culture, the taboos, stigmas, values. And also, ideally
someone from that background, from a cultural and linguistic background, which I
didn't find from the counsellors…so there was a bit of a hesitancy for me because we
have such unique values and there's a lot of fear involved, but eventually got over
that hurdle, conversated [sic], and it wasn't that bad.”

Sometimes there was a gender preference also, based on cultural considerations (for example, for female to
see a female counsellor):

“Look, I just want to talk to someone from the same religion and a woman if that's
okay." Then I was happily given to a woman and she was from the same religion and
it was so welcome then, it was so good. Like she understood everything that I went
through because she went through it once.”
There are Aboriginal and Torres Strait Islander young people who use headspace services, however they
may benefit from more First Nations staff.
Members of Youth Reference Groups noted that headspace offered a range of supports for diverse groups.
They commented that headspace provided inclusive services particularly for LGBTQIA+ young people:

“I would see pamphlets around. Sometimes when they had the booths around I would
see that. But one day I got curious, and I came round and there was a LGBT group.”

There were some concerns from Youth Reference Group participants that young people who fell outside
the age ranges of 12 to 25 fell through service gaps:

“But age wise, because it’s between 12 and 25 if someone calls up and they’re not in
that age range I don’t think very much support’s given, they just say sorry, we can’t
cater to you. Yeah, I’m not sure what happens with that but I’m pretty sure different
issues or concerns that’s supported but if you don’t fit in that age category I don’t
 think there’s much headspace would do.”

Perspectives of young people who do not use headspace

Interviews and focus groups found there is recognition amongst non-headspace users that headspace
services appear to cater well to the LGBTQIA+ young people in the community, and have knowledge of
issues affecting these young people. There was also some indication from non-headspace users from
culturally and linguistically diverse cohorts that they would consider using headspace services, as they are
able to assess the service without parental consent, especially where they encounter cultural stigma related
to mental health support. A key caveat here was the importance of appropriate staff members, for example
that young Muslim women need a female worker, and that the mix of service staff may not always provide
the right support.
Aboriginal and Torres Strait Islander young people who do not use headspace indicated that they thought
there was some variation in the appropriateness of services between locations.
Neurodivergent young people who have not used headspace indicated in interviews and focus groups that
they did not necessarily identify with the service. The neurodiverse flag is not present, and their
interactions with headspace staff did improve their level of trust in the service.
There was also some indication from non-users that they identified with the brand more when they were

younger (high school age), with this dropping off as they got older. Amongst non-headspace users,
there was also very inconsistent understanding of what age groups were eligible for support from
headspace services
Perspectives of other external stakeholders including school and university counsellors,
GPs and PHNs
School and university counsellors identified that in some communities, they saw beneficial impacts for
Aboriginal and Torres Strait Islander young people through informal community outreach to remote
communities by headspace services. The extent to which this was seen as happening varied between
services.
Key staff from PHNs discussed how some services did not provide much outreach as they favoured centre-
based services as part of the hMIF. There was broad agreement that there would be benefit in greater
flexibility in services (for example being provided through the local AMS rather than in the headspace
centre, as this is where some young people feel more comfortable accessing services).

The extent to which headspace provides a culturally appropriate and inclusive

service for young people and their friends and families, including for vulnerable
and diverse population groups and different age groups
Responses indicate that headspace is an appropriate and inclusive service for the general population of
young people (responses range from neutral to strongly agree), and for a number of indicators this was
particularly the case for LGBTQIA+ young people as well, with scores significantly higher than the
general population on six indicators. In contrast, average ratings made by Aboriginal and Torres Strait
Islander young people were statistically significantly lower across all 14 indicators than the satisfaction
levels of the general population of young people accessing headspace.
Data from a range of sources indicates that headspace is broadly effective in providing culturally
appropriate and inclusive services for the general population of young people, and for LGBTQIA+ young
people. However, user satisfaction is significantly lower for culturally and linguistically diverse young
people and for Aboriginal and Torres Strait Islander young people. In contrast, measures of satisfaction
undertaken for this evaluation, including of how welcome young people felt and how respectful services
were of a young person’s culture, gender or faith identity were all positive and in line with results for the
general population of young people using headspace.
At the same time, discussions with young people and other stakeholders again highlighted the need for
staff with particular cultural backgrounds as a key mechanism to providing culturally appropriate care for
young people from that culture. There were also differences between the age when young people felt
headspace was appropriate for them, with young people more likely to see it as a service where they feel
included. At the same time there remain some potential misconceptions about the age groups welcome at
the service, with non-headspace users consistently sharing in focus groups that they were unsure what age
groups headspace services supported, and hesitation from non-users of headspace about how welcoming
and appropriate it is for them.
Overall, there are mixed results from the data and insights gathered through this evaluation about how well
the headspace model effectively provides a culturally appropriate and inclusive model for young people
and their families, with strong satisfaction from the general population and LGBTQIA+ young people, but
significantly lower satisfaction levels on relevant measures from culturally and linguistically diverse and
Aboriginal and Torres Strait Islander young people.

D.14 Enabling young people and their families to access support

where, when and how they want it
D.14.1 Appropriate, accessible and youth friendly support
Table 57 Overview of objectives of headspace for appropriate, accessible and youth friendly support

Objective Short term impacts Medium term impacts

Ensuring young people • Young people feel listened to and • headspace services operate
can access the help they involved in decision-making flexibly as appropriate to the
need in an appropriate, community needs and profile
accessible and youth • Young people and families feel
friendly way their needs and interests are • Local service system provides
 understood and reflected in more youth-friendly, accessible
their local headspace service and inclusive services as a
(participation outcomes) result of learning through
partnerships, shared
• headspace services meet the
professional development etc
expectations of their friends
and family and Youth
Reference Group.

A key element of the headspace program logic is that the services provided are appropriate for young
people. Through providing a positive experience of service, by ensuring young people feel that their needs
and interests are reflected in the services on offer, and that the services adapt to the needs of young people
the overall objectives of the model are supported.
This evaluation examines a range of data and evidence regarding the extent to which headspace is
successful in these domains. Feedback from young people using headspace collected through hAPI
surveys, as well as direct consultation with young people, Youth Reference Group members and staff and
other stakeholders provide evidence of relevance to this evaluation question.

Evidence of the effectiveness of headspace in providing appropriate, accessible and

youth friendly support
Perspectives of young people who use headspace
Young people were asked in the evaluation survey about their experiences with headspace services over
the previous 12 months. Sixty-six per cent of headspace users responding to the survey indicated that
headspace centres ‘always’ made an effort to see them when they wanted.
This result varied with the number of OOS, with the fewer the number of OOS the young person had, the
more likely they were to indicate an answer other than ‘always'.
Young people using headspace described in focus groups and interviews that they found headspace staff
easy to talk to, non-judgmental and relatable, and appreciate that the people who work at headspace can be
quite young but still qualified and experienced.
headspace users interviewed described that they accessed support either face to face by going to a service,
which were well-located and near public transport, or online (mainly due to the pandemic, or distance):

“It was very flexible”

and

“When I was filling out the forms, there were a lot of different options…”

Some users talked about wanting some online resources while waiting for their first appointment (or
between appointments). There is a US meditation app called ‘headspace’ often mistaken for the work of
headspace Australia that at least one user referred to as ‘their’ app (it is not – but indicates that such an app
might be useful):

“But the counselling side of it, I don't access them as much anymore, but I do use
their app a lot. The meditation one, that's a massive proponent of my life to this very
day. [Note: this person was referring to the US Headspace app – guided meditation
and mindfulness]”
Some wanted more online resources:

“Maybe they could make a website and have maybe a course you could do in the
meantime I guess between the sessions, like an online course or something.”

Barriers included opening hours (as users in their 20s were more likely to be at work during the day),
being able to move to another counsellor if they were not the right ‘match’ with the headspace staff
member, the cultural or gender characteristics of the staff member being too different so that they could
not relate, however the actual logistics of appointments were not a problem for the vast majority of
headspace users.
Quite a few interviewees had been to headspace and then to a clinical psychologist later, and the majority
much preferred the therapeutic relationship with the psychologist in private practice, while also
acknowledging that headspace has been useful at the time or pointed them in the right direction. A
minority felt headspace had been of very little use to them and were glad they had ‘moved on’.
In interviews with Youth Reference Groups, young people noted that accessible locations, the high
recognition of the headspace brand for example on social media, outreach activities in schools and the
community and the youth friendly approach to providing help and advocacy contributed to enabling young
people to access services.
Youth reference group participants identified several barriers to accessing support: waiting lists, staffing
shortages, and resourcing:

“Sometimes you can get yourself to a point you need help right now but if you've got
to wait three months you're just not going to be motivated enough and you're
probably just going to go downhill even more.”
and

“biggest issue is staff, I feel like it’s just completely understaffed”

and

“I personally think the staff here are great but I don’t think there is enough again. It
 all comes with being in the remote area”
and

“The amount of sessions can be a bit of a barrier as well, like you know, you have to
decide whether or not you want to – like if you’re going through something you have
to decide whether or not you’re going to see someone soon or you want to push it
back a bit so that you have enough sessions for the end of the year.”

Perspectives of young people who do not use headspace

Feedback from non-headspace users that opening hours predominantly in business hours did not support
young people with full time study and workloads to access services. The views of young people who do
not use headspace were explored in interviews and focus groups, to understand how appropriate and
accessible they see headspace services to be.
There was positive feedback from non-headspace users who have accessed website resources from
headspace. Some non-headspace users also recognised that headspace also has online and phone
counselling services through eheadspace for those who can’t access a service in person. Non-headspace
users saw these examples of telehealth services as important for those who can’t make it to a physical
service.
Non-users of headspace also discussed the location and accessibility of the physical headspace services
near them. Many non-headspace users knew where their local service was, but highlighted that this was
sometimes not accessible from local communities due to travel durations and lack of public transport. In
these discussions, young people identified greater flexibility for outreach services as being potentially
beneficial.
There was some hesitancy from the group around using the service from the public setting of a service, as
they didn’t want to be seen walking through the door. These young people thought the presentation of the
building would draw unwanted attention, and in small communities, young people were concerned about
their privacy.
In contrast, other non-users spoke positively about the bright and vibrant brand of the headspace service
and though this looked welcoming and inviting.
Feedback from non-headspace users was that they thought opening hours were predominantly in business
hours, and that this does not support young people with full time study and workloads to access services.
When discussing accessibility of headspace, non-users also highlighted that they thought that providing
services without cost was an important benefit of the headspace model.
Perspectives of headspace service providers
Survey responses from service and lead agency staff indicate that the majority of people working within
headspace services believe that their service provides services that are youth friendly, appropriate and
accessible with 90 per cent of respondents indicating either ‘very well’ or ‘well’ on this domain.
Figure 51: Responses from service and lead agency survey: how well does your centre provide services that are youth
friendly, appropriate and accessible

Source: KPMG analysis of headspace service and lead agency survey

Notes: A total of 60 responses were received for this question.

When asked to describe enablers and barriers to their service providing youth friendly, appropriate and
accessible services, respondents provided a range of responses. Some identified the youthful, friendly and
welcoming service design as a key enabler, others that the physical site is important, needing to be
accessible for young people, and big enough to support engaging private providers. The flexible model
with centre-based and some outreach services, as well as having multiple referral pathways, strong staff
knowledge and relationships with the local service system and a ‘no wrong door’ approach that aim to
meet the needs of young people were also highlighted.
The role of Youth Reference Groups in service design was also identified as a key enabler, with services
designed by young people for young people and including youth friendly approaches such as ‘walk and
talk’ sessions, sessions held outdoors and experiential learning approaches. Similarly, services noted that
they try to employ younger staff to help make the service more ‘youth friendly’.
Many of the barriers service providers described in response to this question are related to the key
enablers, highlighting the ongoing challenges they face with limited referral pathways in some
communities, waitlists for tertiary mental health services where young people have more complex or acute
needs, and headspace service waitlists impacting accessibility as well. In regional areas, the distance
between towns and the lack of public transport were also raised as barriers to access.
As described in appendix D.4, above, when asked whether responses to this question changed when
considering young people from ‘hard to reach’ groups, providers felt that their headspace service was less
able to support access rates of young people with disability, young people from culturally and
linguistically diverse backgrounds and Aboriginal and Torres Strait Islander young people.
Figure 52: Responses from service and lead agency survey: how well does your centre provide services that are youth
friendly, appropriate and accessible

Source: KPMG analysis of headspace service and lead agency survey

Notes: A total of 60 responses were received for this question.

Perspectives of other external stakeholders including school and university counsellors,

GPs and PHNs
School counsellors identified that the branding and youthful energy of services, as well as headspace’s
social media presence worked to destigmatise use of headspace services. They thought this made it less
intimidating for young people to access headspace compared to other clinical services.
There was consistent feedback from school and university counsellors that often, young people prefer
face-to-face supports when they’re seeking the type of counselling and psychology headspace services
provide. School and university counsellors also identified alternative service formats as being helpful.
Drop-in centres and sessions, outreach into schools where a young person can attend a session with a
headspace clinician at school, and social groups were seen as important services, particularly for ‘hard to
reach’ groups.
School and university counsellors identified that service location was an important aspect of accessibility,
with some indicating they did not refer to headspace as they knew the closest service was not accessible
for high school students who can’t drive. They also noted that where there is a distance to a service, access
requires parental support, which isn’t always what the young person wants, or parents may be
unsupportive.

The extent to which headspace provides appropriate, accessible and youth friendly
support
Overall, a range of perspectives from a range of stakeholders indicate that headspace provides appropriate,
accessible and youth friendly supports, with strong positive responses from young people in surveys and
interviews for these domains. The more contact young people had had with their headspace service, the
more likely they were to rate the experience highly, which is a further positive reflection on the
appropriateness, accessibility and youth friendliness of the headspace model.
Qualitative insights indicate that young people value the rapport built with headspace staff, and the easily
accessed location of their local headspace service. At the same time, for those not accessing headspace,
fear of being stigmatised arose in relation to the central location of headspace service sites and being seen
by others seeking mental health support, while the need to be close to public transport was again
highlighted. Barriers to accessibility were raised by users and non-users, including waiting times and the
centre's opening hours. A lack of flexibility to change counsellors within headspace if they were not the
right ‘match’ with the young person was also raised as an area where headspace could be more 'youth
friendly'. Cultural and gender characteristics of the staff member were again very important for a young
person to feel comfortable.
Other stakeholders had positive views of the youth friendly, appropriate and accessible nature of the
services, with drop-in sessions and outreach highlighted as key enablers.
Evaluation results suggest that headspace is effective in enabling young people to access support where,
when and how they want it, and that it is generally appropriate, youth friendly and accessible, with some
issues around opening hours and waiting times proving a challenge.

D.15 Participation of young people in the design and delivery of

headspace
D.15.1 Participation in the design and delivery of services
Table 58 Overview of objectives of headspace for young people’s participation in the design and delivery of services

Objective Short term impacts Medium term impacts

Ensuring young people • Young people feel listened to and • Local service system provides
can access the help they involved in decision-making more youth-friendly, accessible
need in an appropriate, and inclusive services as a
accessible and youth • Young people and families feel
result of learning through
friendly way their needs and interests are
partnerships, shared
understood and reflected in
professional development etc
their local headspace service
(participation outcomes)

Ensuring young people are actively engaged in the design and delivery of the services they receive is
another key element of the headspace program logic. Through providing a positive experience of service
by ensuring young people feel listened to and involved in decision making the overall objectives of the
model are supported.
To examine the extent to which young people are participating in the design and delivery of services, and
how this relates to their experience of headspace, user satisfaction data was analysed, along with
interviews with headspace users and Youth Reference Group members. Stakeholder consultation as part of
deep dive fieldwork provided additional data and insight as to the perceived success of headspace in these
domains.

Evidence of the effectiveness of headspace in ensuring young people participate in

the design and delivery of headspace
Perspectives of young people using headspace
The views of young people captured in the hMDS young person satisfaction matrix indicate that most are
very satisfied with their experience of being involved in the design and delivery of headspace, with the
majority selecting ‘strongly agree’ or ‘agree’ for the statements:
Figure 53: Young people’s experience of being at headspace, interaction with the staff and the service received

Source: KPMG analysis of hMDS of episodes created within 2015-16 to 2019-20

Notes: Sample includes 379,130 episodes. However, only 136,362 episodes had sufficient data summarising the young person’s
experience of being involved in headspace.

Reference groups discussed a variety of activities and influences they had on headspace practice. Young
people in the reference group spoke about the relaxed and flexible approach of headspace staff that
supports and guides them to make their decisions about how to deal with their mental health issues:

“Being your kind of guide in a way of navigating it…They also have like an informal
kind of approach as well, like my experiences with the IPS have been like we’ve met
for a cup of coffee, it’s kind of relaxed and you’re not in the office talking about what
you’re doing with your life, they’re not just like straight to the point necessarily, like
they’re more open-ended towards what your main goals are and they’ve always been
advocating for if you wanted to change your mind about what kind of – you know,
 they’ve assumed that you’re thinking more for yourself, they’re very supportive of…”
Reference group members in one area participated in a review of the forms young people filled out when
they first presented to headspace. They suggested changes to the forms to ‘make it as easy and
straightforward to fill out as possible’. Making the process simple was especially important for young
people who attended headspace alone:

“when you walk in, as much as it’s a bright and friendly environment, it’s still a
reception area that’s just the same as a doctor or the dentist or whatever, sometimes I
think that can be quite daunting, especially if you’re coming by yourself, like if you
haven’t brought mum and dad and they usually fill out the forms for you.“

Members of another reference group helped to facilitate groups of likeminded people around issues they
felt strongly about, to the support young people on the headspace waiting list in their area. The young
people hoped to start a climate group:

“There’s quite a committed presence of people – young people and people of all ages
..who are dedicated to climate action. When you feel connected to them about an
issue the weight of the issue doesn’t seem so heavy…that’s a positive kind of outcome
of being involved with headspace.”

The inclusion of young people in the design and delivery of headspace services is designed to improve the
service experience of the young person and, where relevant, their families. While the hMDS collects
satisfaction data directly from young people it does not survey family members participating in family and
friend focussed OOS. This prevents analysis of the extent to which including young people in design and
decision making is associated with improved service experience for families.

The extent to which young people participate in the design and delivery of
headspace, and how this influences young people and their families’ experience of
headspace
Young people recognise and value the extent to which they are invited to codesign their service
experience, and rated this highly in satisfaction surveys. The extent to which this translates to improved
experience of headspace for their families is unclear, however, as satisfaction of families attending family
focused sessions is not measured.
Evaluation evidence suggests that the headspace model effectively enables young people to participate in
the design and delivery of headspace, and this is associated with positive experiences of headspace for
young people.

Appendix E :
Effectiveness in
improving mental
health and wellbeing
outcomes
E.1 Overview of effectiveness analysis completed
The analysis of headspace’s effectiveness in improving mental health and wellbeing outcomes are a critical
part of this evaluation. This section of the report provides a high level, plain English summary of the
analysis undertaken and corresponding results, before the detailed, technical analysis is provided in
sections E.3 through to E.8.

E.1.1 Overview of effectiveness methodology

The analysis of improvement in outcomes for young people accessing headspace services is based on a
comparison of clinical scores on intake to headspace service, before any support is received, with the last
clinical score captured for the young person (often at the end of their episode of care).
There are three different clinical outcome measures used in the evaluation to determine the impact of
headspace services. These are known as the K10, SOFAS and MLT. A description of each is contained in
section E.2 below.
The analysis of these outcome measures was also adjusted to make sure that any improvement noted in
outcomes could be attributed to headspace services. There is a method for this, known as ‘regression to the
mean’ or RTM. This method estimates the improvements that would have occurred in outcomes for a
young person, without treatment or support from headspace services.
Analysis was conducted on episodes of care, where the primary issue on intake for young people was
either mental health-related, or situational. The episodes of care also needed to have a minimum of two
OOS completed in them, to ensure there were clinical scores record at two points in time, to enable
comparison.

E.1.2 Overview of effectiveness results

There are a number of key findings across this analysis undertaken:
• For all three outcome measures, improvement in scores were noted for young people using headspace,
both based on observed scores, and also once adjusted for RTM.
• For all outcome measures, there was greater improvement in outcomes for young people the more OOS
they had within their episode of care.
• Once analysis was undertaken to determine if the improvements in outcomes were clinically significant,
it was noted that similar to the previous evaluation of headspace, the majority of young people do not
see a clinically significant change to their outcomes.
o 17.2 per cent of young people have a clinically significant change in their K10 scores.
o 43.6 per cent of young people have a clinically significant change in their SOFAS scores.
o 31 per cent of young people have a clinically significant change in their MLT scores.
• There are also a range of other factors that influence the extent to which young people experience
positive mental health outcomes. These include things such as age, whether the person identifies as
part of a ‘hard to reach’ cohort, such as being from a culturally and linguistically diverse background.
However, the biggest drivers for young people’s outcomes were the number of OOS they received, and
their initial level of psychological distress on intake to headspace. For those who attended two or more
sessions, the greater the level of distress, the greater the improvement recorded in outcomes.
• In addition, individual service factors also contributed to differences in outcomes. However, how this
occurs for each of the three outcomes measures differs, and there are no clear patterns in which
individual service factors are associated with headspace services delivering above-average clinical
outcomes for young people.

Effectiveness of headspace services over time

The analysis also considered the extent to which outcomes are sustained over time. To understand this,
headspace provides a follow up survey to young people three months after their episode of care ended.
These surveys focus on K10 outcomes only. It should be noted that the response rate for this survey was
only 4.6 per cent from all surveys delivered between 2015-16 and 2019-20.
The follow up survey highlights that outcomes achieved during a headspace episode of care are sustained
over the following 90 days. However, the follow up survey response rate is low and likely suffers from
non-response bias . This bias arises when young people who did not respond to the follow-up survey are
203F

systematically different from those who did respond. Further, non-response bias becomes a critical issue
when response rates fall below 70 per cent . Appendix H show that completion of the follow up survey was
204F

not random. For example, 15 to 19 year old young persons were more likely to complete the survey than
their younger counterparts. Further, the young persons with relatively higher initial K10 outcomes and
relatively lower K10 outcomes are more likely to complete the follow up survey. It is likely responses
from the survey are provided by young people who had a significant impact after accessing a headspace
service. Further follow up would improve the reliability of this finding.

E.1.3 Overview of effectiveness at an area-level

Analysis was also undertaken of the impact of headspace services at the ‘area-level’ – or at the PHN level.
There were three ways in which the impacts of headspace services were measured:
• based on the number of headspace services in the area;
• based on the number of headspace clients per 1,000 young people in the area; and
• based on the ratio of headspace OOS to MBS-funded mental health services in the area.
The impact of these factors was considered for:
• mental health related hospitalisations;
• self-harm hospitalisations;
• substance abuse hospitalisations;
• suicide deaths;
• MBS mental health services accessed; and
• Mental health related emergency department presentations.
The effectiveness of headspace in improving area-level outcomes is inconclusive. There is some evidence
that headspace has an effect on some outcomes but the impacts are typically lagged and inconsistent over
time. For example, there is a meaningful improvement in self harm hospitalisations in local areas, however
this is only seen three years after a new headspace service is established. Further, this improvement is only
observed when measuring headspace exposure with number of active services and ratio of headspace
occasion of services to MBS-subsidised mental health services. When measuring exposure with number of
young person clients per 1,000 young persons, this impact disappears.
Also, the introduction of a new headspace service does not have a statistically significant impact on mental
health related hospitalisations, suicide deaths, MBS-subsidised mental health services, and mental health
emergency department presentations.

E.2 How effective is headspace in improving mental health and

wellbeing outcomes?
Table 59 Overview of mental health and wellbeing objectives of headspace

Objective Short term impacts Medium term impacts

Improving mental • Young people accessing • Young people accessing
health and wellbeing headspace services feel more headspace services experience
outcomes for young hopeful for the future improvements (or stability) in
people aged 12 to 25 social and occupational
years -improvements in • Young people accessing
functioning
K10 SOFAS and MLT headspace services feel better
outcome measures able to cope • Young people accessing
headspace services experience
• Young people accessing
improvements in their quality
headspace services gain skills
of life and wellbeing
to better manage their mental
health and wellbeing issues • Family and friends accessing
headspace services have
• Young people accessing
increased capacity to support
headspace services experience
their young person
reduction in symptoms and
levels of psychological distress • Young people report sustained
and increased wellbeing improvements in mental health
• Young people accessing • Young people who receive
headspace services start to work/study, alcohol or other
experience improvement to drug, and/or physical health
their day to day lives assistance are better able to
manage these aspects of their
• Young people accessing
life in the medium- to long-
headspace services receive
term
appropriate support for
physical health, alcohol and
substance use and work and
study needs
• Young people who receive
work/study, alcohol or other
drug and/or physical health
assistance, gain skills to better
manage these aspects of their
lives

E.2.1 Extent to which young people accessing headspace achieve improvements in

mental health and wellbeing
To evaluate the effectiveness of the headspace program in improving outcomes, three factors were
considered:
• Primary issue on intake: what is the primary issue when a young person attends a headspace service?
An improvement in mental health outcomes may not be an appropriate goal for those presenting with
non-mental health issues.
• Occasions of service: how many OOS did a young person receive during an episode of care with a
headspace service? Improvement in mental health outcomes has been found to be positively associated
with number of OOS attended by a young person . 205F

• Improvement in outcomes: did the young person experience any improvement in their mental health
and wellbeing outcomes? If so, was it a clinically meaningful improvement? Without a control group,
how much of an improvement can be attributed to headspace?
These issues are examined below using episodes created during 2019-20 and closed before December
2020. Full details of the dataset used in the subsequent sections, including exclusion criteria, are provided
in Appendix F.

Primary issue on intake

Figure 54 shows that during 2019-20, around two thirds of episodes had “Mental health and behaviour” as
the young person’s primary issue when accessing headspace. The subsequent analysis of the effectiveness
of headspace focuses on these episodes, i.e., episodes with a ‘mental health and behaviour’ or ‘situational’
primary issue during initial presentation. It should be noted that hMDS data collection only allows for
selection of a limited number of presenting issues, which impacts the completeness of this data.
Figure 54: Primary issue during initial presentation per episode during 2019-20
Source: KPMG analysis of KPMG master dataset – closed episodes
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 61,911; Habitual behaviour issues refer to addictions
other than alcohol and drugs. These include technology use (e.g., social media and gaming), gambling and pornography.

Occasions of service
Figure 55 summarises the distribution of the OOS per episode during 2019-20. Thirty-six per cent of all
episodes had one OOS. By contrast, sixty four per cent had at least two OOS. This distribution has been
relatively consistent over the last five years.
Figure 55: Distribution of OOS per episode during 2019-20

Source: KPMG analysis of KPMG master dataset – closed episodes

Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 61,911. Some further episodes of care were not
completed in time to be included in this report, and they likely would have two or more OOS.

Analysis of outcomes
The evaluation adopts a pre-post quasi-experimental methodology for the analysis of improvement in
outcomes at the episode level. Young person outcomes are recorded at intake into headspace and before
multiple OOS across an episode. Intake measures are considered the pretreatment measures; the last
observed outcome measure within an episode is considered the post-treatment outcome measure. A
minimum of two OOS are required for a prepost comparison to be made. See Appendix F for a description
of how the dataset was derived.
As per Section 2.2.3 , this evaluation considered three outcomes available within the hMDS:
1) The Kessler Psychological Distress Scale (K10). The K10 is a 10-item questionnaire intended to yield a
global measure of distress based on questions about anxiety and depressive symptoms that a person
has experienced in the most recent four-week period . The K10 measure is a sum of all responses to
206F

the ten items, producing a value ranging from 10 to 50, with values indicating higher levels of
distress.
2) The Social and Occupational Functioning Assessment Scale (SOFAS). The SOFAS is a global rating of
current social and occupational functioning from zero to 100, with lower values representing lower
functioning. It is a single-item assessment of current functioning, independent of the severity of the
young person’s psychological symptoms . 207F

3) My Life Tracker (MLT). MLT was developed specifically for use in headspace with the purpose of
providing a quality of life measure that better reflected the important areas of life for young people.
The MLT is a five-item measure, where each item is rated on a zero to 100 scale, with 100
representing the highest level of wellbeing in that domain. The MLT takes the average of the five
responses . 208F
Adjusting for regression to the mean
Pre-post evaluations can suffer from a statistical phenomenon called regression to the mean (RTM) . RTM 209F

occurs when a high measurement at one point in time is followed by a lower value on remeasurement,
even in the absence of an intervention. It also accounts for low measures observed during entry into
headspace and adjustment to the mean in the post-measure. Failing to account for RTM risks
overestimating the treatment effect of headspace . 210F

This evaluation accounts for RTM by explicitly calculating an RTM effect using the variation in outcomes
observed between measurement at intake and measurement before the second OOS. In most cases, the first
OOS is an intake assessment rather than an explicit treatment, and as such the first and second
measurements capture natural variation in the outcome measure.
The RTM effect is interpreted as the ‘expected’ change due to natural variation as opposed to change due
to headspace, and is used to adjust the ‘pre’ measure in the prepost evaluation. It is acknowledged that this
methodology provides only a proxy of the likely RTM, and that there may be some treatment effect
associated with the intake assessment. A secondary check of the magnitude of the RTM effect was also
completed using the RTMCI Stata command by Ariel Linden . 211F

RTM can occur both positively and negatively: young people with initially low levels of psychological
distress can regress to worse levels, independent of headspace, just as young people with high initial levels
of psychological distress can regress to improved levels. To incorporate this, the RTM adjustment factor is
estimated for each quintile of initial outcome measures. This approach captures that observed worsening of
psychological distress for young people with low levels of psychological distress on intake is more likely a
result of RTM than headspace.
Table 60 summarises the magnitude of the RTM effect by initial outcome quintile. As expected, the RTM
effect is largest for young people in worst quintiles on intake. For example, the RTM effect for those
young people in the highest quintile of psychological distress on intake (mean K10 of 42), is estimated to
be 2.9 points.
Table 60: Regression to the mean effect by outcome measure

Quintiles of Initial level of K10 RTM effect SOFAS RTM effect MLT RTM effect
distress1 (SE) (SE) (SE)
0.8* 6.5* 6.9*
First
(0.02) (0.05) (0.06)
-0.3* 2.4* 4.1*
Second
(0.02) (0.04) (0.06)
-1.1* 0.3* 2.3*
Third
(0.02) (0.04) (0.06)
-1.8* -2.1* 0.5*
Fourth
(0.02) (0.04) (0.06)
-2.9* -6.1* -2.6*
Fifth
(0.03) (0.06) (0.06)
Source: KPMG analysis of the hMDS with closed episodes created between 2015-16 to 2019-20
Notes: * Signification at 5 per cent. The sample also must have a minimum of two OOS in order to estimate the RTM effects for the
K10, SOFAS, and MLT measures. Further, the sample must include an intake measure, a measure observed during their second
occasion of service, and a final measurement by the end of the episode of care. There are 215,578 episodes to estimate the K10 RTM
effects, 220,964 episodes to estimate the SOFAS RTM effects, and 215,264 episodes to estimate the MLT RTM effects. RTM:
Regression to the mean. SE: Standard errors.

Average improvement in accessing headspace

Table 61 summarises the average changes in the K10 outcome measures by the number of OOS as
observed, and after adjusting for RTM. Improvements in outcomes increased with the number of OOS,
even after adjusted for RTM.
Two meta analyses suggested that the average treatment effect after eight to twelve psychotherapy sessions
is a K10 improvement of around three, which is similar to what is achieved by young people accessing
headspace (see Table 61). The meta analysis reported a standardised mean difference (SMD) of 0.37 for
212F

using psychotherapy. The SMD can be rescaled into the K10 by multiplying the SMD with the standard
deviation of individuals seeking psychological treatment . This estimate can be cited from a representative
213F

observation study such as the one conducted by NovoPsych . NovoPsych reported that the standard
214F

deviation of the K10 score for someone seeking psychological treatment in Australia is 9.1. Thus, the
change in K10 is calculated as 3.37.
Table 61: Average intake, final and change in K10 measurements in young people accessing headspace

Occasions of service Episodes Average intake Average final Average Average

per episode measure (SD) measure (SD) improvement – improvement –
observed (SD) RTM adjusted
(SD)
29.4
1 11,776
(8.6)
29.4 28.3 -1.2 -0.1
2 5,724
(8.5) (8.7) (4.7) (4.6)
29.3 26.9 -2.5 -1.5
3-5 11,978
(8.2) (8.8) (6.0) (5.8)
29.9 26.7 -3.2 -2.2
6-9 6,921
(7.8) (8.7) (6.5) (6.3)
 31.0 27.6 -3.4 -2.1
10+ 3,244
(7.7) (8.7) (7.2) (6.9)
Source: KPMG analysis of KPMG master dataset – Episodes with non-missing services and issue data
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 39,652 including episodes with non-missing intake and/or
final K10 measurements. Improvement is measured by the difference between the last observed measure and the initially observed
measurement. Where indicated, outcome measurements have been adjusted for regression to the mean effects. A negative change
between the final K10 and initial K10 outcome measure indicate better mental health outcomes. SD: Standard deviation.

Table 62 summarises the average changes in the SOFAS outcome measure by the number of OOS as
observed, and after adjusting for RTM. Similar to the K10 outcomes, average improvements in the SOFAS
outcome measure increased with the number of OOS, even after adjusting for RTM.
Table 62: Average intake, final and change in SOFAS measurements in young people accessing headspace

Occasions of service Episodes Average intake Average final Average Average

per episode measurement measurement improvement – improvement –
(SD) (SD) observed RTM adjusted
(SD) (SD)

63.9
1 15,890
(12.4)
64.4 65.6 1.2 0.5
2 6,597
(11.7) (12.4) (10.1) (9.5)
65.0 68.2 3.2 2.7
3-5 13,186
(11.2) (11.8) (11.3) (10.3)
64.9 70.4 5.5 5.0
6-9 7,219
(10.6) (11.8) (12.3) (11.2)
63.6 70.1 6.5 5.5
10+ 3,349
(11.0) (12.0) (13.4) (11.9)
Source: KPMG analysis of KPMG master dataset – Episodes with non-missing services and issue data
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 46,241 including episodes with non missing intake and/or
final SOFAS outcome measures. Improvement is measured by the difference between the last observed measure and the initially
observed measurement. Where indicated, outcome measurements have been adjusted for regression to the mean effects. SD: Standard
deviation.

Table 63 summarises the average changes in the MLT outcome measures by the number of OOS as
observed, and after adjusting for RTM. Like the K10s and SOFAS, average improvements in the MLT
outcome measure increased with the number of OOS, even after adjusting for RTM.
Table 63: Average intake, final and change in MLT measurements in young people accessing headspace

Occasions of service Episodes Average intake Average final Average Average

per episode measures (SD) measures (SD) improvement – improvement –
observed (SD) RTM adjusted
(SD)
11,862 48.8
1
(21.4)
5,752 49.0 53.2 4.2 2.1
2
(21.2) (21.9) (13.7) (13.3)
12,015 49.6 57.9 8.4 6.4
3-5
(20.3) (21.8) (17.4) (16.6)
6,934 48.4 61.5 13.1 11.0
6-9
(19.3) (22.0) (19.8) (18.9)
3,256 45.9 61.0 15.1 12.5
10+
(18.9) (22.4) (21.1) (20.2)
Source: KPMG analysis of KPMG master dataset – Episodes with non-missing services and issue data
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 39,819 including episodes with non missing intake and/or
final MLT measurements. Improvement is measured by the difference between the last observed measure and the initially observed
measurement. Where indicated, outcome measurements have been adjusted for regression to the mean effects. SD: Standard
deviation.

Reliable change and clinically significant change

Reliable change occurs when the change in the outcome measure meets or exceeds the reliable change
index (RCI) as determined by the Jacobson and Truax method . Reliable change is a criterion used to
215F

evaluate whether a change over time of an episode outcome measure (i.e., the difference between the
initial and last observed K10, SOFAS or MLT outcome measure) is considered statistically significantly
greater than a difference that could have occurred due to random measurement error alone. The RCIs for
the K10, SOFAS and MLT indicators are summarised below in Table 64.
Table 64: Reliable significant change index by outcome measure

Gender Age Reliable change index

K10 SOFAS MLT
All All 7 10 18
Source: Rickwood et al. (2015) ; Kwan et al. (2018) ; Kwan & Rickwood (2020)
216F 217F 218F

Clinically significant change occurs when the young people with an outcome measure within the clinical
range when they present to headspace and who mental health improves so that they are no longer in the
clinical range when they closed their treatment episode. The cut-off point separating the two populations is
known as the clinically significant index (CSI) and is also determined by the Jacobson and Truax method . 219F
The CSI cut-off points for the K10, SOFAS and MLT indicators are summarised below in Table 65. In this
analysis, the cut-off points only differ by age and gender for the MLT indicator based on recent literature . 220F

Table 65: Clinically significant change index by gender and age

Gender Age Clinically significant change index

K10 SOFAS MLT

Male Less than 14 years 23.0 69.0 75.5

15 to 17 years 23.0 69.0 68.8

18 to 21 years 23.0 69.0 61.6

22 years and older 23.0 69.0 61.7

Female Less than 14 years 23.0 69.0 68.3
15 to 17 years 23.0 69.0 58.3
18 to 21 years 23.0 69.0 57.2

22 years and older 23.0 69.0 59.3

Source: Rickwood et al. (2015) ; Kwan & Rickwood (2020)
221F 222F

Table 66 shows the proportion of closed episodes within the effectiveness analysis dataset that experienced
a reliable change and clinically significant change.
• Reliable change was observed in 16, 23 and 25 per cent of episodes using the K10, SOFAS and MLT
outcome measures, respectively. That is, 16, 23 and 25 per cent of episodes experienced changes in the
K10, SOFAS and MLT, respectively, that are unlikely to be due to simple measurement unreliability.
• Clinically significant change was observed in 17, 44 and 31 per cent of episodes using the K10, SOFAS
and MLT outcome measures, respectively. Note that it is not possible to assess the clinical
improvement in young people who were in the non-clinical population at intake to headspace, and as a
result these episodes were excluded from the clinically significant change analysis. Consistent with
previous evaluations, a majority of young people within the clinical population did not achieve a
reliable change or a clinically significant change in their mental health outcomes, based on the three
outcome measures considered . 223F

Table 66: Improvement in average outcome measures in young people accessing headspace (per completed episode)

Method K10 SOFAS MLT

(No. of episodes) (No. of episodes) (No. of episodes)
Reliable change 15.5% 22.5% 24.6%
(27,867) (30,351) (27,957)
Clinically significant change 17.2% 43.6% 31.0%
(21,477) (23,569) (21,786)
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures
Notes: See Appendix F for detailed exclusion criteria.

Table 67 presents the proportion of episodes that experienced a reliable change or a clinically significant
change based on raw unadjusted K10, SOFAS and MLT outcome measures.
Table 67: Average Improvement in raw outcome measures in young people accessing headspace (per completed
episode)

Method K10 SOFAS MLT

(No. of episodes) (No. of episodes) (No. of episodes)
Reliable change 21.7% 28.8% 28.1%
(27,867) (30,351) (27,957)
Clinically significant change 18.4% 36.4% 30.5%
(22,312) (18,649) (21,553)
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures
Notes: See Appendix F for detailed exclusion criteria. Improvement is the difference between the last observed outcome measure and
the initially observed measure. Measures are based on raw observed outcome measures.

E.2.2 Effectiveness of headspace in improving outcomes

Approximately 36 per cent of closed episodes during 2019-20 consisted of one single OOS. For those
episodes with at least two OOS, a clinically significant change in the K10 outcome measure was reported
in 17 per cent of cases. A higher proportion of improvement was recorded when using the broader SOFAS
and MLT outcome measures which look at holistic psychosocial functioning, with 44 and 31 per cent of
episodes achieving a clinically significant change, respectively.
Overall, data from the hMDS shows that more engagement and treatment through headspace is associated
with greater improvements in mental health and wellbeing. The improvement in young persons accessing
six or more headspace sessions is on par to that observed from psychotherapy treatments in the literature.
For comparison, a meta analysis within the RAZNCR clinical practice guidelines suggested that the
average treatment effect after eight to 12 psychotherapy sessions for depression is a K10 improvement of
around three, which is similar to what is achieved by young people accessing headspace with at least six
OOS. The challenge remains to increase the share of young persons accessing a greater number of
headspace services.
Overall, headspace is effective at improving outcomes for young people. When young people access six or
more headspace sessions, effectiveness improves.

E.3 Factors associated with the number of occasions of service per

episode
Before evaluating the factors associated with mental health improvements, this reported estimated a
multivariate logistic regression to analyse the variation in probability of a young person receiving two or
more OOS. This is done to better understand the determinants of an episode going beyond the first
occasion of service. This analysis focuses on 49,925 closed episodes, created during 2019-20, from the
‘Episodes with only MH/situational primary issues’ dataset as described in Appendix F.
This is done by estimating the following logistic regression:

where:
• represents an indicator variable if the episode had two or more OOS;
• represents a numerical constant;
• represents a vector of the young person’s characteristics. including age, gender, initial level of mental
distress, Aboriginal and Torres Strait Islander status, culturally and linguistically diverse status and
regionality.

represents the relevant coefficient estimates;

• represents the type of services received. represents the relevant coefficient estimates;
• represents a vector of dummy variables indicating the service where the episode of care was held at.
represents the relevant service fixed effects; and
• represents an error component.
A range of potential explanatory factors are considered:
• Young person socio-demographics such as their age, gender, Aboriginal and Torres Strait Islander status,
culturally and linguistically diverse status, rurality and as well as their initial mental distress on intake.
• The main services the young person received at headspace site.
• Site specific factors. The young persons’ mental health outcome can be influenced by the site they
accessed. This could be because of heterogenous service quality across headspace services, the staff
availability and locally specific unobservable factors affecting mental health wellbeing.
• The regionality of the episode.
The results of the multivariate logistic regression are illustrated in Table 68 and discussed below.

E.3.1 Age
Holding all other factors being equal, young persons younger than 15 years are most likely to receive two
OOS or more. Young people aged older than 24 years are least likely to receive two occasions or more.

E.3.2 Gender
Holding all other factors being equal, there are no statistically significant differences in the probability of
receiving two OOS or more between gender groups.

E.3.3 Priority cohorts

Holding all other factors being equal, Aboriginal and Torres Strait Islander, culturally and linguistically
diverse and LGBTQIA+ cohort status appear to have no meaningful impact of receiving two OOS or
more.

E.3.4 Initial severity

This report hypothesises that the initial severity of the young persons mental health status has an in impact
on their need for mental health services, and the potential range of improvement that treatment can deliver.
If a young person accesses headspace with very high levels of mental distress, it is likely they will require
multiple OOS for adequate treatment.
Holding all other factors being equal, the results show that young persons accessing headspace with low
initial mental distress are least likely to access two or more OOS. Young persons with high or very high
mental distress have the highest probability of receiving two or more OOS.

E.3.5 Main services provided

Holding all other factors being equal, those receiving non-mental health services or only intake/assessment
have relatively lower probabilities of receiving more than two OOS.

E.3.6 Rurality
Holding all other factors being equal, young people residing in major cities are more likely than other
young persons to have two or more OOS than young people residing in inner or outer regional areas. There
are no statistically significant differences between young people in major cities and their counter parts in
very remote or remote areas.
Table 68: Logit regression of receiving two or more OOS
Independent variables Probability of receiving two or more OOS
(1)
Age categories (ref = younger than 15 years)
15 to 19 years old 0.7**
(0.03)

20 to 24 years old 0.6**

(0.04)

Older than 24 years 0.4**

(0.07)

Gender (ref = Male)

Female 1.0
(0.04)

Other 1.1
(0.18)

Aboriginal and Torres 1.0

Strait Islander cohort status (0.05)
LGBTQIA+ cohort status 1.0
(0.05)

Culturally and linguistically diverse cohort status 1.0

(0.07)

Rurality (ref = Major cities)

Inner regional 0.8*
(0.08)
Outer regional 0.8**
(0.09)
Remote or very remote 1.1
(0.40)
Initial distress categories (ref = Low distress)
Moderate distress 1.2*
(0.11)
High distress 1.6**
(0.15)
Very high distress 1.8**
(0.16)
Main services provided (ref = Intake/assessment)
Non-MH services 21.8**
(3.76)
MH services 46.0**
(5.01)

No. of episodes 29,298

Pseudo R2 0.43
Source: KPMG analysis of the ‘Episodes with only MH/situational primary issues’ dataset
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 29,298, where 20,627 episodes were excluded due to
missing young persons characteristics. Coefficients reported are odds ratios. Standard errors in parentheses. For brevity, this report
did not include the site’s coefficient estimates *: Significant at 10 per cent; **: Significant at 5 per cent. MH: Mental health.

E.4 How do outcomes vary by service?

The analyses below highlight the variation in the outcomes across services, based on the same data used in
Section E.2 (see Appendix F for full details), and without accounting for differences in young person’s
demographic characteristics, the type or number of services received at a headspace service or their initial
level of mental distress.
Figure 56 shows that 90 out of 117 headspace services delivered a statistically significant improvement in
the K10 outcome measure, on average, with the remainder reporting an average measure that is
statistically insignificantly different from zero. Five services delivered average improvements greater than
three K10 points, more than double the national average.
Figure 56: Distribution of K10 improvements by headspace service
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – K10 analysis
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 27,867. Improvement is measured by the difference
between the last observed measure and the initially observed measurement. Outcome measurements have been adjusted for regression
to the mean effects. Positive values indicate an improvement in K10 outcomes. There are 117 services with complete initial and final
K10 outcome measures.

Figure 57 shows that 97 out of 118 headspace services delivered a statistically significant improvement in
the SOFAS outcome measure, with two reporting average outcomes statistically significantly below zero.
Figure 57: Distribution of SOFAS improvements by headspace service

Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – SOFAS analysis
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 30,351. Improvement is measured by the difference
between the last observed measure and the initially observed measurement. Outcome measurements have been adjusted for RTM
effects. Positive values indicate an improvement in SOFAS outcomes. There are 118 services with complete initial and final SOFAS
outcome measure.

Figure 58 shows that 109 out of 117 headspace services delivered a statistically significant improvement in
the MLT outcome measure, with eight services reporting average changes in the MLT outcome measure
that are statistically similar to zero.
Figure 58: Distribution of MLT improvements by headspace service
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – MLT analysis
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 27,957. Improvement is the difference between the last
observed and the initially observed outcome measure. Positive values indicate an improvement in MLT outcomes. Outcome
measurements have been adjusted for RTM effects. There are 117 services with complete initial and final MLT outcome measures.

E.4.1 Variation in outcomes by service

Figure 56, Figure 57 and Figure 58 showed that the majority of headspace services delivered a statistically
significant and positive mental health outcome among the sampled episodes. Across all outcome measures,
there are a small proportion of services that deliver either significantly better or worse results than average.
However, it is not clear if these outcomes are a result of the services themselves or are confounded by
additional factors such as the young person’s demographic characteristics, the type or number of services
received at a headspace service or their initial level of mental distress. This is explored in Appendix E.5 in
which a multivariate multi-level linear regression is used to analyse variation in outcomes to better
understand why some episodes of care, and some services, experienced better outcomes than others.
Analysis of variation suggests that the majority of headspace services deliver a positive and statistically
significant improvement in mental health and broader outcomes as measured by the K10, SOFAS and
MLT outcome measures.

E.5 What factors are associated with positive outcomes?

A multivariate multi-level linear regression was used to analyse variation in outcomes to better understand
why some young people experience better outcomes than others, and why on average, some services
delivered a larger improvement in outcomes. A range of explanatory factors were considered:
• Young person factors – socio-demographics such as their age, gender, Aboriginal and Torres Strait
Islander status, culturally and linguistically diverse status, as well as their initial psychological distress
on intake as measured by the K10.
• Occasion of service factors – the number and type of services received during their episode. During an
episode, a young person can receive mental health only services; non-mental health only services
(such as vocational services); or any mixture of mental health and non-mental health services.
• Service specific factors – a young person’s mental health outcome can be influenced by the headspace
service they accessed. This could be because of heterogenous service quality across headspace
services, the staff availability and locally specific unobservable factors affecting mental health
wellbeing.
This is done by estimating the following regression:
where:
• represents the change between the final and initial K10, SOFAS or MLT outcome measures for an
episode . The change between the initial and final observed outcome measures have been adjusted for
RTM;
• represents a numerical constant;
• represents a vector of the young person’s characteristics including age, gender, initial level of mental
distress, Aboriginal and Torres Strait Islander status, culturally and linguistically diverse status and
regionality.

represents the relevant coefficient estimates;

• represents a vector of variables including the episode’s number of OOS and the type of services received.

represents the relevant coefficient estimates;

• represents a vector of dummy variables indicating the service where the episode of care was held at.

represents the relevant service fixed effects; and

• represents an error component.
Although the effectiveness analysis dataset includes 33,394 episodes, only 27,867 episodes had
nonmissing initial and final K10 outcome measure. The regression further dropped episodes with missing
explanatory factors leading to a final sample of 22,348 episodes for the K10 regression.
For the SOFAS improvement regression, only 30,351 episodes had non-missing initial and final SOFAS
outcome measure. The SOFAS regression further dropped episodes with missing explanatory factors
leading to a final sample of 22,254 episodes for analysis.
For the MLT improvement regression, only 27,957 episodes had non-missing initial and final MLT
outcome measures. The MLT regression further dropped episodes with missing explanatory factors leading
to a final sample of 22,333 episodes for analysis.
Table 69 below summaries the impact each explanatory variable had on the RTM adjusted changes in the
K10, SOFAS and MLT outcome measures. Each coefficient represents the impact the variable had, holding
all other factors being equal, on unit changes in the K10, SOFAS or MLT outcome measure. For example,
holding all other factors equal, a young person aged 15 to 19 years would have a K10 improvement that is
0.2 smaller than their counterparts aged younger than 15 years.
Key results of the regression analysis are presented in Figure 60, Figure 61 and Figure 62, and discussed
below. Figure 60, Figure 61 and Figure 62 present the average improvements in the K10, SOFAS and MLT
outcome measure, respectively, for each cohort group.
Table 69: Linear regression of mental health improvements

Independent variables Dependent variable

K10 improvement SOFAS improvement MLT improvement
(1) (2) (3)
Age categories (ref = younger than 15 years)
15 to 19 years old -0.2** 0.3 -0.6**
(0.10) (0.18) (0.29)

20 to 24 years old 0.2 0.9** 0.4

(0.11) (0.20) (0.33)

Older than 24 years 0.4 1.3** -0.1

(0.34) (0.60) (0.98)

Gender (ref = Male)

Female -0.6** -0.0 -1.1**
(0.08) (0.15) (0.25)

Other -0.8** -1.9** -3.1**

(0.31) (0.56) (0.91)

Priority cohorts
Aboriginal and Torres -0.3* -1.0** -1.3**
Strait Islander cohort status
(0.15) (0.26) (0.43)

LGBTQIA+ cohort status -0.7** -0.7** -1.7**

(0.10) (0.18) (0.29)

Culturally and linguistically 0.1 -0.2 -0.3

diverse cohort status
(0.13) (0.24) (0.39)

Rurality (ref = Major cities)

Inner regional -0.09 -0.22 -0.01

(0.10) (0.26) (0.23)

Outer regional -0.3 -0.2 -1.1
(0.24) (0.43) (0.70)

Remote or very remote 0.3 0.5 -0.8

(0.37) (0.66) (1.08)

Initial distress categories (ref = Low distress)

Moderate distress 1.9** -0.2 1.5**
(0.23) (0.41) (0.66)

High distress 2.1** -1.1** 2.4**

(0.20) (0.37) (0.60)

Very high distress 2.8** -2.1** 3.2**

(0.20) (0.37) (0.60)

Number of occasions of service (ref = 1 OOS)

2 OOS 0.7** 0.9** 2.2**
(0.13) (0.23) (0.38)

3 to 5 OOS 1.6** 2.7** 6.1**

(0.12) (0.21) (0.34)

6 to 8 OOS 2.0** 3.8** 9.4**

(0.14) (0.25) (0.41)

9 to 13 OOS 1.9** 4.6** 10.5**

(0.17) (0.30) (0.49)

More than 14 OOS 2.2** 4.8** 10.9**

(0.26) (0.46) (0.75)

Main services provided (ref = Intake/assessment)

Non-MH services 0.2 2.9** 0.3
(0.24) (0.43) (0.70)

MH and non-MH services 0.2 2.0** 0.3

(0.19) (0.33) (0.54)

MH only services 0.4** 2.0** 0.8*

(0.14) (0.25) (0.40)

Observations 22,348 22,254 22,333

R2 0.05 0.07 0.07
Adjusted R2 0.04 0.06 0.07
Source: KPMG analysis of the ‘K10 analysis’ dataset, ‘SOFAS analysis’ dataset and ‘MLT analysis’ dataset
Notes: See Appendix F for detailed exclusion criteria. Number of episodes are subject to missing data on initial and final outcomes
and young persons’ characteristics. Improvement is the difference between the last observed and the initially observed outcome
measure. Standard errors in parentheses. For brevity, this report did not include the site’s coefficient estimates. Improvements have
been adjusted for regression to the mean. *: Significant at 10 per cent; **: Significant at 5 per cent. OOS: Occasions of service; MH:
Mental health.

E.5.1 Young person factors

Age
The impact of age on mental health and wellbeing is dependent on the choice of the outcome measures.
Every age cohort experienced positive improvements in their mental health outcomes, but the
improvement varies across the cohorts. Young people within the 15 to 19 year old age group had the
lowest improvement compared to their younger or older counterparts. The 20 to 24 year old age group had
a statistically larger improvement in K10 outcomes. The results suggest there is a nonlinear relationship
between the young person’s age and their mental health outcomes.

Gender
Males had a statistically larger improvement in K10 and MLT outcomes than female or the non-binary
groups. There was no significant difference between genders in SOFAS outcomes.

Priority cohorts
LGBTQIA+ young people experienced lower improvements than young people who did not identify as
LGBTQIA+ across all measures. By contrast culturally and linguistically diverse cohorts achieved
statistically similar improvements as non-culturally and linguistically diverse cohorts. Improvements in the
SOFAS and MLT outcome measures were statistically significantly lower than the average improvements
in the outcome measure among the Aboriginal and Torres Strait Islander cohort. K10 improvements among
the Aboriginal and Torres Strait Islander cohort were statistically similar to the sample average.
This is in contrast to the observations of headspace service providers, who indicated in the service and lead
agency survey that they thought outcomes would be worse for culturally and linguistically diverse young
people, and they thought that LGBTQIA+ young people would experience higher improvements than
young people from the broader population attending headspace. Providers successfully predicted that
improvements would be significantly lower for Aboriginal and Torres Strait Islander young people,
according to the SOFAS and MLT scores.
Figure 59: Responses from service and lead agency survey: how well does your centre provide services that improve
the mental health and wellbeing of young people?
Source: KPMG analysis of headspace service and lead agency survey
Notes: A total of 60 responses were received for this question.

Initial severity (measured by the K10)

This report hypothesise that the initial severity of the young persons mental health status has an in impact
on their need for mental health services, and the potential range of improvement that treatment can deliver.
If a young person accesses headspace with very high levels of mental distress and if treatment is
successful, then the young person will experience a significant larger improvement in mental health than a
young person accessing headspace with low or mild levels of mental distress.
The results show that improvements in K10 and MLT outcomes were largest among young people
presenting to a headspace service with initially very high levels of mental distress. Of all factors, the initial
severity of psychological distress had the largest impact on the magnitude of the improvement achieved . 224F

By contrast, improvements in the SOFAS were highest among young people entering headspace with low
levels of mental distress and lowest among the cohort with the highest level of mental distress. This is
likely explained by weak correlation between the K10 and SOFAS measures (which is explored later in
Table 71) and the RTM adjustment. A separate model using only observed changes in the SOFAS outcome
measure show no statistically significantly differences in outcomes by initial levels of mental distress.

E.5.2 Service factors

Types of services provided

After controlling for variation in young people’s characteristics, and the headspace service, episodes
treated with only mental health services experienced the largest improvement in the K10 and MLT while
those receiving only intake/assessment services experienced the lowest but still positive improvement.

Occasions of service
After controlling for variation in young people’s characteristics, and the headspace service, there was a
clear relationship between outcomes and the OOS attended, with young people attending more than six
and services achieving better outcomes than those attending five or less; and those attending nine or more
achieving better outcomes again.

E.5.3 Service-specific factors

Individual service factors

The individual service’s contribution was estimated within the analysis using a fixed effects approach.
Further analysis and results are discussed in the subsequent sections below.

E.5.4 Contribution to variation

Shaply decomposition was used to assess the relative contributions of young person, OOS and service
factors to the observed variation in outcomes . 225F

Table 70 suggests that the OOS factors (i.e., the number of OOS and the type of services provided) were
the most important factors in explaining the variance in the young persons’ outcomes, accounting for
between 40 and 65 per cent of variation, depending on outcome. Service factors were next most important
(and marginally most important for the SOFAS), accounting for between 27 and 45 per cent of variation.
Young person factors contributed least to variation in outcomes, particularly for the SOFAS and MLT
outcomes.
Within these factor groupings, the number of OOS, the individual service itself and the initial levels of
mental distress (as measured by the K10) were the most important sub-factors explaining variation in
outcomes.
There is evidence that headspace is delivering meaningful improvements but these are concentrated
episodes with at least six OOS based on the results presented in Figure 60, Figure 61, Figure 62 and Table
70. headspace stands to maximise mental health improvements for young people by ensuring they do not
drop out with only one occasion of service and continue with at least six consecutive OOS.
Table 70: Shaply decomposition by patient, service, regional components
Component K10 SOFAS MLT
Young person factors 29.4% 10.5% 8.4%
Age 1.9% 1.1% 1.1%
Gender 4.4% 1.1% 2.0%
Aboriginal and Torres Strait 0.3% 1.1% 0.5%
Islander young person
LGBTQIA+ young person 4.3% 1.3% 2.3%
culturally and linguistically diverse 0.1% 0.1% 0.1%
young person
Initial level of mental distress 18.4% 5.8% 2.6%
Occasions of service factors 40.3% 44.2% 64.5%
No. of occasions of service 34.6% 31.9% 57.9%
Type of services received 5.7% 12.3% 6.6%
Service-level factors 30.3% 45.3% 27.1%
Total 100% 100% 100%
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – K10 analysis, SOFAS analysis and
MLT analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measure. Number of episodes: 27,867 (K10), 30,351 (SOFAS) and 27,957 (MLT). Improvement is the
difference between the last observed and the initially observed outcome measure. Improvements across the K10, SOFAS and MLT
measures have been adjusted for RTM.
Figure 60: Average improvement in the K10 by young person, OOS and service-level factors

Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – K10 analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measures. Number of episodes: 22,348, where 5,519 out of 27,867 episodes were excluded due to missing
young persons characteristics. Improvement is the difference between the last observed and the initially observed outcome measures.
Positive values indicate an improvement in K10 outcomes. K10 outcomes have been adjusted for RTM.
Figure 61: Average improvement in the SOFAS by young person, OOS and service-level factors

Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – SOFAS analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measure. Number of episodes: 22,254 episodes, where 8,097 out of 30,351 episodes were excluded due to
missing young persons characteristics. Improvement is the difference between the last observed and the initially observed outcome
measures. Positive values indicate an improvement in SOFAS outcomes. SOFAS outcomes have been adjusted for RTM.
Figure 62: Average improvement in the MLT by young person, OOS and service-level factors
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – MLT analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measures. Number of episodes: 22,348, where 5,609 out of 27,957 episodes were excluded due to missing
young persons characteristics. Improvement is the difference between the last observed and the initially observed outcome measures.
Positive values indicate an improvement in MLT outcomes. MLT outcomes have been adjusted for RTM.

E.5.5 Service factors – detailed analysis of variation associated with individual

services
Table 69 summarises the impact of young person specific factors and episode specific factors on an
episode’s mental health outcome, but omitted the average impact a headspace service had on an episodes
outcome measure. Figure 63, Figure 64and Figure 65 below summarise the impact each of the 117 226F

headspace services within our estimable sample headspace service had on an episode’s mental health
improvement in the K10, SOFAS and MLT outcome, after adjusting for young person and occasion of
service differences.
This is done to examine if the services are delivering consistent outcomes, but Figure 63, Figure 64and
Figure 65 suggests there are significantly variations in impacts associated with each headspace service.
The average K10, SOFAS and MLT improvements across all headspace services was 1.5, 3.2 and 7.4 227F

points, respectively.
Figure 63: Distribution of the K10 fixed effects
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – K10 analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measures. Number of episodes: 22,348, where 5,519 out of 27,867 episodes were excluded due to missing
young persons characteristics. Improvement is the difference between the last observed and the initially observed outcome measures.
Positive values indicate an improvement in K10 outcomes. Improvement has been adjusted for RTM. For services with large
confidence intervals (where the upper bounds exceed 5 or the lower bound exceed 3) the sample size range from 2 to 35 episodes.

Figure 64: Distribution of the SOFAS fixed effects

Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – SOFAS analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measure. Number of episodes: 22,254 episodes, where 8,097 out of 30,351 episodes were excluded due to
missing young persons characteristics. Positive values indicate an improvement in SOFAS outcomes. Improvement has been adjusted
for RTM. For services with large confidence intervals (where the upper bounds exceed 10 or the lower bound exceed 10) the sample
size range from 2 to 68 episodes.

Figure 65: Distribution of the MLT fixed effects

Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – MLT analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measure. Number of episodes: 22,348, where 5,609 out of 27,957 episodes were excluded due to missing
young persons characteristics. Positive values indicate an improvement in MLT outcomes. Improvement has been adjusted for RTM.
Improvement has been adjusted for RTM. For services with large confidence intervals (where the upper bounds exceed 10 or the
lower bound exceed -10) the sample size range from 2 to 68 episodes.

Figure 63 shows that there were 13 headspace services with statistically significantly higher improvements
than the average improvement of 1.5, and there were ten headspace services with improvements that were
statistically significantly lower than the average. Figure 64, shows only five out of the 13 services had a
SOFAS improvement above the average of 3.2, with one service with an improvement below the average.
In Figure 65, eight out of the 13 headspace services had MLT improvements higher than the average
improvement of 7.4. The correlation between the service performance across the three outcome measures
is summarised in Table 71. There is a strong positive correlation between service performance in the K10
and MLT improvements, with a value of 0.7, suggesting services that deliver above average improvements
in the K10 also deliver above average improvements in the MLT. By contrast, the correlation is weak
between service performance on the K10 and SOFAS improvements, and the SOFAS and MLT
improvements.
Table 71: Correlation matrix between services fixed effects

K10 SOFAS MLT

K10 1.0

SOFAS -0.1 1.0

MLT 0.7 0.1 1.0

Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – K10 analysis, SOFAS analysis and
MLT analysis
Notes: See Appendix F for detailed exclusion criteria. Sample includes closed episodes that were created during 2019-20; presented
with a mental health or situational primary issue; had no missing main services data; had at least two OOS observations; and had an
initial and final outcome measure. Number of episodes: 27,867 (K10), 30,351 (SOFAS) and 27,957 (MLT). Improvement is the
difference between the last observed and the initially observed outcome measure. Improvements across the K10, SOFAS and MLT
measures have been adjusted for RTM.

E.5.6 Factors associated with positive outcomes

The number of OOS is a key factor associated with positive outcomes. This finding reiterates the
importance of ensuring young people access the requisite number of sessions in order to maximise the
benefit from attending headspace.
A young person’s initial level of mental distress, as measured by the K10, was also an important factor in
explaining variation in outcomes, with those with worse initial distress likely to experience a greater
improvement. While this finding reflects the greater potential for improvement in this cohort, it does
highlight that headspace can be effective at improving outcomes for young people with more severe levels
of distress.
Other young person attributes were not a major driver of variation in outcomes, which highlights that
headspace, in general, makes positive improvements to all young person cohorts. Nonetheless,
LGBTQIA+ and Aboriginal and Torres Strait Islander young people consistently experienced smaller
improvements than young people who do not identify as LGBTQIA+ and young people who do not
identify as Aboriginal and Torres Strait Islander cohorts respectively, which suggests there is potential for
further improvement in maximising outcomes for priority cohorts.
Individual services factors also make a sizeable contribution to the K10, MLT and SOFAS outcomes, with
services that deliver above average results in the K10 also likely to deliver above average results in the
MLT. The specific service factors associated with above average services are investigated in the next
section.

E.5.7 Factors associated with negative or zero outcomes

Not every sampled episode experienced a positive improvement by the closure of their episode of care.
Around 55, 38 and 37 per cent of closed episodes created after June 2019 had a zero or negative mental
health improvement according to the K10, SOFAS and MLT outcome measures, respectively.
As highlighted above, episodes with at least six OOS experienced the highest improvement in their mental
health outcomes. Further higher levels of initial distress (as measured by the K10) is positively correlated
with improvements in the K10, SOFAS and MLT outcomes.
Around 20 per cent of the sampled episodes with non-positive K10 outcomes entered headspace with high
or very high mental distress and received more than six OOS. For episodes with non-positive SOFAS or
MLT outcomes, the proportions are 20 and 18 per cent, respectively. These episodes are a minority of
episodes with reported non-positive outcomes.
The number of OOS, and the individual service itself, are the key drivers of variation in outcomes. Young
people who go on to access at least six to eight OOS achieve the greatest improvement in outcomes.

E.6 What factors are associated with above average headspace

services?
The effectiveness analysis is followed up by second level regression to analyse the relationship between
the service fixed effects and service-level factors. These fixed effects represent the average K10, SOFAS
and MLT improvements delivered for each service as estimated with the coefficients listed under Table 72.
This report estimated another multivariate linear regression to analyse drivers of variation across
headspace services. A range of explanatory factors were considered below:
• Service-specific factors such as maturity and size (as measured by service volumes), share of OOS
funded through the MBS, that are mental health services, and if the headspace service is a headspace
centre or satellite service.
• Geographic factors such as rurality and state of the headspace service, to account for regional differences.
The regression is as follows:
where:

• represents the K10, SOFAS or MLT fixed effects for service ;

• represents a numerical constant;
• represents a vector of service-level explanatory factors including the service’s maturity (in number of
years), the service size (log of OOS), the percentage of OOS funded by the MBS, the percentage of
OOS that is a mental health service, lead agency type, whether the service is a centre, the service’s
state and territory, and the service’s regionality.

represents the relevant coefficient estimates; and

• represents an error component.
The coefficient estimates for this regression are summarised under Table 72. The coefficient estimates are
interpreted as the factor’s impact, holding all other factors constant, on the service’s influence on mental
health improvements.
Summary results are reported in Figure 66, Figure 67 and Figure 68 below along with a discussion
highlighting how a service’s average improvement varies depending on the service-specific and
geographic factors. The estimates presented Figure 66, Figure 67 and Figure 68 highlight the average
mental health improvement delivered by a specific service type.
Table 72: Linear regression of service-specific components on service fixed effects.

Independent variables Service fixed effects on:

K10 SOFAS MLT
(1) (2) (3)
Service maturity (year) 0.003 -0.05 0.02
(0.03) (0.07) (0.09)

Log (Service size) -0.3 1.0** -0.9

(0.19) (0.46) (0.57)

% of OOS funded by -0.01 -0.05** -0.03

MBS
(0.01) (0.01) (0.02)

% of OOS that is a mental 0.02** 0.04 0.07**

health service
(0.01) (0.03) (0.03)

Lead agency type (ref = Other)

Clinical MH service -0.4 0.3 -1.5
(0.40) (0.96) (1.20)

General NGO -0.3 -0.4 -1.1

occasions of service
(0.40) (0.95) (1.19)

Service is a centre 0.9* -0.8 3.5**

(0.48) (1.16) (1.45)

State (ref = ACT)

NSW 1.2 0.2 5.0*
(0.92) (2.22) (2.78)

NT 0.7 0.6 3.3

(1.09) (2.63) (3.29)

QLD 0.9 0.4 4.4

(0.94) (2.26) (2.82)

SA 0.4 -0.0 0.9

(0.95) (2.28) (2.85)

TA -0.1 1.2 0.4

(1.09) (2.61) (3.27)

VIC 0.4 -0.4 2.2

(0.93) (2.23) (2.79)

WA 0.7 0.7 3.7

(0.95) (2.30) (2.87)
Regionality (ref = Major cities)
Inner regional 0.3 0.1 1.4**
(0.22) (0.52) (0.65)

Outer regional 0.2 -0.0 1.2

(0.29) (0.70) (0.88)

Remote or very remote 0.2 4.6** 3.4**

(0.55) (1.33) (1.66)

Observations 117 117 117

R2 0.3 0.3 0.4
Adjusted R2 0.1 0.1 0.3
Source: KPMG analysis of the ‘K10 analysis’ dataset, ‘SOFAS analysis’ dataset and ‘MLT analysis’ dataset
Notes: See Appendix F for detailed exclusion criteria. Number of observations: 117 services. Improvement is the difference between
the last observed and the initially observed outcome measures. Standard errors in parentheses. For brevity, this report did not include
the site’s coefficient estimates. Improvements have been adjusted for RTM. *: Significant at 10 per cent; **: Significant at 5 per cent.
OOS: Occasions of service; MH: Mental health.

E.6.1 Service factors

Service maturity and size

Across all outcome measures, service maturity and size have no statistically significant impact on the
service-specific change in the outcome. The only exception is that bigger services are associated with
better improvements in the SOFAS.

Percentage of occasions of service funded by the MBS

For the SOFAS, a ten percentage point increase in the proportion of OOS funded by the MBS statistically
significantly reduced the SOFAS outcome measure by five points. The more services funded by the MBS,
the lower the SOFAS improvements.

Percentage of occasions of service that are a mental health service

With the K10 and the MLT outcome measures, a ten percentage point increase in the proportion of
occasions that are mental health services statistically significantly increased the outcome measures by two
and seven points, respectively. headspace services with a relatively heavier focus on providing mental
health services are associated with delivering better mental health outcomes.
For the SOFAS, there is no statistically significant relationship between the outcome measure and the
proportion of OOS that are mental health services.

Lead agency type and service type

Figure 66, Figure 67 and Figure 68 show that the lead agency type has no significant and statistical impact
on changes in the outcome measures.
Figure 66 and Figure 67 show that there are no statistically significantly differences in K10 and SOFAS
improvements between headspace centres and satellites. However, Figure 66 show that centres have
statistically significantly higher MLT improvements than their satellite counterparts. This report cannot
conclusively determine if centres and satellites deliver different mental health outcomes. In the sections
below, this report will later show that the type of headspace services explained a minority of the variation
in mental health outcomes.

E.6.2 Geographic factors

Service state or territory

Figure 66 and Figure 67 show that the state or territory the service was based in had no statistically
significant impact on the service-specific change in the K10 or SOFAS outcome measures. The only
exception is Figure 68 which shows that, with the MLT outcome measure, services in New South Wales
have MLT improvements that are statistically significantly higher than average, while services in Victoria
and South Australia have lower than average improvements.

Regionality
The impact of the service’s state or territory on the service’s fixed effects is dependent on the choice of the
outcome measures.
• According to Figure 66, the service’s region has no statistically significant impact on K10 improvements.
• Figure 67 shows that services located in remote areas have the highest effect on changes in the SOFAS
outcome measure.
• Figure 68 shows that MLT improvements were highest among services located in remote areas followed
services in inner regional areas.

Figure 66: Average K10 improvement by service-level factors

Source: KPMG analysis of K10 service-fixed effects derived from section E.5
Notes: See Appendix E.5 for an explanation on service fixed effects, i.e., improvements attributed to the service. Sample includes 117
services. K10 improvement is the difference between the last observed and the initially observed outcome measure. K10
improvement has been adjusted for RTM.
Figure 67: Average SOFAS improvement by service-level factors
Source: KPMG analysis of SOFAS service-fixed effects derived from section E.5
Notes: See Appendix E.5 for an explanation on service fixed effects, i.e., improvements attributed to the service. Sample includes 117
services. SOFAS improvement is the difference between the last observed and the initially observed outcome measure. SOFAS
improvement has been adjusted for RTM.
Figure 68: Average MLT improvement by service-level factors
Source: KPMG analysis of MLT service-fixed effects derived from section E.5
Notes: See Appendix E.5 for an explanation on service fixed effects, i.e., improvements attributed to the service. Sample includes 117
services. MLT improvement is the difference between the last observed and the initially observed outcome measure. MLT
improvement has been adjusted for RTM.

The report uses Shaply decomposition to assess the relative contributions of service-level, state-level, or
regionality factors on the service’s impacts on improving mental health.
Table 73 that the state or territory dummy variables are most important in explaining the variance on the
average impact the service had on the K10 and MLT outcome measures. This is followed by the service’s
characteristics explaining from 29 to 36 per cent of the variance.
However, state-level factors became less important among the SOFAS measure where it only explained
around 18 per cent of variance of the services’ improvement in the SOFAS measure. With the SOFAS,
service characteristics and service regionality explain around 45 per cent and 37 per cent of the SOFAS
variance, respectively.
Table 73: Shaply decomposition by service and regional components

Component K10 SOFAS MLT

Service – level factors 35.6% 45.9% 28.7%
Maturity 1.3% 2.5% 0.9%
Size 3.3% 8.8% 2.4%
MBS funded 6.3% 22.5% 6.2%
Main service OOS 12.2% 5.3% 5.1%
Lead agency type 3.9% 4.3% 4.2%
Service type 8.6% 2.5% 10.0%
Geographic factors 64.4% 54.1% 71.4%
State/territory factors 53.1% 18.2% 50.9%
Regionality factors 11.3% 35.9% 20.5%
Total 100% 100% 100%
Source: KPMG analysis of effectiveness analysis dataset with non-missing outcome measures – K10 analysis, SOFAS analysis and
MLT analysis
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 27,867 (K10), 30,351 (SOFAS) and 27,957 (MLT).
Improvement is the difference between the last observed measure and the initially observed measure. Outcome measurements have
been adjusted for regression to the mean effects.

E.6.3 Factors associated with above average services

From this quantitative analysis, there is little clear evidence of what factors are associated with above
average services. Service size and maturity, type of lead agency, as well as geographic factors, were tested,
however there were few consistent findings. There does appear to be relatively large variation in service
outcomes by state, but these differences were typically not statistically significant, potentially due to the
relatively small sample of services in the analysis.
There were no clear factors associated with above average services other than the proportion of OOS that
are mental health services.

E.7 To what extent are outcomes sustained over time?

To track outcomes over time, headspace sends a follow up survey three months after an episode of care
ends. Within the 2015-16 to 2019-20 dataset (see Appendix F for details), most of the surveys were
collected within 100 days after the last occasion of service as shown in Figure 69. Only the K10 measure
was available as an outcome measure in the follow up survey . 228F

Figure 69: Distribution of follow up survey completion time

Source: KPMG analysis of the follow up analysis dataset

Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 13,839

Table 74 summarises the average K10 outcome measure observed during the start of an episode, the
closure of an episode and the outcome recorded within the follow up survey. Column six of Table 74
shows the observed differences in the K10 outcome measures between the final occasion and the follow
up. Young people continued to experience improvements three months post episode closure, implying a
sustained treatment effect on young people’s mental health condition. Table 74 shows the improvement is
lowest in recent years. This may be because recently created episodes would not have been closed by the
end of the observational period.
Table 74: Average improvement in K10 outcome measures in young people completing the follow up survey

Financial n Average intake Average final Average follow Average improvement from
year (2) measure (SD) measure (SD) up measure (SD) closure – observed (SD)
(1) (3) (4) (5) (6)
2016 3,026 29.84 26.60 24.99 -1.61

(8.32) (8.9) (9.88) (0.24)

2017 3,447 29.70 26.44 25.07 -1.37

(8.29) (9.01) (9.86) (0.23)

2018 3,892 29.99 26.75 25.78 -0.97

(8.32) (8.83) (9.93) (0.21)
2019 1,845 30.38 27.51 26.36 -1.15
(8.32) (8.9) (9.81) (0.31)
2020 1,629 30.34 27.40 26.26 -1.13
(7.64) (8.64) (9.12) (0.31)
Source: KPMG analysis of the follow up analysis dataset
Notes: See Appendix F for detailed exclusion criteria. Number of episodes: 13,839. Column six represents the observed difference
between the follow up K10 and the final K10 outcome measures. A negative change indicates better mental health outcomes. SD:
Standard deviation.

A limitation of the survey is the low response rate. Table 75 shows that under five per cent of all closed
episodes, created within 2015-16 to 2019-20, have returned a follow up survey. Survey completion
depended on the young person’s characteristics such as their age, gender, education level, regionality and
the K10 outcome measure at the start and at the completion of their episode of car . Young persons with a
229F

higher intake K10 outcome measure and a lower final one were more likely to respond to and complete the
follow up survey. This suggests that the follow up response is biased towards young people who benefited
the most from their headspace episode.
Table 75: Number of follow up survey responses
1
Number of episodes Number of responses to follow up survey Response rate
302,861 13,839 4.6%
Source: KPMG analysis of the follow up analysis dataset
1
Notes: See Table 80 under Appendix F for detailed exclusion criteria. This includes all closed episodes with mental health/situational
primary issues during entry from 2015-16 to 2019-20.

E.7.1 The extent to which outcomes are sustained

The follow up survey highlights that outcomes achieved during a headspace episode are sustained over the
following 90 days. However, the follow up survey response rate is low and the sample is likely biased
towards young people that benefited the most from the headspace program. Further follow up would
improve the reliability of this finding.
The follow up survey suggests that outcomes achieved during a headspace episode are sustained over the
following 90 days.

E.8 How effective is headspace in improving mental health and

wellbeing outcomes at an area-level?
E.8.1 Overview of area-level analysis
The previous sections of the report have evaluated the impact of headspace at the young person episode
level using a pre-post quasi-experimental methodology. Limitations of that analysis include the lack of a
control group, and potential to miss some of the broader population-level benefits of headspace. In this
section, another form of a quasi-experimental methodology known as Difference-in-Differences (DID) is
applied to further evaluate the impact of headspace in the location, or area in which it operates, rather than
individual level . DID design makes use of longitudinal data to estimate the effect of headspace by
230F

comparing the changes in outcomes over time between areas. The hypotheses are that areas that have
experienced an increase in headspace services will have a reduction in the number of mental health, self-
harm and substance-abuse related hospitalisations and the number of suicides; and an increase in the
number of Medicare-subsidised mental health-specific services as increasing exposure to headspace can
de-stigmatize the need to seek mental health care, especially outside the headspace program.
A detailed explanation of the methodology and data used to perform the area-level analysis is provided
below, but key points:
• Outcome measures are the rate of mental-health related hospitalisations; intentional self-harm
hospitalisations related hospitalisations; illicit drug and alcohol-related hospitalisations; deaths from
intentional self-harm; and Medicare-subsidised mental health-specific services among 12 to 25 year
olds at the PHN area-level, obtained from the Australian Institute of Health and Welfare (AIHW) and
Services Australia (SA).
• Three headspace metrics were considered for each PHN: the number of headspace services; the number
of headspace clients per 1,000 young person, and the ratio of headspace OOS to MBS-funded mental
health services. A lagged effect of these metrics was also considered for up to three years.

E.8.2 Difference-in-Differences methodology

The DID approach studies how the variation in exposure to headspace across the PHNs influence their
outcomes. This approach can also inform this evaluation of the effect of changes in headspace exposure
(e.g., more headspace services or more headspace clients) rather than just the effect of the existence of
headspace . 231F

This report used the STATA code xtregress to estimate the average treatment effect on the treated PHNs
from observational data by difference in differences (DID) for panel data. The average treatment effect of a
continuous treatment on a continuous outcome is estimated by fitting a linear model with time and PHN
(panel) fixed effects.
These effects represent time-specific impacts on the continuous outcome measures and unobserved PHN-
level characteristics, respectively. The equation used to estimate the regression is as follows:
where:
• represents the PHN-level outcome measure of interest during financial year for PHN . These are
described further in detail in the section below;
• represents an unobserved time-invariant individual effect for PHN;
• represents the intervention variable (described further in detail below) observed during financial year for
PHN . represent the associated coefficient estimates.
• represents a vector of dummy variables indicating the financial year observed. represents the relevant
year fixed effects; and
• represents an unobserved random error component.

E.8.3 Data

PHN-level outcome measures

To examine how a variation in headspace exposure influenced area-level outcomes over time, outcome
measures, aggregated by PHNs, were obtained from the Australian Institute of Health and Welfare
(AIHW) and Services Australia (SA). These data included the population of 12 to 25 year olds from 2008-
09 to 2018-19 and the number of:
• mental-health related hospitalisations;
• intentional self-harm hospitalisations related hospitalisations;
• illicit drug and alcohol-related hospitalisations;
• deaths from intentional self-harm; and
• Medicare-subsidised mental health-specific services among 12 to 25 year olds.
The outcome variables are summarised in Table 76 below.
Table 76: Outcome measures

Outcomes Unit Periods Source

covered
Mental-health related Per 100,000 12 2008-09 to AIHW National Hospital Morbidity
hospitalisations to 25 year olds 2018-19 Database .
232F

Intentional self-harm
hospitalisations (for
Per 100,000 12 2008-09 to AIHW National Hospital Morbidity
233F

brevity, referred to as
to 25 year olds 2018-19 Database .
self-harm
234F

hospitalisations).
Illicit drug and alcohol
related hospitalisations
Per 100,000 12 2008-09 to AIHW National Hospital Morbidity
(for brevity, referred to
to 25 year olds 2018-19 Database .
as substance abuse
235F

hospitalisations)
Deaths data are from AIHW National
Mortality Database.
The Cause of Death Unit Record File data are
provided to the AIHW by the Registries of
Deaths from intentional Births, Deaths and Marriages and the National
Per 100,000 12 2008-09 to
self-harm (for brevity, Coronial Information System (managed by
to 25 year olds 2018-19
referred to as suicides) the Victorian Department of Justice) and
include cause of death coded by the
Australian Bureau of Statistics (ABS). The
data are maintained by the AIHW in the
National Mortality Database. . 236F

Mental health
Per 100,000 12 2013-14 to
emergency department AIHW.
to 25 year olds 2018-19
presentations
Medicare-subsidised
Per 100,000 12 2008-09 to
mental health specific Services Australia.
to 25 year olds 2018-19
services 237F

Source: KPMG 2022

This analysis adjusted the Medicare-subsidised mental health specific-services variable by subtracting the
number of Medicare-subsidised mental health-specific services provided by headspace. This is done to
examine the impact of headspace on mental health-specific services outside the headspace program.
However, the estimated number of headspace provided Medicare-subsidised mental healthspecific services
are only available for 2013-14 to 2018-19.

headspace services
The DID analysis requires an intervention variable to represent the causal link between the headspace
intervention and area-level outcomes. For this analysis, three options are considered:
1. Number of headspace services.
2. Number of headspace clients per 1,000 young persons.
3. Ratio of headspace services to MBS-funded mental health services.
As the benefits from headspace may not be realised immediately, a lagged impact of up to three years for
each of the intervention levers is also considered.

E.8.4 Results

Number of headspace services

Table 77 summarises the impact a new headspace service, and the lagged effect of a new headspace service
one, two and three years ago, had on the outcome measures. There is evidence of a three year lagged effect
of a new service on self-harm hospitalisations within a PHN, with each new service associated with a
lagged reduction in the number of self-harm hospitalisations by 14.1 per 100,000 young persons. Given the
average annual growth in self-harm hospitalisations over the study period was 7 per 100,000 per annum,
this is a meaningful impact. Unfortunately, as this report shows later on, this result is not consistently
derived when using alternative measures of the headspace treatment effect.
The analysis also suggests that each new headspace service is associated with an immediate and lagged
impact on the number of substance abuse hospitalisations. Each new service reduced the number of
hospitalisations by 30 per 100,000 young persons in three years, and by 36 during the current financial
year. This is a meaningful reduction when compared to an average annual growth rate of 9.1 substance
abuse hospitalisations per 100,000 young persons. However, as above, this result is not consistently
derived when using alternative measures of the headspace treatment effect.
There were no statistically significant impacts observed from a new headspace service on the PHN’s
number of mental health related hospitalisations, suicide deaths, MBS-subsidised mental health services,
and mental health emergency department presentations.

Table 77: Difference-in-Difference analysis of the impact of number of headspace services on area-level measures of
mental health

Independent Dependent variables

variables
MH related Self-harm Substance Suicide MBS MH MH ED
1 2 3
hosp. hosp. abuse hosp. deaths services presentations
(1) (2) (3) (4) (5) (6)

Summary statistics
Average in 2018-
1,618 267 503 15 55,009 1,998
19
Annual growth 46 7 9 0.4 2,688 69
No. of services
Services opened -36.2 -8.9 -36.2** 0.2 364.1 142.7
this year
(42.3) (8.3) (15.9) (0.4) (606.8) (90.0)

Services opened 1 -39.0 -10.2 -35.8** 0.6 -180.3 75.2

year ago
(39.0) (8.7) (16.0) (0.4) (566.7) (71.0)

Services opened 2 -42.2 -10.1 -30.6** 0.8 -148.7 13.7

years ago
(35.2) (8.1) (14.5) (0.5) (449.7) (66.0)

Services opened 3 -34.2 -14.1* -30.0** 0.4 33.1 -34.2

years ago
(33.8) (7.3) (14.3) (0.5) (44.2) (33.8)

No. of PHNs 31 31 31 31 31 31
Source: KPMG analysis of PHN-level hospitalisations and Medicare-subsidised MH specific services provided by the AIHW and SA.
No. of services opened estimated with opening dates provided by headspace National.
Notes: * Significant at 10 per cent. ** Signification at 5 per cent; Dependent variables are measured at per 100,000 young persons.
Fixed year effects are estimated but omitted from this table. SH: Self-harm. MH: Mental health. MBS: Medicare Benefits Schedule
(MBS).
1
Also known as ‘illicit drug and alcohol related hospitalisations per 100,000’.
2
Also known as ‘deaths from intentional self-harm per 100,000’.
3
Mental health emergency department presentations are only available from 2013-14 to 2018-19.

Number of headspace clients per 1,000 young people

Table 78 summarises the impact of the number of headspace clients per 1,000 young people in a PHN on
outcomes. There is some evidence that increasing the number of headspace clients is associated with a
reduction in the rate of mental health-related hospitalisations but the association is only significant when
lagged by a year. There is also some evidence of a positive association with the number of headspace
clients and the rate of suicides and mental-health ED presentations but this is potentially reverse causality:
high rates have led to an increase in headspace services. For the remainder of the outcome measures, there
does not seem to be any significant impact of increasing the share of headspace clients as per young person
population.

Table 78: Difference-in-Difference analysis of the number of headspace clients per 1,000 young people on area-level
measures of mental health

Independent Dependent variable:

variables
MH related Self-harm Substance Suicide MBS MH MH ED
hosp. hosp. abuse hosp.1 deaths2 services presentations3
(1) (2) (3) (4) (5) (6)
Summary statistics
Average in 2018-
1,618 267 503 15 55,009 1,998
19
Annual growth 46 7 9 0.4 2,688 69
No. of clients per 1,000 young people
No. of clients this
year -8.2 -0.2 -1.5 0.2* -89.9 25.6*

(4.9) (1.4) (2.6) (0.1) (66.7) (13.1)

No. of clients 1
year ago -12.7*** 0.4 -0.6 0.2 53.1 5.8

(4.3) (1.8) (2.6) (0.1) (50.8) (14.0)

No. of clients 2
years ago -9.1 -2.4 -2.7 0.0 106.6 -4.7

(5.5) (1.7) (2.4) (0.1) (76.2) (20.5)

No. of PHNs 31 31 31 31 31 31
Source: KPMG analysis of PHN-level hospitalisations and Medicare-subsidised MH specific services provided by the AIHW and SA.
No. of services opened estimated with opening dates provided by headspace National.
Notes: * Significant at 10 per cent. ** Signification at 5 per cent; Dependent variables are measured at per 100,000 young persons.
Fixed year effects are estimated but omitted from this table. SH: Self-harm. MH: Mental health. MBS: Medicare Benefits Schedule
(MBS).
1
Also known as ‘illicit drug and alcohol related hospitalisations per 100,000’.
2
Also known as ‘deaths from intentional self-harm per 100,000’.
3
Mental health emergency department presentations are only available from 2013-14 to 2018-19.

Ratio of headspace services to MBS mental health services

Table 79 summarises the impact of the increasing the number of headspace services as a ratio of the
number of MBS mental health services accessed within the PHN. The results show that there is an
immediate and lagged effect of a higher ratio on the rate of mental health related hospitalisations. There is
also a lagged effect on the rate of self-harm hospitalisations and substance abuse hospitalisations per
100,000 young persons by five and six, respectively. A higher ratio is also associated with an increase in
the number of MBS subsidised mental health services by 83 per 100,000 young persons suggesting a
potential de-stigmatisation effect.

Table 79: Difference-in-Difference analysis of headspace intensity on area-level measures of mental health

Independent Dependent variable:

variables
MH related Self-harm Substance Suicide MBS MH MH ED
hosp. hosp. abuse hosp.1 deaths2 services presentations3
(1) (2) (3) (4) (5) (6)

Summary statistics
Average in
1,618 267 503 15 55,009 1,998
2018-19
Annual growth 46 7 9 0.4 2,688 69

Ratio of headspace services to MBS items

Ratio
this year -7.9* 0.2 -1.0 -0.2 -111.9* 16.5

(4.4) (0.9) (1.8) (0.1) (58.0) (14.9)

Ratio
1 year ago -6.0 -3.4 -0.7 0.4 34.1 10.1

(4.9) (3.5) (2.3) (0.4) (40.4) (8.5)

Ratio
2 years ago -13.0*** -4.5*** -6.4** 0.2 83.2** 13.7

(2.9) (1.3) (2.8) (0.1) (39.1) (17.5)

No. of PHNs 31 31 31 31 31 31
Source: PHN-level hospitalisations and Medicare-subsidised MH specific services provided by the AIHW and SA. No. of services
opened estimated with opening dates provided by headspace National.
Notes: * Significant at 10 per cent. ** Signification at 5 per cent; Dependent variables are measured at per 100,000 young persons.
Fixed year effects are estimated but omitted from this table. SH: Self-harm. MH: Mental health. MBS: Medicare Benefits Schedule
(MBS).
1
Also known as ‘illicit drug and alcohol related hospitalisations per 100,000’.
2
Also known as ‘deaths from intentional self-harm per 100,000’.
3
Mental health emergency department presentations are only available from 2013-14 to 2018-19.

E.8.5 The effectiveness of headspace in improving area-level outcomes

The analyses presented above offers inconclusive evidence that headspace is effective at improving
outcomes at a PHN level. There is some evidence that headspace has an effect on some outcomes but the
results are typically lagged and inconsistent over time. In some instances, this is to be expected. For
example, suicide deaths occur in small numbers and are volatile at the PHN level, and headspace plays
only a small part in wider suicide prevention. By contrast, it is reasonable to expect that headspace
services would help to lower mental health hospitalisations. Updating this analysis over time would help to
strengthen the conclusions.
The effectiveness of headspace in improving area-level outcomes is inconclusive. There is some evidence
that headspace has an effect on some outcomes but the impacts are typically lagged, as expected, but
inconsistent over time.

E.9 How is the establishment of alternative service delivery

models assisting headspace to meet its program outcomes?
As outlined in Section 2.3.4 above, there are a range of different headspace services now present within
communities across Australia, and increasing emphasis is being placed on diversifying the headspace
model by the Commonwealth Government. Additional satellite services have been funded to support
young people in smaller communities surrounding headspace centres to offer them face-to-face services
mental health and counselling support.
As intended, the types of services delivered by these alternative models differs to those offered by
headspace centres. The predominant focus of supports is on mental health and counselling, with only two
of the three other core services required to be provided by a satellite service, either directly by staff, or
through linkages with local providers of those supports. This is further explored in Sections 2.5.2 and 2.6.2
above.
There are mixed views from across stakeholders involved in delivering, or working with headspace
services as to the impact of satellite services. As outlined in this report, there is significant positive regard
for headspace services, and communities and stakeholders view any headspace services as a positive
addition to achieving core objectives.
Deep dive stakeholders linked to satellite services either directly or as a parent centre recognised the value
of the work they were undertaking and the contribution headspace, in any form, makes to communities.
However, these stakeholders also indicate that the level of need in their local community warranted a
headspace centre, with increased funding levels and longer services hours to support young people, and
that being able to implement the full headspace model would make the most difference for young people
locally.
Of the six responses received to the headspace service and lead agency survey from satellite or outreach
service respondents, there were no discernible differences in responses received to enablers and barriers
identified, or how well these services are able to support headspace’s objectives. These respondents
indicated similar challenges recruiting appropriate staff, managing wait times for young people, and
challenges with perceived complexity of presenting need. One satellite service respondent indicated that
the small funding amount received by headspace satellites meant they were only able to employ a single
clinician, and for this service in particular, this contributed to wait times.
Similarly, PHNs as commissioners of services indicated a preference for headspace centres to better meet
the needs of local young people through the holistic headspace model.
With respect to clinical outcomes, only a small number (less than five) of alternative models were able to
be analysed in line with criteria established for this analysis. Services were typically excluded from
analysis if they had not been open long enough to move past their establishment phase. This was assumed
to be 12 months for the purposes of this evaluation. Of the services that were able to be analysed in
comparison to headspace centres, there was some indication that improvements in K10 and MyLifeTracker
outcomes were better for young people accessing headspace centres, while there was limited difference
between outcomes for young people based on the SOFAS outcome measure (see Appendix E).

E.9.1 How the establishment of other service delivery models delivers on headspace
program outcomes
The benefits of access to headspace services, in any form, are consistently recognised by all stakeholders.
However, there were mixed views as to the utility of satellites as a type of headspace service without
access to the full headspace model, including the four core service pillars. Initial observations able to be
made about the impact of service types on outcomes for young people also indicates that for two of the
three outcome measures used within this evaluation, headspace centres provided better outcomes.
However, it should be noted that the number of satellite services that were able to be analysed within the
data period for this evaluation was limited. A number of satellite services have since opened, but were
unable to be evaluated at this time, due to their short time in operation. Further evaluation of any
differences in outcomes for young people accessing headspace satellites and other models should be
undertaken once more services have reached full establishment, at least 12 months after they have
commenced operations.
The expansion of headspace services into new communities assists headspace to meet its objectives by
supporting a greater number of young people. However, there is recognition amongst stakeholders that the
full headspace model is preferred over satellite services, and this is supported by clinical outcomes for
young people based on the small number of satellite services able to be observed in this evaluation.

Appendix F :
Inclusion and
exclusion criteria
This section describes data and the inclusion and exclusion criteria for samples used in this report.

F.1 Effectiveness in improving mental health and wellbeing

outcomes
For the evaluation, this report combined OOS reported between 1 July 2019 to 30 June 2020 from the
hMDS main extract, the family and friends survey and the phone intake survey to form the KPMG master
dataset as labelled in Figure 70 below.
Ongoing episodes of care were omitted, and the master dataset filtered to only include episodes that were
created during 2019-20 and closed by December 2020. This forms the ‘Closed Episodes’ dataset labelled
in Figure 70 below. Around 85 per cent of episodes created during 2019-20 were completed by December
2020. This is slightly less than earlier financial years where around 87 per cent of new episodes created
during each financial year completed by December.
Episodes with missing main services and primary issues data are dropped. Episodes with missing main
services and primary issues data were dropped. This forms the ‘Episodes with non-missing services and
issue dataset’ in Figure 70 below.
Episodes that had an initial primary issue other than mental health or situational are dropped to form the
‘Episodes with only MH/situational primary issues’ dataset in Figure 70.
Last, episodes with only one OOS are dropped to create the ‘Effectiveness analysis’ dataset labelled below
in Figure 70 and the data used within Section 3.2.4. Last, episodes with only one OOS were dropped to
create the ‘Effectiveness analysis’ dataset labelled below in Figure 70 and the data used within Section
3.2.4. Further, additional episodes are dropped if they have missing initial and final K10, SOFAS and MLT
outcome measures for their respective analysis. These split off into ‘K10 analysis’, ‘SOFAS analysis’ and
‘MLT analysis’ datasets, respectively, in Figure 70.
Figure 70: Exclusion pathways
Source: KPMG 2022

F.1.1 To what extent are outcomes sustained over time?

headspace sends a follow up survey three months after an episode of care ended to track outcomes over
time. Appendix E.7 uses data from this survey to study the sustained impacts of headspace treatment. Due
to the low response rate, the analysis in this section uses all data from the follow up surveys during 2015-
16 to 2019-2020. Figure 71 shows the inclusion criteria for the sample in this section.
Figure 71: Exclusion criteria for sustained outcome analysis

Source: KPMG 2022

Table 80 shows the response rate of follow up survey from 2015-16 to 2019-20.
Table 80: Follow up survey responses from 2015-16 to 2019-20.
Financial Number of follow up response Number of closed episodes Response rate
year
2015-16 3,026 55,277 5.5%
2016-17 3,447 59,981 5.7%
2017-18 3,892 65,568 5.9%
2018-19 1,845 72,110 2.6%
2019-20 1,629 49,925 3.3%
Total 13,839 302,861 4.6%
Source: KPMG analysis of the follow up analysis dataset and KPMG master dataset

F.1.2 Cost-effectiveness analysis

Figure 72 shows additional exclusion criteria for Section 4. In this section, the analysis includes 112sites
that have been opened on or before the 30 June 2019. During 2019-20, there were 401,325 OOS delivered
in 112 headspace services. The total cost was $123.3 million, so the average cost per OOS was $307.
238F

The QALY gain is calculated for closed episodes created in 2019-20 with non-missing main provider and
with an initial mental health or situational primary issue due to the availability of outcome measures. This
is the ‘Cost utility’ dataset labelled below in Figure 72. There were 39,634 closed episodes with 181,269
OOS, accounting for 45 per cent of the total OOS delivered in 201920. Therefore, the total cost included in
the cost-effectiveness analysis is $41.8 million . 239F

Figure 72: Additional exclusion criteria for cost-effectiveness analysis

Source: KPMG 2022

Appendix G :
Definitions of K10
distress levels
The following thresholds are used to define the four levels of distress by outcome score. For the K10’s, the
ABS K10 outcome groupings and categorisation are used.
Table 81: Definitions of K10 distress level

Initial level of distress K10

Low 10 to 15
Moderate 16 to 21
High 22 to 29
Very high 30 to 50
Source: ABS (2012) 240F

Appendix H : Factors
affecting the
likelihood of
completing the follow
up survey
A multivariate logistic regression is estimated to analyse the probability of a young person completing the
follow up survey. This analysis focuses on 302,861 closed episodes, created from 2015-16 to 2019-20,
from the ‘Episodes with only MH/situational primary issues’ dataset as illustrated in Table 80 in Appendix
F. Due to missing data on young person’s characteristics, the sample size is 243,224 episodes. This is done
by estimating the following logistic regression:

where:
• represents an indicator variable if young person i completed the follow up survey
• represents a numerical constant;
• represents a vector of the young person’s demographic characteristics including age, gender, Aboriginal
and Torres Strait Islander status, culturally and linguistically diverse status, education level, rurality
and main services the young person received at headspace.

represents the relevant coefficient estimates;

• represents a vector of the intake and the final K10 measures.

represents the relevant coefficient estimates;

• represents an error component.
The results show that the completion of the follow up survey was not random, but dependent on young
person’s characteristics, the main service provided and their intake and final K10 outcome measure. The
coefficients reported in Table 82 are odd ratios. Each coefficient implies the odds of completing the follow
up survey compared to the reference group. A coefficient with the magnitude greater than one implies that
the analysed group is more likely to complete the survey than the reference group. For example, the odds
of completing the follow up survey for young persons aged 15 to 19 years old were 1.337 times those of
young person aged under 15 years old. Regarding the impacts of the K10 outcome measure, the odds ratio
for the intake K10 was larger than 1 while the odds ratio for the final K10 and change in K10 (Final K10–
Intake K10) was smaller than 1. Young persons with higher intake measure and lower final outcome or
more decrease in K10 measure were more likely to complete the survey.
Table 82: Logit regression of completing the follow up survey

Probability of completing the follow up survey

Age categories (ref = younger than 15 years)

15 to 19 years old 1.3** 1.3976**
(0.04) (0.0414)
20 to 24 years old 1.0 1.0497
(0.04) (0.0387)
Older than 24 years 0.8** 0.8969
(0.07) (0.0777)
Gender (ref = Male)
Female 1.8** 1.8250**
(0.04) (0.0394)
Non - Binary 2.2** 2.2468**
(0.14) (0.1457)
Culturally and linguistically diverse status (ref = young people who are not culturally and
linguistically diverse)
Culturally and linguistically 1.0
diverse 1.0093
(0.03) (0.0315)
Aboriginal and Torres Strait Islander status (ref = young people who are not Aboriginal and Torres
Strait Islander)
Aboriginal and/or Torres Strait 0.7** 0.7265**
Islander (0.06) (0.0595)
Education level (ref = None)
Year 10 or below 1.4** 1.4364**
(0.20) (0.2087)
Year 11 1.8** 1.8813**
(0.27) (0.2777)
Year 12 2.2** 2.2949**
(0.33) (0.3368)
Certificate 2.1** 2.1900**
(0.32) (0.3270)
Diploma or advanced diploma 2.4** 2.4386**
(0.37) (0.3738)
Bachelor’s degree 2.9** 2.9370**
(0.445) (0.4465)
Postgraduate degree 2.403** 2.3981**
(0.522) (0.5262)
Main services provided (ref = Intake/assessment)
Non-MH services 3.730** 0.9542
(0.283) (0.0732)
MH and non-MH services 5.186** 1.1360**
(0.225) (0.0503)

MH only services 5.646** 1.3505**

(0.201) (0.0496)

Rurality (ref = Major cities)

Inner regional Australia 0.761** 0.7771**
(0.017) (0.0178)

Outer regional Australia 0.822** 0.8277**

(0.027) (0.0270)

Remote Australia 0.477** 0.4973**

 (0.061) (0.0641)

Very remote Australia 1.113 1.1281

(0.413) (0.4211)

K10 outcome measures

Intake K10 measure 1.019**

(0.002)

Final K10 measure 0.981** 0.9981*

(0.001) (0.001)

Change in K10 measure (Final

K10 – Intake K10) 0.9893**
(0.001)
Observations 243,224
Source: KPMG analysis of the follow up analysis dataset
Notes: See Table 80 under Appendix F for details. Number of episodes: 243,224, including all closed episodes with MH/Situational
primary issues during entry from 2015-16 to 2019-20. 59,637 episodes were excluded due to missing young persons’ characteristics.
Coefficients reported are odds ratios. Standard errors in parentheses. *: Significant at 10 per cent; **: Significant at 5 per cent. MH:
Mental health.

Appendix I :
Extrapolation of the
follow up K10
outcome measure
As discussed in appendix E.7, findings on sustained outcomes may be biased due to the low completion
rate of the follow up survey. To account for the missingness and bias stemming from the low response rate
of the follow up survey, the follow up K10 outcome measures for closed episodes are estimated based on
the intake and the final K10 outcomes and young person’s characteristics using the following regression:
where:
• represents the K10 measure at the follow up time for episode i
• represents a numerical constant;
• represents a vector of the young person’s demographic characteristics including age, gender, Aboriginal
and Torres Strait Islander status, culturally and linguistically diverse status, education level, rurality
and main services the young person received at headspace.

represents the relevant coefficient estimates;

• represents a vector of intake and the final K10 measures.

represents the relevant coefficient estimates;

• represents an error component.
The coefficients of the extrapolation model are presented in the below table. The estimation shows that
gender, culturally and linguistically diverse status, Aboriginal and Torres Strait Islander status, the intake
and final K10 measures significantly determine the K10 measure post treatment. In contrast, education
level and rurality do not affect the K10 measure at the 90-day follow up.
Table 83: Extrapolation of the K10 score at the follow up

K10 measure at follow up

Age categories (ref = younger than 15 years)

15 to 19 years old -0.044
(0.21)

20 to 24 years old -0.311

(0.26)

Older than 24 years -0.485

 (0.63)

Gender (ref = Male)

Female 1.113***
(0.157)
Non-binary 2.605***
(0.459)
Culturally and linguistically diverse status (ref = non - culturally and linguistically diverse)
Culturally and linguistically diverse -0.787***
(0.223)
Aboriginal and Torres Strait Islander status (ref = non - Aboriginal and Torres Strait Islander)
**
Aboriginal 0.699
(0.316)
Torres Strait Islander 0.493
(1.317)
Aboriginal and Torres Strait Islander 3.772**
(1.782)
Education level (ref = None)
Year 10 or below 1.167
(1.074)
Year 11 0.552
(1.083)
Year 12 -0.226
(1.075)
Certificate 0.197
(1.091)
Diploma or advanced diploma -0.485
(1.117)
Bachelor’s degree -0.898
(1.104)
Postgraduate degree -1.690
(1.571)
Main services provided (ref = Intake/assessment)
Non-MH services -0.623
(0.559)

MH and non-MH services -0.178

(0.320)
***
MH only services -0.899
(0.266)

Rurality (ref = Major cities)

Inner regional Australia -0.049
(0.165)
Outer regional Australia -0.370
(0.236)
Remote Australia -0.047
(0.948)
Very remote Australia 4.022
(2.670)
K10 outcome measures
Intake K10 measure 0.173***
(0.011)
Final K10 measure 0.570***

(0.010)

Constant 3.883***
(1.096)
Observations 12,962
Source: KPMG analysis of the follow up analysis dataset
Notes: See Figure 71 under Appendix F for details. Number of episodes: 12,962, where 877 episodes were excluded due to missing
young persons’ characteristics. Standard errors in parentheses. *: Significant at 10 per cent; **: Significant at 5 per cent. MH: Mental
health.

Appendix J : Costing
assumptions
The hMDS captures the quantity of OOS that are funded by sources other than the national headspace
grant. The value of these contributions was estimated as a volume-weighted average of the equivalent
average MBS benefit fees, as shown in the table below. The average MBS benefit fees were calculated
from the AIHW ‘Medicare-subsidised GP, allied health and specialist health care across local areas: 2013-
14 to 2018-19’ data based on 2018-19 rates for 15 to 24 yearolds . 241F

Table 84: Volume-weight average of equivalent MBS benefit fees

GP Psychiatrist Clinical Other Allied Other Weighted

psych. psych. health average
1
MBS benefit fees $83 $178 $129 $89 $104 $79

OOS funding source

MBS 14% 1% 29% 38% 4% 14% $100
In-kind 1% 1% 3% 28% 6% 63% $85
contribution
Private payment 1% 0% 33% 43% 2% 21% $100

PHN funding 0% 1% 16% 25% 11% 47% $93

Other federal 0% 2% 3% 10% 5% 80% $84
Other state / 0% 1% 20% 37% 7% 35% $95
EMHSS
Other 1% 3% 5% 34% 1% 55% $89
Missing 17% 2% 15% 28% 2% 36% $92
Source: KPMG analysis of the hMDS dataset and Medicare Benefits Schedule & AIHW Medicare-subsidised GP, allied health and
specialist health care across local areas: 2013–14 to 2018–19 . Psych: Psychologist.
242F

1: Estimated from the average fees reported in the MBS schedule.

Appendix K :
Economic evaluation
parameters and inputs
This section provides further details and assumptions for the parameters generated for the economic
evaluation of headspace. It also includes a summary of the sensitivity analysis scenarios and the proposed
variation to examine how the results change in response.
K.1 Costs
There are no detailed accounts that provides specific and detailed estimates of the cost of directly
providing mental health services by headspace. Instead, this evaluation assumes that, in the base case, 75
per cent of the headspace budget is dedicated to directly delivering mental health services. This
assumption was informed by a deep dive study of six headspace services and considers that the remaining
25 per cent of resources are used for activities that generate benefits not captured in this evaluation. This is
a major source of uncertainty. To compensate, a plausible range of values were defined to explore how the
main results changed over this defined range.
In the base case, it is assumed that 75 per cent of the headspace budget is dedicated to delivering mental
health services. The plausible range was defined as 75 per cent ±15 percentage points.
To reflect the importance of headspace providing services free at the point of delivery to the young person
in need, the base case analysis includes the costs that fall on the users of services. This is referred to in
Table 18 in Section 4.2 as the extended payer perspective accounts for the costs of the sponsors of care
(e.g., government, donors) as well as the direct costs of care incurred by the young person.
The evaluation examines the costs from EOC completed in 2019-20 to align costs with the availability of
outcome data upon treatment completion or episode closure. Table 85 presents the cost calculations for the
cost-effectiveness analysis. During 2019-20, there were 401,325 OOS delivered in 112 headspace services.
The average cost of delivering an OOS was determined as $230 under the assumption that the direct and
indirect costs of providing treatment services account for 75 per cent of the total cost.
Table 85: headspace OOS cost determination

Variable Value
Total cost $123,304,645
Cost attributed to delivering treatment (75%) $92,478,484
Number of OOS 401,325
Costs of delivering an OOS (a) $230
Number of OOS in closed episodes (b) 181,269
Included cost (a)x(b) $41,770,341
Number of closed episodes 49,634
Average cost per closed episodes $842
Source: KPMG analysis of the cost-effectiveness dataset.
Note: OOS occasion of service.

K.2 Consequences
The consequences of not accessing care or accessing receiving a non-MAT EOC are based on a weighted
average of mental health and substance abuse hospitalisation costs. The weights are based on the number
of mental health related hospital separations as recorded by the AIHW (2021) . The costs per separated
243F

were informed with data provided by the Independent Hospital Pricing Authority (2019) . The calculation
244F

of this weighted average hospitalisation cost is presented in Table 86.

Table 86: Determination of mental health and substance abuse hospitalisation costs

DRG groups No. of separations % Cost per separation

Mental health treatment 21,091 22% $1,291
Psychotic disorders 4,570 5% $8,096
Affective and somatoform disorders 7,551 8% $7,460
Anxiety disorders 7,191 8% $6,249
Behavioural disorders 10,214 11% $12,677
Alcohol use disorders 30,429 32% $5,060
Drug use disorders 14,219 15% $6,919
Weighted average per separation $5,745
Source: IHPA (2019); AIHW (2021).
Note: DRG Diagnosis related groups.

The incremental probability of mental health and substance abuse hospitalisation was determined using the
AIHW area-level data described in Appendix E.8:
• There is an estimated reduction of 81 hospitalisations per 100,000 12 to 25 year olds (see Table 77 in
Appendix E.8) relative to the observed incidence rate of 2,279. This suggests the probability of
hospitalisation is 3.6 per cent lower among headspace clients compared to the rest of the target
population.
• The estimated probability of 12 to 25 year olds needing hospitalisation is 2.3 per cent.
• The ratio of headspace clients to the population of 12 to 25 year olds is 2.2 per cent.
• The probability of hospitalisation for young persons not accessing headspace calculated as .
• The probability of hospitalisation for young persons accessing headspace calculated is .
• Thus, the incremental risk of hospitalisation is 2.28% - 2.20% = 0.08 per cent.
Young persons that accessed treatment from headspace in the world with headspace, but do not in the
world without headspace have a 0.08 per cent higher risk of hospitalisation with the expected cost increase
of 0.08% * $5,745 = $4.52 per person not accessing treatment.
K.3 Outcomes
Aligning with the aim of producing results that are most readily suitable to support decision making, the
evaluation captures costs of headspace service provision, and converts mental health outcomes (K10) to
QALYs for the calculation of an ICER which is the standard outcome for expressing value for money of
health policies and interventions. The process of estimating the QALY gain from the treatment at
headspace is outlined below.
9. Calculate the average K10 scores at the start, completion and follow up of an episode.
10. Convert the average K10 scores into utility score AQoL-8D.
11. Calculate QALY gain per episode by linearly extrapolating between points.

K.3.1 Change in K10 measure

Improvements in health outcomes over time are observed for young people receiving treatment and also
for young people not receiving treatment. This is accounted for by estimating the RTM effect as discussed
in Appendix E.2. The gains from receiving a treatment are then considered to be health outcomes
exceeding the outcomes predicted by the RTM effect as illustrated in Figure 73. Further details on this are
available in Appendix E.7.
Figure 73: The average K10 outcome measure for episodes with at least three OOS

Source: KPMG analysis of the cost-effectiveness dataset.

Notes: RTM Regression to the mean.

K10 conversion
The analysis used the algorithms developed by Mihalopoulos et al. (2014) to convert the K10 outcome
measures into Assessment of Quality of Life – Eight Dimension Scale (AQoL-8D), a multi-attribute utility
instrument (MAUI) representing the level of utility at that point in time . The AQoL-8D ranges between
245F

zero and one, where one represents perfect health and zero represents death.
The AQoL-8D was constructed from people with moderate to severe mental health problems, aiming to
achieve sensitivity to the dimensions that are important to people with mental health problems . Given the
246F

nature of the service provided by headspace as well as headspace clients, AQoL – 8D is a more suitable
instrument for the QALY calculation than other more commonly used MAUIs such as EQ-5D and SF-6D . 247F

It is acknowledged that the conversion of K10 score into MAUI may be subject to the sensitivity of the
algorithm, especially when the sample in this study is not a sample of young people.
Hamilton et al (2021, Preprint) is developing Transfer To Utility (TTU) algorithms using a sample of
young people attending Australian primary mental health service . However, the study does not use K10
248F

but converts the K6 measure and the SOFAS measure into AQoL-6D, which are either not fully aligned
with outcome measures used in this report (K6) or not collected in the follow up survey (SOFAS).

QALY gain calculation

QALY gains were calculated as the area beneath the AQOL curve and a line at the ‘pre’ AQOL score at
intake. Both the raw observed QALY and the RTM-adjusted QALY gains were examined. The latter
measure was preferred in the base case due to this approach being conceptually more appropriate.
The QALY gains accounts the treatment outcome observed at the closure of an EOC, mental health
improvements indicated by responses from the follow up survey and benefits that were extrapolated
beyond the time periods observed within the available data.

Extrapolation
The base case reflects mental health benefits of treatment extrapolated over 12 months after the last
observed health outcome data point. This 12-months duration is an assumption based on a literature review
and meta-analysis where a majority of the reviewed studies relied on 12-month follow up data to capture
treatment benefits . The extrapolation assumes a linear decline of the RTM-adjusted benefit from its last
249F

observed value to zero at 12 months.

Figure 74 illustrates how participation in headspace treatment affects the quality of life for young people
attending three or more OOS. The outcome components include:
12. QALY gain up to the completion of an episode (green).
13. QALY gain three months post treatment (blue).
14. Extrapolated QALY gain 12 months from the follow up (yellow).
Figure 74: Mean QALY change for episodes with at least three OOS

Source: KPMG analysis of the cost-effectiveness dataset.

Note: QALY quality-adjusted life year; OOS occasion of service.

Treatment effects as a function of the number of OOS are presented in Section 4.2. The methods used to
estimate the change in the K10 outcomes and the associated results are presented in Appendix E.5. The
analysis assumes patients receiving no treatment received zero gains in mental health outcomes after
adjusting for RTM.
Table 87: Average QALY gain for closed episodes in 2019-20

Number of closed Average QALY gained per episode with benefits up

Episode
episodes to 12 months from the follow up
0.039
3+ OOS (48%) 23,817
(0.0027)
2 OOS (19%) 9,348 0
1 OOS (33%) 16,469 0
Weighted average for base 0.019
49,634
case (0.0008)
Source: KPMG analysis of the cost utility dataset as described in 0.
Notes: Standard errors in parentheses

K.3.2 Comparator
The comparator is broadly defined as the state of the world in which headspace is absent as discussed in
Section 4.2.1. In the base case, it is assumed that 10% of headspace’s closed episodes would seek
alternative treatments in the world without headspace. These episodes are assumed to receive similar
treatment effects as the effects at headspace and to incur treatment service costs. For 90% of headspace’s
closed episodes not receiving treatment in the world without headspace, it is assumed that they would not
receive any treatment effect and have higher probability of hospitalisation, hence, incur the cost of
consequences as discussed above.

Table 88: Average costs and QALYs gained per episode in the world without headspace

Episode Number of episodes Cost per episode QALY gained per episode
No treatment (90%) 44,671 $4.5 0
Treatment (10%)

3+ OOS (5%) 2,381 $260 0.039

2 OOS (2%) 935 $116 0
1 OOS (3%) 1,647 $116 0
Weighted average for base case 49,634 $87 0.002
Source: KPMG analysis of the cost utility dataset as described in 0.

Two elements are used to define the comparator costs: 1) the scheduled fees for the observed mix of initial
appointments; and 2) the Australian Psychological Society national schedule of recommended fees and
item numbers for psychological services for the treatment. It is assumed that for the first and the second
OOS, young people in the world without headspace would seek low cost treatments, which are either bulk-
billed or with low out-of-pocket cost. The cost of these sessions is estimated to include weighted average
MBS schedule fees discussed in Appendix I and 16% out-of-pocket costs to young people . For the third
250F

and following sessions, the cost is assumed to be $260, which is the recommended fee for a 46 to 60
minute consultation . 251F

Table 88 presents the weighted average cost and QALY gained per episode for the comparator in the world
without headspace.

K.3.3 Other parameters of the evaluation

Time horizon
The time horizon for the evaluation is 18 months. This includes the average treatment duration of three
months, the three months follow up data that capture the last measured outcome, and a 12month
extrapolation of gains in mental health outcomes.
The costs captured over this horizon include treatment costs for a variable number of OOS and cost-
consequences in the form of hospital admissions due to mental health needs not being addressed.

Discounting
In the base case analysis, outcomes are not discounted. This is because the 18-month time horizon does not
include substantial long-term costs and effects.
However, when extrapolating and examining benefits beyond the 18-month time horizon in the sensitivity
analyses, a 5 per cent discount rate is applied to benefits accrued.

K.3.4 Summary of evaluation inputs

Table 89 presents a summary of input values used in the economic evaluation. For each parameter, the
point estimate is presented under the ‘Values’ column. The ‘Sensitivity test’ column summarises the range
of values used for a sensitivity analysis which was presented in Section 4.2.
Table 89: Input values used in the economic evaluation

Parameter Values Sources/assumptions Sensitivity testing

Proportions continuing 1 OOS: 33% hMDSC NA
treatment 2 OOS: 19%
3+ OOS: 48%
Average numbers of OOS 3+ OOS: 6.13 hMDS NA
4+ OOS: 7.13
Proportions receiving 10% headspace MBS claims 0% to 20%
treatment in the ‘world data;
without headspace’ headspace Evaluation
scenario Reference Group (ERG)
consensus
Health gains from 1 OOS: nil Analysis of hMDS 2 OOS gives partial
treatment 2 OOS: nil benefit (0.03 QALYs);
3+ OOS: 0.04 QALYs MAT from 4+ OOS (0.04
(MAT) QALYs);
Treatment effect ±20% in
the ‘no headspace’
scenario;
Gains from MAT
diminish over 5 years
(0.09 QALYs)
Cost per OOS headspace: $230 Assumed proportion of headspace: $184, $276
No headspace: $116 headspace budget
(screening) Analysis of visit types No headspace: $198,
No headspace: $260 Australian Psychological $320
(treatment) Society 2020 National
Schedule of
Recommended Fees and
item numbers for
psychological services . 252F

Increase in probability of 0.08% for those not Analysis of headspace Probability also applied
hospital admission in receiving any treatment data to those accessing 1 OOS
case of no MAT or 2 OOS (not MAT)
Cost of hospitalisation $5,745 Weighted average of NA
due to mental health admissions data 253F

Note: OOS; MBS Medicare Benefits Schedule; QALY quality-adjusted life years; MAT minimum adequate treatment; NA not
available

K.3.5 Sensitivity analyses

The economic evaluation included several sensitivity analyses to explore how changes in the value of the
key parameters and inputs impact the main results of the cost-effectiveness analysis. This is done in
response to the uncertainties related to the imperfect data made available to this evaluation.
The following model approaches and assumptions are varied in sensitivity analyses to explore their impact
on the results of the economic evaluation:
• The proportion of current headspace patients that would be receiving treatment in the ‘no headspace
 scenario between zero and 20 per cent. This value was 10 per cent in the base case analysis.
• The proportion of headspace budgets attributable to treatment provision between 60 and 90 per cent. This
value was 75 per cent in the base case analysis.
• Extrapolation of treatment benefits over five years assuming 50 per cent annual benefit decay rate and
applying a discount factor . This value was 12 months decreasing linearly in the base case analysis.
254F

• MAT achieved after four or more OOS . The base case analysis assumed MAT was achieved after three
255F

or more OOS.
• Improvements to mental health outcomes are not adjusted for RTM. In the base case, RTM was
accounted for.
• The effectiveness of services provided outside headspace are assumed to be either 20 per cent more
effective or less effective than equivalent headspace services. The base case assumes equal
effectiveness.
• An EOC with only two OOS gives a partial improvement to mental health outcomes. The base case
assumed there were no benefit.
• Fees for each OOS delivered outside of headspace were set higher (at $320 per OOS) or lower ($198 per
OOS). This was based on the 2020 Australian Psychological Society national schedule of
recommended fees and item numbers for psychological services . The base case analysis assumed
256F

$260 per OOS.

• Evaluation from the payer perspective only, excluding patient out-of-pocket costs. In the base case, out-
of-pocket costs were included.
• Any EOC that did not deliver MAT results in an increased risk of hospital admission. In the base case,
only receiving no treatment at all increased risk of hospital admission.

K.3.6 Limitations
This section identifies the key limitations of the economic evaluation. These limitations centre around the
scope of the economic evaluation and the approaches to cost and outcome estimation. The limitations in
evaluation design and methods stem from the imperfect data available to evaluators. Where possible,
sensitivity analyses were performed to reduce uncertainties related to this.

Understanding headspace funding and costs

The hMDS dataset collates information about funding sources such as the headspace grant, PHN funding
agreements, in-kind contributions, and the MBS. However, indirect costs not funded from the national
headspace grant remain undetermined due to the low response rate to cost specific questions of the service
survey and the large variation in amounts reported from the responses. Given that, the full cost of
delivering headspace activity may be underestimated.
There is no definitive source to determine the amount of funding allocated to specific types of activities
and there are inconsistencies in cost definitions across services. The department records on the national
headspace grant do not separate service provision and indirect costs. In order to determine the costs of
delivering treatment, the evaluation relied on an assumed proportion of the headspace budgets being
allocated to delivering OOS. The assumption was informed by a deep dive analysis of selected headspace
services.

Estimation of incremental benefits

QALY was used as the measure of health outcomes in the cost-effectiveness analysis. The main challenges
of this approach related the lack of a head-to-head control group, the methods for conversion of K10
scores to QALYs not validated in the youth population (noting that the instrument and method are
validated in the adult population) and repeated measures not following up for the full duration of benefit.
The evaluation addressed those by adjusting for regression to the mean and extrapolating benefits beyond
the observed data based on assumptions of benefit duration. Robust analytical methods were applied to
enable this, and sensitivity analyses performed to explore any remaining uncertainty.

Wider societal benefits

The evaluation focuses on those activities and outcomes of headspace activity that were possible to be
quantified and modelled. It is acknowledged that headspace, as an early intervention targeting mental
illness, may lead to additional cost-savings to the healthcare system and benefits to the wider society by
providing services other than treatment and promoting mental health wellbeing. The evaluation addressed
this by providing a clear definition of benefits in scope for the economic evaluation.

Consequences other than hospitalisation

The evaluation equates the implications of not receiving treatment to the expected cost of consequent
hospital admissions. The actual consequences may be broader and include impacts on health and
productivity that were not possible to be modelled.

Appendix L :
Reference List
L.1 Article / report / paper
Australian Government (2020), Closing the Gap National Agreement Preamble, viewed 4 July 2022,
<https://www.closingthegap.gov.au/national-agreement/national-agreement-closing-the-gap/1-preamble>.
Australian Government (2020), Closing the Gap Report, viewed 4 July 2022
<https://ctgreport.niaa.gov.au/sites/default/files/pdf/closing-the-gap-report-2020.pdf>.
Australian Institute of Health and Welfare (2020). People with disability in Australia, viewed 4 July 2022,
<https://www.aihw.gov.au/reports/disability/people-with-disability-in-australia/contents/health/access-to-
health-services>.
Barnett, AG, van der Pols, JC & Dobson, AJ 2004, ‘Regression to the mean: what it is and how to deal
with it’, International Journal of Epidemiology, vol. 34, no. 1, pp. 215–220.
Brinckley, MM., Calabria, B., Walker, J. et al. (2021), Reliability, validity, and clinical utility of a
culturally modified Kessler scale (MK-K5) in the Aboriginal and Torres Strait Islander population. BMC
Public Health 21, 1111.
Commonwealth of Australia, as represented by the Department of Health 2021, Portfolio Budget
Statements 2021–22 Budget Related Paper No. 1.7, Viewed 6 August 2021,<budget-2021-22-portfolio-
budget-statements-budget-2021-22-health-portfolio-budget-statements.pdf>.
Commonwealth of Australia 2017, The Fifth National Mental Health and Suicide Prevention Plan, Viewed
6 August 202, <https://www.mentalhealthcommission.gov.au/getmedia/0209d27b-1873-4245-b6e5-
49e770084b81/Fifth-National-Mental-Health-and-Suicide-Prevention-Plan.pdf>.
Department of Health, 2016, Guidelines for preparing a submission to the Pharmaceutical Benefits
Advisory Committee, Version 5, accessed at https://pbac.pbs.gov.au/content/information/files/pbac-
guidelines-version-5.pdf
Department of Health, 2021, PHN Initial Assessment and Referral (IAR) for Mental Healthcare, accessed
at National IAR State of Play Report 2020 Final.pdf (wqphn.com.au)
Goldman, HH, Skodol, AE & Lave, TR 1992, ‘Revising axis V for DSM-IV: a review of measures of
social functioning’, American Journal of Psychiatry, vol. 149, no. 9, pp. 1148–1156.
headspace National 2020, young people’s experience of telehealth during COVID-19, Viewed 9 August
2021, <https://headspace.org.au/assets/Uploads/Telehealth-Client-Experience-FINAL-8-10-20.pdf,>.
headspace National 2020, headspace staff experience of telehealth during COVID-19, Viewed 9 August
2021, <https://headspace.org.au/assets/Uploads/Telehealth-Staff-Experience-FINAL-8-10-20.pdf>.
headspace National 2020, Coping with COVID: the mental health impact on young people accessing
headspace services, Viewed 11 August 2021, <https://headspace.org.au/assets/Uploads/COVID-Client-
Impact-Report-FINAL-11-8-20.pdf>.
headspace National 2020, Annual Report 2019-20, Viewed 14 August 2021,
<https://headspace.org.au/assets/HSP10755_Annual-Report-2020_FA02_DIGI.pdf>.
headspace National Youth Mental Health Foundation 2020, Response to the Productivity Commission’s
Inquiry into the Mental health Draft Report, Viewed 11 August 2021
https://www.pc.gov.au/__data/assets/pdf_file/0017/251360/sub947-mental-health.pdf.
headspace National 2019, headspace centre young person follow up study, Viewed 11 August 2021,
<https://headspace.org.au/assets/headspace-centre-young-person-follow-up-study-Sept-2019.PDF>.
headspace National 2019, Increasing demand in youth mental health – A rising tide of need, Viewed 23
August 2021, https://headspace.org.au/assets/Uploads/Increasing-demand-in-youth-mentalh-a-rising-tide-
of-need.pdf.
headspace National 2017, headspace Model Integrity Framework – August 2017, Viewed 11 August 2021,
<https://bsphn.org.au/wp-content/uploads/2018/11/headspace-Model-Integrity-Framework-hMIF.pdf>.
headspace National 2015, Service Innovation Project Component 2: Social Inclusion Model Development
Study, Viewed 14 August 2021, <https://headspace.org.au/assets/Uploads/Corporate/Publications-and-
research/HSP201-Service-Innovation-Part-2-FA-LR.pdf>.
headspace National 2013, headspace Psychosocial Assessment for Young People, Viewed 18 August 2021,
<https://headspace.org.au/assets/Uploads/headspace-psycho-social-assessment.pdf>.
Jacobson, NS & Truax, P 1992, ‘Clinical significance: A statistical approach to defining meaningful
change in psychotherapy research.’ Methodological issues & strategies in clinical research. pp. 631648.
Kessler, R.C. , Barker, P.R. , Colpe, L.J. , Epstein, J.F. , Gfroerer, J.C. , Hiripi, E. , Howes, M.J. , Normand,
S-L.T. , Manderscheid, R.W. , Walters, E.E. , Zaslavsky, A.M. (2007). The Kessler psychological distress
scale (K10, K6, K5). Melbourne: Australian Centre for Posttraumatic Mental Health.
Kessler, R.C., Andrews, G., Colpe, et al 2002, ‘Short screening scales to monitor population prevalences
and trends in non-specific psychological distress’, Psychological Medicine, 32, 959-956.
Kwan, B & Rickwood, DJ 2020, ‘A routine outcome measure for youth mental health: Clinically
interpreting MyLifeTracker’, Early Intervention in Psychiatry, vol. 15, no. 4, pp. 807–817.
Kwan B, Rickwood D J and Telford, NR 2018, ‘Development and validation of MyLifeTracker: a routine
outcome measure for youth mental health’, Psychology Research and Behavior Management 2018:11 67–
77.
Linden, A 2013, ‘Assessing regression to the mean effects in health care initiatives’, BMC Medical
Research Methodology, vol. 13, no. 1.
McIntyre, P, on behalf of the World Health Organization 2002, Adolescent Friendly Health Services -An
Agenda for Change, Viewed 6 August 2021,
<https://apps.who.int/iris/bitstream/handle/10665/67923/WHO_FCH_CAH_02.14.pdf?
sequence=1&isAllowed=y>.
Medical Research Council 2019, Developing and evaluating complex interventions, Viewed at <
https://mrc.ukri.org/documents/pdf/complex-interventions-guidance/>.
Mihalopoulos C, Chen G, Iezzi A, Khan MA, Richardson J. Assessing outcomes for cost-utility analysis in
depression: comparison of five multi-attribute utility instruments with two depression-specific outcome
measures. Br J Psychiatry. 2014 Nov;205(5):390-7. doi: 10.1192/bjp.bp.113.136036. Epub 2014 Sep 25.
PMID: 25257063, accessed at https://pubmed.ncbi.nlm.nih.gov/25257063/
Orygen and headspace 2019, Submission to the Productivity Commission’s Inquiry into Mental Health,
Viewed 6 August 2021, <https://www.pc.gov.au/inquiries/completed/mental-health/submissions>.
Rickwood, D., Paraskakis, M., Quin, D., Hobbs, N., Ryall, V., Trethowan, J., McGorry, P. 2018,
‘Australia’s innovation in youth mental health care: The headspace centre model’, Early Intervention in
Psychiatry, Volume 13, Issue 1, p. 159-166.
Rickwood, DJ, Mazzer, KR, Telford, NR, Parker, AG, Tanti, CJ & McGorry, PD 2015, ‘Changes in
psychological distress and psychosocial functioning in young people visiting headspace centres for mental
health problems’, The Medical Journal of Australia, vol. 202, no. 10, pp. 537–542.
Rickwood D., Telford, N., Parker, A.,Tanti, C., and McGorry, P 2014, ‘headspace — Australia's innovation
in youth mental health: who are the clients and why are they presenting?’ The Medical Journal of
Australia, Volume 200, Issue 2, pp. 108-111.
Royal Australian College of General Practitioners (2018), Racism in the healthcare system, Position
statement, viewed 4 July 2022,
<https://www.racgp.org.au/FSDEDEV/media/documents/RACGP/Position%20statements/Racism-in-the-
healthcare-sector.pdf>.
World Health Organization 2012, Making Health services adolescent friendly: Developing national quality
standards for adolescent friendly health services (www.who.int), Viewed 9 August 2021, <
http://apps.who.int/iris/bitstream/handle/10665/75217/9789241503594_eng.pdf;sequence=1>.
The Royal Australian & New Zealand College of Psychiatrists (2021), Position Statement 83, Recognising
and addressing the mental health needs of the LGBTIQ+ population. Viewed 4 July 2022,
<https://www.ranzcp.org/news-policy/policy-and-advocacy/position-statements/mental-health-needs-
lgbtiq>.

L.2 Website reference

Australian Bureau of Statistics 2020, National, state and territory population, March 2020, Population –
Australia (www.abs.gov.au), Viewed 27 August 2021,
<https://www.abs.gov.au/statistics/people/population/national-state-and-territory-population/latest-
release#data-download>.
Australian Bureau of Statistics 2019, Estimates and Projections, Aboriginal and Torres Strait Islander
Australians, 2006 - 2031, Table 5.9 (www.abs.gov.au), Viewed 27 August 2021,
<https://www.abs.gov.au/statistics/people/aboriginal-and-torres-strait-islander-peoples/estimates-and-
projections-aboriginal-and-torres-strait-islander-australians/2006-2031#data-download>.
Australian Bureau of Statistics 2016, The Australian Statistical Geography Standard (ASGS) Remoteness
Structure (Abs.gov.au), Viewed 25 August 2021,
<https://www.abs.gov.au/websitedbs/D3310114.nsf/home/remoteness+structure>.
Australian Bureau of Statistics 2012, 4817.0.55.001 - Information Paper: Use of the Kessler Psychological
Distress Scale in ABS Health Surveys, Australia, 2007-08, www.abs.gov.au, Viewed 30 August 2021,
<https://www.abs.gov.au/ausstats/[email protected]/lookup/4817.0.55.001chapter92007-08>.
Australian Institute of Health and Welfare 2020, Medicare-subsidised GP, allied health and specialist
health care across local areas: 2013–14 to 2018–19, Technical information (https://www.aihw.gov.au/),
Viewed 31 August 2021, <https://www.aihw.gov.au/reports/primary-health-care/medicare-subsidised-
health-local-areas-2019/contents/technical-information>.
Department of Health [2019], The Australian health system (www.health.gov.au), Viewed 6 August 2021,
https://www.health.gov.au/about-us/the-australian-health-system#primary-health-networks
Department of Health [n.d.], Welcome to the PMHC-MDS - Primary Mental Health Care Minimum Data
Set (pmhc-mds.com), Viewed 9 Aug 2021, <https://pmhc-mds.com/>.
headspace National [2021] headspace Family and Friends Satisfaction Survey (headspace.org.au), Viewed
11 August 2021, < https://headspace.org.au/friends-and-family/mental-health/>.
headspace National [2021] Evaluation, Research and Annual Reports (headspace.org.au), Viewed
9 August 2021, <https://headspace.org.au/about-us/evaluation-research-reports/>.
National Mental Health and Suicide Prevention Agreement, 2022, accessed at
https://federalfinancialrelations.gov.au/sites/federalfinancialrelations.gov.au/files/2022-
05/nmh_suicide_prevention_agreement.pdf
Orygen [n.d]. History - Orygen, Revolution in Mind (www.orygen.org.au), Viewed 6 August 2021,
<https://www.orygen.org.au/About/History>.
Public Health Information Development Unit 2021, Social Health Atlas of Australia
(phidu.torrens.edu.au), Viewed 19 August 2021, <https://phidu.torrens.edu.au/social-health-atlases>.
The Centre of Best Practice in Aboriginal & Torres Strait Islander Suicide Prevention, Mental Health
Assessment - CBPATSISP, Viewed 21 June 2021 <https://cbpatsisp.com.au/clearing-house/best-practice-
screening-assessment/mental-health-assessment/>
Who are we, headspace. Viewed 5 June 2022 <https://headspace.org.au/our-organisation/who-we-are/>
World Health Organization 2019, Mental Disorders (www.who.int), Viewed 13 August 2021,
<https://www.who.int/news-room/fact-sheets/detail/mental-disorders>.
World Health Organization Europe [2021] Child and adolescent health - 5S Approach
(www.euro.who.int), Viewed 23 August 2021 < https://www.euro.who.int/en/health-topics/Life-
stages/child-and-adolescent-health/about-child-and-adolescent-health/adolescent-health/5s-approach>.

L.3 Information provided by the department/headspace

headspace Bega 2021, Centre Activity Overview Report Q3 Financial Year 2020-21. Provided to KPMG
by headspace Bega.
headspace Joondalup 2021, Centre Activity Overview Report Q4 Financial Year 2020-21. Provided to
KPMG by headspace Joondalup.
headspace Katherine 2021, Centre Activity Overview Report Q3 Financial Year 2020-21. Provided to
KPMG by headspace Katherine.
headspace National 2021, Centre Snapshot – Your centre at a glance (FY2021) – Mount Isa. Provided to
KPMG by headspace Mount Isa.
headspace National 2021, Tableau Dashboards, Provided to KPMG by headspace National.
headspace National 2020, headspace Model Integrity Framework (hMIF V2), Provided to KPMG by
headspace National.
headspace National 2020, Table of Definitions; headspace Services (DoH Funded) Final. Provided to
KPMG by headspace National.
headspace National 2020, headspace centre services program logic model – July 2020. Provided to KPMG
by headspace National.
headspace National 2019, headspace Primary Program Minimum Data Set_Data Dictionary_V 3.1_July
2019. Provided to KPMG by headspace National.
Youturn 2021, Gympie headspace Statistics July 2021. Provided to KPMG by Youturn.

L.4 Data source provided by department/headspace

headspace Primary Program Minimum Data Set data. Provided to KPMG by headspace National. 2021.
headspace funding data. Provided to KPMG by Department of Health. 2021.

KPMG | 15

©2022 KPMG, an Australian partnership and a member firm of the KPMG global organisation of independent member firms affiliated with KPMG International Limited, a
private English company limited by guarantee. All rights reserved. The KPMG name and logo are trademarks used under license by the independent member firms of the KPMG
global organisation. Liability limited by a scheme approved under Professional Standards Legislation.
Evaluation of the National headspace Program – Final Report
June 2022

Contact us

Trixie Makay
Partner, Health, Ageing and Human Services
+ 61 411 262 346
[email protected]

Andrew Dempster
Principal Director, Health, Ageing and Human Services
+ 61 412 883 667
[email protected]

kpmg.com.au