Average cardiometabolic burden across waves and subsequent difficulties in basic activities of daily living among middle-aged and older adults in China: a longitudinal cohort study

Abstract Background We assessed associations of cardiometabolic burden measured across several survey waves with subsequently confirmed difficulties in basic activities of daily living (ADL), quantified standardized 5-year risks accounting for competing mortality, and explored associations among participants without self-reported heart disease or stroke and after adjustment for current components. Methods We included 9587 adults aged ≥ 45 years from the China Health and Retirement Longitudinal Study (CHARLS) who had no ADL difficulties in 2015. Burden comprised hypertension, diabetes, central obesity, heart disease, and stroke, averaged across at least two valid waves in 2011–2015; scores were prorated when at least four components were assessable. Participants were followed through 2020. Event-specific discrete-time complementary log-log models incorporated inverse probability of censoring weighting, community-clustered inference, and multiple imputation of specific missing values. Joint multinomial models estimated standardized cumulative incidence. Exploratory analyses examined risk-factor burden among participants with negative heart disease and stroke histories at all three waves, and historical burden conditional on the five current components. Two principal supplementary ADL tests underwent Holm adjustment. Results There were 2167 ADL events and 492 preceding deaths. Per standard deviation (0.89 component units), the complementary log-log-scale ratio (CSR) for ADL difficulties was 1.24 (95% confidence interval [CI] 1.19–1.30). Among participants without self-reported heart disease or stroke at all three waves, the CSR per risk-factor component unit was 1.18 (95% CI 1.10–1.27). In participants with complete components at all three waves, the CSR per historical burden unit, adjusted for all five 2015 components, was 1.18 (95% CI 1.06–1.31). Standardized 5-year ADL risks at burden values of 3 and 0 were 35.0% and 19.5%, respectively: a difference of 15.5 percentage points (95% CI 12.1–18.9). Conclusions Higher repeatedly measured cardiometabolic burden was associated with subsequent ADL difficulties. Historical burden retained a conditional association after adjustment for current components. These findings support attention to self-care function during cardiometabolic follow-up, including among people without self-reported heart disease or stroke.