Objectives. This analysis was performed to quantitatively assess the relative risks, associated with underlying cardiovascular disease, incurred in the course of intense competitive sports. Background. Sudden cardiac death during athletic activities is a highly visible event, and controversy persists regarding the true risks associated with participation in sports. Methods. The prevalence of sudden death was assessed in two systematically tabulated groups of endurance runners competing in the annual Marine Corps (1976 to 1994) and Twin Cities (1982 to 1994) marathons, held over a cumulative 30-year period. Results. A total of 215,413 runners completed the races, and four exercise-related sudden deaths occurred, each due to unsuspected structural cardiovascular disease. Three deaths occurred during the race (after 15 to 24 miles [24 to 38.4 km]) and the other immediately after its completion. The ages were 19 to 58 years (average 37), and three were men. Three of the sudden deaths were due to atherosclerotic coronary artery disease (narrowing of two or three vessels) and one to anomalous origin of the left main coronary artery from the right sinus of Valsalva. None of the four runners had prior documentation of heart disease or experienced prodromal symptoms, and two had previously completed three marathon races each. The overall prevalence of sudden cardiac death during the marathon was only 0.002%, strikingly lower than for several other variables of risk for premature death calculated for the general U.S. population. Conclusions. Although highly trained athletes such as marathon runners may harbor underlying and potentially lethal cardiovascular disease, the risk for sudden cardiac death associated with such intense physical effort was exceedingly small (1 in 50,000) and as little as 1/100th of the annual overall risk associated with living, either with or without heart disease. The low risk for sudden death identified in long-distance runners from the general population suggests that routine screening for cardiovascular disease in such athletic populations may not be justifiable.