关键词: 社区健康支持, 老年人, 自评健康状况, 倾向得分匹配

Abstract: Background　To address the challenges brought by accelerated aging，the Chinese government has actively promoted a model improving older people's health，namely，community-based health support model. However，the relationship of different community health support services with the health status of the elderly is unclear，and still needs to be studied systematically. Objective　To explore the association of different community-based health support services with self-rated health of older people，offering ideas for improving such services. Methods　2018 waves of the Chinese Longitudinal Healthy Longevity Survey data were obtained，with a total of 8 860 valid samples included. The relationship of self-rated health status of the elderly with four types of community-based health support services received，including personal daily care，mental health support，home visits and medication delivery，and health education，was analyzed by means of propensity score matching（PSM）. The kernel matching was used to match the processing group with the control group，with quadratic kernel used by default，and bandwidth set at 0.06. The K-nearest neighbor matching and radius matching were used to check the robustness. Results　The average score of self-rated health status of the 8 860 respondents was（3.43±0.90），which was in the general level. The supply of 4 types of community-based health support services：10.01%（887/8 860） of the respondents received community-based personal daily care，35.11%（3 111/8 860） received community-based home visits and medication delivery，14.35%（1 271/8 860）received community-based mental comfort and chatting services for relieving boredom，and 43.16%（3 824/8 860） received community-based health education. The supply of these 4 services showed significant differences across rural and urban areas，as well as eastern，central and western China（P<0.05）. PSM analysis showed that the average self-rated health status score in the processing group was 0.123 points higher than that of the control group（ATT=0.123，P<0.05），indicating that self-rated health status was associated with community-based personal daily care. No significant correlation existed between the self-rated health status and community-based home visits and medication delivery（ATT=0.012，P>0.05）. A significant correlation was found between community-based mental comfort and chatting services for relieving boredom and the self-rated health status. After matching，the ATT value was 0.083（P<0.05）. No significant correlation was found between community-based health education and the self-rated health status（ATT=-0.017，P>0.05）. Heterogeneity test revealed that the associations of self-rated health status with the four health support services in all participants were basically the same as those in subgroups stratified by sex，age，education level and geographical region. Those with higher annual household income，and higher education level living in urban communities of eastern China were more likely to benefit from community-based health support services. Conclusion　The participatory health management service is more beneficial to maintaining or improving the health level of the elderly than passively accepting service. Communities should further enrich the supply mode of essential medical and health services，and accelerate the integration of medical and old-age services. By doing so，the transformation of delivering community-based health support services from a pattern of "regardless of specific health needs of individuals" to the one of "precisely targeting individual specific health needs" will be promoted gradually.

Key words: Community-based health support, Aged, Self-rated health status, Propensity score matching