Leisure activity patterns and health vulnerability among older adults in China: a nationwide analysis of urban-rural differences based on CHARLS data

Background With the deepening of global aging, health vulnerability has become a key indicator of multidimensional health risks in older adults. Existing studies have explored factors influencing health vulnerability but lack systematic analysis of the mechanisms and population heterogeneity of leisure patterns, especially in the context of China’s urban-rural dual structure. This study aimed to investigate the impact of different leisure patterns on health vulnerability among older adults and examine the role of urban-rural differences, educational level, and life satisfaction in moderating these relationships. Methods Data were derived from the 2018 China Health and Retirement Longitudinal Study (CHARLS), including 5,778 older adults aged ≥ 60 years. Systematic cluster analysis was used to classify leisure patterns. Ordered Probit models were applied to assess the effects of leisure patterns on health vulnerability, and stratified regression analyses were conducted to explore population heterogeneity. Results Three distinct leisure patterns were identified: entertainment-oriented (50.23%), social-oriented (41.15%), and beneficial-improvement-oriented (8.62%). Compared with the social-oriented pattern, both entertainment-oriented (β=-0.109, P < 0.01) and beneficial-improvement-oriented (β=-0.185, P < 0.01) patterns significantly reduced health vulnerability, with robust results confirmed by model replacement and variable adjustment. Heterogeneity analyses revealed: (1) In urban and peri-urban areas, entertainment-oriented (β=-0.320, P < 0.01) and beneficial-improvement-oriented (β=-0.319, P < 0.01) patterns reduced vulnerability, whereas in rural areas, these patterns increased vulnerability (β = 0.317, 0.319; both P < 0.01). (2) Among individuals with middle school education or higher, both patterns reduced vulnerability (β=-0.074, -0.074; both P < 0.05), but among illiterate individuals, they increased vulnerability (β = 0.122, 0.121; P < 0.01, P < 0.05). (3) Life satisfaction showed a U-shaped moderating effect: the protective effect was strongest in the high-satisfaction group (β=-0.232, P < 0.01), but reversed to increase risk in the low-satisfaction group (β = 0.199, P < 0.05). Conclusion Participation in entertainment-oriented and beneficial-improvement-oriented leisure activities can reduce health vulnerability among older adults. However, attention should be paid to the moderating roles of urban-rural resource disparities, educational level, and life satisfaction to develop targeted interventions.