Factors influencing the implementation of a home-based rehabilitation program in frail older patients after cardiac surgery: a qualitative study using reflexive thematic analysis

Abstract

Background

Frail older adults undergoing cardiac surgery are at increased risk of functional decline. As part of developing a follow-up program, Nantes University Hospital selected Vivifrail, a structured, multicomponent exercise program designed to improve strength, balance, and mobility. This study aimed to evaluate the barriers and drivers conducive to the adoption of Vivifrail among frail older patients who had recently undergone cardiac surgery.

Methods

We performed a qualitative study using semi-structured interviews between March and December 2024. Eligible patients had recently undergone cardiac surgery in Nantes University Hospital, were aged 75 years and older, with a baseline SPPB score < 9. Patients were interviewed between weeks 5 and 6 post-discharge. Interview transcripts were analyzed using reflexive thematic analysis.

Results

Among 27 eligible patients, 23 were interviewed. The first 9 interviews were conducted using a preliminary version of the guide in order to help us enrich the final version. To ensure methodological rigor and limit interpretation bias, only interviews conducted with the final guide were included in the analysis. As a result, data from 14 interviews were analyzed until data saturation was reached. The average age of participants was 79 years old (± 4), and 71% were men. Median SPPB scores varied between 7 [6–8] points at baseline and 10.5 [9–11] points post-intervention. Three main themes emerged from the interviews. First, patients frequently described post-operative challenges such as impaired physical condition, discomfort, and fatigue. Second, despite these difficulties, the vast majority (11 of 14) demonstrated moderate to strong adherence. Upon returning home, the Vivifrail program served as a framework for patients to start physical activity. The support of healthcare professionals and family also appeared to be decisive. Third, most patients (11 of 14) reported adapting their physical activty according to the recovery of their abilities, particularly through performing activities of daily living, drawing on personal and environmental resources.

Conclusions

The Vivifrail program is a promising approach for guiding frail older adults who have recently undergone cardiac surgery towards establishing a regular physical activity routine. Given the exploratory and qualitative nature of our study, our findings are not representative of the entire frail older population but may be transferable to other similar patients in comparable care settings.

Trial registration

Clinical trial number not applicable. The study was approved by the Nantes Health Ethics Group (GNEDS), Nantes, France (N°24–22-02–273), and registered on OSF: https://doi.org/10.17605/OSF.IO/7HTX2

Supplementary Information

The online version contains supplementary material available at 10.1186/s13102-026-01696-8.

Background

The aging population and the rising prevalence of chronic diseases, particularly cardiovascular, will lead to a growing number of older individuals undergoing cardiac surgery [1]. This population is particularly vulnerable, due to the high frequency of comorbidities and frailty. These factors together increase the risk of functional decline and all-cause mortality by two to three times [2–5].

Timely initiation of cardiac rehabilitation (CR) is critical to counteract this decline, improving physical activity (PA), quality of life, and reducing mortality [6–8]. Despite these benefits, older adults remain under-referred [9, 10], particularly among frail individuals [11]. There are many documented reasons for this, including failure to refer 50% of patients directly to rehabilitation upon discharge, advanced age, and low baseline capacity exercise. Other reasons include geographical and logistical barriers, such as long distance to rehabilitation centers [12, 13]. Home-based rehabilitation programs have emerged to overcome these barriers [14], offering rapid enrolment, flexible schedules, integration into home routines, accessibility, more individualized exercises, and privacy [11]. They appear to be as effective and safe as conventional CR in improving clinical outcomes and patients' quality of life [15].

As part of developing a follow-up program for frail older patients after cardiac surgery at Nantes University Hospital, we selected the Vivifrail program. Recommended as part of the WHO's ICOPE approach, Vivifrail stands out for its comprehensive approach to frailty, addressing physical, functional, cognitive and nutritional dimensions, rather than focusing solely on cardiovascular conditioning [16]. Compared to other programs, it offers a 12-week, structured and individualized exercises program tailored to older adults’ functional status using short performance battery test (SPPB) scores. It includes up to six different programs with specific adaptations for patients at risk of falling; combining daily walking and muscle strengthening, balance work, as well as stretching 3 to 4 times a week [17].

Understanding how patients engage with home-based CR is essential to estimate the true effectiveness of these programs. While these barriers and drivers have been explored in home-based CR post myocardial infarction [18], no studies, to our knowledge, have specifically focused on the frail older population after cardiac surgery. A qualitative methodology seemed particularly appropriate for this exploratory descriptive study, as it allows for a deeper understanding of the barriers and drivers conducive to the adoption of this program among our patients' population. We conducted semi-directive interviews and performed reflexive thematic analysis to identify patient centered mechanisms supporting home-based CR engagement.

Methods

We performed a qualitative study using semi-structured interviews. The results are reported in accordance with the COREQ guidelines [19]. For further details, please see Supplement 1 and 2: COREQ checklist and Standards for reporting qualitative research.

Purposive sampling was used to include all consecutive eligible frail older patients exposed to the Vivifrail program during the study period. This pragmatic approach allowed the inclusion of participants with diverse profiles, reflecting variation in type of cardiac surgery, functional status (as assessed by SPPB scores), and levels of social support (living alone or with family support). This diversity enabled the exploration of a broad range of experiences related to program adoption and perceived barriers. Main inclusion criteria were: 75 years or older, recent cardiac surgery in university teaching hospital in Western France (University Hospital Nantes, France), short performance battery test (SPPB) score < 9/12 at hospital discharge, informed of the study and having given their consent. Non-inclusion criteria were mostly related to inability to take part in the Vivifrail rehabilitation program (i.e.: persistent uncontrolled arrhythmia, hypertension or orthostatic hypotension after surgery, acute endocarditis/pericarditis, acute thromboembolic disease) and/or patient under guardianship or curatorship.

Informed consent to participate was obtained from all of the participants in the study. Patients who agreed to participate performed a standardized assessment (SPPB, ICOPE, handgrip) at discharge, followed by a second assessment halfway through the program (6 weeks post discharge) and a final at the end (12 weeks post discharge). Each patient was given a copy of their specific exercise protocol before being discharged from hospital. This allowed them to track the number of sessions they completed each week.

Interviews were conducted by phone between 5 and 6 weeks after surgery. They form part of the patient's home follow-up and provide information for the outpatient consultations at 6 and 12 weeks. Interviews at weeks 5–6 post-discharge were timed to balance recall and experience. By then, patients had moved past immediate post-operative challenges, allowing early barriers to be identified, while also having enough exposure to the Vivifrail program to reflect on routine formation and integration into daily life. Remote interviews were selected to maintain patient accessibility and streamline organization. All interviews were conducted by the same interviewer, and only the data needed to meet the research objectives were collected.

Interviews were recorded by dictaphone and transcribed for later analysis.

We developed an interview guide based on knowledge of the discipline and a review of literature. To construct the interview guide, we also tested the initial version with 9 patients, as well as two physiotherapists practicing in the cardiology department or at the day clinic. To ensure methodological rigor and limit interpretation bias, only interviews conducted with the finalized and standardized version of the interview guide were included in the analysis. Those preliminary interviews were therefore excluded because they were based on an earlier version of the guide, which could have introduced inconsistencies in the data.

The finalized interview guide covered the following topics:

• Adherence to the rehabilitation program: records the participation rate from hospital discharge to week six to assess the extent to which exercises have been completed.

• Barriers encountered by patients: aims to identify and study any difficulties or obstacles encountered in carrying out the exercises.

• Drivers encouraging patient adherence: questions the factors influencing the implementation and regular continuation of the program.

For further details, please see Supplement 3: Patient Semi-Structured Interview Topic Guide.

Interviews were performed by one researcher (MC, female, MSc candidate), who was not known to the patients and did not work in the Cardiothoracic Surgery Department. The interviewer received training in both qualitative interviewing (i.e. open-ended questioning, active listening, probing without leading participants) and in the Vivifrail program to ensure consistent understanding of the intervention. This study was situated within a contextualist constructivist epistemology, in which participants’ accounts were understood as meaningful construction of their experience, while also being embedded with broader social context. Consistent with this perspective, interview transcripts were analyzed using pragmatic thematic analysis informed by Braun and Clarke’s six-phase framework [20]. Within this approach, themes are not treated as objective entities, but as interpretative patterns developed through active engagement between researchers and data. The analysis included the following phases:

1. Familiarization: Researchers read transcripts multiple times to immerse themselves in the data and note initial ideas.

2. Coding: The transcripts were coded independently by 2 of the coauthors (MC, AK). Meaningful segments were labeled systematically, with codes reflecting both participants’ words and interpretive insights.

3. Theme development: Codes were grouped into preliminary themes, capturing patterns across the dataset.

4. Theme review: Regular meetings were held to compare the data through analyst triangulation, harmonize coding interpretation, and decide by consensus which information would be retained for analysis. Themes were discussed between researchers and cross-checked within the transcripts before being defined into an indexing scheme.

5. Defining and naming themes: Finally, the researchers discussed and synthesized themes, which are reported in this paper. Final themes were clearly defined and named to accurately represent participants’ perspectives while minimizing researcher bias.

Reflexive positioning was maintained through regular team discussion. Prior to data collection, the research team documented their assumptions regarding patient experiences, as well as potential barriers and drivers related to the intervention. These reflections were revisited during team discussions to limit interpretation bias and enhance credibility. This collaborative process allowed the team to compare interpretations and enhance reflexive rigor while maintaining a reflexive awareness of how researchers' perspectives shaped the analysis.

Data saturation was assessed through an iterative process during data collection and analysis. After each set of interviews, the research team reviewed the emerging themes and codes. Saturation was considered achieved when no new themes or subthemes emerged from the data, and when information became repetitive and consistent across participants. In line with reflexive thematic analysis, we did not treat data saturation as a measurable or objective endpoint. Instead, we considered data adequacy in terms of whether the dataset offered sufficient depth, complexity, and variation to meaningfully address the patients’ experience. For further details, please see Supplement 4: Saturation Grid. Patients were informed that illustrative citations from their interview could be used to substantiate scientific publications (after translation), and they all agreed to this.

Results

Population

Between 10/03/2024 and 20/12/2024, we identified 27 eligible patients. One patient refused to participate, stating that he did not follow the exercise booklet and saw no point in talking to us. Two patients could not be reached for interviews despite several reminders, and one agreed to answer, but the call could not be completed because of his dysarthria. We thus obtained 23 interviews. As described above, the first nine interviews were conducted using a preliminary version of the interview guide in order to help us enrich the final version. Therefore, we retained the data from 14 interviews for analysis.

Patient characteristics are presented in Table 1.

Table 1.

Characteristics of subjects

N°	Sex	BMI	Surgery type	Type of physical exercise program	SPPB score (baseline)	SPPB score (post-intervention)	Completion rate	Adherence	Social support
1	F	26.12	CABG × 4	C +	8	11	100	Strong	Yes
2	M	25.34	Bentall + maze	A	3	9	95	Strong	Yes
3	F	21.10	AVR + CABG × 4	C	7	11	0	Low	No
4	M	24.24	TAVI	C	8	8	100	Strong	Yes
5	M	24.68	AVR + CABG × 2	C	7	10	60	Moderate	Yes
6	F	24.65	AVR	B +	4	9	66	Moderate	Yes
7	M	27.76	CABG × 2	C +	7	11	60	Moderate	Yes
8	M	21.53	AVR + MVR	C	8	9	30	Low	No
9	M	25.10	MVR + maze	B +	6	9	100	Strong	Yes
10	M	23.96	MVR + CABG × 4	C	7	12	87	Strong	Yes
11	M	28.44	AVR	C	8	11	80	Strong	Yes
12	M	25.35	CABG × 4	B +	5	9	100	Strong	Yes
13	F	18.13	Myxoma exeresis (left atrium)	B +	6	5	15	Low	Yes
14	M	28.73	CABG × 3	A	7	12	100	Strong	Yes

Depending on the functional level and risk of falls that each participant has, up to six different types of physical exercise programs, A = Severe limitation (SPPB 0–3); B = Moderate limitation (SPPB 4–6); B + = Moderate limitation and fall risk; C = Mild limitation (SPPB 7–9); C + = Mild limitation and fall risk; D = Minimal limitation (SPPB 10–12)

MVR or AVR Mitral or aortic valve replacement, CABG Coronary artery bypass grafting

The mean age was 79 years old (± 4), and 71% were men. The surgeries procedures varied and included mostly coronary artery bypass grafting (50%) and mitral or aortic valve replacement (35%). Frequent comorbidities included hypertension (50%), dyslipidemia (50%), and sleep apnea syndrome (14%). Median SPPB scores varied between 7 [6-8] points at baseline and 10.5 [9-11] points post-intervention. We considered that 5 physical exercise sessions per week represent a 100% adherence rate, as stipulated in the Vivifrail program. Thus, we classified patients into three groups according to their adherence rate (strong 80–100%; moderate 50–80%; low < 50%). The mean completion rate was 71 (± 34) %. We note that 5 patients completed the program in its entirety, whereas 3 made minimal use of it.

Thematic analysis

Three main themes emerged from the interviews. Reflexive engagement informed the coding process, particularly by facilitating a shift from a binary barriers/drivers framework to a conceptual framework comprising three dimensions that interact and vary in their presence during the post-intervention period. This conceptual framework summarizing the data is presented in Fig. 1. The average duration of the interviews was 15 min (range 9 to 23 min).

Fig. 1.

Conceptual framework of the three main themes

Theme 1: post-operative challenges

As part of their cardiac surgery treatment, patients spent an average of fifteen days in hospital, ranging from five days to thirty-one days. Many patients reported starting the rehabilitation program between 0 and 2 additional weeks after returning home, primarily due to their physical condition, discomfort, and fatigue.

The symptoms mentioned are as follows:

• Weakness: “I still have weakness in one leg”(F6); “I no longer had the strength to push myself up with my legs”(M10); “I have no strength left”(M12).

• Pain: “I have chest pain”(F1); “I’m still in pain right now”(M4); “There was still pain after the intervention”(M7).

• Fatigue: “I'm still tired”(F6); “I get tired very quickly”(M8); “It tires me out. I might as well go to bed afterwards”(F13)

• Balance: “I am a little off balance”(M2); “I have poor balance”(M9); “It completely throws me off balance”(F13).

• Dyspnea: “I was more out of breath than usual”(M2).

• Wound healing: “I couldn't do my program right away because the scars weren’t closed”(F6).

Another barrier that is often mentioned is the weather: “There were times when I couldn’t go outside because of bad weather”(M7); “Bad weather for going for a walk (…) is annoying”(M9); “It's raining, and going for a walk for example, isn’t ideal”(M11). These responses appear to be influenced by the patients’ environment. Weather conditions, in particular, seemed to affect patients living in confined spaces, such as small apartments. “Anything involving walking for example, since I live in an apartment, I can't do it, I just can’t, it's not possible, it makes me dizzy, I can't (…) it completely throws me off balance”(F13). Patients who had sufficient space were able to replace outdoor walking sessions with indoor walking or other physical activities.

Theme 2: establishing a routine

Although the patients encountered challenges in carrying out the rehabilitation program, the vast majority demonstrated moderate (n = 3) to strong (n = 8) adherence. Only three patients dropped out of the program due to intensive private physical therapy, difficulty managing their schedule, or extreme fatigue. Upon returning home, the rehabilitation program served as a framework for patients to start physical activity. “I check off each exercise I do every day in detail”(M10); “It’s good because it gives ideas of what kind of movements to do”(M5). Adherence appears to be supported by the positive features of the program, as consistently highlighted by the participants. The main qualifies mentioned include:

• Exercises’ variety: “It gets the arms, legs and stretching involved—everything” (M9); “It was very varied!”(M11); “your little bookset was quite comprehensive”(M7).

• Appropriate difficulty level: “It’s easy to do” (F6); “It's not that it’s very difficult”(M8).

• Low level of constraint: “It's not too demanding” (M12); “It doesn’t require much… it doesn’t take much time”(M14).

While these qualities were appreciated, many patients adapted the program to better fit their personal limitations and needs. For example: “Instead of doing 12 chair rises, I do 10, because sometimes it makes me short of breath, so I do a bit less”(M4); “I did it a bit longer because it worked the legs and thighs”(M5); “I don’t do the exercises all in a row”(F13). These adaptations reflect a sense of ownership and autonomy, suggesting that patient value being active participants in their rehabilitation process.

Conversely, the role of the healthcare system or some professionals was occasionally perceived as restrictive or discouraging: “I was told not to do too much, so I do nothing”(M14); “My cardiologist told me to do less and less”(M5); “The surgeon told me not to lift weights”(F1); “They denied me convalescent center”(F3); “I don’t really have time because of all the medical appointments"(F13).

Theme 3: transfer and maintenance through everyday life

Despite some delay in starting the program, most patients were able to maintain their efforts and remain consistent in performing the exercises. Only three patients dropped out of the program due to fatigue, external medical appointments, and private physiotherapy support. These patients relied on favorable motivational factors, such as:

• Perceived capabilities: “I am improving day by day (…) I am less out of breath than I was”(M2); “I know I am feeling good (…) now I do it as if I never had anything” (M11).

• Visible progress in daily life: “At first it was hard to do the cleaning but now I do it as if nothing had happened”(F1); “The result was there so that means it worked”(F3); “I can go into the garden whereas at first I couldn’t go at all”(M9); “There is a lot of improvement, this is also what my wife notices”(M9).

• Self-efficacy and improvement over time: “What motivated me was wanting to try to get out of it”(M2); “I was doing pretty well so I tried to do things a little more difficult”(M5); “I would like to get back into Olympic shape!”(F6); “I tried to see if I was able to do the exercises”(M7).

The social environment was also an important positive factor for the patients: “I also have my children who are not far from home so that is an advantage”(M11); “My husband helps me with the dishes and things like that”(F1); “my neighbor comes and he does my lawn. So now I go with him when he does it”(M2); “On Thursdays we play cards with the old-timers club”(M8); “When I am with my wife to do shopping I am walking”(M9). While the role of the family thus appears evident for the patients, the role of private practice physiotherapists remain ambivalent: “When I went to the physiotherapist he sayed I could do the exercises so I did some”(F3); “If I hadn’t had physiotherapy I would have done more”(F3); “I have a neighbor next door who has a physiotherapist coming home and it’s the exact same things that I do (…) so I didn’t ask for a physio”(M4); “I do a lot of things by myself so I don’t need it”(F1); “I did a lot of things with him (the physiotherapist)”(M11).

As patients progressed through the program, some began to deviate by substituting exercises with daily life activities, for example: “As I do the cleaning, I consider this being part of the rehabilitation”(F1); “For example, there is an exercise that I did the six first days and then I didn’t do it anymore because I did walking” (M5); “If I did less some of the exercises it’s because I felt I was doing something else to replace”(M5).

Discussion

The purpose of this study was to identify the barriers and drivers conducive to the adoption of the Vivifrail program among frail older patients who had recently undergone cardiac surgery.

The main findings were consistent with data published in the literature. Adoption of regular physical activity through Vivifrail program depends on the patients’ perceived capabilities, perceptions on PA, and previous activity levels. Post-operative physical limitations and fatigue were the main obstacles. On the other hand, motivation was mainly based on regaining abilities, particularly through performing activities of daily living. The Vivifrail program appeared feasible and acceptable, both in terms of the structure it provided to patients as well as the possibilities for adaptation it offered them. The role of healthcare professionals and family support also appeared to be decisive.

Cardiac surgery is often followed by a temporary decline in patients’ functional status. This decline can be attributed to a combination of factors including postoperative pain, discomfort, depressive symptoms, frailty, impaired sleep quality and fatigue [21–23]. Their symptoms may considerably reduce quality of life and initially prevent a return to their pre-procedure functional levels [21]. It can be particularly harmful for older patients as even shorts periods of physical inactivity have detrimental effects on muscle mass and functional capacity [22]. Recovery after cardiac surgery follows a gradual course. The first three months following surgery represent the period during which the most recovery occurs [24], with improvement occurring at different time points depending on the domains assessed [i.e. 1 to 3 weeks: anxiety, depression, mental quality of life, dyspnea; 4 to 6 weeks: 6-min walk test, physical quality of life] [25]. Short-term programs such as Vivifrail have demonstrated efficacy in this context. Studies have shown that 4 to 6 weeks of targeted intervention can reduce frailty, improve functional capacity and autonomy in older adults [17, 26]. These findings highlight the critical importance of initiating CR at the appropriate time, underscoring the need for structured patient support to optimize both functional and psychological outcomes during this pivotal period.

The completion rate observed in our study aligns with those reported in previous home-based rehabilitation programs: 42% in the New Zealand Otago trial and 47% in the LiFE [27, 28]. However, it is somewhat lower than the 79% rate reported by the original Vivifrail study. Several factors may explain this discrepancy. In that study, participants received two telephone calls during the intervention period to reinforce adherence, answer questions, and overcome obstacles. Additionally, family members or caregivers were systematically involved and trained to supervise exercise sessions [17]. In contrast, our protocol did not include intermediate contact between hospital discharge and follow-up. Patients were solely responsible for monitoring their program, although relatives were sometimes informed of the objectives prior to discharge. This design was based on the relatively favorable functional profile of our cohort, characterized by good physical reserves, rapid functional recovery, and minimal limitations on activities [29]. It is particularly noteworthy that only three patients dropped out of the program. In the first case (P3), strong social support – including living with a brother for several weeks a receiving private physiotherapy sessions more than three times per week—was sufficient to counterbalance the challenges of surgery and establish an exercise routine outside the Vivifrail program. In the other two cases (P13, P8), inadequate or even absent social support left them unable to manage physical and organizational constraints. These observations highlight the significant influence of social isolation on program adherence. They underscore the importance of identifying socially isolated patients, providing them with additional support, and encouraging caregiver involvement to enhance home-based physical activities.

In this context, additional resources can be mobilized to support sustainable PA engagement. These include addressing patient beliefs, strengthening motivation, providing social and professional support, as well as using targeted behavioral strategies. The strong emotional and cognitive barriers to PA, such as anxiety or fear—especially the fear of dying during exercise—are a recurrent concern [30, 31]. However, these barriers were rarely mentioned during our interviews. Prescribing a structured program, along with complementary information from a physiotherapist, may have limited patients’ automatic avoidance of exercise [32]. On the other hand, positive health beliefs and favorable attitudes towards PA were often reported by our interviewees. Current literature shows that those factors are associated with higher intentions to be active and better adherence [33–35]. It could therefore be worthwhile to cultivate those drivers using different motivational techniques and resource recommendations dependent on baseline beliefs and awareness of the benefits of PA [31]. During our interviews, it was frequently mentioned that patients were strongly motivated by their perceived capabilities and their improvement over time. These findings are consistent with published data which show that perceived capabilities influence patients’ confidence in their ability to engage in PA, [36, 37] and may enhance the pleasure experienced during PA [38]. This, in turn, can encourage positive attitudes toward PA and promote long-term adherence after CR. Thus, motivational interviewing (MI) could help to define specific, realistic and personalized goals more clearly, encouraging patients to take ownership over their PA journey [39]. Interventions such as motivational interviewing have been shown to promote lasting habits changes when initiated at an early stage of the disease [40]. However, the motivating effect of motivational interviewing may diminish over time, particularly when external stressors accumulate and the perceived benefits of physical activity are no longer directly perceptible [31, 41]. This advocates for early engagement, regular follow-up, and personalization of care pathways to maintain adherence throughout the rehabilitation process.

The strong adherence observed among patients participating in the Vivifrail program suggests that it is a promising and structured tool to guide individuals towards establishing a regular physical activity routine. Vivifrail offers a structured framework, varied exercises, and clear guidance that foster initial engagement and autonomy. However, feedback from patients reveals a range of spontaneous adaptations such as modifying the number of repetitions, sets, or splitting the session. These adaptations raise important questions about the program’s long-term effectiveness and alignment with international physical activity guidelines [42]. On the positive side, patients' ability to adapt the program in real life demonstrates flexibility and self-awareness. This can be interpreted as an indicator of patient empowerment and self-regulation. On the other hand, some adjustments raise concerns. For example, skipping certain exercises entirely or considering walking as a full substitute for specific strength or balance routines brings up a key issue: can one form of activity truly replace another? This highlights the need to clarify, as soon as the program is established, which activities are interchangeable, and which are essential to preserve. The recommendations suggest that older adults engage in multicomponent physical activity, incorporating a combination of aerobic, balance and flexibility training, as well as resistance training two to three times per week [43, 44]. Furthermore, general physical activities alone, such as walking or cycling, are unlikely to reverse frailty or prevent falls [43]. Therefore, it is essential to recognize that these activities cannot substitute for resistance training. Because the benefits of exercise are lost upon cessation, it is crucial to have opportunities to continue with appropriate activity at the end of structured programs. Effective programs should include individualized exercises tailored to daily activities such as sit-to-stand, squats, reaching while standing, or standing with a narrower base of support [44]. Patients may be tempted to replace these exercises with ADL immediately, but they should know that, in order to be effective, exercises must be sufficiently challenging, regularly reviewed, and progressed to maintain an optimal level of difficulty. When individuals withdraw due to concurrent health issues or caring duties, they should be encouraged to return, and programs should be adapted to ensure appropriate difficulty and exercise dose [45, 46].

Another challenge lies in the accuracy of self-assessed physical activity. As noted by Serves, patients may underestimate their activity if they do not account for daily tasks such as housework or gardening. Conversely, others may overestimate their effort by counting light walking as equivalent to structured exercise [18]. In this context, direct measurement of PA (i.e. accelerometers or actigraph) could provide valuable support in future studies [47, 48]. For future research, we recommend using both self-reported and direct measures of activity to more accurately assess patients' ability to self-report and their physical activity levels. The use of hip-worn accelerometer devices should be preferred, as the appropriateness of wear location depends on the activity being captured [49]. In this population, post-operative restrictions affecting upper-limb activities may compromise accuracy of wrist-worn devices, making hip placement more suitable for capturing whole body movement. These devices should require minimal setup, which is particularly appropriate for older adults who may be less familiar with digital technologies, thereby reducing participant burden and improving compliance. A recommended wear period of at least 7 consecutive days, with 8 to 10 h a valid wear time per day, is advised to obtain reliable estimates of PA [49, 50]. In the context of the Vivifrail program, when follow-up is conducted at six weeks, extended wear over multiple weeks is recommended when feasible. Accelerometry data should be triangulated with information on movement type, intensity and purpose as documented by patients in their Vivifrail logbook. Previous research in adults and older populations suggests that the number of valid days needed to achieve reliable estimates of PA decreases as the intensity of the PA increases [50]. Accordingly, for sedentary patients (i.e. Vivifrail A), reliable estimates may be achieved with as few as four valid days, provided at least 6 to 8 h of wear time per day are recorded. Finally, while short term results are encouraging, 70–90% of patients meet the recommended 150 min of moderate to vigorous PA during the first weeks post discharge [33, 51]. However, longer-term data are less reassuring as an estimated 46,6% of myocardial infarction survivors fail to meet PA recommendations one year after rehabilitation [52]. A longer follow-up period of six months, one year or more, appears necessary to better estimate long-term adherence and identify potential barriers that may arise after the cardiac event.

Strengths and limitations

The main strength of this study is that it is the first to focus on this specific population. The representativeness of the cohort, particularly with regard to the distribution of surgical procedures, aligns with other studies on combined cardiac surgery [53]. Participants’ responses were consistent across themes explored in all interviews. The final interviews conducted the standardized interview guide, comprehensively captured the key themes and achieved data saturation indicating that additional interviews (including the preliminary ones) were unlikely to generate new concepts or substantially alter the findings. We therefore emphasize that the exclusion of the nine preliminary interviews does not appear to have affected the overall analysis. On the contrary, it allowed us to enrich our interview guide. The main limitation of our study lies in the use of telephone interviews. Although practical and helpful to ensure continuity of care, this method is susceptible to selection and response bias. Phone follow-ups may offer a practical approach to maintaining contact with patients and offering regular support. Even though the interviewer used active verbal engagement and probing, employing open-ended questions, reflective listening, and prompts to encourage deeper responses, the absence of nonverbal communication, could have hindered the depth of interaction and interpretation [54]. While selection bias is a possibility, it appears limited in this case. The preliminary interviews were part of the same population and did not differ systematically from the included participants (same eligibility criteria, same program exposure). The exclusion was based on methodology, not participant characteristics, so it does not selectively remove certain perspectives. The included sample still reflected variation in surgery type, functional status, and social support, maintaining the representativeness of experiences for the qualitative objectives. Among the final interviewees, only one participant explicitly declined to take part, stating that they had not followed the exercise booklet and did not see the value in being interviewed. Other non-response were due to technical issues: two patients could not be reached, and one was not intelligible during the call. However, the absence of input from this type of non-responding patient, combined with the lack of member checking and the relatively short duration of our semi-structured interviews, may have limited the depth and richness of the data collected, particularly regarding perceived barriers. Our focus on participants’ lived experience, which was predominantly positive, may have limited the spontaneous emergence of more negative experiences. Similarly, patients’ generally positive postoperative course may have contributed to a retrospective bias, as participants who agreed to take part were often favorable toward the intervention, potentially leading them to underreport obstacles or challenges, despite probing by the interviewer. From a behavior change perspective, such barriers and drivers could be interpreted through frameworks such as the COM-B model, which conceptualizes behavior as arising from the interaction of capability, opportunity, and strength of motivation [55]. In this sense, the predominantly positive narratives reported by participants may reflect favorable conditions for engagement with the intervention and the experiences shared during interviews. These factors are important to consider when reflecting on the trustworthiness and transferability of our findings. While the interviews provided sufficient data to identify key themes, the presented interpretive patterns should be understood as reflective of the shared perspectives and analytic choices. They cannot be considered an exhaustive account of all possible patient experiences but may be transferable to other similar patients in comparable care settings.

Implications for practice

Standardized post-discharge guidance: Providing clear, consistent messaging can prevent excessive restriction of physical activity and reduce patient uncertainty about safe exercise. Structured programs such as Vivifrail, offer a framework, varied exercises, and clear guidance that foster initial engagement and autonomy in establishing a regular physical activity routine.

Early and structured follow-up: Motivational interviewing before hospital discharge and during follow-up may reinforce motivation, help identify barriers early and support goal setting tailored to patients’ functional status.

Context-adapted exercise planning: Exercise menus should be flexible to accommodate environmental constraints, such as small apartments, and to include weather contingency alternatives. Patients should be guided on which exercises are essential versus interchangeable to improve their condition and preserve program fidelity.

Family and social support integration: Engaging caregivers through coaching scripts and promoting peer or family involvement can enhance and support habit formation. Combining follow-up visits, enlistment of social support and coordination with physiotherapists can provide ongoing encouragement, monitoring and personalization of exercise routines.

Monitoring and evaluation: Integrating both self-reported activity logs and objective measures (i.e. accelerometers, actigraphs) can improve the accuracy of activity assessments, guide personalized adjustments and facilitate long-term adherence tracking and maintenance.

Conclusions

The Vivifrail program appears to be a feasible approach for guiding frail older adults who have recently undergone cardiac surgery towards establishing a regular PA routine. Its adoption seems to be influenced by the patient’s perceptions and previous activity levels. Post-operative physical limitations and fatigue were frequently reported as obstacles and should therefore be considered. Motivation, on the other hand, appeared to be shaped by perceived capabilities, which could be promoted using different motivational techniques depending on baseline beliefs and awareness of the benefits of PA. The role of healthcare professionals and family support also emerged as an important influence. Future studies should explore direct measurement of PA as well as its long-term maintenance.

Abbreviations

CR

Cardiac Rehabilitation

ICOPE

Integrated Care for Older People

MI

Motivational Interviewing

PA

Physical Activity

SPPB

Short Physical Performance Battery

Authors’ contributions

AK led the conceptualization, methodology, project administration, and resource provision, while CM was responsible for the investigation. Together, AK and CM carried out data curation, formal analysis, drafting, and review/editing of the manuscript. ASB contributed to the conceptualization, methodology, and review/editing of the manuscript.

Funding

Not applicable.

Data availability

Interview transcripts are confidential, but anonymized data can be made available upon written request to the corresponding author.

Declarations

Ethics approval and consent to participate

The observational study was approved by the Nantes Health Ethics Group (GNEDS), Nantes, France. (N°24–22-02–273) and registered on OSF: 10.17605/OSF.IO/7HTX2.

Patients were informed about the study objectives and provided written informed consent to participate in the interviews. All methods were carried out in in compliance with the Helsinki Declaration.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.