The Robustness of the Different Health Measures with Respect to Life Style Choices

Do different health variables measure the same thing? In this paper we investigate the robustness of the effects of life-style choices on (1) self-assessed general health status, (2)problems with undertaking daily tasks and chores, (3) mental health indicators, (4) BMI, (5) the presence of serious long-term health conditions, and (6) mortality. The lifestyle choices we consider are regular exercise, being a smoker and the amount of alcohol consumed. We furthermore distinguish between short-run effects and long-run effects, and estimate both ordinal models and cardinal models. We estimate the models using longitudinal data drawn from the US Health and Retirement Study (HRS) between 1992 and 2002. We find surprisingly large differences in effects of lifestyle on the health measures and a general lack of consistency between our measures. Exercise is found to significantly reduce mortality both in the short and long-run, but it has little effect on stated health or doctor-assessed health measures. Importantly, smoking is found to have a long-run effect on mortality, but smoking improves self-stated health, reduces the problems individuals have with doing daily chores, improves mental health, reduces the number of measured serious illnesses and reduces Body Mass Index (BMI). Finally, we find no short-run or long-run benefits of income or wealth, implying that the effect of wealth and income would have to work via the increased levels of exercise and reduced levels of smoking associated with higher income and wealth levels.