The Social Insurance Institution's Coronary Heart Disease Study is a prospective population study designed to investigate the prevalence, risk factors and incidence of coronary heart disease (CHD) in middle-aged Finnish men and women. The study population consisted of 5 738 men and 5 224 women, aged 30-59 years at entry, drawn from 12 cohorts from south-western, western, central and eastern Finland. The cohorts consisted of whole or random samples of rural or semiurban dwellers or employees of a factory. The participation rate was 90 per cent. The prevalence of symptoms was determined by the Rose questionnaire and abnormalities on resting ECG were coded according to the Minnesota code. Blood pressure, smoking habits, serum cholesterol, triglycerides, postload plasma glucose and obesity were the risk factors analysed at the baseline examination. The mortality of examinees has been followed continuously. This report deals with the main findings at the baseline examination and the mortality follow-up experience in 5 years. The prevalence of typical angina pectoris was 4.4 per cent in men and 5.4 per cent in women. Unequivocal ECG signs of past myocardial infarction were observed in 1.0 per cent of men and 0.3 per cent of women. Other ECG findings suggesting CHD were observed in 9.2 per cent of men and 11.1 per cent of women. The 5-year mortality was 4.3 per cent in men and 0.9 per cent in women. Men with typical chest pain symptoms had a seven-fold risk to die from CHD, compared to men without symptoms. Men with ECG abnormalities compatible with an old infarction had a 19.5-fold and men with other ECG findings suggesting CHD a 7.1-fold risk to die from CHD compared to men without resting ECG abnormalities. Men with ECG findings as the only indicator of CHD had worse survival than men with symptoms as the only indicator of CHD. The value of symptoms and ECG findings as predictors of CHD mortality in women was very low.