Differences between older men and women in the self-rated health-mortality relationship.

PURPOSE The aims of this study were to examine differences between older men and women: (a) in the ability of self-rated health to predict mortality, (b) in the effect of different follow-up periods on the self-rated health mortality relationship, and (c) in the relative importance of self-rated health and self-rated change in health in predicting mortality. DESIGN AND METHODS By using data from the Nottingham Longitudinal Study of Activity and Ageing, the author assessed relationships between self-rated health and self-rated change in health and 4- and 12-year mortality in separate unadjusted and adjusted Cox proportional hazards regression models in men and women. RESULTS The differences between men and women in the hazard ratios for poor self-rated health were not significant, although there were differences in the explanatory factors. The relationship between self-rated health and short-term and long-term mortality was explained by age and health among men. The relationship between self-rated health and short-term mortality was explained by age, physical and mental health, and physical activity among women. The relationship between self-rated health and long-term mortality was explained by age, physical health, and physical activity among women. The relationship between self-rated change in health and short-term mortality was explained by age among men and women. The relationship between self-rated change in health and long-term mortality was explained by age and physical health among men and women. Social engagement was an independent predictor of short- and long-term mortality among men and women in this study. IMPLICATIONS The finding that low self-rated health was not an independent predictor of mortality among men or women, contrary to many, but not all, previous studies, may be related to differences in study design and/or across cultures. Further research investigating relationships between self-rated health and mortality and potential explanatory variables should analyze men and women separately and should consider the length of follow-up period. The benefits of individual physical and social activities in reducing mortality merit further investigation.