Mindfulness

https://doi.org/10.1007/s12671-018-1087-9

ORIGINAL PAPER

The Effectiveness of Mindfulness-Based Cognitive Therapy (MBCT)

in Real-World Healthcare Services
Alice Tickell 1 & Susan Ball 2 & Paul Bernard 3 & Willem Kuyken 1 & Robert Marx 4 & Stuart Pack 5 & Clara Strauss 6 &
Tim Sweeney 7 & Catherine Crane 1

# The Author(s) 2019

Abstract
Depression is common with a high risk of relapse/recurrence. There is evidence from multiple randomised controlled trials
(RCTs) demonstrating the efficacy of mindfulness-based cognitive therapy (MBCT) for the prevention of depressive relapse/
recurrence, and it is included in several national clinical guidelines for this purpose. However, little is known about whether
MBCT is being delivered safely and effectively in real-world healthcare settings. In the present study, five mental health services
from a range of regions in the UK contributed data (n = 1554) to examine the impact of MBCT on depression outcomes. Less than
half the sample (n = 726, 47%) entered with Patient Health Questionnaire (PHQ-9) scores in the non-depressed range, the group
for whom MBCT was originally intended. Of this group, 96% sustained their recovery (remained in the non-depressed range)
across the treatment period. There was also a significant reduction in residual symptoms, consistent with a reduced risk of
depressive relapse. The rest of the sample (n = 828, 53%) entered treatment with PHQ-9 scores in the depressed range. For this
group, 45% recovered (PHQ-9 score entered the non-depressed range), and overall, there was a significant reduction in depres-
sion severity from pre-treatment to post-treatment. For both subgroups, the rate of reliable deterioration (3%) was comparable to
other psychotherapeutic interventions delivered in similar settings. We conclude that MBCT is being delivered effectively and
safely in routine clinical settings, although its use has broadened from its original target population to include people experiencing
current depression. Implications for implementation are discussed.

Keywords Mindfulness-based cognitive therapy . Depression . Implementation . Service delivery . Effectiveness

Electronic supplementary material The online version of this article Depression is a common and debilitating mental health condition.
(https://doi.org/10.1007/s12671-018-1087-9) contains supplementary Globally, it is thought to affect more than 300 million people and is
material, which is available to authorized users. a leading cause of disability-adjusted life years (WHO 2008). Of
those people who access treatment, pharmacological and psycho-
* Catherine Crane logical approaches are most commonly used and are relatively
[email protected]
effective in supporting people to remission (Hollon 2016;
1
Department of Psychiatry, University of Oxford, Oxford, UK
McManus et al. 2016). However, even when acute treatment is
2
successful, people with a history of depression have a high risk of
NIHR CLAHRC South West Peninsula (PenCLAHRC), University
of Exeter Medical School, Exeter, UK
relapse/recurrence that increases with each successive episode: the
3
likelihood is at least 40% after a first episode, 60% after a second,
Tees Esk and Wear Valleys NHS Foundation Trust, Durham, UK
and as high as 90% after a third (Eaton et al. 2008; Moffitt et al.
4
Sussex Partnership NHS Foundation Trust, Sussex, UK 2010; Solomon et al. 2000). Therefore, the clinical management of
5
Oxleas NHS Foundation Trust, Dartford, UK depression encompasses both acute and maintenance treatments.
6
Sussex Partnership NHS Foundation Trust & School of Psychology, Mindfulness-based cognitive therapy (MBCT) was developed
University of Sussex, Sussex, UK as a relapse prevention programme, to help people who are at
7
Nottinghamshire Healthcare NHS Foundation Trust, high risk of depressive relapse/recurrence to learn the skills to
Nottingham, UK stay well in the long term (Segal et al. 2002). It is a psychosocial
 Mindfulness

group-based intervention that comprises training in mindfulness translational stages: basic science, intervention development
meditation and elements of cognitive-behavioural therapy and pilot testing, efficacy trials (in research settings), effec-
(CBT). There is evidence from at least nine clinical trials (n = tiveness trials (in community settings), and implementation.
1258) that MBCT reduces the risk of relapse to depression when They highlighted that little research is being conducted in the
added to usual care, and demonstrates comparable efficacy to later stages of the translational journey, namely effectiveness
maintenance antidepressant medication (Kuyken et al. 2016). in real-world healthcare settings. In a research setting, MBCT
Studies comparing MBCT to closely matched psychological is typically delivered to relatively homogeneous patient
treatments have suggested comparable but not superior efficacy groups in accordance with strict research protocols and with
for relapse prevention (Farb et al. 2018; Manicavasgar et al. a high degree of fidelity to manualised procedures.
2011; Meadows et al. 2014; Shallcross et al. 2015), over a period Conversely, community healthcare providers face real-world
of up to 26-month follow-up (Shallcross et al. 2018). As such, practical constraints and limited resources, which means they
MBCT is included in the clinical guidelines as a recommended adapt treatment manuals to their service needs or broaden the
option for relapse prevention in a number of different countries, population to which the intervention is applied (Onken et al.
including the UK, Netherlands, Canada, Australia and New 2014). Such challenges can reduce a treatment’s effectiveness:
Zealand (Malhi et al. 2015; NICE 2009; Parikh et al. 2016), for example, because key elements are unwittingly removed,
and has been endorsed by the American Psychiatric it is delivered by therapists with less rigorous training, or
Association (Lu 2015). administered to individuals with clinical presentations or
In line with these recommendations, many healthcare services sociodemographic characteristics that differ substantially from
are exploring how to include MBCT in community and public those for which the intervention was developed or in which its
healthcare contexts, as part of the care pathway for people with efficacy was tested (Henggeler 2011; Perepletchikova et al.
recurrent depression. However, real-world mental health services 2007). In line with this, effectiveness studies of psychological
are normally commissioned to address the needs of patients interventions do typically show some decreases in interven-
experiencing acute difficulties, whereas MBCT was developed tion potency when a treatment is translated from research set-
for those who have remitted but are at risk of depressive relapse/ tings into the community (Curtis et al. 2004; Henggeler 2004;
recurrence. This has been a barrier to the implementation of Miller 2005). However, it seems that diminishing effect sizes
MBCT in its original form, as people in remission are less likely may not be inevitable. Some adaptations may be the result of
to access services than those who are experiencing current prob- ‘positive infidelity’: the introduction of well-informed modi-
lems. Some services have responded by adapting MBCT to fit fications to an intervention’s delivery to ensure that it best
their service delivery models (Crane and Kuyken 2013; Rycroft- meets the needs of the population to which it is being applied.
Malone et al. 2017). For instance, there has been a move to widen Therefore, effectiveness research has important implications,
the reach of MBCT to include people experiencing current de- highlighting whether more work is needed to adapt an inter-
pression (Strauss et al. 2014). There were initial concerns that vention to suit the needs of real-world healthcare services, or
MBCT may not be appropriate for this group, because practising whether it can be implemented in its existing format (Demarzo
mindfulness involves processes that could be difficult for those et al. 2015; Dimidjian and Segal 2015). This is critical to fulfil
experiencing an acute depressive episode (e.g. sustained the public health goal of producing treatments that are both
attention, bringing awareness to unpleasant feelings; Strauss effective and implementable (Onken et al. 2014).
et al. 2014). However, meta-analyses have demonstrated the ef- The primary aim of this study was to examine the effec-
ficacy of MBCT for reducing depression symptoms in patients tiveness of MBCT for depression when offered in real-world
with current depression (Hofmann et al. 2010). MBCT has been healthcare settings. We used England as an exemplar, as it is
shown to perform as well as other comparable evidence-based currently one of the countries which has progressed furthest in
treatments such as group CBT (Goldberg et al. 2018; Strauss terms of the formal implementation of MBCT in an integrated
et al. 2014). These effects appear to be maintained at follow- National Health Service (NHS), and providers routinely mon-
up, and robust when accounting for publication bias, study qual- itor the clinical outcomes of patients before and after treatment
ity features and sensitivity analysis (Goldberg et al. 2018). (Clark 2018). Existing clinical data were obtained from five
Therefore, MBCT shows promise as an alternative psychological healthcare services. These services offered MBCT to mixed
treatment for acute depression. groups of people with a history of depression: those in remis-
MBCT is thus increasingly considered suitable for patients sion but at risk of depressive relapse/recurrence and those
with recurrent depression irrespective of their illness stage: in currently experiencing depression. It can be reasonably as-
episode, in partial remission, or in full remission but vulnera- sumed that the target of treatment was different for each of
ble to relapse/recurrence. However, this evidence base reflects these groups, and therefore, we examined separate outcomes
a relatively small number of studies, the majority of which are for each. For service users entering treatment with depression
clinical trials in research settings. A recent review (Dimidjian symptoms in the non-depressed range, our question was
and Segal 2015) mapped existing mindfulness research into its whether MBCT sustained recovery and reduced risk of
Mindfulness

relapse. Service users were not followed up beyond the end of for people with common mental health problems, known as
treatment, so in this subgroup, we conceptualised residual the Improving Access to Psychological Therapies (IAPT)
depression symptoms as a marker for risk of depressive re- Programme (Clark et al. 2009; Clark 2018); Woodpecker
lapse, based on previous studies which show that residual was a secondary care service, and Blackbird was a mixed
symptoms in the non-depressed range are a strong predictor primary and secondary care service. A full description of each
of time to relapse/recurrence, even over long-term follow-ups service can be found in the Supplementary Materials, includ-
(Ali et al. 2017; Judd et al. 1997; Pintor et al. 2004). For ing the nature of MBCT provision and the service’s specific
service users entering treatment with depression symptoms participant inclusion and exclusion characteristics. Table 2
in the clinical range, our question was whether MBCT re- provides a summary of the regional characteristics of each
duced the severity of depression symptoms and led to recov- service, with information relating to the quality of NHS ser-
ery. We also examined depression outcomes as a function of vices, ethnicity, deprivation and prevalence of depression in
demographic characteristics, and in line with Kuyken et al. the adult population.
(2016), we predicted that outcomes would be similar for ser-
vice users irrespective of their age and gender. Finally, be- Design The study used existing clinical data from NHS ser-
cause MBCT provision differed between the services on a vices. Services collected outcome measures as a part of rou-
number of dimensions, including the participant inclusion/ tine clinical practice before and after treatment, although the
exclusion criteria, staff training resources and requirements, timings differed according to the service: Swallow and
and the provision of a full day of mindfulness practice, we Robin collected pre-treatment PHQ-9 scores from the total
conducted exploratory analyses to compare depression out- sample before the course started or in the first session. PHQ-
comes at each service. 9 scores were then collected again in each of the eight
MBCT sessions. For the purposes of the present study, base-
line and post-treatment data were provided, where post-
Method treatment refers to the participant’s last measurement, not
necessarily taken in the final session. As such, the time
Participants interval between baseline and post-treatment varied for each
service user depending on their pattern of attendance at treat-
Five NHS services contributed data from 1554 service users ment. There was no information on pattern of session atten-
who had each taken part in a group-based, face-to-face MBCT dance available, so it was not possible to calculate the length
programme for adults (18+ years). The sample had a mean age of the interval between first and last data collection point.
of 49.37 years (SD = 12.74). Seventy-one percent of the sam- Jackdaw, Woodpecker and Blackbird collected pre-treatment
ple were female, 89% were White British and 59% were PHQ-9 scores either before the course started or in the first
employed. According to scores on the Patient Health session. PHQ-9 scores were collected again in the final ses-
Questionnaire (PHQ-9), before treatment, 53% of service sion offered, at approximately an 8-week interval from the
users were currently depressed and 47% were not depressed. pre-treatment measurement. Service representatives accessed
Table 1 shows the participant characteristics of the overall and anonymised the data to send to the investigators for
sample, each service, and subdivided into groups of services analysis.
users who were currently depressed and non-depressed at en-
try to treatment. Measures

Procedure Sociodemographic Information Sociodemographic measures

were used to determine the baseline characteristics of the sam-
Recruitment We recruited services that offered MBCT in line ple. Each service provided data on the age and gender of
with the manualised programme in content and length (Segal service users. Some services had collected additional informa-
et al. 2002) and routinely collected the PHQ-9 from service tion on ethnicity and employment status, which was reported
users at pre- and post-treatment. Of the MBCT services where available.
approached, five were eligible to participate, which covered
four geographical regions in England and included a range of Session Attendance All but one of the services (Woodpecker)
different types of service. MBCT services took part on the provided information on the total number of MBCT sessions
basis that they would be anonymised, so were given the fol- attended by service users. All services offered the core eight
lowing pseudonyms: Swallow, Robin, Jackdaw, Woodpecker sessions (Segal et al. 2002). Some services also offered a pre-
and Blackbird. Swallow, Robin and Jackdaw were primary treatment orientation session or full day of mindfulness prac-
care services, part of a large-scale programme to make tice. Further information can be found in the Case
evidence-based psychological treatments widely available Descriptions of each service (see Supplementary Materials).
 Mindfulness

Table 1 Baseline characteristics and attendance information for the pooled sample and each service, subdivided into non-depressed (‘No dep’) and
depressed (‘Current dep’) at entry to treatment

Sample n Female % White British % Employed % Age Attendance Dropout %b

M SD Ma SD

Pooled 1554 70.80c 88.85c 59.00c 49.37c 12.74c 6.37 2.39 16.93c
c c c c c
No dep 726 72.81 90.23 64.02 49.68 13.02 6.42 2.35 16.26c
Current dep 828 69.10c 87.36c 52.27c 49.10c 12.50c 6.33 2.42 17.55c
Swallow 150 72.00 76.00 53.02c 44.92 12.09 5.25/8 2.06 20.00
No dep 78 71.79 83.33 51.95c 43.73 11.77 5.06/8 2.20 24.36
Current dep 72 72.22 68.01 54.17 46.21 12.39 5.44/8 1.90 15.27
Robin 508 70.47c 89.78c – 49.01c 11.97c 5.77/8 2.32 20.28c
No dep 245 71.55c 89.06c – 49.65c 12.36c 5.91/8 2.20 17.83c
Current dep 263 69.50c 90.43c – 48.44c 11.60c 5.65/8 2.43 22.48c
Jackdaw 475 70.99c 93.24c 60.90c 50.92 12.95 6.92/9 2.31 14.11
No dep 280 74.05c 92.21c 67.39c 51.23 13.38 6.96/9 2.30 13.93
Current dep 195 66.84c 96.10c 51.56c 50.49 12.32 6.86/9 2.33 14.36
Woodpecker 181 67.40 – – 47.45c 13.58c – – –
No dep 59 74.57 – – 48.56 13.57 – – –
Current dep 122 63.93 – – 46.91c 13.61c – – –
Blackbird 240 72.92 – – 51.25 12.78 7.20/9 2.27 13.75
No dep 64 71.88 – – 51.28 12.89 7.48/9 2.10 10.94
Current dep 176 73.30 – – 51.24 12.77 7.10/9 2.32 14.77

Symbol ‘–’ denotes that no data was obtained

a
Each service offered a different number of MBCT sessions, so at each service the mean number of sessions attended is presented against the total
number offered
b
Dropout refers to the percentage of participants who attended fewer than four MBCT sessions
c
Due to missing data, the sample used to calculate the value in this cell differs from n. Refer to Supplementary Table S1 for the specific sample size

Patient Health Questionnaire for Depression The PHQ-9 was Data Analyses
the primary outcome measure. It is a screening tool designed
to establish a diagnosis of major depression and grade symp- Data Cleaning Before analysis, the data were cleaned. First,
tom severity (Kroenke et al. 2001). The nine-item measure missing data were managed. The reasons for missing data
corresponds to the nine symptoms of depression identified in were administrative oversight or service users missing the
the Diagnostic and Statistical Manual of Mental Disorders, 4th relevant sessions. There were variable amounts of missing
edition (American Psychiatric Association 2000). It is self- sociodemographic data across services and variables. In these
administered by the patient, and each item is rated on a scale cases, summary statistics were calculated using the data avail-
of 0–3, yielding a score of depression severity between 0 and able. Details of the relevant sample sizes are provided in
27. A cutoff score of 10 or above indicates clinically signifi- Table S1 in the Supplementary Materials. With respect to
cant depression symptoms, or ‘caseness’ (Kroenke et al. PHQ-9 scores, those with missing pre-treatment data were
2001). The PHQ-9 has demonstrated high internal reliability excluded from the analysis, as this information was crucial
with Cronbach’s α ranging from 0.86 to 0.89 and high test– for subdividing the sample into those entering treatment above
retest reliability (0.84; Kroenke et al. 2001). It has been vali- and below clinical cutoffs. Missing post-treatment data were
dated in different patient groups and the general population handled using the last observation carried forward (LOCF)
(Kroenke et al. 2001; Martin et al. 2006; Spitzer et al. 1999; method. This provides the most conservative way of dealing
Spitzer et al. 2000). The factor structure of the PHQ-9 has with missing data in relation to the question of symptom re-
been investigated in a number of different populations: in duction, the primary focus of this study, as it assumes no
severe depression, it had two factors (‘affective’, e.g. de- change over time. However, it should be noted that it would
pressed mood, feelings of worthlessness; ‘somatic’, e.g. sleep over-inflate estimates of sustained recovery, particularly if da-
difficulties, appetite changes), which were stable over time up ta were not missing at random but rather patterned by initial
to 12 months (Guo et al. 2017). response to MBCT. The overall percentage of missing post-
Mindfulness

Table 2 Summary of participating MBCT services with their regional characteristics

Pseudonym Region in England Service type Quality ratinga White British %b Deprivation indexc Depression
prevalence %d

Swallow London IAPT Good 60.3 129 7.2

Robin East Midlands IAPT Good 85.0 115 8.9
Jackdaw South East IAPT Good 88.4 137 10.3
Woodpecker South East Secondary care Good 88.4 137 10.3
Blackbird North East Primary and secondary care (mixed) Good 94.1 102 9.8

Each service belonged to an NHS foundation trust: an organisational unit within NHS England providing healthcare services to a particular geographical
area. IAPT = Improving Access to Psychological Therapies programme (primary care)
a
Quality rating refers to the rating given to the NHS foundation trust by the Care Quality Commission
b
White British % was based on the local population obtained from Census data in 2011, averaged across the local authorities covered by the NHS
foundation trust
c
Deprivation index ranges from 1 to 209, where 1 = most deprived. This refers to latest indices of multiple deprivation (IMD) figures from 2015. The
score was averaged across Clinical Commissioning Groups (CCGs) covered by the NHS foundation trust
d
Depression prevalence % was calculated using the practice register aged 18+ in the 2016/17 Community Mental Health profiles, based on the Quality
and Outcomes Framework, NHS Health and Social Care Information Centre (HSCIC). The score was averaged across CCGs covered by the NHS
foundation trust

treatment data was 17.37%. Across the different services, the subgroups of patients above and below the cutoff for current
percentage of missing post-treatment data was as follows: depression at entry to treatment. These subgroups were calcu-
Swallow (0%), Robin (0.79%), Jackdaw (26.95%), lated based on the available clinical information: those scoring
Woodpecker (40.33%) and Blackbird (27.08%). The impact above the clinical cutoff (PHQ-9 ≥ 10) were classed as cur-
of carrying forward the pre-treatment data therefore varied rently depressed, and those below (PHQ-9 < 10) were classed
across the services depending on the amount of missing as without current depression (National Collaborating Centre
post-treatment data and the timing of assessments, and esti- for Mental Health 2018).
mates of sustained recovery may most usefully be interpreted
from Swallow and Robin. Effectiveness: Reliable Change To benchmark the clinical rel-
The data were also checked for accuracy: out-of-range evance of the outcomes on the PHQ-9, the proportion of pa-
values were excluded as they resulted from errors in data tients reporting a reliable change (either an improvement or a
entry. Data were checked for outliers, by inspection of the deterioration) at each service, and for each subgroup was cal-
standardised z-scores for the difference between baseline culated. A change between the first and last measurements is
PHQ-9 and post-treatment PHQ-9 values. Cases with z- considered to be reliable if it exceeds the measurement error of
scores in excess of 3.29 were treated as outliers. The analysis the questionnaire. According to the guidelines laid down by
was run with and without these cases to examine their influ- the IAPT programme, which forms part of the UK NHS, the
ence. Only six cases, distributed across four services, were PHQ-9 has a reliable change index of ≥ 6 (National
identified as outliers. Omission of these values did not alter Collaborating Centre for Mental Health, 2018). Therefore,
the pattern of results in any way. Findings for the full datasets reliable improvement was said to have occurred if an individ-
are reported. ual had decreased in the PHQ-9 between pre- and post-
treatment by 6 points or more, and reliable deterioration was
Effectiveness: Symptom Change Change in the PHQ-9 from said to have occurred if an individual had increased in the
before-to-after MBCT was examined using paired t tests, with PHQ-9 between pre- and post-treatment by 6 points or more.
effect sizes quantified using Cohen’s d. When computing ef-
fect sizes, we did not correct for the correlation between the Effectiveness: Sustained Recovery and Recovery For the sub-
pre- and post-treatment means despite the within-subjects de- group who entered treatment with depression symptoms be-
sign (Morris and DeShon 2002); given that we used the LOCF low the clinical cut-off (in recovery), we reported the number
method to deal with missing data, this would have artificially of patients who had sustained recovery following treatment.
inflated the pre- to post-treatment mean correlations leading to This index quantified the number of people who were below
less conservative estimates of effect size. Effect sizes were the clinical cutoff on the PHQ-9 before treatment and
interpreted in line with Cohen’s (1988) criteria for effect sizes remained below the cutoff following treatment.
(small ≥ 0.20, medium ≥ 0.50 and large ≥ 0.80). This analysis For the subgroup who entered treatment with clinical levels
was conducted for the pooled sample, each service, and for the of depression, we also reported the number of patients who
 Mindfulness

recovered following treatment. This index quantified the num- individuals showing further reliable improvement and 4.13%
ber of people that were above the clinical cutoff on the PHQ-9 showing reliable deterioration in symptoms.
before treatment but were below the cutoff following treat- Of those individuals entering treatment with depression
ment. Finally, we also reported the rate of reliable recovery, scores above the clinical cutoff, n = 828, 40.58% showed a
which is said to have occurred if a patient above the threshold reliable improvement in depression symptoms over the treat-
for depression at entry to treatment, moves below the thresh- ment period and 2.42% showed a reliable deterioration in
old after treatment and also experiences a reduction in PHQ-9 symptoms. In this group, 44.81% were recovered, and
score of 6 points or more. 34.42% were reliably recovered post-treatment and the pre-
to post-treatment effect size for depression symptoms in this
Effects of Age, Gender and Service on Symptom Change To group was statistically significant and large, t(827) = 22.78,
examine whether the change in symptoms from pre- to post- p < 0.001, d = 0.86.
treatment differed with participant age, linear regression anal- Mean pre-treatment and post-treatment depression symp-
yses were conducted of the PHQ-9 change scores (outcome) toms on the PHQ-9 are shown in Table 3, for the overall
on age (treated as a continuous variable). A t test was used to sample and subdivided according to level of depression symp-
compare PHQ-9 change scores between male and female ser- toms at entry. Data on clinical indicators of change are shown
vice users. There were insufficient data to analyse the sample in Table 4, for the overall sample and according to patients’
based on the characteristics of employment status or ethnicity. level of depressive symptoms at entry.
A between-subjects ANOVA was conducted of the PHQ-9
change scores to test whether the change in PHQ-9 from pre Effects of Age, Gender and Service on Symptom
to post treatment differed between services. Change

Linear regression analysis showed no statistically significant

effect of age (p = 0.09) on change in PHQ-9 score, and a t test
Results
showed no evidence of a difference in PHQ-9 change scores
between genders (p = 0.7). A between-subjects ANOVA re-
Session Attendance
vealed a significant effect of service on change in PHQ-9
scores, F(4,1553) = 13.80, p < 0.001. Mean pre-treatment
On average, the sample attended 6.37 sessions of MBCT
and post-treatment depression symptoms on the PHQ-9 are
(SD = 2.39) and 16.93% dropped out of treatment (attended
presented separately for each service in Table 3, and within
less than 4 sessions). Table 1 shows the session attendance of
each service according to patients’ level of depressive symp-
the overall sample, each service, and subdivided into groups
toms at entry. Across the individual services, those service
of services users who were currently depressed and non-
users who were above the clinical cutoff for depression at
depressed at entry to treatment.
entry showed greater pre- to post-treatment reductions in de-
pression scores than those below the clinical cutoff, reflecting
Effectiveness: Symptom Change, Reliable Change, the same pattern of results as the pooled sample, and the
Sustained Recovery and Recovery smaller potential for further positive change in service users
who were already below threshold on the PHQ-9 at treatment
In the pooled sample, comprising all service users irrespective entry.
of whether they were above or below the clinical cutoff on the
PHQ-9 at entry to treatment, mean PHQ-9 score was M =
10.75 (SD = 6.18) at pre-treatment, and reduced to M = 7.81 Discussion
(SD = 6.08) at post-treatment, resulting in a statistically signif-
icant change with a small to medium effect size, t(1553) = Although there are an increasing number of clinical trials ex-
22.46, p < 0.001, d = 0.48. In this pooled sample, 25.16% of amining the efficacy of MBCT for relapse prevention in re-
service users experienced a reliable improvement in depres- current depression and emerging evidence for its efficacy in
sion symptoms from pre- to post-treatment and 3.22% expe- treating acute depressive symptoms, little is known about the
rienced a reliable deterioration. outcomes of MBCT when delivered in real-world healthcare
Of the individuals who entered treatment below the clinical settings, rather than research clinics (Dimidjian & Segal.,
cutoff for depression, n = 726, 95.73% remained in the non- 2015). In the present study, data were pooled from five clinical
clinical range (showed sustained recovery) across the treat- services within the UK NHS to examine patient outcomes.
ment period. There was a small further improvement in resid- The services delivering MBCT were situated in different geo-
ual depression symptoms in this group over the treatment graphical locations, had different models of service delivery,
period, t(725) = 8.21, p < 0.001, d = 0.33, with 7.58% of and reached a relatively mixed sample of patients with respect
Mindfulness

Table 3 PHQ-9 scores at pre- and

post-treatment for the pooled Sample Pre-treatment Post-treatment Mean difference 95% CI Cohen’s d
sample and each service,
subdivided into non-depressed M SD M SD
(‘No dep’) and depressed
(‘Current dep’) at entry to Pooled 10.75 6.18 7.81 6.08 2.94*** [2.69, 3.20] 0.48
treatment No Dep 5.36 2.54 4.38 3.32 0.98*** [0.75, 1.22] 0.33
Current Dep 15.47 4.29 10.81 6.36 4.66*** [4.26, 5.06] 0.86
Swallowa 9.87 6.10 8.37 6.19 1.5** [0.63, 2.37] 0.24
No Dep 4.92 2.81 5.19 3.75 − 0.27 [− 1.09, 0.56] − 0.08
Current Dep 15.24 3.65 11.82 6.49 3.42*** [1.95, 4.89] 0.65
Robina 10.56 6.23 6.83 5.79 3.73*** [3.21, 4.25] 0.62
No Dep 5.30 2.42 4.33 3.80 0.97*** [0.48, 1.45] 0.31
Current Dep 15.47 4.44 9.16 6.34 6.30*** [5.53, 7.08] 1.16
Jackdawa 9.15 5.52 6.88 5.51 2.27*** [1.93, 2.61] 0.41
No Dep 5.37 2.58 3.95 2.74 1.43*** [1.15, 1.70] 0.53
Current Dep 14.56 3.86 11.08 5.76 3.48*** [2.79, 4.17] 0.71
Woodpeckera 12.84 6.49 10.87 6.64 1.96*** [1.34, 2.59] 0.30
No Dep 5.36 2.54 5.07 3.30 0.29 [− 0.51, 1.08] 0.10
Current Dep 16.45 4.36 13.68 6.00 2.78*** [1.96, 3.59] 0.53
Blackbirda 13.28 5.95 9.03 6.26 4.25*** [3.52, 4.98] 0.70
No Dep 6.09 2.38 4.83 2.87 1.27** [0.47, 2.06] 0.48
Current Dep 15.89 4.53 10.56 6.46 5.34*** [4.43, 6.24] 0.96

Where post-treatment PHQ-9 data were missing, the pre-treatment PHQ-9 value was carried forward
a
Pairwise comparisons were used to compare the pre- to post-treatment change in PHQ-9 scores between services
(using Tukey HSD adjustment). There were differences between Robin and Swallow, Jackdaw, and Woodpecker,
as well as between Blackbird and Swallow, Jackdaw and Woodpecker. In summary, Robin and Blackbird did not
differ from one another in PHQ-9 reduction across their whole samples, and in both cases showed significantly
larger pre-post treatment change than Swallow, Jackdaw and Woodpecker
**p < .01; ***p < .001

to presenting problems (acute or recurrent depression or anx- entry to treatment suggests that in routine clinical prac-
iety disorders), and sociodemographic characteristics, tice the use of MBCT has broadened from its original
reflecting the range of service contexts in which MBCT is intention as a relapse prevention intervention, to be used
delivered ‘on the ground’. Thus, this study enabled us to ex- with patients who are symptomatic (see also Rycroft-
amine the real-world outcomes of MBCT on depressive out- Malone et al. 2017, which provides convergent evidence
comes, for those in remission and for those experiencing cur- that services are adapting MBCT). We examined the out-
rent depression, and to compare outcomes as a function of comes of these two groups separately, on the premise
gender and age. In the following sections we discuss the over- that their treatment targets would be different. In the
all findings, their implications for ongoing provision of group who was below the clinical cutoff on the PHQ-9
MBCT within healthcare settings and future research at entry to treatment, the treatment target would be
directions. sustained recovery; 96% sustained recovery across the
Overall, examination of the pooled data provided en- treatment period. Although the LOCF method of dealing
couraging evidence of the acceptability of MBCT when with missing data may have inflated this proportion, the
delivered in the real world. Rates of session attendance rate of sustained recovery at Robin, where there was no
were generally high and rates of drop out from treat- missing data, was 94%, suggesting that any such effect is
ment were relatively low and very similar to those ob- unlikely to be marked. Furthermore, there was a signifi-
served in RCTs of MBCT for relapse prevention in re- cant reduction in residual depression symptoms, d = 0.33,
current depression (e.g. Kuyken et al. 2015; Ma and despite there being a restricted potential for positive
Teasdale 2004; Teasdale et al. 2000; Williams et al. change. Many studies have demonstrated the clinical im-
2014). portance of reducing residual depression symptoms, as
The fact that less than half the sample had depressive they have a large impact on long-term depression out-
symptoms below the clinical cut-off on the PHQ-9 at comes (Rottenberg et al. 2018). In the absence of long-
 Mindfulness

Table 4 Clinical indicators of

change for the pooled sample and Sample Reliable Reliable Recovery % Reliable Sustained
each service, subdivided into non- improvement % deterioration % recovery % recovery %
depressed (‘No dep’) and de-
pressed (‘Current dep’) at entry to Pooled 25.16 3.22
treatment No Dep 7.58 4.13 95.73
Current Dep 40.58 2.42 44.81 34.42
Swallow 16.00 10.00
No Dep 5.13 11.54 91.03
Current Dep 27.78 8.33 43.06 27.78
Robin 34.06 4.72
No Dep 9.39 6.12 93.88
Current Dep 57.03 3.42 59.70 48.29
Jackdaw 17.26 0.63
No Dep 6.79 0.36 98.93
Current Dep 32.31 1.03 36.92 27.18
Woodpecker 18.23 2.76
No Dep 3.39 5.08 93.22
Current Dep 25.41 1.64 22.13 17.21
Blackbird 32.92 1.25
No Dep 10.94 3.13 96.88
Current Dep 40.91 0.57 47.73 36.36

Where post-treatment PHQ-9 data were missing, the pre-treatment PHQ-9 value was carried forward

term follow-up of service users, a reduction in residual (Barnhofer et al. 2009; Chiesa et al. 2015; Eisendrath et al.
symptoms of depression can also be regarded an imper- 2015; Kingston et al. 2007). In the present study, the effect on
fect proxy for reduction in risk of relapse (Ali et al. PHQ-9 symptoms was smaller compared to interventions offered
2017). Although a measure of relapse beyond the end in IAPT services overall for depression cases (1.4; Clark 2018).
of treatment would have been a more appropriate out- However, the rate of reliable improvement (41%) was compara-
come measure for this subgroup, this information was ble to that of CBT in a sample of service users at 103 IAPT
not available from clinical services. However, the pre- services that were above the clinical cutoff on the PHQ-9 at
to post-treatment effect size identified in the present intake (47%; Pybis et al. 2017). Service users in the present study
study was comparable to that of a large RCT (d = 0.35; most commonly took part in MBCT as a second-line treatment
Kuyken et al. 2015), where the main outcome variable after another form of psychological therapy (typically CBT) and
was time-to-relapse over 24 months, and outcomes were had lower PHQ-9 scores at intake than the IAPT samples men-
shown to be comparable between MBCT and mainte- tioned above. Therefore, one might expect a smaller effect size in
nance antidepressant medication. This suggests that this group compared to the IAPT sample that also included those
MBCT in the real world may produce benefits compara- who took part in CBT as a first-line treatment. Nonetheless, an
ble to those found in RCTs in terms of supporting recov- important remaining question is, ‘What approach is most accept-
ery, at least over the limited follow-up periods. able, effective, and cost-effective for different subpopulations of
In the group who entered MBCT with depression (e.g. above people with depression, at different stages in the natural history of
cutoff on the PHQ-9 for caseness), results showed large and depression (e.g. first episode, recurrent depression)?’
significant improvements in depressive symptoms, d = 0.86, Finally, we asked if MBCT was equally helpful for people
and a 34% rate of reliable recovery. In the absence of a of different ages and genders. Consistent with Kuyken et al.
randomised design with a no-intervention control group, it is (2016), neither age nor gender was associated with the degree
not possible to determine the extent to which these improvements of reduction in depressive symptoms during treatment, indi-
in depressive symptoms are attributable to natural recovery rather cating that outcomes do not seem to be influenced by these
than intervention effects. However, it should be noted that the sociodemographic variables.
effect sizes observed are similar to those reported in a study that
documented pre- to post-intervention changes in depressive Interpretation of Service-Specific Outcomes
symptoms in patients randomised to MBCT or treatment as usual
(van Aalderen et al. 2012). Where substantially larger effects Our study examined data from five clinical services which
have been observed, these have been based on small samples differed to some degree in the populations served (in particular
Mindfulness

with respect to clinical eligibility), the intervention delivered who reliably improved was 17 times greater than the proportion
(e.g. number of sessions and adaptations), and the structure of who reliably deteriorated. These figures should be interpreted
the services including the training pathway completed by with some caution since where data were missing at follow-up,
MBCT instructors. There were also marked differences be- baseline data were carried forward. Thus, in the current analysis,
tween services in the availability of post-treatment data. for services where data were collected only at the first and final
Whilst findings were broadly consistent across services, there sessions of treatment, participants experiencing a significant
were also areas of divergence, and the services differed sig- worsening of symptoms and dropping out of treatment would
nificantly from one another clinical outcomes. Attributing be recorded as having no change in symptom levels whereas in
these differences to particular service characteristics is chal- fact symptoms might have worsened. Likewise, in such services
lenging. The measured sociodemographic characteristics of levels of sustained recovery are likely to over-estimate real ef-
service users and their clinical symptoms at baseline did not fects. Robin, the largest unique sample, with negligible missing
show marked differences between services, although it is pos- data, had relatively low rates of reliable deterioration (4.7%),
sible that there were other differences not captured by routine high rates of reliable improvement (34%) and high rates of
data collection. It is also possible that differences between sustained recovery (94%), whereas Jackdaw, in contrast, had
services are related to models of MBCT teacher training, su- very high rates of missing data (and hence baseline data carried
pervision and service delivery. In the absence of data on teach- forward) and showed reliable deterioration rates of 0.63%, reli-
er competency or intervention fidelity, this is speculation, but able improvement rates of 17% and rates of sustained recovery of
it nonetheless raises the important possibility that the way 99%. This pattern of results suggests that the approach we
services offer MBCT can affect outcomes. Future research adopted of carrying forward baseline data where outcome data
might evaluate outcomes over a larger number of clinical ser- were missing data may be suppressing rates of both reliable
vices, varying in patient eligibility requirements, MBCT improvement and deterioration, and exaggerating rates of
teacher training pathways, and models of service delivery, to sustained recovery. Robin, as the largest and most complete sam-
shed light on how patient and service factors interact to deter- ple, probably provides the most robust estimate of reliable dete-
mine outcomes. In addition, an in-depth study of how adapta- rioration, reliable improvement and sustained recovery to be ex-
tions in psychoeducational content or the issues explored in pected from MBCT in routine care settings.
group enquiry in MBCT influence the learning of patients
with different presenting symptoms and clinical histories Accessibility
might shed light on the best way to diversify services to
broader populations without diluting treatment effects. Overall, the rates of engagement suggested that those electing
to start MBCT found it acceptable. However, despite the fact
Safety of MBCT in Real-World Settings that MBCT services were located in relatively ethnically di-
verse communities, the participants attending treatment were
The fact that rates of reliable deterioration are below 5% for the predominantly White. For example, Swallow has a regional
total sample and for subgroups entering above and below clinical ethnicity of 60% White British, compared to 76% of those
cutoff for depression, suggests that MBCT is a treatment option receiving MBCT and providing ethnicity data. Likewise,
which is generally safe, and that reliable deteriorations occur no Jackdaw serves a population that is 88% White British, com-
more frequently after treatment with MBCT than with other pared to 93% of those receiving MBCT and providing ethnic-
psychotherapies. For example, Hansen et al. (2002) report a re- ity data. It is well established that people from Black, Asian
liable deterioration rate of 8.2% in a sample of more than 6000 and minority ethnic backgrounds in the UK face barriers to
individuals receiving various forms of psychotherapy in real- accessing mental health services (e.g. Sashidharan 2003;
world settings in the USA, whilst Crawford et al. (2016) found Memon et al. 2016). It is not clear to what extent the dispar-
that in a survey of over 14,000 UK NHS psychological therapy ities in ethnic composition in our datasets reflect broader struc-
patients, just over 5% reported lasting bad effects of treatment. tural barriers to accessing mental health services by people
Similarly, the observed rates of reliable deterioration are compa- from these ethnic backgrounds, or whether there are particular
rable to those observed in IAPT services for anxiety and depres- issues with access to and acceptability of MBCT. Existing
sion as a whole (e.g. Clark 2018). Results show that those enter- clinical trials of MBCT for relapse prevention in depression
ing MBCT treatment with depression symptoms below the clin- have also focussed largely on Caucasian samples and/or have
ical cutoff are still almost twice as likely to reliably improve not reported information on the ethnicity of participants (e.g.
further as to deteriorate over the course of treatment (although Goldberg et al. 2018; Kuyken et al. 2016). Whilst this aids the
absolute proportions showing both types of change are low), comparison of the results in this dataset with those of previous
even from their low level of baseline symptoms, and in the con- research, it highlights that issues of equality and diversity in
text of a high risk of relapse. Furthermore, of those entering access to MBCT are an area where future research and action
treatment above the clinical cutoff for depression, the proportion is urgently needed.
 Mindfulness

Strengths and Limitations IAPT depression services as a whole (Clark 2018).

Therefore, the current data suggest that whilst MBCT
This study drew existing data from real-world clinical ser- can be safely and effectively delivered to individuals
vices. This is a strength of the work as it provides a naturalistic who are both in remission and symptomatic, we would
picture of the effectiveness of MBCT as it is delivered in suggest it may be best conceptualised as a second-line
healthcare settings, without any experimental interference or treatment. Indeed, many services have chosen to offer
control. However, this also brought with it certain challenges. MBCT in this way. This approach recognises that de-
In particular, there were high rates of missing data in some pression is typically a lifelong, recurrent disorder, and
services, which, as discussed above, may have influenced MBCT can add benefit after other treatments designed
both estimates of positive and negative treatment effects. more specifically to address acute symptoms. In line
Likewise, data on variables such as service-users’ clinical his- with an agenda to increase patient choice, it provides
tory, additional ongoing treatments, instructor competence, an alternative approach for people wishing to better
class size and composition, and service user engagement with manage their depression. Indeed, further RCTs address-
home practice, which might be hypothesised to influence out- ing issues of treatment sequencing, the acute and
comes, were not available. Similarly, we did not have access relapse–prevention effects of MBCT when offered irre-
to information on the groupings of individuals receiving spective of stage of illness, and long-term preventative
MBCT, so in our analysis, we were unable to correct for the effects of MBCT in real-world settings are now re-
intragroup correlations between participants. Previous studies quired. This work would be enriched by broadening
of MBCT have not found significant intragroup dependency, outcomes beyond symptomatology and disorder status
and re-analysis of datasets using hierarchical linear modelling to include functional status, quality of life and living
has been found to reinforce original findings (Williams et al. well with depression.
2008). However, this may not be the case in real-world MBCT
services, and so, we recommend that healthcare services could Acknowledgements We would like to thank the NHS trusts who re-
leased data to contribute to this research, as well as the service users
store information on class grouping so that this can be taken
and NHS teachers whose data contributes to this research.
into account in future studies. Another limitation was that this
study recruited services from NHS England, and so the find- Author Contributions WK, AT and CC conceived the study and devel-
ings only reflect this country’s model of service provision. oped the research protocol. CS, RM, PB, TS and SP supported data
Models of delivery in other countries are likely to differ from acquisition. AT obtained regulatory approvals. AT and SB conducted
the data analyses under supervision of CC and WK. AT and CC drafted
that of England, and so, it is unclear whether the outcomes of
the manuscript. All authors contributed to interpretation of the data and
this study would translate to other forms of MBCT delivery revised the manuscript for important intellectual content.
taking place internationally.
In summary, our findings suggest that MBCT in real- Funding WK, CC and AT are supported by the Wellcome Trust [104908/Z/
world settings is acceptable and effective. However, it is 14/Z and 107496/Z/15/Z]. SB is funded by the National Institute for Health
important to reflect on the fact that MBCT was first Research (NIHR) Collaboration for Leadership in Applied Health Research
and Care South West Peninsula (NIHR CLAHRC South West Peninsula).
designed as a relapse prevention treatment for individ- The views expressed are those of the authors and not necessarily those of the
uals at high risk of depressive relapse to learn skills NHS, the NIHR or the Department of Health and Social Care.
whilst well (i.e. with below threshold levels of depres-
sive symptoms). In line with other recent work (Rycroft- Compliance with Ethical Standards
Malone et al. 2017), services are offering MBCT to a
broader population. In our sample just under 47% of The Health Research Authority gave ethical approval to the University of
Oxford for this study, and it was thereby in accordance with the ethical
service users entered treatment with levels of depressive standards laid down in the 1964 Declaration of Helsinki and its later
symptoms below the clinical cutoff on the PHQ, and amendments. The participating NHS organisations confirmed their capac-
thus might be considered to have been entering MBCT ity and capability to host this research.
for relapse prevention purposes. Encouragingly, in this
subgroup, there were very high rates of sustained recov- Conflict of Interest AT and CC are affiliated with the Oxford
Mindfulness Centre and employed to conduct research on MBCT and
ery, low rates of sustained deterioration a small and related psychological interventions. However, they receive no additional
statistically significant effect on residual symptoms, remuneration for books or training related to MBCT. PB is employed as
which are consistently shown to be the best predictor the Mindfulness Lead for Tees, Esk and Wear Valleys NHS Foundation
of depressive relapse (Bockting et al. 2015). Trust. PB and TS receive payments for MBCT-related training and super-
vision. CS and RM co-lead the Sussex Mindfulness Centre and CS has
However, 53% of our sample entered MBCT with received NIHR funding for research trials evaluating mindfulness-based
acute depressive symptoms, and whilst acceptability interventions. WK is the director of the Oxford Mindfulness Centre. WK
and effectiveness were again encouraging, the effect size receives payments for training workshops and presentations related to
for outcomes, d = 0.86, whilst large, is smaller than for MBCT and donates all such payments to the Oxford Mindfulness
Mindfulness

Foundation, a charitable trust that supports the work of the Oxford Dimidjian, S., & Segal, Z. V. (2015). Prospects for a clinical science of
Mindfulness Centre. WK was until 2015 an unpaid Director of the mindfulness-based intervention. American Psychologist, 70(7), 593.
Mindfulness Network Community Interest Company and gave evidence Eaton, W. W., Shao, H., Nestadt, G., Lee, B. H., Bienvenu, O. J., & Zandi,
to the UK Mindfulness All Party Parliamentary Group. SB and SP declare P. (2008). Population-based study of first onset and chronicity in
that they have no conflicts of interest. major depressive disorder. Archives of General Psychiatry, 65(5),
513–520.
Informed Consent The study used previously collected and anonymised Eisendrath, S. J., Gillung, E., Delucchi, K., Mathalon, D. H., Yang, T. T.,
patient data. Formal consent from patients is not required for this type of Satre, D. D., et al. (2015). A preliminary study: Efficacy of mindfulness-
study. based cognitive therapy versus sertraline as first-line treatments for ma-
jor depressive disorder. Mindfulness, 6(3), 475–482.
Open Access This article is distributed under the terms of the Creative Farb, N., Anderson, A., Ravindran, A., Hawley, L., Irving, J., Mancuso,
Commons Attribution 4.0 International License (http:// E., et al. (2018). Prevention of relapse/recurrence in major depres-
creativecommons.org/licenses/by/4.0/), which permits unrestricted use, sive disorder with either mindfulness-based cognitive therapy or
distribution, and reproduction in any medium, provided you give cognitive therapy. Journal of Consulting and Clinical Psychology,
appropriate credit to the original author(s) and the source, provide a link 86(2), 200–204.
to the Creative Commons license, and indicate if changes were made. Goldberg, S. B., Tucker, R. P., Greene, P. A., Davidson, R. J., Wampold,
B. E., Kearney, D. J., & Simpson, T. L. (2018). Mindfulness-based
Publisher’s Note Springer Nature remains neutral with regard to juris- interventions for psychiatric disorders: A systematic review and me-
dictional claims in published maps and institutional affiliations. ta-analysis. Clinical Psychology Review, 59, 52–60.
Guo, B., Kaylor-Hughes, C., Garland, A., Nixon, N., Sweeney, T.,
Simpson, S., et al. (2017). Factor structure and longitudinal mea-
surement invariance of PHQ-9 for specialist mental health care pa-
References tients with persistent major depressive disorder: Exploratory struc-
tural equation modelling. Journal of Affective Disorders, 219, 1–8.
Ali, S., Rhodes, L., Moreea, O., McMillan, D., Gilbody, S., Leach, C., Hansen, N. B., Lambert, M. J., & Forman, E. M. (2002). The psychother-
et al. (2017). How durable is the effect of low intensity CBT for apy dose-response effect and its implications for treatment delivery
depression and anxiety? Remission and relapse in a longitudinal services. Clinical Psychology: Science and Practice, 9(3), 329–343.
cohort study. Behaviour Research and Therapy, 94, 1–8. Henggeler, S. W. (2004). Decreasing effect sizes for effectiveness studies-
American Psychiatric Association. (2000). Diagnostic and statistical manual implications for the transport of evidence-based treatments:
of mental disorders (4th ed., text rev.). Washington, DC: Author. Comment on Curtis, Ronan, and Borduin (2004). Journal of
Barnhofer, T., Crane, C., Hargus, E., Amarasinghe, M., Winder, R., & Family Psychology, 18(3), 420–423.
Williams, J. M. G. (2009). Mindfulness-based cognitive therapy as a Henggeler, S. W. (2011). Efficacy studies to large-scale transport: The
treatment for chronic depression: A preliminary study. Behaviour development and validation of multisystemic therapy programs.
Research and Therapy, 47(5), 366–373. Annual Review of Clinical Psychology, 7, 351–381.
Bockting, C. L., Hollon, S. D., Jarrett, R. B., Kuyken, W., & Dobson, K. Hofmann, S. G., Sawyer, A. T., Witt, A. A., & Oh, D. (2010). The effect
(2015). A lifetime approach to major depressive disorder: The con- of mindfulness-based therapy on anxiety and depression: A meta-
tributions of psychological interventions in preventing relapse and analytic review. Journal of Consulting and Clinical Psychology,
recurrence. Clinical Psychology Review, 41, 16–26. 78(2), 169.
Chiesa, A., Castagner, V., Andrisano, C., Serretti, A., Mandelli, L., Hollon, S. D. (2016). The efficacy and acceptability of psychological
Porcelli, S., & Giommi, F. (2015). Mindfulness-based cognitive interventions for depression: Where we are now and where we are
therapy vs. psycho-education for patients with major depression going. Epidemiology and Psychiatric Sciences, 25(4), 295–300.
who did not achieve remission following antidepressant treatment. Judd, L. L., Akiskal, H. S., & Paulus, M. P. (1997). The role and clinical
Psychiatry Research, 226(2), 474–483. significance of subsyndromal depressive symptoms (SSD) in unipo-
Clark, D. M., Layard, R., Smithies, R., Richards, D. A., Suckling, R., & lar major depressive disorder. Journal of Affective Disorders, 45(1–
Wright, B. (2009). Improving access to psychological therapy: 2), 5–17.
Initial evaluation of two UK demonstration sites. Behaviour Kingston, T., Dooley, B., Bates, A., Lawlor, E., & Malone, K. (2007).
Research and Therapy, 47(11), 910–920. Mindfulness-based cognitive therapy for residual depressive symp-
Clark, D. M. (2018). Realising the mass public benefit of evidence-based toms. Psychology and Psychotherapy: Theory, Research and
psychological therapies: The IAPT Programme. Annual Review of Practice, 80(2), 193–203.
Clinical Psychology, 14, 159–183. Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9:
Cohen, J. (1988). Statistical power analysis for the behavioral sciences Validity of a brief depression severity measure. Journal of General
(2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates. Internal Medicine, 16(9), 606–613.
Crane, R. S., & Kuyken, W. (2013). The implementation of mindfulness- Kuyken, W., Hayes, R., Barrett, B., Byng, R., Dalgleish, T., Kessler, D.,
based cognitive therapy: Learning from the UK health service expe- et al. (2015). Effectiveness and cost-effectiveness of mindfulness-
rience. Mindfulness, 4(3), 246–254. based cognitive therapy compared with maintenance antidepressant
Crawford, M. J., Thana, L., Farquharson, L., Palmer, L., Hancock, E., treatment in the prevention of depressive relapse or recurrence
Bassett, P., et al. (2016). Patient experience of negative effects of (PREVENT): A randomised controlled trial. The Lancet,
psychological treatment: Results of a national survey. The British 386(9988), 63–73.
Journal of Psychiatry, 208(3), 260–265. Kuyken, W., Warren, F. C., Taylor, R. S., Whalley, B., Crane, C.,
Curtis, N. M., Ronan, K. R., & Borduin, C. M. (2004). Multisystemic Bondolfi, G., et al. (2016). Efficacy of mindfulness-based cognitive
treatment: A meta-analysis of outcome studies. Journal of Family therapy in prevention of depressive relapse: An individual patient
Psychology, 18(3), 411. data meta-analysis from randomized trials. JAMA Psychiatry, 73(6),
Demarzo, M. M. P., Cebolla, A., & Garcia-Campayo, J. (2015). The 565–574.
implementation of mindfulness in healthcare systems: A theoretical Lu, S. (2015). Mindfulness holds promise for treating depression.
analysis. General Hospital Psychiatry, 37(2), 166–171. Monitor on Psychology, 46(3), 50.
 Mindfulness

Ma, S. H., & Teasdale, J. D. (2004). Mindfulness-based cognitive therapy generic counselling in the treatment of depression: Evidence from
for depression: Replication and exploration of differential relapse the 2nd UK National Audit of psychological therapies. BMC
prevention effects. Journal of Consulting and Clinical Psychology, Psychiatry, 17(1), 215.
72(1), 31. Rottenberg, J., Devendorf, A. R., Kashdan, T. B., & Disabatom, D. J.
Malhi, G. S., Bassett, D., Boyce, P., Bryant, R., Fitzgerald, P. B., Fritz, K., (2018). The curious neglect of high functioning after psychopathol-
et al. (2015). Royal Australian and New Zealand College of ogy: The case of depression. Perspectives on Psychological Science,
Psychiatrists clinical practice guidelines for mood disorders. 13(5), 549–566.
Australian and New Zealand Journal of Psychiatry, 49(12), 1087–1206. Rycroft-Malone, J., Gradinger, F., Griffiths, H. O., Crane, R., Gibson, A.,
Manicavasgar, V., Parker, G., & Perich, T. (2011). Mindfulness-based Mercer, S., et al. (2017). Accessibility and implementation in the
cognitive therapy vs cognitive behaviour therapy as a treatment for UK NHS services of an effective depression relapse prevention pro-
non-melancholic depression. Journal of Affective Disorders, 130(1– gramme: Learning from mindfulness-based cognitive therapy
2), 138–144. through a mixed-methods study. Health Services and Delivery
Martin, A., Rief, W., Klaiberg, A., & Braehler, E. (2006). Validity of the Research, 5(14), 1–190.
brief patient health questionnaire mood scale (PHQ-9) in the general Sashidharan, S. (2003). From outside to inside: Improving mental health
population. General Hospital Psychiatry, 28(1), 71–77. services for black and minority ethnic communities in England.
McManus, S., Bebbington, P., Jenkins, R. & Brugha, T. (2016). Mental Mental Health Review Journal, 8(3), 22–25.
health and wellbeing in England: Adult psychiatric morbidity survey Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002). Mindfulness-
2014. Leeds, UK: NHS Digit. based cognitive therapy for depression: A new approach to
Meadows, G., Shawyer, F., Enticott, J., Graham, A., Judd, F., Martin, P., et al. preventing relapse. New York: Guilford Press.
(2014). Mindfulness-based cognitive therapy for recurrent depression: Shallcross, A., Gross, J., Visvanathan, P., Kumar, N., Palfrey, A., Ford, B.,
A translational research study with 2-year follow-up. Australian & New et al. (2015). Relapse prevention in major depressive disorder:
Zealand Journal of Psychiatry, 48(8), 743–755 m. Mindfulness-based cognitive therapy versus an active control condition.
Memon, A., Taylor, K., Mohebati, L. M., Sundin, J., Cooper, M., Journal of Consulting and Clinical Psychology, 83(5), 964–975.
Scanlon, T., & de Visser, R. (2016). Perceived barriers to accessing Shallcross, A., Willroth, E., Fisher, A., Dimidjian, S., Gross, J.,
mental health services among black and minority ethnic (BME) Visvanathan, P., & Mauss, I. (2018). Relapse/recurrence prevention
communities: A qualitative study in Southeast England. BMJ in major depressive disorder: 26-month follow-up of mindfulness-
Open, 6(11), e012337. based cognitive therapy versus an active control. Behaviour
Miller, W. R. (2005). Motivational interviewing and the incredible shrink- Therapy, 49(5), 836–849.
ing treatment effect. Addiction, 100(4), 421–421.
Solomon, D. A., Keller, M. B., Leon, A. C., Mueller, T. I., Lavori, P. W.,
Moffitt, T. E., Caspi, A., Taylor, A., Kokaua, J., Milne, B. J., Polanczyk,
Shea, M. T., et al. (2000). Multiple recurrences of major depressive
G., & Poulton, R. (2010). How common are common mental disor-
disorder. American Journal of Psychiatry, 157(2), 229–233.
ders? Evidence that lifetime prevalence rates are doubled by pro-
Spitzer, R., Kroenke, K., & Williams, J. (1999). Validation and utility of a
spective versus retrospective ascertainment. Psychological
self- report version of PRIME-MD: The PHQ Primary Care Study.
Medicine, 40(6), 899–909.
Journal of the American Medical Association, 282(18), 1737–1744.
Morris, S. B., & DeShon, R. P. (2002). Combining effect size estimates in
meta-analysis with repeated measures and independent-groups de- Spitzer, R. L., Williams, J. B., Kroenke, K., Hornyak, R., & McMurray, J.
signs. Psychological Methods, 7(1), 105. (2000). Validity and utility of the PRIME-MD patient health question-
The National Collaborating Centre for Mental Health. (2018). The improving naire in assessment of 3000 obstetric-gynecologic patients: The PRIME-
access to psychological therapies manual. Retrieved from: https://www. MD Patient Health Questionnaire Obstetrics-Gynecology Study.
england.nhs.uk/wp-content/uploads/2018/06/the-iapt-manual.pdf. American Journal of Obstetrics and Gynecology, 183(3), 759–769.
Accessed 25 Oct 2018 Strauss, C., Cavanagh, K., Oliver, A., & Pettman, D. (2014).
National Institute of Health and Care Excellence [NICE] (2009). Mindfulness-based interventions for people diagnosed with a cur-
Depression: The treatment and management of depression in adults rent episode of an anxiety or depressive disorder: A meta-analysis of
(update). London: Author. randomised controlled trials. PLoS One, 9(4), e96110.
Onken, L. S., Carroll, K. M., Shoham, V., Cuthbert, B. N., & Riddle, M. Teasdale, J. D., Segal, Z. V., Williams, J. M. G., Ridgeway, V. A.,
(2014). Reenvisioning clinical science: Unifying the discipline to im- Soulsby, J. M., & Lau, M. A. (2000). Prevention of relapse/
prove the public health. Clinical Psychological Science, 2(1), 22–34. recurrence in major depression by mindfulness-based cognitive ther-
Parikh, S. V., Quilty, L. C., Ravitz, P., Rosenbluth, M., Pavlova, B., apy. Journal of Consulting and Clinical Psychology, 68(4), 615.
Grigoriadis, S., et al. (2016). Canadian Network for Mood and van Aalderen, J. R., Donders, A. R. T., Giommi, F., Spinhoven, P.,
Anxiety Treatments (CANMAT) 2016 clinical guidelines for the Barendregt, H. P., & Speckens, A. E. M. (2012). The efficacy of
management of adults with major depressive disorder: Section 2. mindfulness-based cognitive therapy in recurrent depressed patients
Psychological treatments. Canadian Journal of Psychiatry, 61(9), with and without a current depressive episode: A randomized con-
524–539. trolled trial. Psychological Medicine, 42(5), 989–1001.
Perepletchikova, F., Treat, T. A., & Kazdin, A. E. (2007). Treatment Williams, J. M. G., Crane, C., Barnhofer, T., Brennan, K., Duggan, D. S.,
integrity in psychotherapy research: Analysis of the studies and Fennell, M. J., et al. (2014). Mindfulness-based cognitive therapy
examination of the associated factors. Journal of Consulting and for preventing relapse in recurrent depression: A randomized dis-
Clinical Psychology, 75(6), 829. mantling trial. Journal of Consulting and Clinical Psychology,
Pintor, L., Torres, X., Navarro, V., Matrai, S., & Gastó, C. (2004). Is the 82(2), 275.
type of remission after a major depressive episode an important risk Williams, J. M. G., Russell, I., & Russell, D. (2008). Mindfulness-based
factor to relapses in a 4-year follow up? Journal of Affective cognitive therapy: Further issues in current evidence and future research.
Disorders, 82(2), 291–296. Journal of Consulting and Clinical Psychology, 76(3), 524–529.
Pybis, J., Saxon, D., Hill, A., & Barkham, M. (2017). The comparative World Health Organization. (2008). The global burden of disease: 2004
effectiveness and efficiency of cognitive behaviour therapy and update. Geneva: WHO.