1500 men (aged 30-55 years) volunteered in a trial of early recognition of cardiovascular disease. History and clinical as well as biochemical data were obtained and compared with results of exercise ECGs. In the group as a whole abnormal ECGs were obtained in 6.2%, in a subgroup of those without exercise-dependent pain in the thorax, without hypertension and noraml resting ECG it was 3.8%, while in the presence of one or several of these risk factors it was as high as 17.8%. It is concluded that in any screening programme it is not reasonable to perform exercise ECGs in asymptomatic persons without risk factors, because in this group there is likely to be only a small percentage of abnormal findings and the number of false-positive ones is higher than that or correct positive ones. In connection with the known risk factors and risk indicators it would be best to define the risk group and, using selective indications, to go step-by-step from exercise ECG to additional invasive diagnostic measures such as coronary angiography.