Association between diversified social interaction and health among older adults in China: a longitudinal analysis by interaction type and frequency.

Abstract

Background: Among the social determinants of health, social interaction is considered a modifiable factor and an essential component of the global active ageing strategy. This study examined the associations of various types and frequencies of social interaction with health outcomes among older adults in China, while accounting for potential simultaneity and heterogeneity biases.

Methods: This study used data from the Chinese Health and Retirement Longitudinal Study (CHARLS), a five-wave Panel survey conducted in 2011, 2013, 2015, 2018, and 2020, comprising 38,420 observations from 7,864 individuals aged 60 years and above. Social interaction was categorised into three types: leisure-based individual interaction, community-based organisational interaction, and responsibility-driven caregiving interaction, to capture its diversity. Descriptive statistics were used to summarise health status and social interaction. Generalised estimating equation regression models were used to examine the associations between one- or two-wave-lagged social interaction and health outcomes (self-rated health, mental health, cognitive function, and diagnosed diseases). Random-effects estimation addressed individual-level heterogeneity. The 2SLS model was applied to assess potential bidirectional associations between interaction frequency and health, followed by endogeneity test.

Results: Social interaction was associated with more favourable health outcomes among older adults, particularly in the medium to long term. One-wave-lagged interaction was linked to better self-rated health (β = 0.014, P < 0.05), lower mental distress (β=-0.232, P < 0.05), and better cognitive function (β = 0.233, P < 0.001), with no significant association with diagnosed disease. Leisure-based and community-based interactions corresponded to more favourable physical and mental health, whereas responsibility-driven interactions were associated with better cognition but also greater mental distress. Interaction frequency was positively associated with health, and was higher among those with better access to facilities and public transport. Living with children or a spouse, being employed and having a higher income tended to report more favourable health outcomes.

Conclusion: Active social interaction, regular participation in leisure activities, organized social activities, and informal social interactions are associated with more favourable health outcomes among older adults. Policies should prioritize supportive environments and age-friendly community renovations, while families and society should strengthen internal and external support systems to help foster active and healthy aging.