The relationship of cardiovascular risk factors measured at different ages to prediction of all-cause mortality and longevity.

The objective was to study the relationship of cardiovascular risk factors measured at different ages to all-cause mortality in a population-based sample of middle-aged Italian men over a 30-year period. Two rural populations of men aged 40-59 were examined in 1960 (n=1712, participation rate 98.8%). Cardiovascular disease risk factors measured at year 0, 10 and 20 were used for the multivariate prediction of all-cause mortality during 30, 20 and 10 years, respectively. In men aged 40-59 with follow-up of 30 years, age, systolic blood pressure, serum cholesterol, cigarette smoking, body mass index, arm circumference (inverse), vital status of father and mother, forced expiratory volume (inverse), arcus senilis and xanthelasma were strong, significant predictors of mortality. In men aged 50-69 followed for 20 years, age, systolic blood pressure, cigarette smoking, vital capacity (inverse), body mass index and arcus senilis were strong, significant predictors of mortality. In men aged 60-79 followed 10 years only age, arm circumference (inverse) and arcus senilis were significant predictors of mortality (respiratory measurements were not available). On two occasions body mass index and serum cholesterol showed curvilinear relationships with mortality (U-shaped). It is concluded that cardiovascular disease risk factors are strong predictors of all-cause mortality. Their predictive power (relative risk) declines with age. Some factors, such as body mass index and serum cholesterol level, show curvilinear relationship to mortality with greater age.