Abstract P303: Time-Varying Healthy Lifestyle Profile and All-Cause Mortality Among the Oldest-Old in China

Abstract

Background: Association between an individual modifiable lifestyle with mortality is well established. However, influences of combined lifestyles on mortality risks remain inconclusive among adults aged 80 years or older (“oldest-old”). We investigated the association of time-varying healthy lifestyle profile (HLP) with mortality among Chinese oldest-old adults. Methods: We pooled 36,021 oldest-old adults from first seven waves of the Chinese Longitudinal Healthy Longevity Survey (1998-2014), with follow-up to latest wave (2018) (range 0-21.04 years). Lifestyle information was collected at each wave. Nonsmoking, no regular alcohol intake, regular exercise, and simplified healthy eating index of 11-18 were defined as healthy lifestyle. We assigned a score of 1 for healthy and 0 for nonhealthy lifestyle and summed the four scores to a HLP index ranging 0-5 (representing least to most healthy) to reflect joint lifestyle measurements. We used both a continuous time-to-event data set up with study enrollment as time origin (SE model) and a discrete time-to-event data set up with age 80 as time origin. We used Cox proportional hazard model with time-varying covariates and complementary log-log regression to examine hazard ratios (HRs) of mortality, adjusting for demographics, self-reported diseases, activities of daily living, and cognitive function. Results: During 110897.40 person-years of follow-up, 25,511 deaths were recorded. Median death age was 5 years older in most versus least healthy group. In SE model with least healthy group as reference, HRs (95% CIs) for mortality were 0.90 (0.86-0.95), 0.83 (0.79-0.87), 0.74 (0.70-0.78), 0.54 (0.49-0.58), respectively for HLP of 1-4 groups (p-trend<0.001) ( Table ). Dose-response relationships were consistent for HRs of mortality over age since 80. Conclusion: Healthier lifestyle profiles were associated with significant reductions in all-cause mortality risks among Chinese oldest-old, suggesting the importance of life-long adherence to HLP into advanced old age.