BACKGROUND

Aims: The aim of the study was to compare the appropriateness of different obesity indicators in the assessment of coronary heart disease (CHD) risk. Methods: The study cohort included 11,510 Finnish men and women aged 25 to 64 year at baseline who participated in a cardiovascular disease risk factor survey in 1987 or 1992. At baseline, data on smoking and diabetes were recorded, blood pressure, body mass index (BMI), waist circumference (WC), and waist to hip ratio (WHR) were measured, and serum total and high-density lipoprotein (HDL) cholesterol were determined. A follow-up was done to the end of 1997. Death or diagnosed event from CHD was used as an outcome variable. Results: At baseline, BMI was the best explaining variable for systolic and diastolic blood pressure (DBP) and for total cholesterol, whereas WC was the best explaining variable for HDL cholesterol, among both men and women. During the follow-up, WHR was the best predictor of CHD incidence. However, after the adjustment for other CHD risk factors none of the obesity indicators remained statistically significant. In both sexes, BMI was a statistically significant predictor of CHD incidence among subjects with DBP lower than the mean. Among men, a similar interaction was seen between DBP and WC. Conclusion: WHR was the best predictor of CHD incidence in our data. Abdominal obesity has an effect on CHD incidence independently of general obesity