This was a parallel (1:1), open-label, prospective, randomized controlled trial that adhered to the CONSORT (Consolidated Standards of Reporting Trials) guidelines (Multimedia Appendix 1) [25]. In addition to evaluating the intervention quantitatively, we qualitatively explored the feasibility of home-based rehabilitation and the use of assistive devices for promoting functional independence among patients after stroke and their family caregivers. This study was conducted at the National Institute of Neurosciences & Hospital (NINS&H) in Dhaka, Bangladesh, between March 2025 and August 2025. As NINS&H is the only specialized referral hospital in Bangladesh, most of the patients with stroke are usually discharged within 15 days to 1 month due to overcrowding/high bed occupancy. Therefore, we included patients who were admitted within 5-14 days of the onset of their stroke and discharged within 1 month from the hospital. Patients with stroke and their family caregivers, both male and female individuals, aged ≥18 years, were included. Patients were enrolled irrespective of their time (first or any recurrent) and type (ischemic or hemorrhagic) of stroke, with a modified Rankin scale (mRS) score of 1-4. Patients who had a physician’s advice for rehabilitation therapy and needed assistive devices for ADL were included. It is evident that within this time (1-2 weeks), if rehabilitation is initiated, the patient experiences a better prognosis [26]. We conveniently enrolled only participants living in the Dhaka district to ensure follow-up visits with minimal dropout.

Sample size was generated using statistical power of 80% at α=.05 (2-tailed), with an estimated medium effect size of 0.75 [27]. We used this effect size as we had sufficient power to detect a meaningful effect. The sample size was calculated to be 29 for each arm (a total of 58 for both arms). Considering a 10% dropout rate, the estimated total sample size was 64 (32 in each arm). We used G*Power software (version 3.1.9.4; Heinrich Heine University Düsseldorf) for Windows for this calculation.

Participants were randomized equally into intervention and control groups (CGs; n=32 in each group) using a simple random technique by computer-generated numbers, regardless of their age and sex. Participants were sequentially allocated to the intervention and CGs based on the randomization chart. Randomization was done, and the random allocation sequences were kept by a researcher who was not involved in this study to minimize bias.

At first, patients who were admitted to the hospital were assessed by physicians and the research team to identify any physical impairment. The physiotherapist also assessed the range of motion and notified the physician who required rehabilitation based on their disabilities. After checking the eligibility criteria and verifying with the physician, research assistant (RA) nurses explained the study purpose, ensured home visits for follow-up, and obtained written informed consent from patients and their caregivers. The patient was assessed for functional independence using the Functional Independence Measure (FIM) by RA nurses to evaluate their level of functioning. They also assessed the patient’s self-efficacy, social participation, and caregivers’ burden.

After allocating study participants to either the intervention group (IG) or the CG, the RA nurses conducted a face-to-face interview to collect data about the sociodemographic characteristics and health conditions of stroke survivors and their family caregivers. At baseline (T0), the functional independence measurement, self-efficacy, social participation, and care burden of the family were assessed. The RA nurses periodically followed up with the patients of the IG through home visits (twice a month) and teleservice (twice a month) to check their motivation and cultural values to adjust to the intervention devices and health booklet.

In the community, the nurse visited the home setting to observe the home environment and adjust to patients’ needs. If required, they suggested modifications to home settings. Moreover, they assessed the way of implementing assistive devices for performing ADL and rehabilitation for improving muscle strength. Nurses demonstrated the use of assistive devices if needed. They also demonstrated the exercise procedure using the health booklet and motivated patients and family caregivers to maintain the daily record book.

To prevent participant contamination, RA nurses obtained consent and collected data in 2 different rooms. They also maintained different schedules for the IG and CG. As the same RA nurses collected data from both groups, they strictly maintained confidentiality to share the intervention procedures with the CG. For home-based data collection, RA nurses were divided into the IG and CG.

To evaluate the feasibility and chronological changes of adjustment of home-rehabilitation and assistive devices qualitatively, RA nurses interviewed participants through some open-ended questions and observed participants of the IG every time they contacted them. The interview was conducted for 20-30 minutes by RA nurses based on observation points and interview questions during home-based follow-up. The observation and interview points were as follows: “home environment, adjustment, motivation and use of assistive devices, lifestyle changes pattern with risk factors, level of required assistance after using assistive devices, adherence to rehabilitation regimen, mental status, communication, and social participation” after having a stroke (Multimedia Appendix 3).

To compare data between the 2 groups at different time points, we conducted a per-protocol set analysis to see the actual efficacy of the intervention. For baseline data, descriptive statistics were used after the normality test. The normal data distribution was examined using Shapiro-Wilk’s test, histograms, box plots, and normal Q-Q plots for both the control and IGs, with P>.05. We used chi-square or the Fisher exact test to compare the IG and the CG using categorical variables, which were reported as frequencies and percentages. Since our data were not normally distributed, we performed the Mann-Whitney U test to report the continuous variables as the mean (SD). The mean differences between the 2 groups were compared by an independent t test at each time point (T0 and T1). In addition, to assess changes in outcomes within the groups over time, a 2-way repeated-measure ANOVA was used. The level of significance was set at P<.05. We conducted generalized linear models considering covariates such as age, residence, and mRS to examine the influence of additional rehabilitation activities (home-based and CRP). We used SPSS (version 31.0; IBM Corp). For the qualitative analysis, after transcribing and translating, we converged and coded the statements. Then we used Dedoose software (version 10.0.59; Socio Cultural Research Consultants, LLC), and the results were categorized into themes and subthemes.

This study was conducted following the ethical principles of the Declaration of Helsinki after getting ethical approval from the institutional review board of NINS&H (approval number IRB/NINS/2024/392), Dhaka, Bangladesh. After an explanation of the study objective, procedure, risk, and benefit, written consent was obtained from participants and their caregivers. Patients’ privacy and confidentiality were maintained throughout the study, and withdrawal was ensured without any consequences. Data were deidentified and assigned unique identification numbers for analysis, and no personal information was used in any publication or presentation resulting from this research. Data were stored and accessible by the principal investigator on a password-protected personal computer.

A total of 273 participants were evaluated to determine their eligibility for the study. Among them, 209 of 273 did not meet the inclusion criteria; 156 of 273 resided out of the study area, 528 of 273 had an mRS score of 5, and 1/273 refused to participate in this study. Therefore, a total of 64 patients were enrolled equally in both groups (IG: n=32 and CG: n=32); 47 (73%) completed (IG: n=23 and CG: n=24) the study, and were analyzed. Among the participants who did not complete the study (n=17), 16/64 (25%) died during the first month of intervention (IG: n=8 and CG: n=8), and 1/64 (1.5%) was lost to follow-up (Figure 2).

The mean age of participants was 55.6 (SD 14.4) years, with a range of 20-82 years. Most of the participants were male (n=52, 81%) and resided in urban areas (n=55, 86%). A higher proportion of rural residents was noted in the CG (n=7, 22% vs n=2, 6%) compared to the IG.

Overall, no significant differences were observed between the IG and CG across sociodemographic and clinical characteristics and for comorbidity, indicating successful randomization (Table 1).

Preintervention FIM scores were compared with sociodemographic and clinical characteristics in Multimedia Appendix 4. The participants’ age, religious status, residence, tobacco smoking, and mRS were significantly associated with FIM (all P<.05).

This study evaluated the feasibility of a nurse-led rehabilitation education with an assistive device to improve functional independence for self-care management, self-efficacy, social engagement, and reduction of caregiver burden among patients with poststroke disabilities. Over the 3-month study period, there was improvement after intervention; however, no significant differences were observed in any of the outcomes between the intervention and CGs.

The participants’ average age was 50 (range 20-82) years, which is consistent with the demographics of patients with stroke in LMIC, where strokes typically occur in early life compared to high-income countries [2]. Among the participants, male individuals were higher, which indicates that stroke is more common in men worldwide [37]. Therefore, it is essential to explore the underlying factors and comorbidities prevalent among male patients after stroke in that age group.

The majority of individuals came from families with moderate income or were employed, and educational attainment was consistent between groups. According to previous studies, socioeconomic and educational characteristics have an impact on stroke outcomes because decreased likelihood of receiving rehabilitation services and a worse long-term prognosis are associated with lower income and educational attainment [38].

The most prevalent comorbidities were diabetes mellitus and hypertension, which were found in almost a third of both groups, which are recognized risk factor profiles for stroke in Southeast Asian populations [39]. In the IG, we observed more ischemic strokes compared to the CG. As the prognosis of ischemic stroke is limited after intervention, this might affect our intervention, making it nonsignificant. Moreover, compared with ischemic strokes, hemorrhagic strokes are typically associated with higher early morbidity and mortality, resulting in more deaths, as reported in a previous study [40]. Despite persistent comorbidities, participants described adaptation and symptom control over time. This indicates resilience and coping strategies that help patients after stroke to survive with multimorbidity [7,41].

We did not find any difference in FIM scores over time between the groups. However, if we observe the trend, the FIM scores improved at each time point for both groups. In the IG, we assumed that the FIM scores were increased due to our intervention. We explored the cause of the improvement in FIM scores in the CG and found that they received rehabilitation services from external sources (home-based and CRP), which we could not inhibit due to ethical considerations. CRP provides rehabilitation services with a very highly trained professionals using all conventional therapies (eg, physiotherapy, occupational, speech, and language) in our study areas. Therefore, we can consider our intervention to be effective in 1 arm. This study emphasizes the importance of adequate rehabilitation intervention for patients with poststroke disabilities. Other studies also explained that poststroke disability can be improved after health education and the use of assistive devices [42]. Though our short (3 months) study duration shows FIM improvement in individual groups, one study showed no significant differences in FIM evolution when compared to usual care alone during the short period of time (<10 weeks) [43]. Improvements in mobility, balance, and ADL performance reflect global evidence that structured poststroke rehabilitation fosters early recovery within 3-6 months [44]. The progression from dependence to modified independence, as reported by participants, reinforces the importance of consistent physiotherapy and guided home exercises [9,45]. Our study findings highlight the significance of comprehensive standard care following a poststroke disability.

With respect to self-efficacy, social participation, and caregiver burden, we did not find any difference between the groups over time. To improve the self-efficacy of patients with stroke after discharge, we provided organized interaction and instruction that helped patients feel more confident and develop the skills they need to manage the aftermath of stroke. As in the CG, some of the patients received group interaction, education on health care and rehabilitation, and a community reintegration program from the CRP; therefore, we found improvement of self-efficacy in both groups. Recent research supports the importance of fostering self-efficacy through community services and standard care in patients, along with natural psychological adaptation [44].

The improvement of social participation and the decrease of caregiver burden were also observed in the IG and in the CG by the influence of CRP services. Despite prominent functional improvements, many patients after stroke report limitations in their ability to participate in the community, which raises concern [46]. Caregivers gradually adjust and become more confident as in the poststroke period, patients regain functioning ability, and families gain experience over time [47]. However, the persistently high levels of burden reported in recent surveys suggest that more focused and direct interventions, such as social support and behavioral education programs, may be effective [47].

The qualitative findings demonstrate how structured therapy promotes progressive recovery of patients after stroke, demonstrating significant gains in functional independence, adherence, and social reintegration. Additionally, patient narratives enhance comprehension by clarifying contextual and personal factors influencing rehabilitation involvement.

Over time, patients demonstrated more motivation and continued use of assistive devices, which is consistent with earlier studies that focused on gradual adaptation to rehabilitation services [44,48]. However, participants acknowledged that financial and caregiving burdens highlight socioeconomic challenges common in low-resource settings [17,42]. These results are consistent with research indicating that long-term rehabilitation success depends critically on patient motivation and family support [18].

This study shows that nurse-led rehabilitation enhances psychological well-being and quality of life through perceived physical improvement and increased social engagement [44]. Self-efficacy’s involvement in fostering recovery is supported by the expressed sense of hope and autonomy [21].

In resource-limited environments, barriers such as inadequate infrastructure, caregiver burden, and financial restraints reduce sustained engagement in rehabilitation [17,49]. These insights highlight the necessity of nurse-led, family-centered, community-based rehabilitation programs that tackle socioeconomic and clinical challenges [21].

Overall, the results show that, even in low-income settings, structured rehabilitation helps stroke survivors recover both physically and psychologically. To improve long-term adherence and social reintegration, future interventions should include systems of social support and monetary assistance.

This is the first nurse-led randomized controlled trial in Bangladesh to improve patients’ ADL with disability by providing rehabilitation intervention. This study represents a proper allocation of participants through successful randomization. We provided assistive devices, education booklets, and follow-up through home visits. We carefully assessed patients and provided the appropriate assistive devices, ensuring that their usage was accurately measured. As this is a feasibility study, we observed that both the device and nurse education were well accepted by patients and family caregivers, indicating that this program could be feasibly implemented in the Bangladesh context.

Nevertheless, limitations must be acknowledged. High early mortality and attrition reflect the acute severity of illness in the study population. It causes underpowering of our study sample size. We conducted our study for a short period (3 months) of time with a small sample, which made it very difficult to interpret the findings. The study was conducted only in urban areas, which did not reflect the scenario of the whole country, which may limit generalizability. Further studies with multiple sites are needed to prevent participant contamination. More accurate measurements, rather than FIM, should be implemented to assess the influencing factors for self-care management. We also need to reconsider the mRS level (mRS-4) as there is a more severe disability that limits physical improvements.

In conclusion, this is the first nurse-led pilot randomized controlled trial aimed at improving the functional abilities of patients with poststroke disabilities in Bangladesh. However, our rehabilitative intervention (3 months) showed no impact on FIM, self-efficacy, social participation, and caregivers’ burden for patients after stroke; obvious improvements were observed at each time point (baseline, monthly, and end line) of evaluation for both IG and CG. We assumed that this improvement occurred in the IG due to our intervention, and for the CG, improvement was observed due to the influence of rehabilitation received from home-based and institutional services. The result suggested that a comprehensive rehabilitation intervention is essential for the physical improvement of patients with poststroke disabilities. A future long-term interventional study with a large sample size will be guaranteed for better outcomes, considering all external influences to understand the underlying factors for the improvement of rehabilitative patients after stroke. Strengthening primary care facilities can be a strategic way for the community integration of patients with stroke disability. A government-led, systemic approach is essential to translate policy commitments into equitable access to rehabilitation services for all patients after stroke.