VALUE OF SOME FACTORS FOR PROGNOSIS OF MORTALITY OF WORKING AGE POPULATION (PROSPECTIVE EPIDEMIOLOGIC-STUDY)

Forty six hundred fifty four men aged 20 to 59 and 863 women aged 35 t o 54 were examined by means of standard epidemiological methods and fo llowed up for 11 and 8 years, respectively. Associations between 16 fa ctors and mortality were estimated by calculating relative and attribu table risks, by Cox regression model and construction of prognostic sc ore XP. Coronary heart disease status at entry, age, systolic blood pr essure, smoking, occupational status ''blue collar work'' and total ch olesterol were found predictive of coronary heart disease mortality in men. Systolic blood pressure and age were significant for cerebrovasc ular mortality. Systolic blood pressure, coronary heart disease status at entry, age, and smoking were positively while occupational status ''white collar work'' - negatively related to cardio-vascular mortalit y. Variables predictive of cancer mortality were: age, ''blue collar w ork'', number of cigarettes per day ((positive relationship) and body mass index (negative relationship). Alcohol consumption and smoking we re predictors of mortality from external causes. The prognostic scores XP obtained in this study made it possible to calculate 11-year risk for coronary heart disease, cerebro-vascular disease and coronary hear t disease mortality in men aged 35 - 59 residing in Tallinn at individ ual and population level. It was also possible to calculate by means o f X beta the potential reduction of coronary heart disease, cerebro-va scular disease and coronary heart disease mortality which would occur after decrease of a certain risk factor level. The number of end point s in the female cohort was too small for performing multivariate analy sis. Univariate analysis showed the significance of systolic blood pre ssure (positive relationship) and educational status (negative relatio nship) for total mortality; smoking was significant for cancer mortali ty.