The aims of this study were to determine the prognosis for women with different symptoms and signs indicating ischaemic heart disease and to relate entry characteristics to events of ischaemic heart disease and stroke, and to overall mortality during a 12-year follow-up period. A prospective cohort study was started as a cross-sectional investigation in 1968-69 (1462 participants aged 38-60, participation rate 90.1%). The same sample was re-studied in 1980-81 (1154 participants, participation rate 78.9% of those studied in 1968-69). In addition, a clinical series comprising all women of similar age in Gothenburg with myocardial infarction during the years 1968-70 (47 women) was followed up for 12 years. The 12-year overall mortality rate for women with initial myocardial infarction in the clinical series was 45%. The 12-year overall mortality rates for women who at the time of the initial study either were considered to have angina pectoris or showed electrocardiographic changes indicating ischaemic heart disease at rest or at work were 10%, 17% and 10% respectively (expected figures 7%, 12% and 10%). Twenty-three women (1.6%) developed myocardial infarction during the follow-up period (8 fatal, 15 non-fatal). New symptoms of angina pectoris were recorded in 56 women (4.0%), new electrocardiographic changes indicating ischaemic heart disease in 73 women (6.0%), and new signs of stroke in 13 women (0.9%). Altogether 75 women (5.1%) died during the follow-up period. These five end-points were taken into consideration. Women with previously untreated arterial hypertension were offered control visits during the follow-up period and were prescribed antihypertensive drugs when clinically indicated. In this group of women, hypertension was not a predictor for any of the end-points. No other systematic intervention was carried out. Increased abdominal adiposity, increased serum triglycerides and low peak expiratory flow were independent predictors of at least three of the end-points studied: myocardial infarction, stroke and death. Increased abdominal adiposity was also an independent predictor of angina pectoris. Initial diabetes was an independent predictor of both myocardial infarction and of death. Low energy intake predicted myocardial infarction and electrocardiographic changes indicating ischaemic heart disease. High serum gastrin levels predicted myocardial infarction. Low education and high degree of mental disorder were independent predictors of angina pectoris. Physical inactivity at leisure and physical inactivity at work were independent predictors of stroke and of death, respectively.(ABSTRACT TRUNCATED AT 400 WORDS)