Previous research suggesting a link between psychosocial variables and survival after bone marrow transplant (BMT) has been limited by: a) retrospective assessment of psychosocial variables; and b) failure to concurrently examine a comprehensive set of disease, treatment, and demographic variables potentially related to post-BMT survival. The present study prospectively assessed psychosocial variables (depressed mood, functional quality of life, and mental adjustment to cancer) that have been linked to survival after BMT and/or malignant disease. Study participants (N = 42) received allogeneic BMT for either acute or chronic leukemia. Analyses using Cox proportional hazards regression indicated that quality of bone marrow graft match was the only disease, treatment, or demographic variable significantly associated with post-BMT survival (p = .05). Addition of psychosocial variables to a multivariate Cox regression model including quality of graft match suggested that an attitude toward cancer characterized by "anxious preoccupation" (p = .008), as well as poorer functional quality of life (p = .052), were each independently associated with poorer post-BMT survival. Further research is necessary to identify the mechanisms by which psychosocial variables could contribute to post-BMT survival.