Heterogeneity and the determinants of multidimensional health transition among Chinese middle-aged and older people: a personcentered approach

Linglong Ye¹Zhengman Wu¹Yi-Wei Kao²Jianping Zhu¹Ben-Chang Shia³Mingchih Chen³Lei Qin^4*

• ¹Xiamen University, Xiamen, Fujian Province, China
• ²Ming Chuan University, Taipei, Taiwan
• ³Fu Jen Catholic University, New Taipei, Taiwan
• ⁴University of International Business and Economics, Beijing, China

Introduction: Previous health transition patterns studies only focused on biological and psychological dimensions, and overlooked social dimension. By combining biological, psychological, and social health dimensions, this study aimed to examine heterogeneous transition patterns and assessed their determinants among Chinese middle-aged and older adults. Methods: Four waves of longitudinal data in 2011-2012, 2013, 2015, and 2018 from the China Health and Retirement Longitudinal Study was adopted. A total of 6161 adults aged ≥45 years with no missing data on three dimensions were included. The repeated-measures latent class analysis as a person-centered approach was used to estimate transition patterns, followed by multinomial logistic regression to assess determinants. Results: This study highlighted five health transition groups, including “social participation slightly improving followed by deterioration” (SP-ID, 23.62%), “cognitive status deteriorating gradually and lacking social participation” (CS-DG&L-SP, 17.97%), “ADLs deteriorating followed by a slight improvement and lacking social participation” (ADLs-DI&L-SP, 9.14%), “died in the follow-up period” (DIFP, 11.65%), and “sustainedly healthy” (SH, 37.62%). Using the SH group as the reference group, sex, education, job, type of residence, and region have different impact on different patterns. Men were more likely to be in the SP-ID and DIFP groups, while women were more likely to be in the CS-DG&L-SP and ADLs-DI&L-SP groups. Being not in marriage was related to the SP-ID, CS-DG&L-SP, and DIFP groups. Those living in the central region and those having smoking and drinking habits less tended to fall into the SP-ID group. The retired, non-smokers, and drinkers even less tended to be in the CS-DG&L-SP group. Those living in the eastern region, those drinking alcohol appropriately, and those with higher income less tended to be in the ADLs-DI&L-SP group. Being uninsured by public insurance and having lower income were associated with the DIFP group. High education levels, working in non-agricultural sectors, and living in urban areas were associated with the SH group. Conclusion: Integrating the social dimension with physical and psychological dimensions enhanced our understanding of the heterogeneous health transition patterns of middle-aged and older people. These findings provide valuable evidence for promoting healthy aging targets for different groups of the aging population.