The family doctor team integrated peer support groups, as one type of social prescribing models, can effectively address the fundamental challenges posed by social factors in the chronic disease prevention and control. With advantages such as strong sustainability, it alleviates the tension between the shortage of rural health workforce resources and the low self-management adherence among chronic disease patients, holding the potential to further unlock the effectiveness of the county health system reform, thereby. However, research on the sustainability mechanisms of the intervention effects remains limited within the social context of rural China.

To conduct an in-depth analysis of the multiple pathways and dynamic processes through which this intervention model exerts a sustained impact on the self-management behaviors of chronic disease patients in real rural social settings.

An embedded single-case study was adopted, with the health mutual aid group project in T Village, F Province, from December 2022 to June 2024 as the primary unit of analysis. Secondary units of analysis included key actors and specific types of self-management behaviors. The study was grounded in an integrated framework of social cognition theory and related theories, utilizing questionnaires, in-depth interviews, and non-participant observation for data collection. Crisp-set qualitative comparative analysis (csQCA) served as the core analytical method to identify pathways. Large language models (LLMs) and retrieval-augmented generation (RAG) were employed during the process to assist in preliminary analysis. One outcome (R0) and six causal conditions (CV1-CV6) were defined.

(1) The consistency for the perceived harmfulness of non-health behavior (CV1), perceived usefulness of health behavior (CV2), perceived ease of practicing health behavior (CV3), and localized doctor-patient social support (CV4) met or exceeded 90%. (2) The overall consistency and coverage for the three pathways of sustained self-management behavior among rural residents were 80.00% and 93.33%, respectively: The primary pathway had a consistency of 83.25% and coverage of 46.67%, including CV1-CV4 and localized neighborhood social support (CV6) ; Secondary pathway A had a consistency of 81.82% and coverage of 30.00%, including CV1-CV2 and CV4-CV6; Secondary pathway B had a consistency of 71.43% and coverage of 16.67%, including CV1-CV4 and CV5.

This model, through collaboration between family doctor teams and localized social support networks—including village doctors, families, and neighbors—can continuously enhance perceived benefits and ease of self-management behavior, bridge the "digital divide", and ensure the sustainability of intervention effects. Therefore, in rural chronic disease prevention and control, it is feasible to explore the integration of essential health services under formal institution with the development of localized social support networks. Specifically, in remote areas such as mountainous regions, family doctor teams can conduct institutionalized regular visits, stimulate volunteers by natural village units, and leverage their localized social support networks to establish mutual aid groups for chronic disease patients. These groups can regularly conduct interactive health education and guide health-oriented daily interactions and mutual assitune practice.