Coronary risk factors predicting coronary and other causes of death in fifteen years.

A multivariate analysis employing the multiple logistic function model has been performed for the prediction of coronary heart disease (CHD) deaths and of other causes of death as function of 14 coronary risk factors measured at entry examination in the pool of two Italian rural population samples, made of 1712 men aged 40-59 at entry and followed for 15 years. A limited number of factors--namely age, serum cholesterol, blood pressure, smoking habits, forced expiratory volume, diabetes--yielded significant coefficients variously associated each other in the different solutions. They were able to provide a satisfactory discrimination between cases and non cases, not only for CHD but also for other end-points including strokes, cancer and lung cancer in particular, chronic bronchitis, and all causes of death. A suggestive prediction of violent causes of death was possible thank to a significant coefficient attributable to blood pressure. Serum cholesterol was significantly contributory only in the prediction of CHD. The multipotentiality of some factors is stressed in view of the planning of community prevention programs directed towards several chronic conditions at the same time.