AIMS: Prospective epidemiological studies demonstrate an increase in coronary heart disease mortality in women beginning at values of body mass index > or = 22 kg. m(-2). However, the metabolic basis for this observation has not been adequately studied in women. Our aim was to examine the association between body mass index, metabolic coronary heart disease risk factors and a predicted 10-year coronary heart disease risk score in a large occupational cohort of women in the U.K. METHODS AND RESULTS: We carried out a cross-sectional survey of cardiovascular risk factors in 14 077 women, aged 30-64 years. The main outcome measures were systolic and diastolic blood pressure, serum total cholesterol, HDL cholesterol, total cholesterol/HDL cholesterol ratio, LDL-cholesterol, triglycerides, apolipoprotein A1, apolipoprotein B, lipoprotein(a), fasting blood glucose and a predicted 10-year coronary risk score. Across seven categories of body mass index, i.e. < 20, 20-, 22-, 24-, 26-, 28- and > or = 30 kg. m(-2), there were highly significant age-adjusted increases in the risk factors (all P < 0.001), except for a decrease in HDL cholesterol and ApoA1 (all P<0.001) and no relationship with lipoprotein(a) (P = 0.05). Based on a multifactorial 10-year coronary heart disease risk estimate, odds ratios for being in the highest quintile of risk for each category of body mass index, were 1 (< 20 kg. m(-2)), 0.91, 1.56, 2.18, 2.97, 3.83 and 4.21 (> or = 30 kg. m(-2)). CONCLUSION: The significant rise in metabolic coronary heart disease risk at 22 kg. m(-2)observed in this study is consistent with prospective epidemiological studies in women which have reported an increase in coronary heart disease mortality starting at 22 kg. m(-2). However, body mass index was a poor discriminator of women at different levels of coronary heart disease risk. The primary goal of weight loss in individuals should be the correction of dysmetabolism, irrespective of the level of body mass index.