Objective   To examine social region-specific characteristics and transition-differentiation of driving factors for multidimensional health poverty (MHP) in China for providing a reference to precise poverty alleviation and effective governance of health poverty.

  Methods  We collected nationwide data from four rounds of the China Health and Retirement Longitudinal Survey (CHARLS) conducted in 2011, 2013, 2015, and 2018 and other relevant data from the China Statistical Yearbook and the China Health and Family Planning Statistical Yearbook of 2012, 2014, 2016, and 2019 and from the Global Burden of Disease Network in 2019. Based on a self-established theoretical framework of MHP associated with three dimensional components of health capability-right-risk, we selected 13 indicators covering the three dimensional components as the driving factors of MHP and used geographic detectors to explore the characteristics of independent and coupled effect of the driving factors and analyze differences in spatial distribution of the driving factors.

  Results  Compared with those in 2011, 2013, and 2015, the ranking of three health capability dimension factors (gross domestic production GDP per capita, urbanization rate, and annual hospitalization rate of residents) increased in 2018; while the ranking of three health right dimension factors (per capita expenditure on both social security and employment and the number of public health education programs) decreased; the effect of health right empowerment on preventing the occurrence of health poverty increased. The analysis on differences in spatial distribution of driving factors during 2011 – 2018 showed that urban unemployment rate was a main driving factor for MHP in eastern region; while, the main driving factor for MHP in central region was the concentration of particulate matter less than 2.5 μm in aerodynamic diameter (PM_2.5) three years ago and the MHP was driven jointly by multiple factors with regard to health capability-right-risk. The analyses on interactive effect of the driving factors revealed bivariate interactive enhancement effect on MHP among GDP per capita and urbanization rate as factors of health capability dimension and government health expenditure ratio, per capita expenditure on social security and employment, and the number of beds in health care institutions versus per 1 000 population as factors of health right dimension; the joint effect of driving factors regarding to health capability and health right reduced the occurrence MHP; the interaction between PM_2.5 concentration three years ago (a health risk indicator) and two health capability indicators (annual hospitalization rate of residents and elderly dependency ratio in 2011/2013/2018) also showed bivariate interactive enhancement effect on MHP.

  Conclusion  The prevalence of MHP is of spatially and temporally heterogeneous pattern across social regions in China and is driven by the coupling of multidimensional factors, suggesting that comprehensive strategies need to be developed for effective health poverty management.