[Estimation of risk as a function of risk factors. II.--Application to the multivariate analysis of coronary heart disease in a middle aged male population (author's transl)].

The annual incidence of coronary heart disease (CHD) in a 8000 male professionnal group, aged 42-53, is computed from the model of LELLOUCH and al. described in the first part of this work. It is shown that each of five risk factors (cholesterol, blood pressure, cigarettes consumption, diabetes history and abnormal electrocardiogram) brings its own contribution to the CHD risk, which may be estimated by an exponential function of a linear combination of the levels of the five factors. "Risk tables" useful in current practice are described.