A follow-up study was conducted to clarify the relationship between physical-strength level and risk of death from all causes and from cancer and cardiovascular disease. The 7286 persons who were examined at seven health-promotion centers throughout Japan between 1982 and 1987 were followed up. By January 1992, 6259 persons (85.9%) had been contacted by questionnaire. They included 3117 men (49.8% of all subjects studied) (average age 53.6 years at baseline, SD = 9.0 years, range 40-84 years), and 3142 women (50.2%) (average age 54.5 years at baseline, SD = 8.5 years, range 40-85 years). The follow-up period for each person averaged 6.1 years, for a total of 38,253 person-years. During this period, 155 deaths were reported. At baseline, five physical-strength tests (grip strength, side step, vertical jump, standing trunk flexion, and sit-ups) were performed. Five clinical laboratory tests (thickness of skinfold, blood sugar, total serum cholesterol, percent vital lung capacity, and blood pressure) were also conducted. The examinees were questioned about smoking status (current smoker, nonsmoker, and ex-smoker). Men with thicker skinfold [relative risk (RR) = 2.11] and higher levels of blood sugar (RR = 1.89) had an excess risk of death from all causes. Men with higher serum cholesterol (RR = 5.08), thicker skinfold (RR = 4.54), and elevated blood pressure (RR = 2.33) had an excess risk of death from cardiovascular disease. In women, no relationship was seen between clinical laboratory tests and an excess risk of death. Men exhibiting lower values for side step (RR = 2.43), vertical jump (RR = 2.37), sit-ups (RR = 1.93) and grip strength (RR = 1.92) also had an excess risk of death from all causes. Furthermore, men with lower heights for vertical jump (RR = 5.51) had an excess risk of death from cardiovascular disease. After adjustment for skinfold thickness, blood sugar, total serum cholesterol, blood pressure, percent vital lung capacity and smoking status, men with a lower level of side step, vertical jump, and grip strength had an excess risk of death from all causes. No such relationship was seen between physical-strength level and an excess risk of death in women. It is concluded that a low level of physical strength might be significantly correlated with subsequent health outcomes in men.