Background Adherence to home-based exercise is essential for functional recovery after stroke, yet non-adherence is common and poorly explained. The Information–Motivation–Behavioral Skills (IMB) model has been widely applied in health behavior research but rarely tested in stroke rehabilitation. This study examined behavioral determinants of exercise adherence and tested the mediating role of self-efficacy based on the IMB framework. Methods A cross-sectional survey was conducted among 549 ischemic stroke survivors with limb dysfunction recruited from rehabilitation clinics in Harbin, China. Participants completed validated measures of exercise adherence, stroke-related knowledge, motivation for exercise, perceived social support, and self-efficacy. Structural equation modeling (SEM) was used to test hypothesized IMB pathways while controlling for demographic and clinical covariates. Results The final model demonstrated good fit ( χ 2 /df = 2.675, RMSEA = 0.055, CFI = 0.949). Self-efficacy had the strongest direct effect on adherence ( β = 0.37, p 0.001) and significantly mediated the effects of stroke knowledge (indirect effect = 0.119, p 0.001), personal motivation (0.078, p 0.001), and social support (0.056, p 0.001). Motivation also had a direct effect on adherence ( β = 0.25, p 0.001), while knowledge and social support influenced adherence only indirectly via self-efficacy. Conclusion Self-efficacy is a central behavioral mechanism underlying home exercise adherence after stroke. The IMB framework offers a useful foundation for designing adherence-enhancing interventions that build confidence, reinforce progressive mastery, and integrate social support in rehabilitation planning.
Li et al. (Thu,) studied this question.