Spatial variation in self-reported arthritis/rheumatism among middle-aged and older Chinese adults: an exploratory CHARLS ecological analysis

Arthritis and rheumatic conditions represent a major source of disability among middle-aged and older adults worldwide. Using data from the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative cohort of middle-aged and older Chinese adults, this secondary analysis explored the spatial distribution of self-reported arthritis/rheumatism and its ecological associations with environmental exposure indicators across China. This secondary ecological analysis was conducted using data from the CHARLS. A total of 25,390 participants aged ≥ 45 years were included. Data collected included self-reported arthritis/rheumatism, provincial environmental indicators, and partial dietary data. Multivariate linear regression was used to analyze linear associations between environmental indicators and provincial prevalence of self-reported arthritis/rheumatism. Exploratory chi-square tests were subsequently conducted to compare environmental indicators across prevalence categories. The prevalence of self-reported arthritis/rheumatism in the cohort was 33%, exceeding that of hypertension (30%) and digestive system diseases (23%). The crude proportion of participants reporting physician-diagnosed arthritis/rheumatism ranged from 16% in Zhejiang to 59% in Chongqing. In the primary multivariable analysis, none of the provincial environmental or dietary indicators showed significant linear associations with arthritis/rheumatism prevalence after adjustment for provincial age and sex composition. Exploratory ecological analyses identified differences in the categorical distributions of ozone concentration (O 3 -8 H-90per), electrical conductivity, permanganate index, ammonia nitrogen, and total nitrogen across provincial prevalence categories. In descriptive comparisons, provinces with lower arthritis/rheumatism prevalence exhibited higher ambient ozone concentrations than high-prevalence provinces, whereas the remaining environmental differences should be considered exploratory. This secondary ecological analysis demonstrated substantial geographic variation in the prevalence of self-reported arthritis/rheumatism across mainland China. Although the primary multivariable analyses did not identify significant linear associations between provincial environmental indicators and arthritis/rheumatism prevalence, exploratory ecological analyses identified several regional environmental patterns that warrant further investigation. Given the ecological design and the potential for ecological fallacy, these findings should be interpreted as hypothesis-generating rather than evidence of causal environmental effects.