To predict the effects of using exercise testing to screen healthy persons for coronary artery disease.
The publications of the Coronary Artery Surgery Study were the principal sources of data for the cost-effectiveness analysis; we also used data from the Veterans Administration Cooperative Study of stable angina and the publications of the European Coronary Surgery Study Group.
We used studies that provided the data that our decision model required--life expectancy and probabilities of outcomes in persons who have or do not have coronary artery disease.
We did not use a structured method for abstracting data. We tested the susceptibility of our conclusions to poor quality of data by substituting a wide range of values for a variable in the decision model, and by calculating life expectancy and costs if screening was used routinely or not used.
We assumed that persons with an abnormal exercise test would have arteriography and that persons with severe coronary artery disease would have bypass surgery. When there were no suitable published data for the model, we made assumptions that favored screening. The model predicts that screening would increase the life expectancy of 60-year-old men at average risk by at most 12 days. Sixty-year-old men with no risk factors for coronary artery disease would derive less benefit, as would women and younger men.
The effect of exercise testing is too small to justify doing this procedure routinely in healthy persons. If coronary bypass surgery is found to prolong life in asymptomatic persons as much as it does in angina pectoris, screening older men with risk factors for coronary artery disease may prove to be worthwhile.