Factors Associated with Exercise Adherence Across Settings Among Stroke Survivors in China: A Systematic Review of Cross-Sectional Studies

Background/Objectives: Exercise is an important component of rehabilitation programmes for stroke survivors, and adherence is key to ensuring the effectiveness of rehabilitation. However, adherence declines over time, especially as stroke survivors transition from in-hospital to post-discharge settings. The present review synthesised evidence on factors associated with exercise adherence among stroke survivors in China across two settings, using the socio-ecological model. This could provide a basis for developing effective intervention strategies to improve exercise adherence. Methods: A literature search was conducted via China National Knowledge Infrastructure, Wanfang, Weipu, PubMed, and Web of Science from inception to 30 December 2025. Only cross-sectional studies that examined factors associated with exercise adherence were included. Studies were screened against the inclusion and exclusion criteria. Data were extracted, categorised, and analysed using a semi-quantitative approach. The review protocol was not prospectively registered but was later registered on the Open Science Framework. Results: Twenty-three studies were included, of which 13 focused on the in-hospital setting and 10 on the post-discharge setting, published between 2013 and 2025, involving 7557 stroke survivors aged 18 to over 75 years, with a mean quality assessment score of 85.33%. Exercise adherence was generally at a medium level in both settings, with a slightly lower adherence index after discharge. During hospitalisation, 34 intrapersonal and five interpersonal factors were identified, while 29 intrapersonal, five interpersonal, and one environmental factor were identified after discharge. Sufficient evidence showed that educational level, functional independence, and self-efficacy were positively associated with exercise adherence across both settings. During hospitalisation, social support was also positively associated with adherence, while age and stroke severity were negatively associated. After discharge, marital status, knowledge, and attitude were positively associated, while time since stroke onset and balance impairment were negatively associated. Conclusions: Previous studies mainly focused on intrapersonal factors, followed by interpersonal factors, while environmental factors received limited attention. Future studies should examine potentially modifiable factors at all three levels and determine whether addressing these factors supports exercise adherence among stroke survivors during hospitalisation and after discharge.