The number of allogeneic BMTs performed worldwide continues to increase. In addition, the characteristics of patients, donors, and selected treatments are changing. In BMT for leukemia during the 1980s there was a marked increase in transplants for early disease, the use of unrelated donors, and utilization of preparative regimens without radiation. Although treatment-related mortality declined during this period, there was only a modest decrease in relapse rates, indicating the need for more effective antileukemia strategies. The IBMTR collects data from many centers, and, as a consequence, it is uniquely suited to examine clinical situations in which patient accrual at a single institution would be insufficient for a study to be performed. In such analyses, BMT was shown to be an effective treatment for Ph1-positive ALL. Patients tended to have earlier relapses and lower probabilities of LFS than Ph1-negative ALL but the differences were not statistically significant. BMT was also shown to be effective in patients with acute leukemia failing to ever go into remission, most of whom would die within the first 6 months following diagnosis. The IBMTR is a premier example of international scientific collaboration. Its success is a consequence of the desire of investigators throughout the world to combine their clinical data for statistical analysis in order to accelerate and improve patient care.