Abstract
BACKGROUND: Social participation is a core indicator of successful rehabilitation and quality of life for stroke survivors. However, significant heterogeneity exists in their participation levels, which challenges the development of standardized nursing interventions. This study aimed to identify the latent profiles of social participation among stroke survivors and explore the specific determinants associated with these distinct subgroups. METHODS: Convenience sampling method was used to select stroke survivors from neurology and rehabilitation departments of multiple Grade-A tertiary hospitals in Chenzhou City, Hunan Province. The Chinese version of the Impact on Participation and Autonomy Questionnaire (IPA), Fatigue Severity Scale (FSS), Connor-Davidson Resilience Scale-10(CD-RISC-10), and Social Support Rating Scale (SSRS) were used as survey tools. Latent profile analysis was conducted to identify subtypes based on social participation patterns, and multinomial logistic regression was applied to investigate factors associated with these profiles. RESULTS: A total of 313 stroke survivors completed the questionnaire, of which 300 were valid, with a validity rate of 95.85%. A three-class model was identified as the best-fitting latent profile structure, characterized by the following patterns: High Participation (33.3%), Moderate Participation (46.3%), and Low Participation (20.4%). Variations in social participation patterns were significantly associated with stroke duration, fatigue severity, psychological resilience, current employment status, and number of stroke episodes. CONCLUSION: Stroke survivors' social participation is heterogeneous, highlighting the need for health professionals to consider profile characteristics and specialty-specific factors. It is crucial to develop tailored interventions that are responsive to key modifiable factors such as psychological resilience, fatigue severity, and employment status to mitigate participation restrictions and enhance rehabilitation outcomes among this population. CLINICAL TRIAL NUMBER: Not applicable.