The Healthy Women Study has evaluated risk factor changes as women go through the menopause. In 274 women, measures of atherosclerosis in the coronary, aorta and carotid arteries were obtained at 8 years postmenopause using electron beam tomography and carotid ultrasound. The purpose of this study was to evaluate the association between baseline premenopausal risk factors and subclinical disease in various vascular beds and with an aggregate atherosclerosis index based on all subclinical measures. The prevalence of measurable disease for the coronaries, carotids and aorta was 45, 55 and 72%, respectively. Disease measures across these three locations were significantly correlated (P<0.001). Premenopausal LDL, triglycerides, blood pressure (BP) and smoking were associated with each measure of subclinical atherosclerosis, and were strikingly associated with the number of elevated disease measures. Among 126 women taking hormone replacement therapy (HRT), there was a strong positive association between systolic BP and subclinical atherosclerosis (P<0.001). No such relationship was seen among women not on HRT and this interaction was significant (P=0.048). An index of subclinical atherosclerosis rather than measurement at a single arterial site may be the best estimate of disease risk. Such a measure should be evaluated for its prognostic value along with individual measures of subclinical disease. Women with lower premenopausal risk factors did not develop high levels of subclinical atherosclerosis, suggesting that atherosclerosis is a preventable disease in women.