The ability of self-rated health to predict mortality among community-dwelling elderly individuals differs according to the specific cause of death: data from the NEDICES cohort.

BACKGROUND The biomedical and psychosocial mechanisms underlying the relationship between self-rated health (SRH) and mortality in elderly individuals remain unclear. OBJECTIVE To assess the association between different measurements of subjective health (global, age-comparative, and time-comparative SRH) and cause-specific mortality. METHODS Neurological Disorders in Central Spain (NEDICES) is a prospective population-based survey of the prevalence and incidence of major age-associated conditions. Data on demographic and health-related variables were collected from 5,278 subjects (≥65 years) in the baseline questionnaire. Thirteen-year mortality and cause of death were obtained from the National Death Registry. Adjusted hazard ratios (aHR) for SRH and all-cause and cause-specific mortality were estimated by Cox proportional hazard models. RESULTS At baseline, 4,958 participants (93.9%) answered the SRH questionnaire. At the end of follow-up, 2,468 (49.8%) participants had died, of whom 723 (29.2%) died from cardiovascular diseases, 609 (24.7%) from cancer, and 359 (14.5%) from respiratory diseases. Global SRH independently predicted all-cause mortality (aHR for 'poor or very poor' vs. 'very good' category: 1.39; 95% confidence interval (CI): 1.15-1.69). Analysis of cause-specific mortality revealed that global SRH was an independent predictor for death due to respiratory diseases (aHR for 'poor or very poor' vs. 'very good' category: 2.61; 95% CI: 1.55-4.39), whereas age-comparative SRH exhibited a gradient effect on the risk of death due to stroke. Time-comparative SRH provided small additional predictive value. CONCLUSIONS The predictive ability of SRH for mortality largely differs according to the specific cause of death, with the strongest associations found for respiratory disease and stroke mortality.