The aim of this prospective epidemiologic study is to examine the relationships of physical activity (PA) and physical fitness (PF) to heath in a defined population of men and women in the middle to later years. Health is broadly defined and includes measures of cause-specific and all-cause mortality, non-fatal diseases (hypertension and diabetes), and functional capability. The population under study is 28,600 men and 9,400 women examined 1970- 1987 in a preventive medicine clinic. This cohort will contribute 343,000 person-years of follow-up for analyses, which by conservative estimate should yield 1,343 deaths for analysis. The database includes demographic, behavioral, and clinical data from the baseline clinical examinations; health habit and health status measures from mail surveys; and mortality information from official death certificates. Analyses will focus on baseline PA and PF as possible correlates of health during follow-up. The extensive data base permits control for possible confounding variables such as smoking status, blood lipids, body composition, blood pressure, and faMily history. Approximately 14,000 patients have two clinic examinations, with follow-up after the second examination estimated to be 121,000 person-years. In this sub-group, analyses will be done on change in PA and PF, as related to health. The strengths of the proposed research include the extensive with carefully designed measures on both PA and PF, the magnitude of follow-up, a relatively large number of women, and the ability to examine baseline PA and PF and changes in these variables for relation to health.