Eric T C Lai1,2, Ruby Yu1, Jean Woo1,2. 1. Jockey Club Institute of Ageing, Chinese University of Hong Kong, Hong Kong SAR. 2. Institute of Health Equity, Chinese University of Hong Kong, Hong Kong SAR.
Abstract
BACKGROUND: Social gradients of self-rated health (SRH) of older people are evident in various settings. However, it is not clear whether improving older people's sense of community (SoC) could mitigate the social gradient. METHODS: People aged above 60 in five residential districts of Hong Kong were sampled using multistage sampling (n = 1,793). SoC was measured using the validated eight-item Brief Sense of Community Scale, with each item on a five-point Likert scale, forming a score from 8 to 40. SRH was considered as dichotomous (poor/not poor). Socioeconomic status (SES) was operationalised as monthly income, highest education attainment and self-rated disposable income (defined as whether the older person feel he/she has sufficient income). Causal mediation analysis using four-way decomposition was used to assess whether SoC mediates/moderates the association of SES and poor SRH. RESULTS: A social gradient of poor SRH by all measures of SES was observed [adjusted relative risk (RR) per standard deviation income increase = 0.92; 95% confidence interval (95% CI) 0.88-0.97; RR comparing lowest to highest education= 1.77; 95% CI: 1.48-2.11; RR comparing very insufficient to very sufficient disposable income = 1.74; 95% CI: 1.48-2.05]. Causal mediation analysis showed that SoC interacts with the association of education and SRH, with higher the SoC, stronger the education gradient. CONCLUSIONS: Our findings showed a social gradient of SRH in the older population in Hong Kong. This relationship was moderated by SoC, for which higher SoC is related to stronger SES-SRH gradient.
BACKGROUND: Social gradients of self-rated health (SRH) of older people are evident in various settings. However, it is not clear whether improving older people's sense of community (SoC) could mitigate the social gradient. METHODS:People aged above 60 in five residential districts of Hong Kong were sampled using multistage sampling (n = 1,793). SoC was measured using the validated eight-item Brief Sense of Community Scale, with each item on a five-point Likert scale, forming a score from 8 to 40. SRH was considered as dichotomous (poor/not poor). Socioeconomic status (SES) was operationalised as monthly income, highest education attainment and self-rated disposable income (defined as whether the older person feel he/she has sufficient income). Causal mediation analysis using four-way decomposition was used to assess whether SoC mediates/moderates the association of SES and poor SRH. RESULTS: A social gradient of poor SRH by all measures of SES was observed [adjusted relative risk (RR) per standard deviation income increase = 0.92; 95% confidence interval (95% CI) 0.88-0.97; RR comparing lowest to highest education= 1.77; 95% CI: 1.48-2.11; RR comparing very insufficient to very sufficient disposable income = 1.74; 95% CI: 1.48-2.05]. Causal mediation analysis showed that SoC interacts with the association of education and SRH, with higher the SoC, stronger the education gradient. CONCLUSIONS: Our findings showed a social gradient of SRH in the older population in Hong Kong. This relationship was moderated by SoC, for which higher SoC is related to stronger SES-SRH gradient.