Psychological Disposition, Health-related Mediators, and Longevity Among Chinese Older Adults

Background: Psychological disposition has been linked with health outcomes and longevity; however, existing research has yielded inconsistent evidence. This study aims to investigate the associations between psychological disposition and all-cause mortality among older Chinese individuals, and to assess potential mediating effects of health-related factors.

Methods: Utilizing data from seven waves of the Chinese Longitudinal Healthy Longevity Survey (CLHLS), this community-based, prospective cohort study included Chinese adults aged 65 and above who were recruited from 1998 to 2014 and followed until 2018. The main outcome was all-cause mortality. Psychological disposition scores were measured with a 7-item scale. Categorical psychological disposition indexes were generated based on quartiles of continuous scores. Kaplan-Meier method was employed to fit survival curves. Cox proportional hazard models were established to evaluate the relationships between categorical or continuous psychological disposition and mortality, adjusting for sociodemographic characteristics and selected health-related factors including lifestyle behavior factors, self-rated health, cognitive function, functional disability, and chronic diseases. Restricted cubic spline analyses were utilized to estimate linear relationships between continuous psychological disposition and mortality. The indirect effects of psychological disposition on all-cause mortality attributable to selected health-related factors were quantified through calculating the percentage excess risk mediated (PERM). Subgroup analyses were carried out to determine whether the associations differed by gender and age. Sensitivity analyses were performed to assess the robustness of the study.

Results: A total of 21,891 participants (56.5% female, 80.6% oldest-old) were followed over a cumulative 105,033.6 person-years. Participants who were male and aged 65-79 years old had a higher likelihood of being classified into higher quartiles of psychological disposition indexes. Dose-response associations were delineated between psychological disposition and all-cause mortality. The Cox regression adjusted for sociodemographic characteristics showed hazard ratios of highest quartile compared to lowest quartile as 0.75 (95%CI: 0.72, 0.79), 0.80 (95%CI: 0.76, 0.83), and 0.81 (95%CI: 0.78, 0.85) for total, positive, and negative psychological disposition indexes, respectively. After adjusting for the selected health-related factors, the associations were mediated by 49.5%, 45.0%, and 61.8% for total, positive, and negative psychological disposition indexes, respectively. The associations remained significant only at the 3rd and 4th quartiles after being fully adjusted. The most potent mediators were self-rated health and cognitive function. Gender subgroup analysis showed that among males, overall mediation in the association was 57.2% and 53.3% for total and positive psychological disposition, respectively; while among females, overall mediation was 42.0% and 38.3% for total and positive psychological disposition, respectively. Gender difference in mediation effect was inconsequential for negative psychological disposition. The age subgroup analysis showed no significant overall mediation but revealed variations in the extent of mediation for cognitive function (PERM from 1.2% to 3.1% among younger-old; PERM from 38.8% to 43.4% among oldest-old). The mediating effects of functional disability and chronic diseases were more potent among the younger-old. The associations between continuous total and negative psychological disposition, but not positive disposition, and mortality were linear. Sensitivity analyses showed consistent findings.

Conclusions: The findings suggested dose-response associations between psychological disposition and all-cause mortality, which were largely attributable to health-related factors, notably self-reported health and cognitive function. Tailored interventions targeting cognitive function enhancement for the oldest-old, and preventive measures addressing disability and chronic diseases for the younger old are essential for fostering successful ageing across diverse age groups.