To understand the time relatedness of the various causes of death after heart transplantation, a study was undertaken to examine the relative importance of each of these causes of death across time, as well as between eras. The analysis, which was in the competing outcomes domain, involved 326 patients undergoing primary heart transplantation between November 1981 and June 1994, of which 122 patients died. Two eras were considered--an earlier era (1981 to 1986) and a later era (1987 to 1994). During the later era, the estimated survival rate at 1 year and 5 years was 85% and 69%, respectively. During the first posttransplantation year, patients in the earlier era were three times more likely to die of early graft failure than in the later era, but the difference may have been due to chance (p > 0.2). During the first posttransplantation year, patients were three times as likely to die of acute rejection in the earlier era (p = 0.02). Patients surviving 3 or more years in both eras had a similar (approximately 3%) likelihood of death from coronary artery disease over the ensuing 4 years. In the later era, the risk of dying of malignancy (2.5%), infection (3.2%), or miscellaneous other causes (4.1%) was as likely after 3 years as dying of coronary artery disease (3.2%). Competing outcomes analysis is a useful means of portraying the distribution of multiple time-related causes of death. In the current era, the deleterious long-term effects of immunosuppression (malignancy and infection), as well as coronary artery disease, threaten long-term survival.