Sage Journals: Discover world-class research

Abstract

Objective

Digital health interventions for behaviour change are usually complex interventions, and intervention developers should ‘articulate programme theory’, that is, they should offer detailed descriptions of individual intervention components and their proposed mechanisms of action. However, such detailed descriptions often remain lacking. The objective of this work was to provide a conceptual case study with an applied example of ‘articulating programme theory’ for a newly developed digital health intervention.

Methods

Intervention Mapping methodology was applied to arrive at a detailed description of programme theory for a newly developed digital health intervention that aims to support cardiac rehabilitation patients in establishing heart-healthy physical activity habits. Based on a Predisposing, Reinforcing, and Enabling Constructs in Educational Diagnosis and Evaluation (PRECEDE) logic model of the problem, a logic model of change was developed. The proposed mechanisms of action were visualised in an acyclic behaviour change diagram.

Results

Programme theory for this digital health intervention includes 4 sub-behaviours of the main target behaviour (i.e. habitual heart-healthy physical activity), 8 personal determinants and 12 change objectives (i.e. changes needed at the determinant level to achieve the sub-behaviours). These are linked to 12 distinct features of the digital health intervention and 12 underlying behaviour change methods.

Conclusions

This case study offers a worked example of articulating programme theory for a digital health intervention using Intervention Mapping. Intervention developers and researchers may draw on this example to replicate the method, or to reflect on most suitable approaches for their own behaviour change interventions.

Keywords

Behavior and behavior mechanisms Cardiac rehabilitation Cardiovascular diseases Digital health Health promotion Public health Secondary prevention Telemedicine

Introduction

Ischaemic heart disease is a major cause of mortality and morbidity, accounting for 16% of global deaths^1,2 and representing the most frequent cause for disability-adjusted life years in age groups 50 years and older.³ Regular heart-healthy physical activity (PA) is a crucial behaviour for the primary and secondary prevention of ischaemic heart disease and other cardiovascular diseases (CVDs). For example, a recent study demonstrated that every increase in PA by 500 metabolic equivalent of task minutes per week results in a 14% and 7% risk reduction in mortality in the secondary and primary prevention of CVD, respectively.⁴ Therefore, exercise-based cardiac rehabilitation is a central component of secondary CVD prevention. Exercise-based cardiac rehabilitation is a multi-disciplinary programme targeting multi-factorial cardiovascular risk factors with a specific focus on physical inactivity.⁵ As a general recommendation, patients with CVD should carry out aerobic PA on at least 3 days per week at moderate to high intensity, and muscle strength training twice per week. PA should provide an energy expenditure of 1000 to 2000kcal/week.⁶

Importantly, patients with CVD should establish ongoing (and life-long) PA habits, to reduce their risk of recurring cardiovascular events and death, and to improve their physical functioning and quality of life. However, this remains a major challenge. In the 2017 EUROASPIRE V survey of 8261 patients with CVD who were interviewed at median 1.1 years following an acute coronary event, only 34% of patients reported performing regular PA (i.e. ≥ 30 min on average 5 times a week); only 35% reported performing planned PA to increase physical fitness; 42% of patients did not perform regular PA and had no intention to do so in the next 6 months; and 46% of patients did not recall having received personal advice on PA.⁷ Moreover, a systematic review of 26 randomised controlled trials (8593 patients) showed that completion of a cardiac rehabilitation programme did not result in a clear positive impact on sustained heart-healthy PA behaviour compared with patients who did not participate in such a programme.⁸ Thus, there is an urgent need to develop effective support for CVD patients’ behaviour change and habit formation for regular heart-healthy PA.

Digital health interventions have the potential to offer suitable solutions that contribute to addressing this problem. For example, in a recent systematic review of 14 randomised controlled trials (1497 patients), digital health interventions showed positive effects for promoting PA following completion of a cardiac rehabilitation programme (pooled effect size assessed as standardised mean difference (SMD) = 0.35; 95% confidence interval (CI) 0.02–0.70; p = 0.04).⁹ Moreover, several systematic reviews support patient safety of digital health interventions in cardiac rehabilitation, for example the studies by Wongvibulsin et al.¹⁰and Antoniou et al.¹¹ However, it is notable that digital health interventions for CVD secondary prevention are typically very heterogeneous – including building blocks such as websites, text messages, wearable sensing devices, smartphone apps and virtual reality programmes – and little information is available on their underlying programme theory; that is, what intervention components are expected to lead to which changes.⁹ This runs contrary to methodological recommendations and reporting guidelines for behaviour change interventions, such as the United Kingdom Medical Research Council's framework for developing and evaluating complex interventions,¹² the Template for Intervention Description and Replication (TIDieR) checklist,¹³ the Workgroup for Intervention Development and Evaluation Research (WIDER)¹⁴ and others.^15,16

Digital health interventions that aim to support behaviour change are often complex interventions, in the sense that they may combine a number of components and target a number of behaviours.¹² Developers of complex interventions should ‘articulate programme theory’¹⁷; that is, they should describe all individual intervention components (‘active ingredients’) and their proposed mechanisms of action, and they should make transparent the rationale(s) for intervention components and proposed mechanisms of action, including underpinning theories and/or empirical evidence. Such an explicit description of the complex intervention and its (proposed) workings can be helpful to communicate the importance and function of intervention components to others,¹³ and to select appropriate evaluation designs that capture the relevant processes and outcomes.¹² At a wider scientific and societal level, detailed intervention descriptions support transparency, reproducibility and the reduction of waste in research.^18,19 However, despite available guidance, the articulation of programme theory for complex behaviour change interventions remains lacking.^20,21 It may be presumed that the reasons for this include lack of time and resources, lack of awareness, and lack of methodological knowledge, but there is little data to explain why this problem persists. However, one contributing factor may be the limited availability of applied (‘worked’) examples that demonstrate how this can be realised in practice. Such applied examples can arguably raise awareness, offer methodological guidance, and possibly reduce the cost and effort required for intervention descriptions, by providing practical examples to relate to.

In this article, we therefore describe a conceptual case study with an applied example of ‘articulating programme theory’ for our newly developed digital health intervention aktivplan. aktivplan is a web-based application designed to support cardiac rehabilitation patients in establishing regular heart-healthy PA behaviour following completion of a cardiac rehabilitation programme. In our example, we draw on Intervention Mapping (IM), an established approach for the systematic development, evaluation and implementation of behaviour change interventions²² to make transparent the (proposed) active ingredients and mechanisms of action underlying aktivplan. We hope that this conceptual case study and worked example will raise awareness of the importance of methodologies for articulating programme theory, such as IM, and will support developers of complex behaviour change interventions in applying similar approaches to developing and reporting their own work. This work did not involve any recruitment or data collection from research participants or patients and therefore did not require participant/patient consent.

Methods

We present a conceptual (descriptive) case study with an applied example, describing the retrospective application of IM to articulate programme theory for an existing digital health intervention. This work was conducted in spring 2023 at the Ludwig Boltzmann Institute for Digital Health and Prevention in Salzburg, Austria.

The aktivplan digital health intervention

The aktivplan digital health intervention was developed by our team at the Ludwig Boltzmann Institute for Digital Health and Prevention in Salzburg, Austria, to address the abovementioned challenge of establishing regular heart-healthy PA habits in patients with CVD following completion of a cardiac rehabilitation programme. Starting in summer 2020, we applied a user-centred design process, working with cardiac patients and involving experts from software design, human computer interaction, cardiac rehabilitation and psychology.²³The development process was iterative and included a comprehensive initial needs analysis and repeated rounds of formative user testing and heuristic evaluation, to arrive at a mature prototype which is currently under evaluation in a clinical pilot study (https://www.clinicaltrials.gov/study/NCT06025526). For illustration, a previous user study with seven cardiac rehabilitation patients²³ demonstrated average (satisfactory) usability of aktivplan according to the System Usability Scale²⁴ (mean 65.4, SD 17.3, possible score range 0–100). In the dimensions of the Unified Theory of Acceptance and Use of Technology questionnaire,²⁵ aktivplan achieved mean scores of 2.6 (SD 0.6) for performance expectancy, 3.5 (SD 0.3) for effort expectancy, 3.5 (SD 0.7) for facilitating condition, 2.9 (0.7) for behavioural intention to use, and 1.4 (SD 0.5) for technology anxiety (possible score range 1–5), demonstrating overall fair to good usability and acceptability of the tool.

A digital planning calendar for regular heart-healthy PA and exercise is a central component of the aktivplan digital health intervention. Cardiac rehabilitation healthcare professionals (HCPs) access this planning calendar via the aktivplan website (view for HCPs, Figures 1(a) and 1(b).

Figure 1.

The healthcare professional's view of the aktivplan application: (a) overview of the healthcare professional's caseload, with traffic light display indicating patient adherence; and (b) physical activity planning page with calendar view for individual patient.

In clinical practice, the aktivplan planning calendar is introduced to the cardiac rehabilitation patient by an HCP trained in exercise prescription for cardiac rehabilitation (e.g. an exercise physiologist, sports scientist or physiotherapist) towards the end of a cardiac rehabilitation programme. During a PA counselling session lasting approximately one hour, the HCP and patient discuss and devise a PA plan for the patient to carry out during the weeks following completion of the cardiac rehabilitation programme. During this counselling session, the HCP applies a participatory communication style that supports principles of shared decision making.²³ For this, the HCP can use an optional integrated conversation guide that provides prompts and example phrases for shared decision making. The HCP enters the agreed-upon PA plan into the planning calendar via the aktivplan website (view for HCPs, Figures 1(a) and 1(b), assists the patient in setting up the aktivplan app on their personal device, and talks the patient through the features of the app.

Patients can view the planning calendar via the aktivplan app on their smartphone, tablet or computer (patient view, Figure 2(a)). They can check-off planned physical activities and exercise sessions to indicate that these have been completed. When checking off completed activities, patients indicate their rate of perceived exertion for the activity (Borg rating of perceived exertion scale, Figure 2(b)). The patient can also enter additional unplanned physical activities and exercise sessions. Via the aktivplan app, the patient can access short videos with exercise instructions (Figure 2(c)) and export a report documenting all performed physical activities and exercise sessions. This report can be used for the patient's own feedback and reflexion, or it can be handed to the HCP for their information and documentation. The HCP can access the planning calendars of all patients under their care via the aktivplan website and can monitor patients’ adherence with the agreed-upon PA plan on a traffic light display (Figure 1(a)).

Figure 2.

The patient's view of the aktivplan application: (a) physical activity plan; (b) confirming completion of a planned activity; (c) short video with exercise instruction; and (d) automated messages.

Further functions of the aktivplan digital health intervention are automated messages to the patient, delivering reminders of planned physical activities and exercise sessions, motivating and humorous messages, and evidence-based information (Figure 2(d)); personalised goal setting, with the patient's individual timed goals displayed in the aktivplan app (Figure 2(a)); and the possibility for the HCP to message the patient via the app (push notification), to give the patient feedback on their performance and to make contact, for example if adherence is low. At follow-up appointments with the HCP, the patient can reflect on their documented physical activities, exercise sessions and achievements, and discuss the further plan, which can then again be added to the aktivplan planning calendar by the HCP.

Intervention Mapping

IM is a six-stepped approach (Figure 3) for the systematic development, implementation and evaluation of health promotion programmes.²⁶ The six steps are each broken up into smaller tasks. Upon completion, these smaller tasks provide a basis from which the subsequent step is started.

Figure 3.

The six steps of Intervention Mapping.

The first step of IM starts by forming a planning group that consists of the programme planners, community members and programme beneficiaries among others. This mirrors the group of contributors in a user-centred design process, for example, patients and experts from different disciplines in our development of the aktivplan digital health intervention. A diverse planning group ensures the problem is looked at from multiple stakeholder perspectives, which can enhance the effectiveness of the intervention. Hereafter, a needs assessment is conducted using the PRECEDE (Predisposing, Reinforcing, and Enabling Constructs in Educational Diagnosis and Evaluation) model developed by Green at al.²⁷ The context for the intervention is described, and programme goals are established. The final product of the first step of IM is a logic model of the problem.

The second step comprises of stating behavioural and environmental outcomes, specifying performance objectives (POs) and change objectives (COs), selecting determinants for the behavioural and environmental outcomes, and constructing matrices of change. POs are the sub-behaviours necessary to perform a desired target behaviour, and COs are specific changes needed at the determinant level to achieve those sub-behaviours. These COs result from crossing POs with their related personal determinants. The final product of the second step is a logic model of change.

The third step consists of generating programme themes, components, scope and sequence. Matching the previously generated COs with theory-based and evidence-based behaviour change methods and selecting or designing practical applications to deliver the change methods conclude the third step.

The fourth step starts off by refining programme structure and organisation, whereafter plans for programme materials are drafted. Programme messages, materials and protocols are developed in collaboration with designers. The final task of the fourth step is to pre-test, refine and produce programme material. After the fourth step, the development of the intervention programme is finalised.

Step 5 relates to the implementation of the programme. It is a repetition of the first three steps of IM and starts off by identifying programme adopters, implementers and maintainers, for whom outcomes, POs and COs are developed in terms of programme use. This step is finalised by constructing matrices of change and designing implementation interventions.

The sixth and final step of IM concerns the evaluation of the programme. The evaluation is split into effect and process evaluation. First, questions in regard to the evaluation are formed. Then, indicators and measures for assessment are developed, the evaluation design is specified and finally, the evaluation plan is completed.

Within each IM step, many planning questions will be raised. These are answered using the core processes of IM. Core processes are a helpful and systematic way to answer planning questions during intervention development, supporting a complete understanding of the problem, which is imperative for finding effective and appropriate solutions.^26,28 The six core processes need to be conducted in order and are as follows: (1) posing questions, (2) brainstorming potential answers, (3) reviewing findings from empirical literature, (4) reviewing theories, (5) assessing and addressing needs for new data, and (6) developing a working list of answers.²⁶

Applying Intervention Mapping to articulate programme theory for aktivplan

The six steps of IM as described in section Intervention Mapping lay out the entire sequence of developing and evaluating a health promotion programme (or behaviour change intervention) anew, from initial conceptualisation to intervention development, implementation planning, and formative and summative evaluation. Orienting ourselves on this IM sequence, our aktivplan digital health intervention could be positioned at the completion of step 4 – because the intervention has been fully developed, pre-tested and refined following a user-centred co-design methodology – but having ‘skipped’ the systematic conceptualisation and description of its underlying theory and evidence that would usually be conducted in IM steps 1, 2 and 3. We therefore set out to retrace these missing steps and ‘retro-fit’ the programme theory for aktivplan, using the usual IM steps and core processes.

First, the target population and main behavioural outcome were defined. Because user-centred design methodology includes a thorough needs assessment and close involvement of intended users (cardiac patients and rehabilitation professionals), we had rich data to inform the PRECEDE model and to devise the logic model of the problem (IM step 1).

Before moving on to IM step 2, we sought to ascertain all psychological determinants addressed in aktivplan. For this, we defined all distinct aktivplan features (which correspond to practical applications in IM) and mapped these to behaviour change methods, related personal determinants and linked behaviour change theories.²⁶ This linking of intervention features with theory-based and evidence-based behaviour change methods usually takes place in IM step 3, ‘working forwards’ by devising a new intervention based on the logic model of change that was defined in IM step 2. In our case, however, it was more suitable at this point to ‘work backwards’ from our existing intervention, to most efficiently contextualise the programme features that have been incorporated in aktivplan.

We then completed the tasks of IM step 2. Based on the different aktivplan features, the main behavioural outcome, and known associations between certain behaviours, determinants and the main behavioural outcome, we defined the POs, determinants and COs, and we constructed a matrix (logic model) of change.

Lastly, we combined all these aspects in an acyclic behaviour change diagram (ABCD) matrix. ABCD matrices assist in more clear and transparent communication of processes in intervention development.²⁹ The ABCD matrix provides a complete overview and enables the creation of a graphical representation of the ‘logic chain’ between any single aktivplan feature and its underlying theory and evidence in relation to the desired PO(s) and the main behavioural outcome.

Results

The logic model of the problem

Figure 4 shows the logic model of the problem which the aktivplan digital health intervention intends to address. This specifies the target (at-risk) population, defined as ‘patients who had an acute cardiac event and who are attending a cardiac rehabilitation programme’. To present a holistic picture of the problem, we have also included two environmental agents in our logic model of the problem, that is, HCPs in cardiac rehabilitation and people in the patient's interpersonal environment. (Environmental agents are individuals who through their own behaviour can influence the behaviour of the target population, at the level of interpersonal interaction with the target population, or at an organisational, community or societal level. While there are many environmental agents who can impact our defined target population, we do not include environmental agents in the following, as they are not the focus of the aktivplan intervention.)

Figure 4.

Logic model of the problem.

Based on the logic model of the problem, we define the main programme goal (the main behavioural outcome) as ‘following discharge from a cardiac rehabilitation programme, cardiac patients independently and regularly carry out the recommended amount of heart-healthy PA over months and years’.

aktivplan features mapped to behaviour change methods

The aktivplan application incorporates several behaviour change techniques to achieve the desired behavioural outcome. Table 1 provides an overview of all aktivplan features, the behaviour change methods these features employ, the related determinants, and underlying psychological theories.

Table 1.

aktivplan features mapped to behaviour change methods, determinants and theories.

Feature	Behaviour change method	Related determinant	Related theories
Physical activity planner	Shared decision making	Motivation³⁰ Behavioural capability³¹	Social Cognitive Theory Theory of Planned Behaviour Self-Determination Theory Health Belief Model^32–35
Physical activity planner	Tailoring	Self-efficacy^{31, 36, 37}	Trans-Theoretical Model Precaution Adoption Process Model Protection Motivation Theory Communication-Persuasion Matrix^38–41
Goal setting	Goal setting	Motivation³⁰ Skills⁴⁴ Self-efficacy^{30, 45, 46}	Goal Setting Theory Theories of Self-Regulation^{42, 43}
Adherence visualisation	Self-monitoring	Self-efficacy⁴⁵ Skills⁴⁴	Theories of Self-Regulation⁴³
Active minutes
Perceived exertion rating
Messages with evidence-based information	Arguments	Outcome expectations⁴⁷	Communication-Persuasion Matrix Elaboration Likelihood Model^{41, 48}
Reminder messages	Feedback	Awareness^{42, 49, 50} Motivation³⁰	Trans-Theoretical Model Theories of Learning Goal Setting Theory Social Cognitive Theory^{42, 49–52}
Adherence monitoring by HCP	Monitoring by others	Motivation³⁰	Self-Determination Theory³⁴
Personal messages from HCP to patient	Feedback	Knowledge⁵³ Self-efficacy^{30, 45, 46}	Theories of Learning Goal Setting Theory Social Cognitive Theory^{42, 49, 50}
Personal messages from HCP to patient	Reinforcement	Motivation³⁰ Self-efficacy^{30, 45, 46}	Theories of Learning Social Cognitive Theory^{49, 50, 54}
Exercise videos	Modelling	Self-efficacy^{30, 36, 45, 46}	Goal setting Theory Theories of Self-Regulation^{49, 50}
Timing of implementation	Teachable moment	Readiness for behaviour change*	Trans-Theoretical Model⁵²
Physical activity counselling session	Guided practice	Skills⁵⁵ Self-efficacy⁴³	Social Cognitive Theory Theories of Self-Regulation^{43, 50}
Physical activity counselling session	Active learning	Skills⁵⁵	Social Cognitive Theory Elaboration Likelihood Model^{50, 56}

Of note, the intervention does not aim to change readiness for behaviour change, but rather the intervention is purposively timed to coincide with a moment when patients are likely to be receptive to behaviour change.

HCP, healthcare professional.

As one important aspect of the application, aktivplan facilitates a collaborative dialogue between HCPs and patients through shared decision making. This is achieved by including a ‘shared decision making conversation guide’ in the aktivplan website (HCP's view) which serves as a reminder for HCPs to articulate key aspects of shared decision making during the PA counselling session (e.g. stating that the patient should express and decide their preferences regarding the PA plan, etc.). Shared decision making empowers the patient to actively include and combine their personal goals and preferences with the HCP's professional expertise when discussing the PA plan.^{6, 23, 57} Moreover, shared decision making facilitates desired behaviour change through goal setting and public commitment, in addition to increasing motivation of the patient.^23,58 Accordingly, the recent European Association of Preventive Cardiology position paper on exercise prescription in cardiovascular rehabilitation explicitly recommends to incorporate shared decision making when devising PA plans with patients with CVD.⁶

Within this collaborative relationship between HCP and patient, aktivplan leverages the reputable authority of HCPs to support and motivate the patient. The patient gains validation and support from a trustworthy source by including professional expertise and guidance within the app, which enhances the desired behaviour change.⁵⁹

Additionally, the aktivplan application puts emphasis on tailoring the PA plan to the individual patient. A lack of tailoring has been shown to have a negative impact on cardiac rehabilitation attendance.⁶⁰ Furthermore, the possibility to tailor or individualise PA within interventions has been found to facilitate engagement in PA among cardiac patients.⁶¹ In the same systematic review, the authors concluded that tailoring also increased the patients’ self-efficacy in managing bothersome symptoms that could be barriers to PA. This was the case for both home-based and centre-based interventions.⁶¹

Goal setting is another major factor of the aktivplan application. Patients can add both individually chosen goals and goals formulated in collaboration with their HCPs. Goal setting positively influences motivation and its use enhances PA levels.^62,63 Through the visualisation of their daily, weekly and monthly active minutes and adherence to their planned PA, patients can self-monitor their behaviour. Self-monitoring is known as an effective behaviour change method that increases self-efficacy and promotes the maintenance of PA after discharge from a cardiac rehabilitation programme.⁶⁴

aktivplan uses readily composed messages containing evidence-based information on PA and its health benefits. Additionally, patients receive reminders to enter their PA, if they were due a session and have not yet entered it to the app. These automated messages provide patients with relevant information whilst simultaneously serving as reminders of the importance of PA. Furthermore, the HCPs can monitor all their registered patients’ adherence and initiate personal messages. These messages can serve as opportunities for the HCPs to provide feedback and reinforcement. Feedback and reinforcement have been found to be effective methods to tackle psychological determinants related to PA such as positive attitude and outcome expectations towards PA, increased self-efficacy to perform exercises, motivation and problem solving skills to overcome barriers to PA.^65,66

Furthermore, aktivplan provides users easy access to training videos containing exercises and short workouts that can be completed at home with limited or no additional equipment. These training videos are an example of the behaviour change method modelling, which is known to enhance exercise-related self-efficacy.⁶⁷

In terms of its implementation in the cardiac rehabilitation patient pathway, aktivplan utilises the time window during which patients with CVD are often observed to be receptive to behaviour change (‘teachable moment’), because of their recent experience of a significant or even life-threatening medical episode that has prompted referral to a cardiac rehabilitation programme.^68,69

In addition to the individual effectiveness of these behaviour change methods, in their systematic review on lifestyle modification programmes for patients with coronary heart disease, Janssen et al.⁷⁰ reported on the synergetic effects of behaviour change methods, including goal setting, self-monitoring, planning and feedback, in changing lifestyle behaviours. All these methods are effectively incorporated into the aktivplan application, for example, feedback on one's own performance (adherence visualisation, active minutes and perceived exertion rating) to facilitate self-monitoring; and guided practice and active learning during the PA counselling session to reinforce the use of these features.

The logic model of change for aktivplan

We defined four POs for the target population. In sum, these four sub-behaviours contribute to the main behavioural outcome of the target population:

PO.1 Patients plan their weekly PA sessions in the aktivplan application with their HCP^{61, 71};

PO.2 Patients complete their minimum recommended amount of weekly PA;

PO.3 Patients monitor their PA using the aktivplan app^{8, 61, 62, 64, 72, 73} and

PO.4 Patients adapt PA plans according to their own capabilities in collaboration with their HCP.⁶¹

After defining these POs, we conducted literature searches on the personal determinants associated with these behaviours. Thereafter, we conducted a comparative analysis between the identified determinants and those addressed by the aktivplan application. COs were consecutively developed for all overlapping determinants. The evidence-based justification for selected determinants and the cross-tabulation of POs, determinants and COs can be found in the online supplemental file, under the tabs ‘Justification of determinants’ and ‘CO matrix – patients’, respectively.

The acyclic behaviour change diagram for aktivplan

The ABCD in Figure 5 gives a complete overview of all POs, COs, determinants, behaviour change methods, and practical applications (i.e. the corresponding features of aktivplan). This makes transparent the intended – or hypothesised – mechanisms of action of the aktivplan digital health intervention, linking through from the behaviour change methods to the main behavioural outcome.

Figure 5.

Acyclic behaviour change diagram (ABCD) for the aktivplan digital health intervention.

For example, the aktivplan feature ‘PA planner’ utilises the behaviour change methods tailoring and shared decision making. Both methods target the determinants behavioural capability and self-efficacy to influence PO.1 (patients plan their weekly PA session in the aktivplan application with their HCP), which is a contributing sub-behaviour to achieve the main behavioural outcome. The corresponding COs (BC.1 and SE.1 in Figure 5) articulate how the success of this feature can be observed, that is, the impact on behavioural capability can be observed by patients demonstrating their ability to plan their weekly PA (BC.1), and the impact on self-efficacy can be observed by patients expressing confidence in planning their weekly PA (SE.1).

The ABCD matrix in the online supplemental file (tab ‘ABCD – patients’) gives a tabular presentation of the content in Figure 5.

Discussion

We applied the IM approach to ‘articulate programme theory’ for our digital health intervention aktivplan, to make transparent the (proposed) active ingredients and mechanisms of action underlying the intervention. Working backwards from the mature intervention prototype, we re-traced IM steps 1, 2 and 3 and ‘retro-fit’ a comprehensive logic model of change for aktivplan. The logic model of change includes four POs (i.e. sub-behaviours of the main target behaviour), 8 personal determinants (i.e. factors that influence these sub-behaviours) and 12 COs (i.e. changes needed at the determinant level to achieve the sub-behaviours), which are linked to 12 distinct features of the aktivplan digital health intervention, and their 12 underlying behaviour change methods (Table 1). This creates a detailed map of all individual components of the aktivplan intervention and their proposed or hypothesised influence on a cardiac patient's personal determinants, sub-behaviours and finally the main target behaviour of regular heart-healthy physical activity (Figure 5).

Practical use of the logic model

As already stated, at a wider scientific and societal level, detailed intervention descriptions support transparency, reproducibility and the reduction of waste in research.^18,19 At a more immediate level, one practical use of the aktivplan logic model of change is to arrive at a thorough understanding of the intervention ourselves. Because aktivplan features emerged organically, based on needs analysis and decision making in the user-centred iterative design process, the work of explicitly drawing apart all individual features and developing a logic model around them helped us to reflect on what is – and what is not – part of our aktivplan intervention. Notably, for example, the formulation of implementation intentions as a behaviour change method is finding increasing application in physical activity promotion,⁷⁴ but it had not explicitly been considered in aktivplan. Similarly, social support has been associated with greater leisure-time physical activity in older adults,⁷⁵ but aktivplan did not include any features to increase or leverage social support. For us as intervention developers, this makes apparent further development opportunities, that is, aspects which could be addressed in future iterations of aktivplan or in separate new interventions. It also helps us reflect on why certain aspect were not incorporated in the development, for which there can of course be valid reasons such as deliberate design decisions to avoid overloading the application with features and components.

Another practical use for our logic model is to support our communication about aktivplan components to other stakeholders such as researchers, rehabilitation providers and prospective users (both healthcare professionals and patients).¹³ The logic model serves to describe the specific functions of aktivplan features and their importance, by linking each with their underlying psychological rationales and behaviour change methods. Anecdotally, the graphical representation of the ‘logic chain’ (ABCD, Figure 5) has proven particularly helpful to relate the workings of this behaviour change intervention to colleagues from the biomedical domain, by drawing a parallel with biomedical pathway models such as the coagulation cascade. For potential users who wish to select a suitable digital tool from a range of options, such a depiction of which aspects of behaviour change are (not) addressed within the aktivplan intervention can be helpful to meet the specific needs of individual patients or service user groups. Importantly, ABCDs may be regarded as ‘stand alone’ tools and complementary to intervention development frameworks such as IM. ABCDs offer intervention developers an operational means to help optimise transparent development and reporting of interventions by visualising the active components of behaviour change interventions.²⁹ The causal logic that is depicted in ABCDs, while well-reasoned and supported by available scientific evidence, of course represents the intervention developer's proposition of mechanisms of action of the intervention and requires verification, for example through data collected in an evaluation study.

With regard to the evaluation of the aktivplan intervention, the logic model offers a comprehensive scaffold against which to select appropriate evaluation designs that capture the relevant processes and outcomes.¹² As the COs defined in the logic model essentially present targets for evaluation outcomes, the model can be used to inform the selection of outcome measures for specific COs, or the development of instrument items for specific COs if no existing measures are available. Additionally, the behaviour change methods and determinants incorporated in the logic model signpost towards associated outcome measures (e.g. self-efficacy questionnaires) which may be appropriate to use. Changes in these outcome measures, in combination with automatically logged usage data for the different features of the tool, provide a descriptive indication of the relative impact of individual features and may also allow for more complex statistical assessment through mediation analysis.⁷⁶ Moreover, this can support the use of objective and subjective measures to capture user engagement, and the use of large amounts of real-time data generated by a digital health intervention⁷⁷ to test and advance models of behaviour change, as specified in recent recommendations for developing and evaluating digital interventions to promote behaviour change in health and healthcare.⁷⁸

Possible disadvantages of developing a detailed logic model

There are some considerations which might preclude developing such a detailed logic model as presented in this article. The required expertise in the IM methodology or the required time resources might not be available. From a design perspective, such an analytic dissection of the intervention as presented in this example might take away from the more creative ‘designerly intuition’ and ‘abductive reasoning’ that is characteristic for co-design approaches.⁷⁹ In the retrospective development of the logic model, initial design intentions might be misrepresented, and intervention components mapped inappropriately. Similarly, one could get lost in the details of the logic model and lose sight of a clear design focus and vision. In any case, intervention developers should work up a type of logic model for their digital health intervention, and approaches that require less detailed description and justification of psychological mechanisms and theories than in the example presented here are available (see section Other approaches for the retrospective development of a logic model).¹⁷

User-centred iterative co-design meets Intervention Mapping

As demonstrated in this work, user-centred design and IM can both be integrated in the development of digital health interventions for behaviour change, building on the strengths of both methodologies, each underpinned by complementary principles. User-centred design, with stakeholder involvement and iteration as key tenets, emphasises the involvement of end-users at every stage of the design process, fostering a deep understanding of user needs and preferences.⁸⁰ This approach aligns closely – and is often used in conjunction – with co-design, which involves stakeholders, including users, in the creative process to ensure that the outcomes are both usable and meaningful.⁸¹ IM, a systematic approach for health promotion programme planning, encompasses six iterative steps as detailed above: needs assessment, formulation of change objectives, selection of theory-based methods and practical strategies, programme design, adoption and implementation planning, and evaluation planning.²² Similarities between user-centred design/co-design and IM are their iterative nature and shared commitment to a systematic, user- (or human-)focused approach. Both methodologies highlight the understanding of user context and needs, albeit with different emphasis – user-centred design/co-design through direct user involvement and iterative testing, and IM through a structured analysis of the problem and possible solutions together with the target population and other relevant stakeholders. Both approaches complement each other in their theoretical underpinnings and scope. While user-centred design and co-design are grounded in design theory and primarily focus on product usability and satisfaction, IM initiated from the field of health promotion and encompasses a broader range of considerations, including the effectiveness of interventions in changing (health) behaviours. This is critical in the context of digital health interventions, where the convergence of these approaches can leverage the strengths of each – user-centred design/co-design enhancing user engagement and adherence in close interplay with digital health intervention feature design, and IM ensuring the intervention's theoretical robustness and practical effectiveness in behaviour change.^{82, 83}

Other approaches for the retrospective development of a logic model

As stated above, there are approaches to constructing a logic model for a complex intervention that require less detailed description and justification of psychological mechanisms and theories than in the example presented here. O'Cathain et al.¹⁷ describe a more basic structure for a logic model of a complex intervention, which starts from ‘principles’ and links sequentially to ‘actions to consider’, ‘intervention’, ‘outputs’, ‘short term effects’ and ‘long term effects’. A similarly basic logic model structure is described in the frequently cited ‘Logic Model Development Guide’ by the Kellogg Foundation⁸⁴ which takes ‘resources/inputs’ as a starting point and links sequentially to ‘activities’, ‘outputs’, ‘outcomes’ and ‘impact’. Diagrammatic representations of such logic models are often seen in scientific publications of complex interventions, for example, in Cook et al.⁸⁵ Nevertheless, for a digital behaviour change intervention even such a basic logic model will need to include reference to behaviour change methods and/or theory.¹³ To achieve this in retrospect, an approach applied in the systematic review by Patterson et al.⁸⁶ can be appropriate, by which descriptions of digital health interventions according to the TIDieR checklist were mapped against a comprehensive classification of behaviour change methods and theories – in this case the behaviour change technique taxonomy developed by Michie et al.⁸⁷ As described in section Practical use of the logic model, an ABCD may also be developed as a ‘stand alone’ tool, to offer a visual representation in support of a logic model that has been developed via a framework other than IM.²⁹

Another approach for ‘reverse-engineering’ a logic model for an existing intervention has been described by Walker et al.¹⁶ Termed ‘Evidence-Based Intervention (EBI) Mapping’, this approach is similar to our example, as it also applies IM methodology retrospectively. However, the primary motivation for EBI Mapping is to support uptake of interventions with proven efficacy or effectiveness (EBIs) in practice, whereas our approach prioritises systematically analysing existing interventions regardless of their evidence base to understand their components and logic, offering necessary insights for their evaluation, potential adaptation or enhancement. In EBI Mapping, the suggestion is that a detailed description of an EBI's goals, components and theory of change will help potential adopters judge and decide whether the EBI could be suitable for their particular population and context. Walker et al. define four stepwise tasks – identifying the who, what, why and how of an EBI – to arrive at a logic model that lists practical applications, change methods, determinants, health behaviour, health outcome and health problem. In contrast to our example, EBI Mapping does not include an ABCD with its detailed specification of POs and COs and its depicted (proposed) causal linkage between individual behaviour change methods, applications, COs, determinants and POs; and the suggested use of EBI Mapping is mainly to inform potential adopters of an EBI about suitability of the intervention and adaptation needs for their particular setting.¹⁶

Strengths and limitations

A strength of our work is the comprehensive and practical description of the retrospective application of IM to ‘articulate programme theory’, providing readers with sufficient detail to replicate the approach. Moreover, because IM is a well-established and actively taught methodology, readers seeking further guidance on the application of IM may draw on current scientific publications and educational resources about IM. Another strength of our approach is its compatibility with user-centred design/co-design, enabling synergies between this prevalent paradigm in modern design thinking and the psychologically and theoretically informed underpinning logic of a behaviour change intervention that is generated through IM.

We acknowledge limitations to our work. The approach may be limited by lack of access to intervention contents and materials, especially when the development of the logic model is conducted by persons who are outside the intervention development team and who may rely on scientific publications and other publicly accessible information about the intervention. Differing conceptual understandings and definitions of terms may add to this difficulty. Another limitation concerns the possibility that the delineation of intervention features, determinants, behaviour change methods and theories may not be obvious or straightforward, and decisions may to some extent require judgement calls or group consensus. We recommend that an attempt is made to create a suitable stakeholder group, ideally including representatives from the intervention development team, so that access to intervention contents and materials is granted and accounts of the ‘history’ of intervention development can be provided. This stakeholder group may then constitute the forum in which concepts, definitions and decisions are discussed and agreed upon.

While the example presented in this article only addressed the behaviour of the at-risk populations (i.e. patients who had an acute cardiac event), a logic model may also extend to include various environmental agents. However, the roles of environmental agents can sometimes overlap with those of implementers (i.e. the persons involved in the adoption, implementation and maintenance of an intervention),²⁶ creating potential for lack of conceptual clarity in this respect.¹⁶

Recent literature calls for introducing more flexible and dynamic logic models to incorporate elements of interaction between an intervention and the context within which it is deployed^{15, 88} including considerations of potential intervention harms and their mechanisms of action.⁸⁹ We acknowledge that this is not accounted for in our example.

Conclusion

It is important to make transparent the indented – or hypothesised – workings of digital health interventions, by ‘articulating programme theory’. This article offers a conceptual case study and a worked example, applying IM to retrospectively develop a detailed logic model of an existing, co-designed digital behaviour change intervention for physical activity in cardiac rehabilitation. Intervention developers and researchers may draw on this example to replicate the method, or to reflect on most suitable approaches for ‘articulating programme theory’ for their own interventions. Publications of similar case studies and worked examples from other clinical use cases can enrich and support researchers’ practice of articulating programme theory.

Supplemental Material

sj-xlsx-1-dhj-10.1177_20552076241260974 - Supplemental material for Making it transparent: A worked example of articulating programme theory for a digital health application using Intervention Mapping

Supplemental material, sj-xlsx-1-dhj-10.1177_20552076241260974 for Making it transparent: A worked example of articulating programme theory for a digital health application using Intervention Mapping by Tamika A. Marcos, Rik Crutzen, Veronika Leitner, Jan D. Smeddinck, Eva-Maria Strumegger, Daniela Wurhofer and Stefan T. Kulnik in DIGITAL HEALTH

Footnotes

Contributorship

STK and RC conceived the work. TAM developed the logic model with support from STK and RC. VL, JDS, E-MS and DW peer reviewed and validated the logic model. TAM and STK wrote the first draft of the manuscript; and E-MS, RC and JDS contributed to the manuscript draft. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Guarantor

STK

ORCID iD

Stefan T. Kulnik

Supplementary material

Supplementary material for this article is available online.

References

Wang

. GBD Demographics collaborators. Global age-sex-specific fertility, mortality, healthy life expectancy (HALE), and population estimates in 204 countries and territories, 1950–2019: a comprehensive demographic analysis for the global burden of disease study 2019. Lancet 2020; 396: 1160–1203.

Di Cesare

, et al. World heart report 2023: Confronting the world's number one killer. Geneva, Switzerland: World Heart Federation, 2023.

Guidetti

, et al. Inconsistencies across regional strategies. Eur J Phys Rehabil Med 2014; 50: S0140–S6736.

Jeong

S-W

, et al. Mortality reduction with physical activity in patients with and without cardiovascular disease. Eur Heart J 2019; 40: 3547–3555.

Visseren

, et al. ESC Guidelines on cardiovascular disease prevention in clinical practice: developed by the task force for cardiovascular disease prevention in clinical practice with representatives of the European Society of Cardiology and 12 medical societies with the special contribution of the European association of preventive cardiology (EAPC). Eur J Prev Cardiol 2022; 29: 5–115.

Hansen

, et al. Exercise intensity assessment and prescription in cardiovascular rehabilitation and beyond: why and how: a position statement from the secondary prevention and rehabilitation section of the European association of preventive cardiology. Eur J Prev Cardiol 2022; 29: 230–245.

Kotseva

, et al. Lifestyle and impact on cardiovascular risk factor control in coronary patients across 27 countries: results from the European Society of Cardiology ESC-EORP EUROASPIRE V registry. Eur J Prev Cardiol 2019; 26: 824–835.

ter Hoeve, N., et al. Does cardiac rehabilitation after an acute cardiac syndrome lead to changes in physical activity habits? Systematic review. Phys Ther 2015; 95: 167–179.

Heimer

, et al. Ehealth for maintenance cardiovascular rehabilitation: a systematic review and meta-analysis. Eur J Prev Cardiol 2023; 30: 1634–1651.

10.

Wongvibulsin

, et al. Digital health interventions for cardiac rehabilitation: systematic literature review. J Med Internet Res 2021; 23: e18773.

11.

Antoniou

, et al. Safety and long-term outcomes of remote cardiac rehabilitation in coronary heart disease patients: a systematic review. Digital Health 2024; 10: 20552076241237661.

12.

Skivington

, et al. A new framework for developing and evaluating complex interventions: update of medical research council guidance. Br Med J 2021; 374: n2061.

13.

Hoffmann

, et al. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide. Br Med J 2014; 348. doi:10.1136/bmj.g1687

14.

Abraham

, et al. Enhancing reporting of behavior change intervention evaluations. JAIDS J Acquir Immune Defic Syndr 2014; 66: S293–S299.

15.

Mills

Lawton

Sheard

. Advancing complexity theory in health services research: the logic of logic models. BMC Med Res Methodol 2019; 19: 55.

16.

Walker

, et al. Evidence-Based intervention (EBI) mapping: a systematic approach to understanding the components and logic of EBIs. BMC Public Health 2022; 22: 1300.

17.

O'Cathain

, et al. Guidance on how to develop complex interventions to improve health and healthcare. BMJ open 2019; 9: e029954.

18.

Altman

Simera

. A history of the evolution of guidelines for reporting medical research: the long road to the EQUATOR network. J R Soc Med 2016; 109: 67–77.

19.

Chalmers

Glasziou

. Avoidable waste in the production and reporting of research evidence. Lancet 2009; 374: 86–89.

20.

Ryan

, et al. Incomplete reporting of complex interventions: a call to action for journal editors to review their submission guidelines. Trials 2023; 24: 176.

21.

Johnston

. Improving the reporting of behaviour change interventions. The European Health Psychologist 2014; 16: 181–189.

22.

Kok

. A practical guide to effective behavior change: how to apply theory- and evidence-based behavior change methods in an intervention. Eur. health Psychol 2014; 16: 156–170.

23.

Wurhofer

, et al. Investigating shared decision-making during the use of a digital health tool for physical activity planning in cardiac rehabilitation. Front. Digit. Health 2024; 5: 1324488.

24.

Brooke

. SUS: a retrospective. J Usability Stud 2013; 8: 29–40.

25.

Hoque

Sorwar

. Understanding factors influencing the adoption of mHealth by the elderly: an extension of the UTAUT model. Int J Med Inf 2017; 101: 75–84.

26.

Bartholomew Eldredge

, et al. Planning health promotion programs: an intervention mapping approach. 4 ed. San Francisco, CA: Jossey-Bass Inc, 2016.

27.

Green

, et al. Health program planning, implementation, and evaluation: creating behavioral, environmental, and policy change. Baltimore, MD: John Hopkins Univ. Press, 2022.

28.

Ruiter

RAC

Crutzen

. Core processes: how to use evidence, theories, and research in planning behavior change interventions. Front Public Health 2020; 8: 247.

29.

Metz

Peters

Crutzen

. Acyclic behavior change diagrams: a tool to report and analyze interventions. Health Psychol Behav Med 2022; 10: 1216–1228.

30.

Kim

, et al. A survey of the perceptions of barriers to and facilitators of cardiac rehabilitation in healthcare providers and policy stakeholders. BMC Health Serv Res 2022; 22.

31.

Rhodes

, et al. Mediators of physical activity behaviour change interventions among adults: a systematic review and meta-analysis. Health Psychol Rev 2021; 15: 272–286.

32.

Bandura

. Social cognitive theory: an agentic perspecitve. Annu Rev Psychol 2001; 52: 1–26.

33.

Ajzen

. The theory of planned behavior. Organ Behav Hum Decis Process 1991; 50: 179–211.

34.

Ryan

Deci

. Self-Determination theory and theory and the facilitation of instrinsic motivation, social development, and well-being. Am Psychol 2000; 55: 68–78.

35.

Janz

Becker

. The health belief model: a decade later. Health Educ Q 1984; 11: 1–47.

36.

Ghisi

and, et al.

. Behavior determinants among cardiac rehabilitation patients receiving educational interventions: an application of the health action process approach. Patient Educ Couns 2015; 98: 612–621.

37.

Murray

Rodgers

. The role of socioeconomic Status and control beliefs on frequency of exercise during and after cardiac rehabilitation. Appl Psychol Health Well Being 2012; 4: 49–66.

38.

Lustria

, et al. Computer-tailored health interventions delivered over the web: review and analysis of key components. Patient Educ Couns 2009; 74: 156–173.

39.

Weinstein

Sandman

Blalock

. The precaution adoption process model. In: Health behavior and health education: theory, research, and practice. 4th ed. Jossey-Bass, CA, US: San Francisco, 2008, pp.123–147.

40.

Werrij

, et al. Message framing. In: Abraham

Kools

(eds) Writing health communication: an evidence-based guide. London, United Kingdom: Sage, 2012, pp.134–143.

41.

McGuire

. Inpuit and output variables currently promising for constructing persuasive communications. In: Rice

Atkin

(eds) Public communication campaigns. Thousand Oaks, CA: Sage, 2001, pp.22–48.

42.

Latham

Locke

. New developments in and directions for goal-setting research. Eur Psychol 2007; 12: 290.

43.

Creer

. Self-management of chronic illness. In: Boekaerts

Pintrich

Zeidner

(eds) Handbook of self-regulation. San Diego, CA: Academic Press, 2000, pp.601–629.

44.

Dahlan Tabah

, et al. Factors influencing stroke patient adherence to physical activity: a systematic review. J. Gerontol. Geriatr 2020; 68: 174–179.

45.

Han

Won

. Association between Social Support and Physical Activity in Patients with Coronary Artery Disease: Multiple Mediating Roles of Self-Efficacy and Autonomous Motivation. Healthcare (Basel) 2022; 10.

46.

Klompstra

Jaarsma

Stromberg

. Self-efficacy mediates the relationship between motivation and physical activity in patients with heart failure. J Cardiovasc Nurs 2018; 33: 211–216.

47.

Bohlen

, et al. A systematic review and meta-analysis of the outcome expectancy construct in physical activity research. Ann Behav Med 2022; 56: 658–672.

48.

Petty

Wegener

. Attitude change: multiple roles for persuasive variables. In: Gilbert

Fiske

Lindzey

(eds) The handbook of social psychology. Boston, MA: McGraw-Hill, 1998, pp.323–390.

49.

Kazdin

A.E

., Behavior modification in applied settings. Waveland Press, 2012.

50.

Kelder

Hoelscher

Perry

. How individuals, environments, and health behaviors interact. Health Behavior: Theory, Research, and Practice 2015; 159: 144–149.

51.

Godden

Baddeley

. Context-dependent memory in two natural environments: on land and underwater. Br J Psychol 1975; 66: 325–331.

52.

Prochaska

Velicer

. The transtheoretical model of health behavior change. Am J Health Promot 1997; 12: 38–48.

53.

Williamson

T.M.

, et al., Promoting adherence to physical activity among individuals with cardiovascular disease using behavioral counseling: a theory and research-based primer for health care professionals. Prog Cardiovasc Dis, 2021. 64: p. 41–54.

54.

McSweeney

F.K.

Murphy

E.S.

, The Wiley Blackwell Handbook of Operant and Classical Conditioning. 2014.

55.

Fishbein

Ajzen

, Predicting and changing behaviour: the reasoned action approach. 1st ed. New York: Psychology Press, 2010. 538.

56.

Petty

Barden

Wheeler

. The elaboration likelihood model of persuasion: Developing health promotions for sustained behavioral change. Emerging theories in health promotion practice and research, 2009. 2: p. 185–214.

57.

Bonneux

, et al. The SharedHeart Approach: Technology-Supported Shared Decision Making to Increase Physical Activity in Cardiac Patients, in Pervasive Computing Technologies for Healthcare. Cham, Switzerland: Springer International Publishing AG, 2022, p. 469–488.

58.

Bonneux

, et al. Theory-informed design guidelines for shared decision making tools for health behaviour change. in Persuasive Technology: 16th International Conference, PERSUASIVE 2021, Virtual Event, April 12–14, 2021, Proceedings. 2021. Springer.

59.

Cialdini

. Influence: Science and practice. Needham Heights, MA: Allyn & Bacon, 2001.

60.

Dafny

, et al. Cardiac rehabilitation, physical activity, and the effectiveness of activity monitoring devices on cardiovascular patients: an umbrella review of systematic reviews. Eur Heart J Qual Care Clin Outcomes 2023; 9: 323–330.

61.

Platz

Kools

Howie-Esquivel

. Benefits, facilitators, and barriers of alternative models of cardiac rehabilitation: a qualitative systematic review. J Cardiopulm Rehabil Prev 2022; 43: 83–92.

62.

Knittle

, et al. How can interventions increase motivation for physical activity? A systematic review and meta-analysis. Health Psychol Rev 2018; 12: 211–230.

63.

Swann

, et al. Updating goal-setting theory in physical activity promotion: a critical conceptual review. Health Psychol Rev 2021; 15: 34–50.

64.

Graham

, et al. Systematic review of interventions designed to maintain or increase physical activity post-cardiac rehabilitation phase II. Rehabil Process Outcome 2020; 9: 1179572720941833.

65.

Amirova

, et al. Informing behaviour change intervention design using systematic review with Bayesian meta-analysis: physical activity in heart failure. Health Psychol Rev 2023; 17: 456–484.

66.

Pradal-Cano

, et al. Using Mobile applications to increase physical activity: a systematic review. Int J Environ Res Public Health 2020; 17: 8238.

67.

Rajati

, et al. Self-efficacy strategies to improve exercise in patients with heart failure: a systematic review. ARYA Atheroscler 2014; 10: 319–333.

68.

Locke

. Putting the ‘teachable moment’in context: a view from critical health psychology. J Health Psychol 2023; 28: 3–16.

69.

Brust

, et al. Exploring timing and delivery of lifestyle advice following an acute cardiac event hospitalization: the cardiac patient’s perspective. Patient Educ Couns 2024; 124: 108279.

70.

Janssen

, et al. Lifestyle modification programmes for patients with coronary heart disease: a systematic review and meta-analysis of randomized controlled trials. Eur J Prev Cardiol 2013; 20: 620–640.

71.

Tummers

, et al. Determinants of physical activity behaviour change in (online) interventions, and gender-specific differences: a Bayesian network model. Int J Behav Nutr Phys Act 2022; 19: 155.

72.

Chase

. Systematic review of physical activity intervention studies after cardiac rehabilitation. J Cardiovasc Nurs 2011; 26: 351–358.

73.

Room

, et al. What interventions are used to improve exercise adherence in older people and what behavioural techniques are they based on? A systematic review. BMJ Open 2017; 7: e019221.

74.

Warner

, et al.

What makes implementation intentions (in) effective for physical activity among older adults?

Br J Health Psychol 2022; 27: 571–587.

75.

Smith

and, et al.

. The association between social support and physical activity in older adults: a systematic review. International Journal of Behavioral Nutrition and Physical Activity 2017; 14: 1–21.

76.

MacKinnon

, et al. Statistical mediation analysis in psychological research. 2023.

77.

Short

, et al. Measuring engagement in eHealth and mHealth behavior change interventions: viewpoint of methodologies. J Med Internet Res 2018; 20: e292.

78.

Michie

, et al. Developing and evaluating digital interventions to promote behavior change in health and health care: recommendations resulting from an international workshop. J Med Internet Res 2017; 19: e232.

79.

Kulnik

, et al. A gift from experience: co-production and co-design in stroke and self-management. Design for Health 2019; 3: 98–118.

80.

Norman

D. A.

Draper

S.W.

, User Centered System Design: New Perspectives on Human-Computer Interaction. 1986.

81.

Sanders

EB-N

Stappers

. Co-creation and the new landscapes of design. CoDes 2008; 4: 5–18.

82.

Lyon

Koerner

. User-Centered design for psychosocial intervention development and implementation. Clinical Psychology: Science and Practice 2014; 21: 180–120.

83.

Yardley

, et al. The person-based approach to intervention development: application to digital health-related behavior change interventions. JMIR 2016; 17: e30.

84.

Kellogg

. Logic Model Development Guide. East Battle Creek, MI: W.K. Kellogg Foundation, 2004.

85.

Cook

, et al. Communities in charge of alcohol (CICA): a protocol for a stepped-wedge randomised control trial of an alcohol health champions programme. BMC Public Health 2018; 18: 1–11.

86.

Patterson

, et al. Behaviour change techniques in cardiovascular disease smartphone apps to improve physical activity and sedentary behaviour: systematic review and meta-regression. International Journal of Behavioral Nutrition and Physical Activity 2022; 19: 1–14.

87.

Michie

, 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.

88.

Ebenso

, et al. Dealing with context in logic model development: reflections from a realist evaluation of a community health worker programme in Nigeria. Eval Program Plann 2019; 73: 97–110.

89.

Bonell

, et al. ‘Dark logic’: theorising the harmful consequences of public health interventions. J Epidemiol Community Health 2015; 69: 95–98.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB