WP1: Cross-sectional population survey
To guide the development of the training intervention and to identify potential opportunities for improving the quantity and quality of PA advice, WP1 aims to assess national representative data on “how” GP advice is currently provided to people with IHD, and to explore specific individual person characteristics that might be associated with the receipt of and wish for such advice.
The primary outcome of WP1 is to determine the proportion of people in the population of Germany aged ≥ 35 years with self-reported IHD, who have had at least one GP consultation since the last remembered IHD-event (e.g., myocardial infarction, stent insertion), and who report having received GP-delivered advice on PA. This will be achieved by assessing the receipt of the 3As method (ask/assess, advise, assist). Secondary outcomes include exploring IHD-patients’ general wish for receiving GP advice on PA including desired content. Possible associations with sociodemographic, socioeconomic, and health-related individual characteristics (e.g., age, sex, body mass index (BMI), PA level) will be explored.
By means of computer-assisted personal interviews (CAPI) – conducted by a market research institute since June 2023 –, a sample of approximately 1,000 persons aged ≥ 35 years with self-reported IHD who had visited a GP since the last remembered IHD-event will be interviewed in a cross-sectional representative national household survey. Respondents will be selected by using a nationwide multistage, multi-stratified dual frame design: a composition of random stratified sampling (50% of the sample) and quota sampling (50% of the sample).
Based on a standard formula for cross-sectional surveys [32], the target size of 1,000 respondents with IHD will allow an estimation of the actual proportions of GP advice on PA with an absolute error of about 5%. This sample size also allows for some subgroup analyses with sufficient numbers of cases per subgroup.
The lifetime prevalence of IHD is about 9% in those aged ≥ 40 years in Germany [2]. The present study will include people aged ≥ 35 years. We also estimate a relatively conservative probability of 80% visiting a GP since the last remembered IHD-event. Older national survey data suggests that just over one third of people with IHD aged ≥ 65 years have received PA advice by a GP during the past year [26]. However, GP-delivered advice on other health behaviour occurs less often, and younger people might be even less likely to receive advice. The study therefore estimates a more conservative proportion of 20% of people with IHD who received GP-delivered PA advice including all elements of the 3As method (ask/assess, advise, assist). Based on this, about 14,000 individuals aged ≥ 35 years will be needed to identify the target sample. For this purpose, four established questions on IHD will be used which have been applied in several population health surveys before (e.g.,[2]), asking for a history of angina pectoris, myocardial infarction, bypass surgery or coronary stent insertion. A modified version of the validated single-item question of Milton et al. will be used to record the current PA level [33]. The CAPI questionnaire on primary and secondary outcomes was developed in a multi-professional team (e.g., GPs, psychologists, epidemiologists, public health researchers) and pre-tested in patients with IHD (the full questionnaire translated into english has been published at Open Science Framework (OSF) [34]).
Analyses WP1
A detailed study and analysis protocol will be published at OSF prior to the statistical analyses. Descriptive statistics will be used to analyse prevalence data on primary and secondary outcomes. Non-adjusted and adjusted multinomial regression analyses will be used to explore associations between respondents’ characteristics (e.g., age, sex, BMI) and receipt of GP advice as well as with the expressed wish for receiving such advice. Face-to-face data sampling usually produces few missing data. If missing data occur to a relevant extent, non-response analyses will be conducted and the application of multiple imputation methods will be considered.
Current status
At the time of submission of this protocol, data collection has already been completed. The analyses are still being conducted.
WP2: Qualitative study on GPs’ and patients’ attitudes, experiences, and needs
In WP2, attitudes, motivation and experiences of GPs with the delivery of PA advice to their people with IHD will be explored to understand barriers and facilitators to routinely delivering such advice. In addition, we aim to identify GPs’ needs concerning the concept, content, and conditions for an optimal brief training on the delivery of PA advice. Complementarity, attitudes, acceptability, needs, and barriers of people with IHD towards the receipt of PA advice in the general practice setting will also be assessed, what corresponds to the contents of the COM-B Model ( [35], definition see below). For this purpose, individual problem-centred interviews [36] and focus group discussions [37, 38] with GPs and people with IHD, respectively, will be conducted based on interview and focus group guides. These guides were developed in a multi-professional study team, including patient representatives and GPs, and have been pre-tested.
In order to achieve a diverse range of professional experience, sociodemographic characteristics, general practice characteristics, and interest in PA among GPs, and of sociodemographic, socioeconomic, and health-related characteristics, as well as interest in PA among people with IHD, these groups will be selected using a purposive sampling strategy [39]. For maximum structural variation, case contrast strategies are provided, which also serves to assure the quality [40]. The process of sampling is supported by the use of a short questionnaire. This questionnaire and the interview and focus group guides have been published at OSF [41].
A number of four to six focus groups each with GPs and patients, respectively, with a maximum of ten to twelve participants per group, and eight to twelve individual interviews each with GPs and patients are expected to be sufficient. Data collection will continue until reaching data saturation [42]. GPs will be recruited through addressing regional quality circles of GPs (association of GPs who meet regularly to share and reflect on their everyday practice [43]), representatives of practice and GP networks, the teaching and research practice network of the study institute, and from the NRW General Practice Research Network [44]. Patients will be contacted via self-help or rehabilitation groups in the Rhine-Ruhr region. In addition, study information in health care facilities and public institutions will be used to reach people who are preferably not members of self-help groups or cardiac sports groups. Experienced moderators from the study institute will conduct the focus group discussions. Data will be collected in an iterative cyclical process. Saturation marks the point in the iterative process of data collection and evaluation at which additional data collection does not lead to any further gain in knowledge according to the research question. The interviews and group discussions will be audio-recorded and transcribed verbatim, according to simple scientific transcription rules [45]. Postscripts document the atmosphere, the conversation process, interactions, specifics, and disruptions.
Analyses WP2
The content-analytical data evaluation of the audio-recorded and verbatim transcribed pseudonymised data will be carried out using the software MAXQDA [46] in a content-structuring procedure [47] by a multi-professional study team (e.g., GPs, patient representatives, psychology, social science, public health). Content-structuring qualitative content analysis is, in addition to evaluative and type-forming qualitative content analysis, a procedure of category-based methods for the systematic analysis of qualitative data [47]. These evaluation procedures on category-based methodology is a language-related, rule-guided systematic scientific method with the aim of pragmatic reduction of complexity [47].
Current status
At the time of submission of this protocol, the data collection has been completed. The analyses of GP data have also been completed, analyses of patient data are still being conducted.
WP3: Systematic development of the GP training
WP3 focuses on the synthesis of results from WP1 and WP2 to inform the systematic development of a brief, single-session, tailored GP training on the delivery of PA advice to people with IHD according to the 3As method.
In a strategy workshop, results of WP1 and WP2 will be compared and connected, using visual representations (e.g., acyclic behaviour change diagrams [48]), discourse and expert consensus techniques.
The interdisciplinary study team, comprising experts in research on behaviour change, health services and public health, will develop a training manual in collaboration with GPs and patient representatives, with provisions for future adjustments for sustainability, such as online booster modules and good practice videos. The COM-B (capability, opportunity, and motivation-behaviour) model [35] will serve as theoretical framework for the development of the training, and the behaviour change taxonomy (BCT) will be used to describe active elements of the training [28]. According to COM-B, the interplay between capability (C), opportunity (O), and motivation (M) influence behaviour (B), and a specific behaviour in turn influences these factors [35]. The training will be developed to address “capability and motivation” by providing knowledge and practical skills to deliver PA advice. By using the 3As (ask/assess, advise, assist) method as a very brief, less time-consuming method of delivering advice, which can be more easily integrated into daily practice, the training will also aim to influence “opportunity”. The definition and description of potentially active training elements according to the BCT [28] will also be part of WP3. Training materials and case vignettes for simulated role-plays using simulation patients will be developed in collaboration with the “communication in medical education” team of the Heinrich-Heine University. Experienced GPs and peer-trainers from the study institute will review the training manual and didactic methods.
Current status
At the time of submission of this protocol, a first draft of a training manual has been developed as described above.
WP4: Accompanying health economic research
WP4 aims at developing a (web-based) discrete choice experiment (DCE) which will identify preferences regarding IHD patients’ outcomes on increased PA, and to develop a questionnaire that accurately assesses health care utilisation in people with IHD. Findings of WP4 will thus help to inform future RCTs by providing the most preferred patient-relevant outcomes for PA based interventions and to inform future health economic evaluations. The development of the DCE will adhere to recent recommendations [49].
First, a literature review and eight to twelve qualitative semi-structured individual interviews with people with IHD based on an interview guide will be conducted to identify relevant outcomes on increased PA. Second, these identified outcomes will be discussed with experts (e.g., people with IHD, GPs, cardiologists) in a joint workshop. In addition, the findings on relevant patient outcomes will be considered in the process of training refinement described in WP5. The experts will select the most important outcomes that will be used as attributes in the DCE. Alongside, a health care utilisation questionnaire will be developed based on a literature review and previous work by members of the study team [50].
The DCE will be pilot tested in people with IHD of the pilot study (see WP5, N = 120), and evaluated in patients of the control group of the main cluster randomised controlled trial (cRCT) (WP6, N = 300). The health care utilisation questionnaire will be pilot tested in patients of the control group of the main cRCT. IHD patients will receive a printed version of the questionnaires as well as a link enabling them to fill in a web-based version, alternatively. This will help to estimate the acceptance of different modes of assessment and to choose the optimal mode of assessment for the evaluation of the DCE which will follow in WP6 (n = 300). To avoid influences on the main study outcomes of the cRCT in WP6, patients will be asked to complete the DCE and health care utilisation questionnaires at home, following the collection of the primary and secondary outcomes of the cRCT.
Analyses WP4
Transcribed individual interviews will be analysed by qualitative content analysis. Outcomes will be identified according to the suggested criteria for patient-relevant outcomes by Nano et al. [51]. Preference weights of the DCE will be derived by a conditional logit regression model and by latent class analysis considering preference heterogeneity. Health care utilisation data will be analysed descriptively by means, standard deviations, and distributions of quantities.
Current status
At the time of submission of this protocol, WP 4 is still ongoing.
link