Association of multimorbidity and mortality risk in senior adults: a population-based cohort study

Background This study aimed to evaluate the association of three multimorbidity indicators with mortality risk among senior adults, and compare their predictive performance on mortality risk. Methods This prospective cohort study selected 162,958 participants aged ≥ 65 years from Yuexiu Ageing and Health Cohort. Information on eleven system diseases was extracted; three multimorbidity indicators (condition count, multimorbidity patterns, and multimorbidity index) were created. Hazard ratio (HR) with 95% confidence intervals (CI) was calculated using Cox proportional hazard model after adjustment for confounders. The C-statistic, integrated discrimination improvement (IDI), and net reclassification improvement (NRI) were used to compare the performance of multimorbidity indicators. Results 15,525 deaths were identified during a median of 4.79 years of follow-up. Compared with participants with no multimorbidity, those with multimorbidity had a 1.56-fold risk of all-cause mortality. Every one condition count increment was associated with a 17% increased risk of all-cause mortality. Three multimorbidity patterns labeled as Patterns I, II, and III were extracted and were significantly associated with the increased mortality risk, with HR being 1.97, 1.41, and 1.44 for Patterns I, II, and III respectively. Every 1-unit increment of multimorbidity index was associated with an 18% increased mortality risk. The multimorbidity index outperformed both multimorbidity pattern (IDI = 0.007, NRI = 0.0055), and condition count in predicting mortality (IDI = 0.003, NRI = 0.0046). Conclusions Three multimorbidity indicators were all associated with the increased mortality risk in community-dwelling older Chinese. The multimorbidity index had slightly better predictive performance for mortality than other two indicators.