③

Presseau et al. Implementation Science (2019) 14:102

https://doi.org/10.1186/s13012-019-0951-x

METHODOLOGY Open Access

Action, actor, context, target, time (AACTT):

a framework for specifying behaviour
Justin Presseau1,2,3* , Nicola McCleary1,2, Fabiana Lorencatto4, Andrea M. Patey1, Jeremy M. Grimshaw1,2,5 and
Jill J. Francis6

Abstract
Background: Designing implementation interventions to change the behaviour of healthcare providers and other
professionals in the health system requires detailed specification of the behaviour(s) targeted for change to ensure
alignment between intervention components and measured outcomes. Detailed behaviour specification can help
to clarify evidence-practice gaps, clarify who needs to do what differently, identify modifiable barriers and enablers,
design interventions to address these and ultimately provides an indicator of what to measure to evaluate an
intervention’s effect on behaviour change. An existing behaviour specification framework proposes four domains
(Target, Action, Context, Time; TACT), but insufficiently clarifies who is performing the behaviour (i.e. the Actor).
Specifying the Actor is especially important in healthcare settings characterised by multiple behaviours performed
by multiple different people. We propose and describe an extension and re-ordering of TACT to enhance its utility
to implementation intervention designers, practitioners and trialists: the Action, Actor, Context, Target, Time (AACTT)
framework. We aim to demonstrate its application across key steps of implementation research and to provide tools
for its use in practice to clarify the behaviours of stakeholders across multiple levels of the healthcare system.
Methods and results: We used French et al.’s four-step implementation process model to describe the potential
applications of the AACTT framework for (a) clarifying who needs to do what differently, (b) identifying barriers and
enablers, (c) selecting fit-for-purpose intervention strategies and components and (d) evaluating implementation
interventions.
Conclusions: Describing and detailing behaviour using the AACTT framework may help to enhance measurement
of theoretical constructs, inform development of topic guides and questionnaires, enhance the design of
implementation interventions and clarify outcome measurement for evaluating implementation interventions.
Keywords: Behaviour, Framework, Behaviour specification, TACT, Behaviour change, Health professional behaviour

* Correspondence: [email protected]
1
Clinical Epidemiology, Ottawa Hospital Research Institute, Ottawa, Canada
2
School of Epidemiology and Public Health, University of Ottawa, Ottawa,
Canada
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Presseau et al. Implementation Science (2019) 14:102 Page 2 of 13

an intervention’s effect and (e) ultimately facilitate evi-

Contributions to the literature
dence synthesis. A generalisable framework may help to
! Behaviour change is a fundamental outcome of interest in ensure consistency in the description and specification
implementation science and underpins change at multiple of behaviour in implementation research.
health care levels, but behaviour is often not clearly specified
In the mid-twentieth century, after social psychologists
identified that scores on attitude measures were not as-
! Detailed behaviour specification helps to clarify evidence-
sociated with actual behaviour [6], Fishbein [7] proposed
practice gaps, clarify who needs to do what differently,
that the low predictive validity of attitude measures
identify barriers and enablers, design interventions and could be addressed by assessing attitude to an action
provides an indicator of what to measure to evaluate an (e.g. voting for a specific political party) rather than
intervention’s effect on behaviour change assessing attitude to the target of that action (e.g. the
! Existing frameworks (TACT) do not clarify the Actor(s) political party). While now taken for granted, this focus
(e.g. clinicians) performing the behaviour, an important on the action led to a paradigm shift in attitude-
distinction given multiple Actors and multiple Actions of
behaviour research and was a key principle underlying
the Theory of Reasoned Action. Extending the approach,
interest in implementation science
Ajzen [8] proposed the Theory of Planned Behaviour for
! We propose an extension: the Action, Actor, Context, Target,
predicting and explaining human behaviour in a specific
Time (AACTT) framework for behaviour specification context at a specific time. Together, these ideas gave rise
! We provide practical tools to aid in AACTT-specification for a to the specification of behaviour according to what be-
range of implementation study designs came known as TACT: Target, Action, Context, Time
[9]. Similarly, Michie and Johnston [10] proposed that
when behaviours are described in terms of what, who,
Background when, where and how, they are more actionable and
Innovations in health research have the potential to im- hence more likely to be performed.
prove health but harnessing this potential requires that Clear specification of the behaviour is a key though often
effective innovations be translated into routine health- overlooked first step in conducting implementation re-
care. Unfortunately, evidence-practice gaps continue to search for a range of study objectives, such as identifying
be documented: medicines are inappropriately pre- influences (barriers and enablers, and determinants) on
scribed [1], patient safety practices are not enacted [2] behaviour or designing implementation interventions to
and harmful practices persist [3]. Sub-optimal clinical support behaviour change among stakeholders in the
practices (over, under and misuse of tests/treatments) health system. Despite half a century of guidance on behav-
result in avoidable morbidity and mortality [4]. Such iour specification, research is frequently published in which
gaps in care are consistent across countries and clinical the behaviour is poorly specified. A systematic review of 67
areas, leading some to suggest that health research is ‘all reports of behaviour change interventions found that the
breakthrough, no follow through’ [5]. Implementation Action domain was clearly specified in 69% of reports, and
science emerged in response to this, focusing on the that all components of the TACT framework were
rigorous scientific study and development of a cumula- described in only 5 (7.5%) reports [11]. Poor specification
tive evidence-base for how best to address evidence- makes it difficult to measure behaviour and behaviour
practice gaps. A foundational requirement of implemen- change. Clear specification facilitates strong compatibility
tation is the need for someone (usually more than one between the behaviour under investigation and the theore-
person or group), somewhere (from organisational lead- tical constructs that predict that behaviour, which enhances
ership through to those providing direct patient care) prediction (cf. the principle of compatibility) [12, 13].
doing something (usually more than one thing) differ- Consider the following description of a potential
ently. In short, taking up new evidence requires health- guideline-recommended clinical behaviour for primary
care providers and other health system stakeholders to care practitioners: ‘For people with diabetes, record a
change their behaviour. blood pressure reading in the patient’s medical records’
Given the centrality of behaviour in implementation (Example 1). Using the TACT framework to unpack this
science, there is a need for describing behaviour as recommendation, the two specified components are the
clearly as possible. Doing so may help to (a) clarify Target (people with diabetes) and the Action (record a
evidence-practice gaps, (b) clarify the various people and blood pressure reading in the patient’s medical record).
groups at different levels that need to do something dif- While a seemingly straightforward description, it is not
ferently, (c) identify modifiable barriers and enablers and clear who should do the behaviour and when and where
design implementation interventions to address them, it should take place. Leaving this implicit introduces am-
(d) provide an indicator of what to measure to evaluate biguity that may undermine change efforts as well as
Presseau et al. Implementation Science (2019) 14:102 Page 3 of 13

measurement of whether or not the behaviour has been While Action and Actor are important, specification of
performed. Context and Time allows the responder to keep these
Drawing upon Fishbein and Ajzen’s [9] early advice, a elements in mind when answering a questionnaire,
clearer, more actionable specification could be: ‘For responding in an interview or changing their behaviour.
people with diabetes (Target), record a blood pressure Behaviours are inherently tied to the time and place in
reading in the patient’s medical records (Action) in the which they occur, and thus clarification of these ele-
primary care clinic (Context) when they attend for their ments provides an opportunity to situate analyses of bar-
annual diabetes review (Time)’ (Example 2). Example 2 riers and enablers and intervention design within the
specifies further components of the behaviour: Time contexts that behaviours take place. Recent theoretical
(when patients attend) and Context (primary care clinic). advances emphasise the role of associative processes in
This enhanced specification corresponds to the TACT behaviour [15], including automatic processes [16–19].
framework, but still lacks a fundamental component: Specification of contextual and temporal cues in ques-
who (i.e. which person or people on the primary care tionnaire items, interview topic guides and observational
team) is responsible for performing the Action. Further- tools may increase the validity of responses, especially if
more, Example 2 arguably involves a sequence of the behaviour has an element of automaticity. Context
discrete Actions (taking the blood pressure reading, and time may also be important for identifying when
accessing the patient’s medical records, entering the and where it is appropriate to perform an Action, thus
blood pressure reading into the record). It may also in- informing implementation efforts.
clude ancillary behaviours such as inviting the patient to When considering the behaviour of individuals and
attend the clinic for the annual review. These actions teams as they deliver health care, a further refinement of
may be performed by different primary care staff (e.g. the ‘Target’ domain is appropriate. ‘Target’ is often ex-
physician, nurse, administrator) to support the focal be- plained as (performing) an Action to someone, i.e. who
haviour of interest. This dimension of specification is the behaviour is targeted at. However, current models of
not included in the TACT framework, which assumes healthcare delivery place a focus on patients as active
that the individual is performing the behaviour for them- participants in their health care, and thus as a collabor-
selves. In implementation research, individuals often ator with the healthcare professional. Hence, rather than
perform a behaviour for someone else’s benefit (i.e. the performing an Action to a passive recipient, the health-
Target, such as a patient). We propose to expand the care professional may act with or for the patient. Thus, it
TACT framework to guide specification of behaviour in is recommended that researchers frame (i) Action and
terms of not only Target, Action, Context and Time but (ii) Target as (i) doing what? (ii) to, for, with or on behalf
also Actor—the person(s) who will perform the Ac- of whom?
tion(s). By clarifying the Actor, the Action then becomes In doing so, the Actor-Target relationship need not
clearer and more specific, allowing for clarification of only reflect a healthcare professional-patient relation-
complex behaviours (or sequences of behaviour) in ship. As demonstrated in Fig. 3, for healthcare profes-
terms of different Actions performed by different Actors sionals working in a team, one healthcare professional’s
in the health care setting at different times (i.e. prepara- (Actor A) behaviour (Action A) may be for the benefit
tory and sequential Actions). of another healthcare professional (Target A), enabling
the latter’s subsequent behaviour (Action B). Such hori-
AACTT: an expanded framework for specifying behaviour zontal sequences within teams also apply to specifying
We propose the AACTT framework (Action, Actor, the behaviour of vertical sequences of behaviour within
Context, Target, Time) for specifying behaviour. Re- the health system, where a policymaker’s (Actor A) be-
arranging the order of domains in the framework reflects haviour (Action A) sets the stage for a healthcare ad-
a more easily defined sequence for specifying behaviour ministrator (Target A/Actor B) to perform a behaviour
than TACT that naturally begins with the Action and within their role (Action B) that benefits the healthcare
who performs it. professional (Target B/Actor C) and enables them to
Although it may sometimes seem obvious who is to provide care (Behaviour C) to benefit their patients
perform the Action, for behaviours that are performed (Target C). Thus, a given Actor’s Target can also be
by healthcare professionals or teams for, with or on be- another Actor in the system.
half of their patients, specification of the Actor is par- AACTT provides common elements that can be used
ticularly helpful. Indeed, healthcare delivery behaviours for consistent description and specification of behaviour.
have been described as ‘collective behaviours’, suggesting By extension, AACTT can be used to describe the
that role confusion may be a barrier to performance [14] sequence of multiple behaviours of multiple Actors at
that could be illuminated by careful specification of the different levels of the organisation required to enact
behaviour at the outset using AACTT. change. For instance, in the case of promoting hand
Presseau et al. Implementation Science (2019) 14:102 Page 4 of 13

Table 1 AACTT framework definitions and examples

AACTT Definition1 Examples
domains
Action A discrete observable behaviour Prescribing antihypertensives, providing a referral to a specialist,
washing hands, setting a policy
Actor The individual or group of individuals who perform (or Primary care physician, pharmacist, social worker, resident,
should/could) the Action administrator, middle manager, head of unit, policymaker
Context The physical, emotional or social setting in which the Examination room, doctor’s office, outside a patient room, in a
Actor performs (or should/could) the Action boardroom, stressful vs. calm situation, when patients’ relatives
are present or not
Target The individual or group of individuals for/with/on behalf of whom Patient with diabetes and blood pressure above 140/80 mmHg,
the Actor performs the Action patient wanting to quit smoking
Time The time period and duration that the Actor performs the At annual review, next time a patient visits, every week, over the
Action in the Context with/for the Target next 6 months
1
from Francis et al (2004)

hygiene in hospital, AACTT provides a means for clari- overcome the modifiable barriers and enhance the enablers)
fying the behaviour of those engaging in hand sanitizing and Step 4 (how can behaviour change be measured and
behaviour, but also the leadership in the organisation understood). We selected French’s process model to guide
whose policy-enacting behaviour sets the stage for mid- the demonstration of AACTT’s application given that its
dle management to engage in procurement behaviour to foundational step involves clarifying the behaviour.
provide hand sanitizing stations and gels, through to However, AACTT aligns with other process models (e.g.
maintenance staff engaging in refilling behaviour to Knowledge-to-Action Framework [21], the UK MRC
ensure sanitizing gel is available for the healthcare framework for developing and evaluating complex interven-
providers. Each behaviour by these organisational Actors tions [22] and process evaluations [23]). At each step, we
is required for healthcare providers to engage in hand- identified how the AACTT framework could be applied
sanitizing activities. Rather than making implicit as- and provide published examples across a range of
sumptions about such a sequence of behaviours or methodological approaches which align with those
describing them as separate organisational factors, the commonly used in implementation research [24].
AACTT framework helps to unpack the complexity and
clarify the responsibility of all behaviours in organisa- Results
tional health settings, providing a clear opportunity for Step 1—Using AACTT to identify who needs to do what,
behavioural approaches to inform organisational change. differently, when and where
The level of granularity or aggregation in the specifica- The AACTT framework can be used at this foundational
tion of each AACTT domain should be defined by what step of implementation research to ensure coverage across
is measurable, useful and practical for the given applica- key behaviours (Actions), stakeholders (Actors, Targets) in
tion, to ensure practical utility. the particular settings (Time, Context) of desired perform-
In summary, we propose a new framework: AACTT ance. This helps ensure that subsequent steps remain con-
(Action, Actor, Context, Target, Time—see Table 1 for sistent with the AACTT-specified behaviour(s) established
definitions and examples) to allow for the careful delin- at the outset. Application of the AACTT framework at this
eation of ‘who does what; to, for or with whom; when; step can help to identify which individuals at which levels
where?’ [10]. Herein, our aim is to demonstrate how the of an organisational hierarchy need to do something differ-
AACTT framework can be used within the main steps ently to implement an evidence-based practice (or indeed
of an implementation research process and to provide a de-implement an outdated or non-evidenced practice).
simple tool that implementation researchers and practi- Actors often include healthcare professionals, but may also
tioners can use to apply the AACTT framework to involve their colleagues within a team, administrative staff
specify the behaviour(s) of stakeholders. and middle and upper management whose own Actions fa-
cilitate the clinical behaviour of the health professional, and
Methods ultimately patients and citizens behaviour. One can extend
We used French’s four-step implementation process model Actors to people at multiple levels in a healthcare system,
[20] to exemplify how AACTT can be applied to each of e.g. organisational leadership, policy makers and political
the key steps of the implementation process: Step 1 (clarify- Actors, whose own behaviours may be centrally important
ing who needs to do what differently), Step 2 (using a the- to ensuring that the health professional can engage in a
oretical framework, which barriers and enablers need to be given guideline-recommended behaviour under investiga-
addressed), Step 3 (which intervention components could tion. The AACTT framework can be employed to unpack
Presseau et al. Implementation Science (2019) 14:102 Page 5 of 13

Fig. 1 AACTT specification tool for a single Action, with worked example applied to improving hand hygiene

complex organisational and hierarchical inter-relationships In specifying behaviour using the AACTT framework, it
into the multiple observable, measurable (and changeable) may become clearer where specifically evidence-practice
Actions of individuals at each level. This can provide the gaps exist (i.e. specifically whose behaviour in what
basis for clarifying whose behaviour to focus on in a given settings and time) rather than broader generalisations of
investigation. Figures 1, 2 and 3 provide generalisable work- evidence of gaps in care typically available to justify a
sheets with a worked example that can be used by imple- focused implementation diagnostic effort.
mentation researchers and practitioners, on their own or as
part of a stakeholder engagement activity, to clarify who Step 2—Using the AACTT framework to inform
needs to do what, differently (see also Additional file 1 for investigations to identify barriers and enablers
blank versions of each worksheet). Having an AACTT-specified behaviour(s) can inform
The implication of identifying who needs to do what more focused investigations into the barriers and drivers
differently also suggests a need to clarify a gap in care. of Actions for each Actor. A key benefit to specifying
Presseau et al. Implementation Science (2019) 14:102 Page 6 of 13

Fig. 2 AACTT specification for focal and ancillary Actions of multiple Actors, Contexts and Times, with worked example applied to improving
hand hygiene

the Actor using this framework is that it helps to iden- specification for each Actor to promote role clarity) or
tify which specific agents one should engage with in acceptable/desirable (in which case a given setting may
eliciting their barriers and enablers to performing the find it useful to allocate clear responsibility).
Action; focusing consideration down specifically on
those who are or could perform the Action. This may Interviews/focus groups An AACTT-specified behav-
help in qualitative and quantitative approaches to in- iour provides greater focus to an interview or focus
vestigating barriers and enablers, including interviews/ group to maximise the likelihood that responses are
focus groups to generate qualitative data and reflective of the specific Actions of the Actors situated in
questionnaire-based quantitative data operationalizing a time and place, which may also help with recall. Speci-
theoretical constructs hypothesised to correlate with fying the Actor up front helps to inform recruitment so
implementation behaviour. At the initial stages of that the respondents are those actually tasked with the
inquiry, it may not yet be clear who all these potential Action (as opposed to others speaking on their behalf).
Actors and Actions are but AACTT specification al- Topic guides can then be designed to ensure that inter-
lows this to be further clarified and be made explicit. view and focus group prompts are consistently focused
Initial broad specification at a higher level (e.g. primary on the AACTT-specified behaviours of the respondents
care staff including nurses and physicians) may help to (Actors). For instance, a broadly specified behaviour
recognise the potential for multiple Actors to under- such as ‘improving blood pressure prescribing for pa-
take the same Action, and to explore during the barrier tients with type 2 diabetes’ lacks AACTT-specificity and
elicitation whether this diffusion of responsibility is thus interview/focus group questions aiming to unpack
problematic (and thus may benefit from greater views on readiness to change [25], beliefs about
Presseau et al. Implementation Science (2019) 14:102 Page 7 of 13

Fig. 3 AACTT specification tool for team-based Actions and Actors with variable Target and consistent Context and Time, with worked examples
applied to diabetes care

consequences and social influences [26, 27] or imple- This is particularly aided when using theoretical
mentation climate and culture [28] may remain at a level frameworks to inform interview topic guide develop-
that does not clarify how those views translate into un- ment, such as when using the Theoretical Domains
derstanding who needs to do what differently and why Framework [26, 27], COM-B within the Behaviour
they do or do not. In contrast, while ‘increasing the dose Change Wheel [29] or Consolidated Framework for
of existing antihypertensive medication (Action) by a Implementation Research [28, 30, 31] to elicit views
family physician (Actor) during annual diabetes review about barriers and enablers that are rooted in a given
(Context) for their patients with type 2 diabetes (Target) domain or construct. An AACTT-specified behaviour
when their blood pressure is above 140/80 mmHg can also help a directed content analysis of theory-
despite previous management (Time)’ provides an informed interviews by providing the capacity to code
admittedly longer description, the specificity allows barriers and enablers described by respondents directly
much greater insight into the factors that may determine linked to a given AACTT-specified behaviour, to ensure
this behaviour than might otherwise be missed by a that the barriers identified are in fact related to the
broader specification. Describing the AACTT upfront Actions and Actors of interest. Importantly, AACTT-
also allows similar topic guides to be tailored to different specification ensures that barriers and enablers are spe-
Actors who may be engaging in the same or different cifically those linked to a particular Actor and behaviour
Actions. in the time and context of performance, rather than
Presseau et al. Implementation Science (2019) 14:102 Page 8 of 13

vague and broad barriers/enablers. This facilitates more Step 3—Using AACTT to inform selecting and
targeted intervention development in Step 3. operationalizing implementation intervention
components to address barriers and enablers
Questionnaires The original TACT framework was Best practice in implementation intervention develop-
established to inform the careful development of ques- ment involves selecting evidence-based techniques and
tionnaire items to operationalise constructs from the strategies that address the barriers and enablers identi-
Theory of Reasoned Action, and then the Theory of fied [39], and clarifying who delivers and receives the
Planned Behaviour that were used to predict behaviour intervention, how, when, where and how much [40].
using cross-sectional and prospective designs. This ap- Implementation intervention developers have a range of
proach, known as the principle of compatibility, involves tools at their disposal for selecting and specifying inter-
developing questionnaire items tailored to the specific ventions: high-level descriptors of implementation stra-
Target, Action, Context and Time to ensure consistency tegies such as those described in Cochrane’s Effective
between the items measuring theoretical constructs and Practice and Organisation of Care (EPOC) [41] and Ex-
the subsequent behaviour being predicted while ensuring pert Recommendation for Implementing Change (ERIC)
rigour in operationalisation of the theoretical constructs. [42] taxonomies and the Behaviour Change Wheel [29],
The TACT-informed principle of compatibility was as well as fine-grained descriptors of techniques such as
widely adopted in the 1980s and 1990s when testing of those proposed in Behaviour Change Techniques Tax-
the Theory of Reasoned Action and Theory of Planned onomy [43], Intervention Mapping [44] or emerging
Behaviour grew in popularity in social and health psy- tools for linking behaviour change techniques to mecha-
chology. At the turn of the century, these social cogni- nisms of action [45]. Regardless of the strategies and
tion models began to be adapted and adopted for use in techniques, it can help to know specifically whose
implementation research to predict the behaviour of (Actor) behaviour (Action) is being targeted for change
healthcare professionals. Francis and colleagues devel- with the strategies/techniques, and where (Context) and
oped specific guidance for developing Theory of Planned when (Time) the behaviour is expected to be performed,
Behaviour-based questionnaires for use to understand for/with whom (Target). This may help in narrowing
and predict health care professional behaviour [32]. This and prioritizing amongst potential strategies and tech-
highly cited and influential guidance provided a descrip- niques. By having AACTT-specified behaviours defined,
tion of how the TACT framework could be applied in many of these decisions are made clearer and help to
this setting and influenced subsequent rigorous develop- operationalise the strategies themselves. For instance,
ment of Theory of Planned Behaviour questionnaires to clearly articulated AACTT-specified behaviour can (i)
understand how its constructs explain variability in the inform selection of performance indicators provided to
behaviour of health care professionals (see Godin et al primary care physicians as part of an Audit and Feed-
2008 for a review [33]). Such methods were also then back (A&F) intervention, (ii) ensure that the A&F inter-
adopted to operationalise other theories, while to date this vention is directed to the correct Actors and (iii) that it
has largely been confined to theories of behaviour [34], reflects the context and times of performance and (iv)
there is an opportunity to consider how AACTT- that the data reflect the Target patients.
specification can be integrated into the operationalisation A multicomponent intervention involving multiple
of questionnaire items for other theories, models and behaviours and Actors can also be aided by AACTT spe-
frameworks used in implementation science and in particu- cification. For instance in a hospital setting, an interven-
lar those for which there is concerted effort to developed tion to improve hand hygiene may involve the focal
measures such as Consolidated Framework for Implemen- behaviour of sanitizing hands using alcohol-based gel
tation Research [35], Organizational Readiness for Change (Action) performed by nursing staff, surgeons, residents
[36] and Normalization Process Theory [37]. Huijg and col- (Actors), as well as ancillary behaviours such as purchas-
leagues ([38], see their Table 5) developed and established ing hand sanitiser (by administrators) and checking/
the discriminant content validity of questionnaire items refilling dispensers (by cleaning staff) that are necessary
designed to assess domains from the Theoretical Domains in supporting the focal behaviour. Other settings, such
Framework in such a way to allow any Action, Actor, Con- as primary care, might similarly involve a focal Action of
text, Target, Time to be integrated into the measure, and initiating a new medication or deprescribing another as
Eccles and colleagues operationalized questionnaires con- well as ancillary Actions, each with respective Actors
sistent with AACTT for a range of theories of behaviour (e.g. family doctors, nurse practitioners, nurses, adminis-
([34] see the Additional file 2 in Huijg et al. for examples of trative staff, practice managers). Community settings
the questionnaire items that they developed). Opportunities could also involve other professional groups (e.g. social
present themselves for similar adaptations for other models, workers), parents/carers and teachers. For example, re-
theories and frameworks. garding implementing dietary menu guidelines in day
Presseau et al. Implementation Science (2019) 14:102 Page 9 of 13

care (nursery) services [46], the action could be ‘prepar- of the behaviour of interest as well as ancillary be-
ing food consistent with childhood dietary guidelines’, haviours that necessarily support the behaviour of interest
the actor could be ‘cooks’, the context ‘in the day care but in themselves are not sufficient.
centre kitchen’, the target ‘children attending day care’
and the time ‘at lunch every week day’. Regardless of the Intervention process evaluation In addition to evaluat-
apparent complexity of the implementation intervention ing whether an implementation intervention is effective,
being developed, the AACTT framework can provide it is important to understand the mechanisms through
transparency and clarity in terms of whose behaviour is which this effect occurred both in terms of changing the
targeted by the strategies in the intervention. targeted mediating constructs (mechanisms of change)
and assessing delivery and receipt as designed (fidelity).
Step 4—Using AACTT to specify how behaviour change AACTT-specified behaviours provide the same advan-
can be measured and understood tage as in Step 2 for informing the design of qualitative
The application of the AACTT framework in previous or quantitative assessment of mechanisms of change
steps ensures compatibility with the measures used to alongside outcome assessment, such as in theory-based
assess change in Step 4. Implementation researchers are process evaluations [47–49]. Using AACTT to inform
often interested in both whether an intervention changes the wording of process measures that operationalise the
behaviour (outcome) as well as explaining the (theory- theoretical constructs targeted for change provides mea-
informed) mechanisms through which change occurs. sures of mediators tied directly to the behaviour the
intervention is targeting for change. This provides more
Intervention outcome evaluation Irrespective of the direct correspondence between mechanism and out-
design used to evaluate an implementation intervention, comes. In this instance, AACTT-specification provides
a measure or indicator of behaviour is often central to greater measurement sensitivity.
demonstrating change. In some instances, outcome AACTT-specification also allows for more careful as-
measurement is embedded within available routinely col- sessment of fidelity (and adaptation) of delivery and re-
lected data, such as prescribing, ordering and referring, ceipt of an implementation intervention by providing
or specific process data collected locally or as part of lar- structure and transparency in terms of who should re-
ger scale initiatives (e.g. Quality and Outcomes Frame- ceive the intervention (Actor) and which Action(s) under
work in the UK) and national audits. When such data which circumstances (Time, Context) for which Targets.
are accessible, they provide a pragmatic, low-burden This specificity and transparency can clarify what and
means of evaluating implementation interventions. How- whom to track to assess fidelity and adaptation. Table 2
ever, this often involves having to balance consideration provides worked examples of AACTT specifications
of pragmatism with those of measurement specificity. across study designs.
While the likelihood is low that routinely collected data
perfectly correspond with the Action performed by the Discussion
targeted Actor in the Context and Time for the Target Herein, we introduced the AACTT framework that can
designed for the intervention, having an AACTT- be used to inform the careful specification of behaviour
specification allows clarity of the degree (or not) of this for implementation research and practice and provided a
correspondence. While it is true that in randomised de- generalisable worksheet to facilitate use of the frame-
signs, any additional ‘noise’ introduced by pragmatic work (worked examples in Figs. 1, 2 and 3, blank work-
outcome measures would at least be balanced by the sheets in Additional file 1). AACTT can be applied
randomisation, the more ‘noise’ (i.e. error variance) the across key steps of implementation research and practice
less power to detect change in the actual outcome of advocated by process models [20–22] to transparently
interest, which has implication for sample size calcula- define and measure behaviour(s) in terms of who per-
tions. Thus, even in randomised designs, AACTT- forms them, for/with whom, when and where. AACTT
specification provides an opportunity to establish the formalises a natural progression of the TACT framework
degree of correspondence between the targeted behav- developed by Fishbein and popularised in social and
iour(s) and the indicators of behaviour available to assess health psychology for more direct application to behav-
the degree of noise (error) in the outcome. iour where the individual is performing a behaviour for
In evaluations where no routinely collected data are someone else’s (i.e. Target) benefit. This has direct appli-
available, outcome measures are sometimes developed cation not only in implementation science applied to
for the intervention evaluation itself. In such instances, healthcare settings but also public health–, social
the added advantage of AACTT specification is that it welfare– and family-based settings in which someone
can directly guide which data to collect and provides full performs an Action for someone else (e.g. school- and
control over what ‘counts’ as performance both in terms family-focused interventions) [46, 50].
 Presseau et al. Implementation Science

Table 2 Worked examples of AACTT-specified behaviours across study designs

Study Interviews Questionnaire Intervention Cluster randomised Fidelity assessment Mechanism of change (process
design (qualitative) [55] (quantitative) [34] development [56] trial [57] (process evaluation) [58] evaluation) [59]
Action Managing back pain Prescribing additional Provision of sexual Examining feet yourself Providing behavioural support for Advising patient to make an
without X-ray antihypertensive drugs counselling group and/or referring for foot smoking cessation (detailed by appointment for retinal screening
(2019) 14:102

sessions exam component behaviour change within the next 12 months

techniques)
Actor Chiropractors General practitioners Cardiac rehabilitation General practitioner and Trained specialist stop smoking Family doctors
healthcare staff nurse advisor
Context Private clinics Practice clinic Hospitals in the Practice clinic English stop-smoking service clinic Examination room in family practice
(Canada, USA with Republic of Ireland in East London and North England
an HMO)
Target Patient with acute Patients with type 2 diabetes Patients aged 18+ Patients with type 2 Smokers trying to quit Specific patient scenario: 57-year-old
low back pain whose blood pressure (BP) is with cardiovascular diabetes woman with type 2 diabetes on metformin
5 mmHg above a target of disease 500 mg, non-smoker, no other medication,
140 mmHg systolic BP or A1C < 7%, BMI 25, BP 125/75, normal foot
80 mmHg diastolic BP even exam, attended retinal screening over
following previous management 12 months ago
Time During patient visit Over the next 12 months During Phase III In the last 12 months Over four weekly sessions During annual diabetes checkup
cardiac rehabilitation
Page 10 of 13
Presseau et al. Implementation Science (2019) 14:102 Page 11 of 13

Using the Theory Comparison and Selection Tool (T- (dose), implementation outcomes affected and justifica-
CaST) and checklist [51], the AACTT framework is de- tion. While some of the Proctor et al. domains share
signed to be usable (includes relevant domains, has been similar or overlapping nomenclature with AACTT (i.e.
developed so that key stakeholders can use it, we provide Actor, Action, Time), the scope of intended application
steps for its application and methods for promoting its of each framework and its domains differs. As exempli-
application across a range of possible studies and an ex- fied in Table 1, in the AACTT framework ‘Actor’ refers
planation for how the domains influence each other), to the individuals or groups performing the behaviour
testable/valid (can form the basis for testable hypoth- that could be the recipient of an implementation strat-
eses, includes face-valid explanations and has been used egy, the ‘Action’ refers specifically to the behaviour being
in empirical studies), applicable (focuses on a key imple- performed by the Actor and is the object of change and
mentation outcome, can be applied across a range of ‘Time’ refers to when that behaviour is performed. Thus,
methods and across a range of analytical levels, popula- whereas the seven dimensions proposed by Proctor et al.
tions and conditions and is generalisable across disci- focus on detailing the delivery of an implementation strat-
plines) and is likely to be acceptable (to key stakeholders, egy within an implementation intervention, AACTT do-
and is the historical evolution of a framework rooted in a mains focus uniquely on detailing behaviour(s). Thus, in
particular discipline). Thus, in principle it fulfils all the the context of an implementation intervention that would
criteria for use of a framework by implementation science be specified with Proctor et al.’s dimensions, the behaviour
researchers and practitioners, though its actual usability, specified using AACTT would be an outcome that the
testability, applicability and acceptability will ultimately be (now well-specified) implementation strategy aims to
determined through application of the tool across a range change. Thus, AACTT is designed for specifying be-
of types of implementation research [52]. haviour as the object of change, rather than specifying
Michie and Johnston made a call for making clinical implementation strategies designed to bring about
practice guidelines more specific by specifying ‘what’, that change.
‘who’, ‘when’, ‘where’ and ‘how’ [10]. While sharing some
similarities with AACTT, there are three important
differences that distinguish AACTT and underscore its Conclusion
potential added utility. First, the ‘who’ in Michie and The careful description of outcomes and measures is of
Johnston’s recommendations refers to the ‘Actor’ but central importance to the development of a cumulative
does not make any mention of ‘to, for or with whom’ and generalisable science. The AACTT framework pro-
the action is performed (i.e. the ‘Target’ in AACTT, vides a means for establishing the core elements of
which may be a patient, a healthcare team member or behaviour targeted for change. The framework can be
other organisational actor). Second, the ‘where’ is spe- applied to describe the behaviour of Actors at multiple
cific to a physical location, whereas ‘Context’ in AACTT organisational levels. The framework facilitates the sys-
can refer to a broader set of contexts that include the tematic description of who needs to do what differently,
physical location but could also include a context that is understanding what may stop or help them from doing
internal to the actor (e.g. emotional context of a stressful so, how to support them to address barriers to change
versus a calm situation) or the social context (e.g. when and how to demonstrate that such support worked.
patients’ relatives are present). Third, in proposing AACTT is compatible with any theory, model or frame-
AACTT as an extension to Fishbein and Ajzen’s TACT work in which behaviour is the focus of inquiry and thus
framework, we are deliberate in ensuring cumulative has the potential to be integrated alongside key theoretical
theory and methods development as it extends to approaches used in the field. Given on-going efforts to
applications within implementation science. systematise the measurement of theoretical constructs
Existing calls for better specification and reporting [54], and description of intervention components [41–43],
have focused on the description and labelling of imple- there is a need for ensuring that the object of change, i.e.
mentation intervention strategies (e.g. with taxonomies behaviour, is equally as carefully described and under-
of change strategies such as the ERIC [42] and BCT [43] stood. This will ultimately contribute to a greater capacity
taxonomies) and on the wider components of interven- to synthesise research over time.
tions (e.g. with checklists such as TIDIER [40]). Within
implementation science, Proctor, Powell and McMillen
Supplementary information
[53] proposed seven domains for specifying implementa- Supplementary information accompanies this paper at https://doi.org/10.
tion strategies per se, including who delivers the strategy 1186/s13012-019-0951-x.
(actors), how they deliver the strategy (actions), what
and to whom the strategy is directed (action target), the Additional file 1. Blank worksheets for single-actor and multiple-actor/
action AACTT specification.
sequence of strategy delivery (temporality), intensity
Presseau et al. Implementation Science (2019) 14:102 Page 12 of 13

Abbreviations 10. Michie S, Johnston M. Changing clinical behaviour by making guidelines

AACTT: Action, Actor, Context, Target, Time; A&F: Audit and Feedback; COM- specific. BMJ. 2004;328:343–5.
B: Capability, opportunity, motivation, behaviour; EPOC: Effective Practice and 11. Miller M. How well is the target behaviour specified in reports of behaviour
Organisation of Care; ERIC: Expert Recommendation for Implementing Change; change interventions for health care professionals? A systematic review of
TACT: Target, Action, Context, Time; T-CaST: Theory Comparison and Selection Tool reporting practice [Master’s thesis]: City University of London; 2015.
12. Ajzen I, Fishbein M. Attitude-behavior relations: a theoretical analysis and
Acknowledgements review of empirical research. Psychol Bull. 1977;84:888–918.
We thank Ian D. Graham for thought provoking discussions on the order in 13. Siegel JT, Navarro MA, Tan CN, Hyde MK. Attitude–behavior consistency, the
which to present the framework’s domains. principle of compatibility, and organ donation: a classic innovation. Health
Psychol. 2014;33:1084–91.
Authors’ contributions 14. Eccles MP, Hrisos S, Francis JJ, Steen N, Bosch M, Johnston M. Can the
JJF conceived the idea and all authors substantively developed the idea. JP collective intentions of individual professionals within healthcare teams
and JJF drafted the manuscript. All authors contributed to refining the predict the team’s performance: developing methods and theory.
manuscript. All authors read and approved the final manuscript. Implement Sci. 2009;4:24.
15. Strack F, Deutsch R. Reflective and impulsive determinants of social
behavior. Personal Soc Psychol Rev. 2004;8:220–47.
Funding
16. Gardner B, Lally P, Wardle J. Making health habitual: the psychology of
JMG is supported by a Tier 1 Canada Research Chair in Health Knowledge
‘habit-formation’ and general practice. Br J Gen Pract. 2012;62:664–6.
Transfer and Uptake and a CIHR Foundation grant for related work.
17. Potthoff S, Rasul O, Sniehotta FF, Marques M, Beyer F, Thomson R, et al. The
relationship between habit and healthcare professional behaviour in clinical
Availability of data and materials practice: a systematic review and meta-analysis. Health Psychol Rev.
Not applicable. 2019;13:73–90.
18. Potthoff S, Presseau J, Sniehotta FF, Breckons M, Rylance A, Avery L.
Ethics approval and consent to participate Exploring the role of competing demands and routines during the
Not applicable. implementation of a self-management tool for type 2 diabetes: a theory-
based qualitative interview study. BMC Med Inform Decis Mak. 2019;19:23.
Consent for publication 19. Presseau J, Johnston M, Heponiemi T, Elovainio M, Francis JJ, Eccles MP,
Not applicable. et al. Reflective and automatic processes in health care professional
behaviour: a dual process model tested across multiple behaviours. Ann
Competing interests Behav Med. 2014;48:347–58.
JP is an associate editor and JMG and FL are on the editorial board for 20. French SD, Green SE, O’Connor DA, McKenzie JE, Francis JJ, Michie S, et al.
Implementation Science but were not involved in the management or Developing theory-informed behaviour change interventions to implement
decisions made for this manuscript. evidence into practice: a systematic approach using the Theoretical
Domains Framework. Implement Sci. 2012;7:38.
Author details 21. Graham ID, Logan J, Harrison MB, Straus SE, Tetroe J, Caswell W, et al. Lost
1 in knowledge translation: Time for a map? J Contin Educ Health Prof.
Clinical Epidemiology, Ottawa Hospital Research Institute, Ottawa, Canada.
2
School of Epidemiology and Public Health, University of Ottawa, Ottawa, 2006;26:13–24.
Canada. 3School of Psychology, University of Ottawa, Ottawa, Canada. 22. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M.
4 Developing and evaluating complex interventions: the new Medical
Centre for Behaviour Change, University College London, London, UK.
5
Department of Medicine, University of Ottawa, Ottawa, Canada. 6School of Research Council guidance. BMJ. 2008;337:a1655.
Health Sciences, City University of London, London, UK. 23. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process
evaluation of complex interventions: Medical Research Council guidance.
Received: 25 July 2019 Accepted: 22 October 2019 BMJ. 2015;350:h1258.
24. Brown CH, Curran G, Palinkas LA, Aarons GA, Wells KB, Jones L, et al. An
overview of research and evaluation designs for dissemination and
References implementation. Annu Rev Public Health. 2017;38:1–22.
1. Barnett ML, Linder JA. Antibiotic prescribing for adults with acute bronchitis 25. Weiner BJ. A theory of organizational readiness for change. Implement Sci
in the United States, 1996-2010. JAMA. 2014;311:2020. [Internet]. 2009;4:67.
2. Erasmus V, Daha TJ, Brug H, Richardus JH, Behrendt MD, Vos MC, et al. 26. Michie S, Johnston M, Abraham C, Lawton R, Parker D, Walker A, et al.
Systematic review of studies on compliance with hand hygiene guidelines Making psychological theory useful for implementing evidence based
in hospital care. Infect Control Hosp Epidemiol. 2010;31:283–94. practice: a consensus approach. Qual Saf Health Care. 2005;14:26–33.
3. Gonzalez-Urzelai V, Palacio-Elua L, Lopez-de-Munain J. Routine primary care 27. Atkins L, Francis JJ, Islam R, O’Connor D, Patey A, Ivers NM, et al. A guide to
management of acute low back pain: adherence to clinical guidelines. Eur using the Theoretical Domains Framework of behaviour change to
Spine J. 2003;12:589–94. investigate implementation problems. Implement Sci. 2017;12:77.
4. Institute of Medicine. Committee on Quality of Health Care in America. 28. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC.
Crossing the quality chasm: a new health system for the 21st century. Fostering implementation of health services research findings into practice:
Washington DC: National Academies Press; 2001. a consolidated framework for advancing implementation science.
5. Woolf SH. Unhealthy Medicine: All breakthrough, no follow-throw. Implement Sci. 2009;4:50.
Washington Post [Internet]. 2006; Available from: https://www. 29. Michie S, van Stralen MM, West R. The behaviour change wheel: a new
washingtonpost.com/archive/opinions/2006/01/08/unhealthy-medicine- method for characterising and designing behaviour change interventions.
span-classbankheadall-breakthrough-no-follow-throughspan/d3b838a9- Implement Sci. 2011;6:42.
db0b-4b23-a410-c5de185f51d0/ 30. Birken SA, Powell BJ, Presseau J, Kirk MA, Lorencatto F, Gould NJ, et al.
6. LaPiere RT. Attitudes vs. actions. Soc Forces. 1934;13:230–7. Combined use of the Consolidated Framework for Implementation
7. Fishbein M, editor. Attitude and the prediction of behaviour (Chapter 51). In: Research (CFIR) and the Theoretical Domains Framework (TDF): a systematic
Readings in attitude theory and measurement. New York: Wiley; 1967. review. Implement Sci. 2017;12:2.
p. 477–492. 31. Kirk MA, Kelley C, Yankey N, Birken SA, Abadie B, Damschroder L. A
8. Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process. systematic review of the use of the Consolidated Framework for
1991;50:179–211. Implementation Research. Implement Sci. 2015;11:72.
9. Fishbein M, Ajzen I. Defining and predicting behaviour (Chapter 2). In: 32. Francis JJ, Eccles MP, Johnston M, Walker A, Grimshaw J, Foy R, et al.
Predicting and changing behavior: the reasoned action approach. New Constructing questionnaires based on the theory of planned behaviour: a
York: Psychology Press; 2010. p. 29–73. manual for health services researchers]. Centre for Health Services Research:
Presseau et al. Implementation Science (2019) 14:102 Page 13 of 13

University of Newcastle upon Tyne; 2004. Available from: http://openaccess. 52. Hull L, Goulding L, Khadjesari Z, Davis R, Healey A, Bakolis I, et al. Designing
city.ac.uk/1735/ high-quality implementation research: development, application, feasibility
33. Godin G, Bélanger-Gravel A, Eccles MP, Grimshaw J. Healthcare and preliminary evaluation of the implementation science research
professionals’ intentions and behaviours: a systematic review of studies development (ImpRes) tool and guide. Implement Sci. 2019;14:80.
based on social cognitive theories. Implement Sci. 2008;3:36. 53. Proctor EK, Powell BJ, McMillen JC. Implementation strategies:
34. Eccles MP, Hrisos S, Francis JJ, Stamp E, Johnston M, Hawthorne G, et al. recommendations for specifying and reporting. Implement Sci. 2013;8:139.
Instrument development, data collection, and characteristics of practices, 54. Lewis CC, Mettert KD, Dorsey CN, Martinez RG, Weiner BJ, Nolen E, et al. An
staff, and measures in the Improving Quality of Care in Diabetes (iQuaD) updated protocol for a systematic review of implementation-related
Study. Implement Sci. 2011;6:61. measures. Syst Rev. 2018;7:66.
35. Kegler MC, Liang S, Weiner BJ, Tu SP, Friedman DB, Glenn BA, et al. 55. Bussières AE, Patey AM, Francis JJ, Sales AE, Grimshaw JM. Identifying factors
Measuring constructs of the consolidated framework for implementation likely to influence compliance with diagnostic imaging guideline
research in the context of increasing colorectal cancer screening in federally recommendations for spine disorders among chiropractors in North
qualified health center. Health Serv Res. 2018;53:4178–203. America: a focus group study using the Theoretical Domains Framework.
36. Helfrich CD, Li Y-F, Sharp ND, Sales AE. Organizational readiness to change Implement Sci. 2012;7:82.
assessment (ORCA): development of an instrument based on the Promoting 56. McSharry J, Murphy PJ, Byrne M. Implementing international sexual counselling
Action on Research in Health Services (PARIHS) framework. Implement Sci. guidelines in hospital cardiac rehabilitation: development of the CHARMS
2009;4:38. intervention using the Behaviour Change Wheel. Implement Sci. 2016;11:134.
37. Rapley T, Girling M, Mair FS, Murray E, Treweek S, McColl E, et al. Improving 57. Presseau J, Mackintosh J, Hawthorne G, Francis JJ, Johnston M, Grimshaw JM,
the normalization of complex interventions: part 1 - development of the et al. Cluster randomised controlled trial of a theory-based multiple behaviour
NoMAD instrument for assessing implementation work based on change intervention aimed at healthcare professionals to improve their
normalization process theory (NPT). BMC Med Res Methodol. 2018;18:133. management of type 2 diabetes in primary care. Implement Sci. 2018;13:65.
38. Huijg JM, Gebhardt WA, Crone MR, Dusseldorp E, Presseau J. Discriminant 58. Lorencatto F, West R, Christopherson C, Michie S. Assessing fidelity of
content validity of a theoretical domains framework questionnaire for use in delivery of smoking cessation behavioural support in practice. Implement
implementation research. Implement Sci. 2014;9:11. Sci. 2013;8:40.
39. Michie S, Johnston M, Francis JJ, Hardeman W, Eccles M. From theory to 59. Grimshaw JM, Presseau J, Tetroe J, Eccles MP, Francis JJ, Godin G, et al.
intervention: mapping theoretically derived behavioural determinants to Looking inside the black box: results of a theory-based process evaluation
behaviour change techniques. Appl Psychol. 2008;57:660–80. exploring the results of a randomized controlled trial of printed educational
40. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. messages to increase primary care physicians’ diabetic retinopathy referrals
Better reporting of interventions: template for intervention description and [Trial registration number ISRCTN72772651]. Implement Sci. 2014;9:86.
replication (TIDieR) checklist and guide. BMJ. 2014;348:g1687.
41. Effective Practice and Organisation of Care (EPOC). EPOC Taxonomy. Oslo: Publisher’s Note
Norwegian Knowledge Centre for the Health Services; 2002. Available at: Springer Nature remains neutral with regard to jurisdictional claims in
https://epoc.cochrane.org/epoc-taxonomy. published maps and institutional affiliations.
42. Powell BJ, Waltz TJ, Chinman MJ, Damschroder LJ, Smith JL, Matthieu MM,
et al. A refined compilation of implementation strategies: results from the
Expert Recommendations for Implementing Change (ERIC) project.
Implement Sci. 2015;10:21.
43. Michie S, Richardson M, Johnston M, Abraham C, Francis JJ, Hardeman W,
et al. The behavior change technique taxonomy (v1) of 93 hierarchically
clustered techniques: building an international consensus for the reporting
of behavior change interventions. Ann Behav Med. 2013;46:81–95.
44. Kok G, Gottlieb NH, Peters G-JY, Mullen PD, Parcel GS, Ruiter RAC, et al. A
taxonomy of behaviour change methods: an Intervention Mapping
approach. Health Psychol Rev. 2016;10:297–312.
45. Centre for Behaviour Change. Theory and Techniques Tool. Theory and
Techniques Tool. 2019. Available from: https://theoryandtechniquetool.
humanbehaviourchange.org/
46. Seward K, Wolfenden L, Wiggers J, Finch M, Wyse R, Oldmeadow C, et al.
Measuring implementation behaviour of menu guidelines in the childcare
setting: confirmatory factor analysis of a theoretical domains framework
questionnaire (TDFQ). Int J Behav Nutr Phys Act. 2017;14:45.
47. McIntyre SA, Francis JJ, Gould NJ, Lorencatto F. The use of theory in process
evaluations conducted alongside randomized trials of implementation
interventions: a systematic review. Transl Behav Med. 2018:iby110; Available
from: https://doi.org/10.1093/tbm/iby110/5208274
48. Presseau J, Grimshaw JM, Tetroe JM, Eccles MP, Francis JJ, Godin G, et al. A
theory-based process evaluation alongside a randomised controlled trial of
printed educational messages to increase primary care physicians’ prescription of
thiazide diuretics for hypertension [ISRCTN72772651]. Implement Sci. 2016;11:121.
49. Grimshaw JM, Zwarenstein M, Tetroe JM, Godin G, Graham ID, Lemyre L, et al.
Looking inside the black box: a theory-based process evaluation alongside a
randomised controlled trial of printed educational materials (the Ontario
printed educational message, OPEM) to improve referral and prescribing
practices in primary care in Ontario. Canada. Implement Sci. 2007;2:38.
50. Wolfenden L, Yoong SL, Williams CM, Grimshaw J, Durrheim DN, Gillham K,
et al. Embedding researchers in health service organizations improves
research translation and health service performance: the Australian Hunter
New England Population Health example. J Clin Epidemiol. 2017;85:3–11.
51. Birken SA, Rohweder CL, Powell BJ, Shea CM, Scott J, Leeman J, et al.
T-CaST: an implementation theory comparison and selection tool.
Implement Sci. 2018;13:143.