Layan Fessler1⍰, Silvio Maltagliati2, Philippe Meyer3, Axel Finckh4, Stéphane Cullati5, David Sander67, Malte Friese8, Reinout W. Wiers9, Ata Farajzadeh10, Christophe Luthy11*, Philippe Sarrazin1 and Boris Cheval6712*
1Univ. Grenoble-Alpes, SENS, F-38000 Grenoble, France
2Université Bretagne Sud, LP3C – EA1285 – Laboratoire de Psychologie : Cognition, Comportement, Communication, 56100, Lorient, France
3Division of cardiology, University Hospitals of Geneva, Geneva, Switzerland
4Division of Rheumatology, Department of Medicine, Geneva University Hospitals, Geneva, Switzerland
5Population Health Laboratory, Department of Community Health, University of Fribourg, Fribourg, Switzerland
6Swiss Center for Affective Sciences, University of Geneva, Geneva, Switzerland
7Laboratory for the Study of Emotion Elicitation and Expression (E3Lab), Department of Psychology, University of Geneva, Geneva, Switzerland
8Department of Psychology, Saarland University, Saarbrucken, Germany
9Department of Psychology and center for Urban Mental Health, University of Amsterdam, Amsterdam, The Netherlands
10School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa, Ottawa, Ontario, Canada
11Division of General Medical Rehabilitation, University Hospitals of Geneva, Geneva, Switzerland
12Univ Rennes, École normale supérieure de Rennes, VIPS2, F-35000 Rennes, France
*Corresponding Author’s Contact Information
Corresponding authors: Hôpitaux Universitaires Genève, Rue Gabrielle-Perret-Gentil 4 1205 Genève, Suisse; christophe.luthy{at}hug.ch; École normale supérieure de Rennes, Campus de Ker Lann, 35170 Bruz, France; boris.cheval{at}ens-rennes.fr
Authors contributions. LF: Conceptualisation, investigation, formal analysis, writing-original draft; SM: Methodology-creation of the tasks, writing-review and editing; PM; Resources, writing-review and editing; AF: Writing-review and editing; SC: writing-review and editing; DS: writing-review and editing; MF: writing-review and editing; RWW: writing-review and editing; AF: Conceptualisation, writing-review and editing; CL: Resources, writing-review and editing; PS: Supervision, writing-review and editing; BC: Conceptualisation, supervision, writing-original draft.
medRxiv preprint DOI: https://doi.org/10.1101/2025.02.27.25323019
Posted: March 01, 2025, Version 1
Copyright: ⍰ Layan Fessler is now affiliated to Univ. Lille, CNRS, UMR 9193 – SCALab – Sciences Cognitives et Sciences Affectives, F-59000 Lille, France
Abstract
Objectives Promoting regular physical activity (PA) is essential in cardiac rehabilitation; yet many patients exhibit low levels of PA. In January 2022, the Improving Physical Activity (IMPACT) trial, a randomised controlled trial at the University Hospital of Geneva, was launched to promote PA in cardiac patients by targeting automatic approach tendencies towards exercise-related stimuli through a cognitive bias modification (CBM) intervention. This article examines the limited acceptance of this intervention, identifies potential barriers, and proposes strategies to improve future implementations.
Design Retrospective acceptance evaluation of a pre-registered clinical trial.
Setting The intervention was conducted in a cardiac rehabilitation centre in Switzerland.
Participants Sixty-eight cardiac rehabilitation patients (Mage = 57.76 ±□10.76 years, 87% male).
Intervention Patients received 12 CBM sessions over 6 weeks, designed to target approach-avoidance tendencies to exercise-related stimuli and improve PA levels.
Primary and secondary outcome measures Acceptance was assessed using behavioural (e.g., enrolment and engagement rates), cognitive (e.g., perceived effectiveness), and emotional (e.g., affective evaluation) indicators.
Results Of the 352 patients initially required, only 68 (19%) were enrolled. Among these 68, 17% completed the minimum number of CBM sessions, and 7% completed accelerometer-based PA measures during the week following discharge. These behavioural indicators of low acceptance were complemented by patients’ verbal reports, which showed cognitive (e.g., scepticism about the task relevance and perceived effectiveness of the intervention) and emotional (e.g., feelings of boredom and disinterest) barriers.
Conclusion The IMPACT trial’s inability to recruit and retain sufficient participants hindered the evaluation of its effectiveness in promoting PA among cardiac patients. These challenges highlight the importance of addressing patients’ perceptions of the task’s relevance and the satisfaction derived from it in order to improve intervention acceptance. Recommendations for improving behavioural, cognitive, and emotional acceptance are discussed, with the aim of increasing the effectiveness of rehabilitation programmes.
Strengths and limitations of this study
- Developing innovative intervention to promote PA in cardiac rehabilitation is essential.
- The intervention involved a computerised intervention targeting automatic precursors of PA behaviours
- Low enrolment and engagement rates limit the ability to assess the effectiveness of the intervention.
- The study identifies key barriers to acceptance, including cognitive (e.g., scepticism) and emotional (e.g., lack of motivation) factors, which hinder engagement.
- The retrospective, unplanned, and unsystematic nature of the measures may introduce recall bias, reducing the validity and accuracy of the reported indicators.
Introduction
Promoting physical activity (PA) in patients with cardiovascular diseases is critical, as regular PA is associated with significant health benefits, including reduced risks of mortality, less frequent cardiac events and hospitalisations, and improved quality of life.1–3 Despite these benefits, many patients exhibit low levels of PA, and previous interventions have failed to produce sustained behavioural changes.4–8 Therefore, it is essential to design, implement, and evaluate innovative interventions aimed at promoting PA behaviour in this population. Along with more reflective factors (e.g., intention to be physically active), recent research highlights the potential of targeting automatic precursors of PA, such as automatic action tendencies, to enhance interventions effectiveness.9 Automatic tendencies, grounded in dual-process accounts of human behaviour,10–12 refer to the automatic preparation of behavioural schemata, directing the organism’s action towards or away from an object.13 While extensively studied in psychopathology, – where approach tendencies are linked to addictions and avoidance tendencies to anxiety disorders14 –, emerging evidence suggests that they also play an important role in health behaviours, including PA,15–19 as they influence individuals’ inclination to engage in or to avoid effortful activities. Interventions based on Cognitive Bias Modification (CBM) may be particularly beneficial for patients recovering from an acute health event, such as a stroke or heart attack. This population is often more susceptible to fear of PA,20 which can result in avoidance behaviours.21 However, to our knowledge, no randomised controlled trials (RCTs) have been conducted to evaluate the efficacy of such interventions in patients with cardiovascular diseases. The Improving Physical Activity (IMPACT) trial was designed to address this critical gap in the literature.22
The main purpose of this article is to report the acceptance of the IMPACT trial, a CBM intervention in cardiac rehabilitation patients. We then proposed strategies to improve effectiveness of future interventions.
Methods
The IMPACT Trial
The protocol of the IMPACT trial, including its theoretical background and planned analyses, has been detailed elsewhere.22 In brief, the IMPACT trial was designed to target approach-avoidance tendencies towards exercise-related stimuli using a CBM intervention for patients enrolled in a cardiac rehabilitation program. Similar CBM interventions have shown promise in reducing certain unhealthy addictive behaviours such as alcohol consumption and smoking;23–26 and in promoting healthy behaviours such as healthy eating and PA.25,27–30
The IMPACT trial took place at the University Hospital of Geneva, in Switzerland and was approved by the Ethics Committee of Geneva Canton, Switzerland (reference number: CCER2019-02257). It initially consisted of a training programme of 12 sessions over 3 weeks (i.e., an average of 4 sessions per week) using a Visual Approach/Avoidance by the Self Task (VAAST31). However, logistical constraints – particularly a limited number of research assistants – restricted the participant pool to ambulatory patients and extended the programme duration to 6 weeks. Consequently, the intervention was adapted to 12 sessions over 6 weeks (i.e., 2 sessions per week). In the VAAST, patients responded to the format (i.e., portrait vs landscape format) of images depicting PA or sedentary behaviours by pressing a button on a touchscreen tablet to virtually move forward or backward in the image, thereby approaching or avoiding the stimuli. Patients were instructed to approach the image when it appeared in a portrait format and to avoid the image when it appeared in a landscape format, with the rule being counterbalanced across participants. Participants completed the VAAST in the presence of research assistants, not privately in front of a private computer over the Internet.
In the intervention group, patients received training in which 90% of images depicting PA were presented in the approach format (and 10% in the avoidance format), while 90% of images depicting sedentary behaviours were presented in the avoidance format (and 10% in the approach format). In contrast, the comparator group (placebo; sham controlled) approached and avoided PA and sedentary behaviours equally often (50% in the approach format and 50% in the avoidance format), so they were neither specifically trained to approach PA nor to avoid sedentary behaviours, compared to patients in the intervention group. At the end of the 6 weeks of rehabilitation, patients were asked to wear an accelerometer over 1 week to measure time spent in moderate-to-vigorous PA, which served as the primary outcome of the IMPACT trial. The main hypothesis was that the CBM intervention group would show higher levels of PA during the week following rehabilitation discharge compared to the placebo intervention group.
The Acknowledgement of the Failure
The IMPACT trial faced significant challenges, particularly regarding the acceptance of the intervention among patients, which is defined as a posteriori pragmatic evaluation of an activity (e.g., an intervention).32 As a low acceptance significantly hinders the success of an intervention, we deemed important to better understand the reasons underlying the low acceptance of our intervention. In this context, the Theoretical Framework of Acceptabilitya (TFA33) posits that acceptance can be assessed through behavioural indicators (e.g., attrition), cognitive indicators (e.g., scepticism about the task relevance and perceived effectiveness of the intervention), and emotional indicators (e.g., feelings of boredom, disinterest, or frustration). Understanding and addressing these behavioural, cognitive, and emotional dimensions of acceptance is critical to improving the design and implementation of interventions. Therefore, the main objective of the present article is to report on the low acceptance of the IMPACT intervention in cardiac rehabilitation patients, investigate potential reasons for these challenges, and propose strategies to improve the potential of future interventions.
Measures
In this Methods section, we describe only the behavioural, cognitive and emotional indicators of acceptance of the IMPACT trial. The methodology used to assess demographic, anthropometric and psychological outcomes is described elsewhere.34 It is important to note that the objective of this report was defined after the trial was halted due to insufficient participant engagement. As a result, qualitative indicators of acceptance were not included in the initial protocol (see Cheval et al.22). Feedback reports were compiled by research assistants during and after the intervention based on verbal reported from patients, which were not recorded or documented at the time. This reliance on the research assistants’ recollection of patients’ verbalisation may introduce recall bias, limiting the validity and accuracy of these indicators.
Behavioural Indicators
Non-Enrolment Rate
The non-enrolment rate was defined as the percentage of patients who attended the initial meeting with the research assistants and provided consent to participate in the IMPACT trial.
CBM Training Engagement Rate
The engagement rate in the CBM intervention was defined as the percentage of enrolled patients in the IMPACT trial (i.e., who agreed to participate in the IMPACT trial) who completed the minimum number of training sessions (i.e., 3) over the 6 weeks.
Engagement Rate in Post-Intervention PA Measures
PA engagement was defined as the percentage of patients with sufficient engagement to CBM training who provided complete PA measures in the week following the rehabilitation programme. PA was measured using an accelerometer (ActiGraph GT3X, Pensacola FL, USA) worn on the hip. PA measures were considered valid if the device was worn for at least 10 waking hours per day for at least four days, including one weekend day.35,36
For each of the 3 indicators, we compared the sample size achieved with the planned sample size of 352 to test our primary hypothesis. This was done to estimate the difference between the planned and actual sample sizes. For all these indicators, results were also described separately by group (intervention and placebo).
Cognitive Indicators
Perceived Effectiveness of the Intervention
Prior to the start of the intervention, perceived effectiveness of the intervention was assessed using the following item: “To what extent do you think that your PA behaviours will improve as a result of training on the computerised task?”22 Participants answered on a scale from 1 (not at all) to 5 (totally). During the intervention period, research assistants documented any spontaneous patient feedback related to perceived effectiveness of the intervention, with particular attention to verbal evaluations of the approach-avoidance training task.
Burden
Prior and during the intervention period, the burden of the intervention was assessed using patients’ spontaneous feedback. Research assistants documented any spontaneous patient feedback related to the perceived amount of effort that was required to participate in the intervention, with particular attention to verbal evaluations of the task perceived difficulty or demands.
Emotional Indicators
Attitudes Towards the Intervention
Prior and during the intervention period, affective attitudes towards the intervention (i.e., the extent to which participants perceived the intervention as something unpleasant or pleasant, boring or fun) were assessed using patients’ spontaneous feedback. Research assistants documented any spontaneous patient feedback related to their feelings about the intervention, with particular attention to verbal ratings of the level of boredom-fun and disinterest-interest towards the CBM task.
Resources Constrains
Resource limitations, such as insufficient equipment and a shortage of research assistants, were documented as potential barriers to the effective implementation of the intervention.
Results
Descriptive results of the sample characteristics are detailed elsewhere.34 Briefly, the sample was predominantly composed of older male adults (86.76% male; Mage = 57.76 ± 10.76 years) with acute coronary syndrome and a moderate intention to engage in PA (M = 5.51±1.34 on a scale of 1 to 10) at the start of the cardiac rehabilitation programme.
Behavioural Indicators
In terms of non-enrolment rate, over the first 18 months after the trial launch, 262 patients attended the initial meeting during which the intervention was described. Of them, 68 patients (26%) agreed to participate in the IMPACT protocol. Regarding engagement in the CBM training, 43 out of the 68 patients who started the protocol (63%) completed the minimum required three training sessions over six weeks. As for engagement in the accelerometer-based measures of PA, 17 of the 43 patients (40%) who completed sufficient CBM sessions provided valid accelerometers data. As a result, relative to the 252 participants screened for enrolment, 27% (n = 68) were enrolled, 17% (n = 43) completed the CBM training, and only 7% (n = 17) provided PA data during the week following discharge from the rehabilitation programme (see Figures 1 and 2)b. Overall, the results showed a high attrition rate across the entire protocol, with 75% of enrolled participants (i.e., 51/68) withdrawing from the study before the end of the protocol.
Figure 1.Flowchart of the Study
Abbreviations: CBM, cognitive bias modification; PA, physical activity. The expected sample size expected in Cheval et al.22 is shown in blue.
Figure 2.Study Progress Metrics
Abbreviations: Screened, patients who were screened and found eligible for the study; Enrolled, patients who were enrolled in the study; Complete CBM, participants who completed the minimum required CBM training (i.e., three sessions); Complete PA, participants who provided complete physical activity data during the week following discharge from the rehabilitation programme. Percentages were calculated on the screened patients.
Additional analyses were performed to examine the power achieved with a final sample of 17 patients (n = 10 patients in the intervention group and n = 7 in the placebo group). For a two-tailed t-test, an alpha of .05, an effect size of d = 0.35, and a total sample size of 17, we could expect a power of 10% (Figure S1).
We now present the results stratified by group (placebo and intervention). Thirty-five patients (51.5% of the total number of enrolled participants) were assigned to the intervention group. Regarding engagement in the CBM training, 25 out of the 35 patients who started the protocol (71%) completed the minimum required 3 training sessions over 6 weeks. When it came to engagement in the accelerometer-based measures of PA, only 10 of the 25 patients who completed the CBM training (40%) provided valid data. As a result, relative to the planned sample size of 176 patients per group, the intervention group reached 20% (n = 35) of the targeted enrolment, 14% (n = 25) for engagement in the CBM training, and a mere 6% (n = 10) for PA data in the week following discharge from the rehabilitation. Concurrently, 33 patients (48.5% of the total number of enrolled participants) were assigned to the placebo sham-controlled group. Regarding engagement in the CBM training, 18 out of the 33 patients who started the protocol (55%) completed the minimum required 3 training sessions over 6 weeks. When it came to engagement in the accelerometer-based measures of PA, 7 out of the 18 patients who completed the CBM training (39%) provided valid data. As a result, relative to the planned sample size of 176 patients per group, the placebo group only reached 19% (n = 33) of the targeted enrolment, 10% (n = 18) for engagement in the CBM training, and a mere 4% (n = 7) for collecting PA data in the week following discharge from the rehabilitation
Cognitive Indicators
Prior to the start of the intervention period, patients demonstrated a moderate perceived effectiveness of the intervention (M = 3.09 on a scale ranging from 1 to 5; Table S1 and Figure S2). Research assistants documented several patient complaints regarding the intervention perceived effectiveness, with many expressing scepticisms through comments such as, “I really don’t see how this task can help me to be more active,” “Why do we have to do stupid things when we are old?” Additionally, patients were reluctant to take on extra tasks during their rehabilitation, likely due to the demanding protocol (see Supplementary Material 3 for details of the rehabilitation protocol).
As for those who delivered the intervention, despite our willingness to integrate them, the research assistants were not fully integrated within the healthcare team (e.g., they did not attend weekly seminars or receive formal introductions as part of the clinical staff). This lack of integration may have reinforced the perception of the intervention as an additional burden for the patients, distinct from the regular rehabilitation program. This separation likely increased the perceived burden not only for patients but also for healthcare professionals, who may have considered it an extra task outside their standard clinical responsibilities. Specifically, medical staff – including medical doctors and physiotherapists – responsible for reminding patients about the intervention reported challenges, stating, “We don’t always have time to talk to the patients about the intervention with everything else we have to do as part of the rehabilitation programme.”
Emotional Indicators
Additional feedback from the patients highlighted significant issues related to boredom and disinterest. Patients frequently commented on the monotony of the tasks, stating, “The task is long and boring,” and “I’m not interested in doing anything on a tablet.” Furthermore, the visual presentation of the intervention was criticised, with remarks such as, “The images are depressing; it would be nicer with more colour.” As the intervention progressed, boredom seemed to increase as feedback about monotony became more frequent. It is possible that the novelty of the intervention, a key factor in maintaining interest and curiosity,37 wore off over time, contributing to the increased feelings of boredom.
Resources Constrains
A total of 8 tablets were available to deliver the CBM intervention to all patients, which created logistical challenges as the devices were occasionally in use when potential participants were available for recruitment. In addition, the 3 research assistants, who were students at the University of Geneva, were not available full-time, which limited their ability to recruit participants consistently.
Discussion
The main purpose of this report was to examine the limited acceptance of the IMPACT trial, a CBM intervention in cardiac rehabilitation patients, to identify the underlying factors and, ultimately, to propose strategies to improve future interventions. We observed a low enrolment, with only 27% of the patients screened agreeing to participate, and low adherence, as 17% of participants completed the intervention and just 7% provided PA data. Key barriers to acceptance were identified, including cognitive factors, such as scepticism about the task’s relevance and the perceived effectiveness of the intervention, and emotional factors, such as feelings of boredom and disinterest. These findings suggest that such an intervention could be difficult to implement in a clinical context without significant changes. However, it is important to interpret these results within the context of the unplanned and unsystematic nature of the measurements, which were based on memorable feedback that research assistants recalled from their interactions with patients.
Difficulty with Patient Enrolment: Reasons and Recommendations for Future Studies
The first major challenge we faced was patient enrolment, a critical factor in the success of clinical trials.33,38 In clinical research, studies are often abandoned due to insufficient recruitment,38 and a clinicians survey identified it as the most significant difficulty in conducting clinical research.39 Our initial goal was to recruit a sample of 352 patients to adequately test our primary hypothesis. This number included 252 patients, with a buffer to account for a potential loss to follow-up of 10–20% 1-week after the end of the intervention. To maintain sufficient statistical power for the 1-year follow-up, we also anticipated a dropout rate of 30–40%, resulting in a planned sample size of 352. However, after 18 months, we were only able to recruit 68 patients, which was only 19% of the expected sample size and 26% of the patients screened for enrolment. This significant shortfall meant that, even before considering acceptance issues, we lacked the statistical power necessary to accurately test our hypothesis.
One possible cognitive indicator of this low enrolment and acceptance was the perceived burden of the intervention protocol. Patients may have seen it as a research initiative rather than an integral part of their rehabilitation programme, and may have perceived it as an additional constraint. This lack of seamless integration into their daily health care routines may have diminished its perceived relevance and importance, further discouraging participation and reducing their willingness to prioritise the CBM programme within their rehabilitation schedules.
Additionally, the low acceptance of the intervention may be further suggested by participants’ moderate intention to engage in PA (i.e., the primary objective of the intervention) and their moderate perception of its effectiveness. These findings suggest that, even before starting the intervention, participants lacked a strong motivation to be physically active and were not fully convinced of the intervention’s potential to help them achieve this goal. Beyond the task itself, some patients expressed scepticism about the intervention’s effectiveness and were discouraged by its digital format, particularly the use of touchscreen tablets. This perceived lack of relevance of the task probably contributed to the difficulties in recruiting sufficient participants for the IMPACT trial. For those who agreed to participate in the protocol, resource constraints further complicated recruitment. The limited availability of tablets – only 8 in total – often meant that devices were unavailable when patients were ready to engage. Additionally, the absence of research assistants in the department at all times hindered consistent communication and support, which may have further discouraged participation. Collectively, these indicators contributed to patients’ low acceptance of the intervention.
Recommendations for potential improvements in enrolment and acceptance are primarily based on the characteristics identified above (see Table 1 for a summary). To optimise patient enrolment, we suggest that future studies integrate the protocol as a central, albeit optional, component of the patient’s rehabilitation treatment.40 It is essential that healthcare professionals working with the patients, such as medical doctors, physiotherapist, and nurses actively engage in the study protocol. Research assistants should be fully integrated into the healthcare team and recognised as essential members by both the healthcare professionals and the patients. For example, ensuring that all training sessions are well integrated into patients’ schedules can help reduce the perception of additional burden. Notably, the protocol should be presented in such a way that patients feel that participation requires minimal effort and time.33,40
Table 1.Potential Reasons for Failure and Recommendations for Future Studies
Difficulty with Patient Acceptance: Reasons and Recommendations for Future Studies
The second challenge encountered in the IMPACT trial was patients’ acceptance to the CBM intervention. Of the 68 patients enrolled in the study, only 43 completed the six weeks of training, and only 17 provided valid accelerometer data for the week following the rehabilitation programme. This high attrition rate (75%), combined with the already limited enrolment, resulted in a severely underpowered study, with an estimated power of 10% rather than the planned 90%. However, it is worth noting that such high attrition rates are not uncommon in studies evaluating CBM interventions targeting health behaviours such as gambling,41 smoking,42,43 or alcohol consumption.44 Attrition rates reported in these studies range from 43%44 to 90.1%.41 For instance, Snippe et al.41 discontinued their study due to a 90.1% attrition rate after three years of data collection. Importantly, many of these studies were conducted online and without supervision, which likely compounded the challenge of maintaining participants’ motivation.41 By contrast, in our study, the CBM tasks were supervised by research assistants, suggesting that even direct oversight may not fully mitigate challenges to engagement.
One possible explanation for the low acceptance observed in the trial lies in cognitive and emotional indicators. Patients reported difficulty understanding how completing the task could influence their PA behaviour. Additionally, patients expressed that the task felt long and boring and that the images were depressing. The combination of these cognitive (i.e., perceived effectiveness, task irrelevance) and emotional (i.e., boredom) barriers highlights the importance of designing CBM tasks that are perceived as both engaging and relevant.33,45 Regarding perceived uselessness, novel digital interventions, such as ours, may be perceived as ineffective in promoting behaviour change – in the absence of direct experiences with the targeted behaviour (e.g., practicing PA in a park, under the supervision of a trained coach).This low perceived usefulness could stem from a general underestimation of the impact of conditioning procedures on behaviour, coupled with an overestimation of individuals’ ability to consciously self-regulate and control their actions.9,46 Addressing this issue is particularly challenging because one of the core assumptions of digital interventions is their ability to influence behaviour without participants needed to fully understand or be aware of the underlying mechanisms.9 However, this assumption has been challenged by Van Dessel et al.,47 who suggest that the effectiveness of such interventions may also rely on participants’ awareness of the intervention’s goals, thereby engaging more reflective processes.
To increase engagement to training sessions, it is critical to ensure that the subjective value of the task outweighs its associated costs by addressing concerns of boredom and perceived lack of relevance (Table 1). One promising approach is to incorporate meaningful goals and consequences into the intervention.47 For instance, in the Antecedents Behaviours Consequences (ABC) training task, participants move an avatar away from unhealthy choices (e.g., fatty foods, cigarettes, alcohol) and towards healthier alternatives, while receiving positive feedback on personally relevant goals and outcomes, such as weight loss or improved health.48 Recent applications of ABC training in the context of addiction49 and PA28 have shown promise. Thus, this type of task may reduce the cognitive and emotional indicators of low acceptance of the IMPACT trial.
Perceived boredom can be attributed to at least two key factors: the repetitive nature of the task and the lack of intrinsically engaging content. While reducing the number of repetitions is challenging – especially since we had already reduced the duration of the CMB while ensuring an adequate dose of intervention – alternative strategies could be explored. For example, the Heartphone mobile application, developed by Conroy and Kim,50 exemplifies such a strategy. This application pairs images of PA with positive affective stimuli whenever users interact with their smartphone, fostering an automatic association between PA and positive feelings. This method offers frequent exposure to intervention content without placing additional burden on participants.
To further mitigate boredom, it is important to recognise that highly repetitive tasks can be disengaging, even over short periods of time.51 One promising solution is to incorporate more intrinsically motivating tasks. For example, the ABC training task can include animations and “vitality gauges” that can increase or decrease depending on the action (e.g., approaching or avoiding PA).28 The gamified nature of this task enhances its intrinsic value compared to traditional approach-avoidance tasks.52,53 Additionally, emerging technologies such as virtual reality offer opportunities to create more engaging and intrinsically motivating CBM tasks.54,55 Consequently, exploring tasks that naturally resonate with participants and leverage automatic responses to exercise-related stimuli represents a valuable avenue for future research. However, it is also crucial to ensure that participants, regardless of their age, socio-economic background, or health conditions, possess adequate skills to engage with digital interventions.56 To address this challenge, the co-production of the protocol by patients, health professionals, and researchers, alongside conducting pilot studies prior to full implementation, can help ensure that interventions are accessible, engaging, and effective for the target population.57,58
Conclusion
This article highlights the critical challenges in developing and implementing CBM-based PA interventions in cardiac rehabilitation patients. Despite these obstacles, our findings offer valuable insights into potential barriers and strategies for future interventions. Effective PA promotion requires approaches that are not only evidence-based but also acceptable, engaging, feasible, and integrated into both patients’ lives and health care routines. Addressing these challenges is the only way future interventional research could unlock the potential of PA in improving patients’ life.
Supporting information
Supplementary Materials[supplements/323019_file02.docx]
Data Availability
Deidentified data, data management, analysis codes and research materials have been made publicly available on Zenodo
Availability of Data and Materials
Deidentified data, data management, analysis codes and research materials have been made publicly available on Zenodo (https://doi.org/10.5281/zenodo.14937262).
Funding
L.F. is funded by the French Ministry of Higher Education, Research and Innovation. B.C. is supported by an Ambizione grant (PZ00P1_180040) from the Swiss National Science Foundation (SNSF).
Competing interests
None declared
Acknowledgements
The authors are grateful to Arbnor Hasani, Auriane Emeline Micheli, Zoë Alicia Piazza of the University of Geneva, and Anaïs Quossi of the University of Lyon for their implication on data collection. The authors are grateful to all study participants who volunteered, as well as the medical staff at Geneva University Hospitals for their assistance.
Footnotes
- a The TFA uses the term “acceptability” to refer to the evaluation of an intervention before, during, and after its implementation.33 However, some authors propose a distinction between the a priori evaluation of a task (acceptability) and the actual use of the task (acceptance).32 In this report we have used only the term “acceptance” to ensure internal consistency and readability. This choice was made because patient feedback was primarily reported during the intervention period, not before.
- b Of the planned sample size of 352 patients, we reached 19% (n = 68) of the target enrolment, 12% (n = 43) for engagement in the CBM training, and a mere 5% (n = 17) for PA data in the week following discharge from the rehabilitation programme
This pre-print is available under a Creative Commons License (Attribution 4.0 International), CC BY 4.0, as described at http://creativecommons.org/licenses/by/4.0/
References
58.Leon AC, Davis LL, Kraemer HC. The role and interpretation of pilot studies in clinical research. Journal of Psychiatric Research. 2011;45(5):626–629. doi:10.1016/j.jpsychires.2010.10.008
1.Haseler C, Crooke R, Haseler T. Promoting physical activity to patients. BMJ. 2019;366:l5230. doi:10.1136/bmj.l5230
2.Anderson L, Thompson DR, Oldridge N, et al. ExerciseLJbased cardiac rehabilitation for coronary heart disease. Cochrane Database of Systematic Reviews. 2016;(1)doi:14651858.CD001800.pub3
3.Dibben GO, Faulkner J, Oldridge N, et al. Exercise-based cardiac rehabilitation for coronary heart disease: a meta-analysis. European Heart Journal. 2023;44(6):452–469. doi:10.1093/eurheartj/ehac747
4.Kotseva K, Wood D, De Bacquer D, et al. EUROASPIRE IV: A European Society of Cardiology survey on the lifestyle, risk factor and therapeutic management of coronary patients from 24 European countries. European Journal of Preventive Cardiology. 2016;23(6):636–648. doi:10.1177/2047487315569401
5.Naito K, Izawa KP, Maeda N, Kasai Y, Iwama H. Characteristics of physical activity and sedentary behavior in patients undergoing outpatient cardiac rehabilitation. Scientific Reports. 2024/10/16 2024;14(1):24217. doi:10.1038/s41598-024-75362-9
6.Murray JM, Brennan SF, French DP, Patterson CC, Kee F, Hunter RF. Effectiveness of physical activity interventions in achieving behaviour change maintenance in young and middle aged adults: A systematic review and meta-analysis. Social Science & Medicine. 2017/11/01/ 2017;192:125–133. doi:10.1016/j.socscimed.2017.09.021
7.Tang L, Patao C, Chuang J, Wong ND. Cardiovascular risk factor control and adherence to recommended lifestyle and medical therapies in persons with coronary heart disease (from the National Health and Nutrition Examination Survey 2007-2010). American Journal of Cardiology. Oct 15 2013;112(8):1126–32. doi:10.1016/j.amjcard.2013.05.064
8.Grimmett C, Corbett T, Brunet J, et al. Systematic review and meta-analysis of maintenance of physical activity behaviour change in cancer survivors. International Journal of Behavioral Nutrition and Physical Activity. 2019/04/27 2019;16(1):37. doi:10.1186/s12966-019-0787-4
9.Larsen JK, Hollands GJ. Targeting automatic processes to reduce unhealthy behaviours: a process framework. Health Psychology Review. 2022/04/03 2022;16(2):204–219. doi:10.1080/17437199.2021.1876572
10.Conroy DE, Berry TR. Automatic affective evaluations of physical activity. Exercise and Sport Sciences Reviews. 2017;45(4):230–237. doi:10.1249/JES.0000000000000120
11.Cheval B, Boisgontier MP. The Theory of Effort Minimization in Physical Activity. Exercise and Sport Sciences Reviews. Jul 1 2021;49(3):168–178. doi:10.1249/JES.0000000000000252
12.Hofmann W, Friese M, Wiers RW. Impulsive versus reflective influences on health behavior: a theoretical framework and empirical review. Health Psychology Review. 2008/09/01 2008;2(2):111–137. doi:10.1080/17437190802617668
13.Friese M, Hofmann W, Wiers RW. On taming horses and strengthening riders: Recent developments in research on interventions to improve self-control in health behaviors. Self and Identity. 2011/07/01 2011;10(3):336–351. doi:10.1080/15298868.2010.536417
14.Loijen A, Vrijsen JN, Egger JIM, Becker ES, Rinck M. Biased approach-avoidance tendencies in psychopathology: A systematic review of their assessment and modification. Clinical Psychology Review. 2020;77:101825. doi:10.1016/j.cpr.2020.101825
15.Cheval B, Sarrazin P, Pelletier L. Impulsive approach tendencies towards physical activity and sedentary behaviors, but not reflective intentions, prospectively predict non-exercise activity thermogenesis. Plos One. 2014;9(12):e115238. doi:10.1371/journal.pone.0115238
16.Cheval B, Sarrazin P, Isoard-Gautheur S, Radel R, Friese M. Reflective and impulsive processes explain (in)effectiveness of messages promoting physical activity: A randomized controlled trial. Health Psychology. 2015;34(1):10–19. doi:10.1037/hea0000102
17.Cheval B, Sarrazin P, Pelletier L, Friese M. Effect of retraining approach-avoidance tendencies on an exercise task: A randomized controlled trial. Journal of Physical Activity & Health. 2016;13(12):1396–1403. doi:10.1123/jpah.2015-0597
18.Wang C, Zhang G. Examination of the Consistency Effect between Physical Activity Levels and Approach-Avoidance Behaviors. Psychology of Sport and Exercise. 2025/02/13/ 2025:102825. doi:10.1016/j.psychsport.2025.102825
19.Wang Y-B, Zhang Z, Mao Z-X. You don’t know why you (don’t) exercise: The relationship between automatic processes and physical activity (or sedentary behavior): A meta-analysis. Scandinavian Journal of Psychology. 2024;65(3):501–510. doi:10.1111/sjop.12991
20.Goubran M, Farajzadeh A, Lahart IM, Bilodeau M, Boisgontier MP. Physical activity and kinesiophobia: A systematic review and meta-analysis. medRxiv. 2024:2023.08.17.23294240. doi:10.1101/2023.08.17.23294240
21.Witthöft M, Hiller W. Psychological approaches to origins and treatments of somatoform disorders. Annual Review of Clinical Psychology. 2010;6:257–283.
22.Cheval B, Finckh A, Maltagliati S, et al. Cognitive-bias modification intervention to improve physical activity in patients following a rehabilitation programme: protocol for the randomised controlled IMPACT trial. BMJ open. 2021;11(9):e053845. doi:10.1136/bmjopen-2021-053845
23.Boffo M, Zerhouni O, Gronau QF, et al. Cognitive Bias Modification for Behavior Change in Alcohol and Smoking Addiction: Bayesian Meta-Analysis of Individual Participant Data. Neuropsychology Review. 2019;29(1):52–78. doi:10.1007/s11065-018-9386-4
24.Jones EB, Sharpe L. Cognitive bias modification: A review of meta-analyses. Journal of Affective Disorders. 2017;223:175–183. doi:10.1016/j.jad.2017.07.034
25.Van Dessel P, Hughes S, De Houwer J. Consequence-Based Approach-Avoidance Training: A New and Improved Method for Changing Behavior. Psychological Science. 2018;29(12):1899–1910. doi:10.1177/0956797618796478
26.Wiers RW, Eberl C, Rinck M, Becker ES, Lindenmeyer J. Retraining Automatic Action Tendencies Changes Alcoholic Patients’ Approach Bias for Alcohol and Improves Treatment Outcome. Psychological Science. 2011;22(4):490–497. doi:10.1177/0956797611400615
27.Aulbach MB, Knittle K, Haukkala A. Implicit process interventions in eating behaviour: a meta-analysis examining mediators and moderators. Health Psychology Review. 2019;13(2):179–208. doi:10.1080/17437199.2019.1571933
28.Maltagliati S, Sarrazin P, Muller D, et al. Improving physical activity using a single personalized consequence-based approach-avoidance training: Effects on self-reported behaviors, attitudes, and choices. Psychology of Sport and Exercise. 2024:102565. doi:10.1016/j.psychsport.2023.102565
29.Preis MA, Zellerhoff M, Brockmeyer T. Approach bias modification training to increase physical activity: A pilot randomized controlled trial in healthy volunteers. Journal of Health Psychology. 2020;26(13):2470–2486. doi:10.1177/1359105320913936
30.Cheval B, Sarrazin P, Pelletier L, Friese M. Effect of retraining approach-avoidance tendencies on an exercise task: A randomized controlled trial. Journal of Physical Activity & Health. 2016;13(12):1396–1403. doi:10.1123/jpah.2015-0597
31.Rougier M, Muller D, Ric F, et al. A new look at sensorimotor aspects in approach/avoidance tendencies: The role of visual whole-body movement information. Journal of Experimental Social Psychology. 2018;76:42–53. doi:10.1016/j.jesp.2017.12.004
32.Alexandre B, Reynaud E, Osiurak F, Navarro J. Acceptance and acceptability criteria: a literature review. Cognition, Technology & Work. 2018;20(2):165–177. doi:10.1007/s10111-018-0459-1
33.Sekhon M, Cartwright M, Francis JJ. Acceptability of healthcare interventions: an overview of reviews and development of a theoretical framework. BMC Health Services Research. 2017;17(1):88. doi:10.1186/s12913-017-2031-8
34.Fessler L, Tessitore E, Craviari C, et al. Motivational and emotional correlates of physical activity and sedentary behavior after cardiac rehabilitation: an observational study. BMC Sports Science, Medicine and Rehabilitation. 2024;16(1):209. doi:10.1186/s13102-024-00997-0
35.Evenson KR, Terry Jr JW. Assessment of Differing Definitions of Accelerometer Nonwear Time. Res Q Exerc Sport. 2009/06/01 2009;80(2):355–362. doi:10.1080/02701367.2009.10599570
36.Matthews CE, Hagströmer M, Pober DM, Bowles HR. Best practices for using physical activity monitors in population-based research. Medicine & Science in Sports & Exercise. 2012;44(1 Suppl 1):S68–76. doi:10.1249/MSS.0b013e3182399e5b
37.Erdemli A, Audrin C, Sander D. Is interesting knowledge a reward? An integrative model of emotion, reward and appraisal frameworks of epistemic curiosity. 2024;doi:10.31234/osf.io/kz2a7
38.Fogel DB. Factors associated with clinical trials that fail and opportunities for improving the likelihood of success: A review. Contemporary Clinical Trials Communications. 2018;11:156–164. doi:10.1016/j.conctc.2018.08.001
39.Cullati S, Courvoisier DS, Gayet-Ageron A, et al. Patient enrollment and logistical problems top the list of difficulties in clinical research: a cross-sectional survey. BMC Medical Research Methodology. 2016;16(1):50. doi:10.1186/s12874-016-0151-1
40.Conley S, O’Connell M, Linsky S, et al. Evaluating Recruitment Strategies for a Randomized Clinical Trial with Heart Failure Patients. Western Journal of Nursing Research. 2020;43(8):785–790. doi:10.1177/0193945920970229
41.Snippe L, Boffo M, Galvin H, et al. How We Lost 90% of Participants on a Bad Bet: Results from a Pilot Randomized Controlled Trial on Cognitive Bias Modification in Problem Gamblers. Journal of Gambling Studies. 2024;40(2):521–554. doi:10.1007/s10899-023-10263-6
42.Wen S, Larsen H, Boffo M, et al. Combining Web-Based Attentional Bias Modification and Approach Bias Modification as a Self-Help Smoking Intervention for Adult Smokers Seeking Online Help: Double-Blind Randomized Controlled Trial. JMIR Ment Health. 2020;7(5):e16342. doi:10.2196/16342
43.Wittekind CE, Lüdecke D, Cludius B. Web-based Approach Bias Modification in smokers: A randomized-controlled study. Behaviour Research and Therapy. 2019;116:52–60. doi:10.1016/j.brat.2018.12.003
44.Wiers RW, Houben K, Fadardi JS, van Beek P, Rhemtulla M, Cox WM. Alcohol Cognitive Bias Modification training for problem drinkers over the web. Addictive Behaviors. 2015;40:21–26. doi:10.1016/j.addbeh.2014.08.010
45.Maltagliati S, Sarrazin P, Fessler L, Lebreton M, Cheval B. Why people should run after positive affective experiences instead of health benefits. Journal of Sport and Health Science. 2022;13(4):445–450. doi:10.1016/j.jshs.2022.10.005
46.Strack F, Deutsch R. Reflective and Impulsive Determinants of Social Behavior. Personality and Social Psychology Review. 2004;8(3):220–247. doi:10.1207/s15327957pspr0803_1
47.Van Dessel P, Hughes S, De Houwer J. How Do Actions Influence Attitudes? An Inferential Account of the Impact of Action Performance on Stimulus Evaluation. Personality and Social Psychology Review. 2019/08/01 2019;23(3):267–284. doi:10.1177/1088868318795730
48.Wiers RW, Van Dessel P, Köpetz C. ABC Training: A New Theory-Based Form of Cognitive-Bias Modification to Foster Automatization of Alternative Choices in the Treatment of Addiction and Related Disorders. Current Directions in Psychological Science. 2020;29(5):499–505. doi:10.1177/0963721420949500
49.Pan T, Szpak V, Laverman J, et al. ABC-Training for Alcohol Use During a Voluntary Abstinence Challenge: A Randomized Controlled Trial. International Journal of Mental Health and Addiction. 2024/11/04 2024;doi:10.1007/s11469-024-01409-7
50.Conroy DE, Kim I. Heartphone: Mobile evaluative conditioning to enhance affective processes and promote physical activity. Health Psychol. 2021;40(12):988–997. doi:10.1037/hea0000886
51.Cummings ML, Gao F, Thornburg KM. Boredom in the Workplace:A New Look at an Old Problem. Human Factors. 2016;58(2):279–300. doi:10.1177/0018720815609503
52.Cugelman B. Gamification: What It Is and Why It Matters to Digital Health Behavior Change Developers. JMIR Serious Games. 2013;1(1):e3. doi:10.2196/games.3139
53.Schakel L, Veldhuijzen DS, Middendorp Hv, et al. The effects of a gamified approach avoidance training and verbal suggestions on food outcomes. Plos One. 2018;13(7):e0201309. doi:10.1371/journal.pone.0201309
54.Reer F, Wehden L-O, Janzik R, Tang WY, Quandt T. Virtual reality technology and game enjoyment: The contributions of natural mapping and need satisfaction. Computers in Human Behavior. 2022;132:e107242. doi:10.1016/j.chb.2022.107242
55.Lee J, Kim J, Choi JY. The adoption of virtual reality devices: The technology acceptance model integrating enjoyment, social interaction, and strength of the social ties. Telematics and Informatics. 2019;39:37–48. doi:10.1016/j.tele.2018.12.006
56.Chalabaev A, Cheval B, Maltagliati S, Saoudi I, Sniehotta FF. Beyond Individual Cognitions: Time for Intervention Science to Focus on Health Context and Audience. Journal of Physical Activity and Health. 2023;20(6):465–470. doi:10.1123/jpah.2023-0072
57.Smith B, Williams O, Bone L, Collective tMSWC-p. Co-production: A resource to guide co-producing research in the sport, exercise, and health sciences. Qualitative Research in Sport, Exercise and Health. 2023;15(2):159–187. doi:10.1080/2159676X.2022.2052946