This study demonstrates the importance of analyzing survival by cause of death in order to achieve a better understanding of the prognostic indicators involved. It further emphasizes the need for analysis of risk factors in both univariate and multivariate models, and the danger of making judgements based on premature analysis of data on follow-up after heart transplantation. Survival following transplantation is characterized by the major hazards of early death due to infection and rejection and late graft loss due to coronary occlusive disease (COD). This study summarizes the first-graft survival experience for 323 transplant patients at Papworth Hospital, and assesses a number of potential risk factors for (1) early mortality, (2) late mortality from COD, and (3) development of COD. The potential risk factors considered for all hazards are donor and recipient age, sex, blood group, and matching of these factors; donor cause of death and recipient immunosuppression; inotropic support; waiting time; preoperative diagnosis and previous cardiac surgery; ischemic time; and extubation time. In addition, for development of, and graft loss from, COD, perioperative rejection and cytomegalovirus infection; hypertension at discharge; and cholesterol, triglycerides, and lipids at two years were assessed as risk factors. Advances in immunosuppression were observed to have increased overall survival rates and decreased mortality from infection, rejection, and COD, as well as decreasing morbidity from COD. Fatal rejection was found to be more likely in female recipients, recipients over 40 years, recipients of grafts from donors over 30 years old, patients who were transplanted for valvular heart disease, and patients who waited less than three months for their transplant. Male recipients of female donor organs were more likely to lose their grafts as a result of COD. Patients older than 50 and hearts from donors older than 40 conferred a high risk of development of and loss from COD. Patients transplanted for ischemic heart disease were more likely to develop COD. High cholesterol, low HDL, high LDL, and high triglycerides at two years after transplant showed some evidence of high risk for the subsequent development of COD, although these relationships are not statistically significant at this stage. Contrary to other recent studies, cytomegalovirus infection was not found to be a risk factor for the development of COD.