Factors that influence the effect of coronary artery bypass graft surgery (CABG) on employment were examined in 1427 2-year survivors from entry into the nonrandomized Mayo Clinic portion of the National Heart, Lung, and Blood Institute's prospective Coronary Artery Surgery Study. Five hundred seventy-nine patients (41%) were treated medically, 769 (54%) were treated surgically and 79 (5%) were initially treated medically but later had CABG. All groups were similar at entry with respect to age and percent working, retired or quitting work for cardiac reasons. There was a similar decline in employment from entry to 2 years in all three groups. The number who quit work for cardiac reasons was twice the number of previously disabled workers who returned to work with all three modes of therapy. Univariate and multivariate analyses were used to test the independent prognostic values of a number of variables for subsequent employment in the largest homogeneous surgical subset--496 nonretired males. Postoperative angina was the most powerful factor reducing postoperative employment. Postoperative employment was also lower if the patient was not working before surgery, was a laborer, or older than age 55 years (all p less than 0.01). Although the level of nonwork physical activity was generally increased after CABG, there was no relationship between physical activity and employment. This study confirms that employment declines after CABG, but three of the four most powerful prognostic variables were not related to surgery.