Eric T C Lai1, Hung Chak Ho2, Suzanne C Ho3, Jean Woo4. 1. Jockey Club Institute of Ageing, Chinese University of Hong Kong, New Territories, Hong Kong; Institute of Health Equity, Chinese University of Hong Kong, New Territories, Hong Kong. Electronic address: etclai@cuhk.edu.hk. 2. Department of Urban Planning and Design, The University of Hong Kong, Pokfulam, Hong Kong. 3. Jockey Club School of Public Health and Primary Care, Chinese University of Hong Kong, New Territories, Hong Kong. 4. Jockey Club Institute of Ageing, Chinese University of Hong Kong, New Territories, Hong Kong; Institute of Health Equity, Chinese University of Hong Kong, New Territories, Hong Kong.
Abstract
OBJECTIVES: It is inconsistent in the literature on whether inequalities of health in older age widen or narrow over time. We assessed the associations of socioeconomic status (SES), physical functioning, and mortality in an older age cohort in Hong Kong. DESIGN: Longitudinal cohort study. SETTING AND PARTICIPANTS: We recruited 2032 older adults aged 70+ in 1991 to 1992 and followed them for 10 years. METHODS: SES was operationalized as education, baseline individual income, and longest-held occupation in lifetime. Physical functioning was measured by Barthel's Index for activities of daily living (ADL), from which disability was defined as ADL score <20. Mortality data were obtained from the Death Registry. Bayesian joint modeling with 2 sub-models, mixed-effect, and Cox proportional hazard model, were used to respectively model the associations of SES and disability, and SES and mortality, accounting for selection by mortality. RESULTS: Education and income at baseline were not clearly related to disability, but those with lower education level and income at baseline tended to have their risks increased with time. Older adults who had been mostly economically inactive or unemployed in their lifetime had higher risk of disability [odds ratio 3.24; 95% credible interval (95%CrI) 1.29 to 7.97], and such risk increased over time. For mortality, older adults with no schooling were at higher risk compared with those with secondary education or above (hazard ratio 1.25; 95%CrI 1.00 to 1.57). Income at baseline and longest-held occupation in lifetime were not clearly related to mortality. CONCLUSIONS AND IMPLICATIONS: We observed inequalities of health of older adults in Hong Kong that widened as they age. Community and medical interventions targeting the older adults with the lowest SES would be important to prevent their more rapid decline in physical functioning.
OBJECTIVES: It is inconsistent in the literature on whether inequalities of health in older age widen or narrow over time. We assessed the associations of socioeconomic status (SES), physical functioning, and mortality in an older age cohort in Hong Kong. DESIGN: Longitudinal cohort study. SETTING AND PARTICIPANTS: We recruited 2032 older adults aged 70+ in 1991 to 1992 and followed them for 10 years. METHODS: SES was operationalized as education, baseline individual income, and longest-held occupation in lifetime. Physical functioning was measured by Barthel's Index for activities of daily living (ADL), from which disability was defined as ADL score <20. Mortality data were obtained from the Death Registry. Bayesian joint modeling with 2 sub-models, mixed-effect, and Cox proportional hazard model, were used to respectively model the associations of SES and disability, and SES and mortality, accounting for selection by mortality. RESULTS: Education and income at baseline were not clearly related to disability, but those with lower education level and income at baseline tended to have their risks increased with time. Older adults who had been mostly economically inactive or unemployed in their lifetime had higher risk of disability [odds ratio 3.24; 95% credible interval (95%CrI) 1.29 to 7.97], and such risk increased over time. For mortality, older adults with no schooling were at higher risk compared with those with secondary education or above (hazard ratio 1.25; 95%CrI 1.00 to 1.57). Income at baseline and longest-held occupation in lifetime were not clearly related to mortality. CONCLUSIONS AND IMPLICATIONS: We observed inequalities of health of older adults in Hong Kong that widened as they age. Community and medical interventions targeting the older adults with the lowest SES would be important to prevent their more rapid decline in physical functioning.