Objective : To examine the association between hospital procedure volume and treatment outcomes following allogeneic bone marrow transplantation (allo-BMT). Methods : Out of 1,050 patients who received allo-BMTs between 1998 and 2000 in 21 Korean hospitals, 752 with first allo-BMT and complete data were included in this study. Study subjects were divided into the following three groups according to cumulative hospital experience of all-BMTs during the study period: low (<30 cases), medium (30-49) and high ($geq$50 cases) volume. Patient outcome was defined as early survival at day 100 and one-year survival. Multiple logistic regression analyses were performed to examine the association between hospital experience and survival at day 100 and one year. Results : When the low volume group was defined as the reference group, the adjusted relative risks (RR) of survival at day 100 for the high volume group were 2.46(95% CI, 1.13-5.36) for all patients, 2.61(1.04-6.57) for those with leukemia, and 2.20(0.47-10.32) for those with aplastic anemia. For one-year survival, adjusted RR for the high volume group were 2.52(1.40-4.51) for all patients, 1.99 (1.01-3.93) for leukemia, and 6.50(1.57-26.80) for aplastic anemia. None of the RR for the medium volume group was statistically significant. Patient factors showing significant relationship with survival were donor-recipient relation, human leukocyte antigen matching status, time from diagnosis to transplant, and disease stage. Conclusions : The study results suggest that the cumulative experience of hospitals in providing allo-BMT is positively associated with patient survival.