Psychosocial Factors and Overall Survival After Autologous and Allogeneic BMT

Various psychosocial factors have been assessed in BMT patients (pts), but there are few reports of a comprehensive psychosocial evaluation pre-BMT predicting outcomes post-BMT. We retrospectively analyzed 204 consecutive first adult (≥18 years) autologous (n = 108) or allogeneic (n = 96) BMT pts treated from 7/05 to 5/08 to analyze BMT outcomes in relation to a comprehensive psychosocial evaluation. 200/204 (98%) had a psychosocial inventory assessment collected by in-person interview with a BMT social worker during their routine pre-BMT evaluation. The psychosocial inventory had questions regarding employment, marital status, education, financial concerns, transportation and local lodging needs, degree of understanding of diagnosis and treatment, tobacco, alcohol and drug use, importance of and comfort with religious beliefs, and personal or family suicidal ideation or attempt. Median age was 53 years (range 18–74), 62% were male, 93% were White, diagnoses included leukemia (n = 73), lymphoma (n = 71), myeloma (n = 36), hematologic disorders (n = 17), and solid tumors (n = 3), donors were autologous (n = 106), related (n = 41), or unrelated (n = 53). Median follow-up in the survivors was 1.1 years (range 0.1–3). By univariate analysis, significant predictors of decreased 1-year overall survival (OS) in allogeneic BMT pts were KPS ≤ 80 (47% vs. 75%, p = 0.003), marital status (married 39% vs. never married 69% vs divorced 90%, p = 0.016), not living alone pre-BMT (48% vs. 100%, p = 0.012) and smoking status (ever 42% vs. never 67%, p = 0.07). Marital status was significantly related to age with median 53 vs. 36.5 vs. 46.7 years of married vs. never married vs. divorced pts. By univariate analysis, the only significant predictor of decreased 1-year OS in autologous BMT pts was living >30 minute drive from the BMT center and needed local lodging (72% vs. 93%, p = 0.006). For allogeneic BMT, multivariate analysis demonstrated KPS ≤ 80 (RR = 4.6, 95% CI 1.4–15, p = 0.01), married compared to divorced (RR = 5.9, 95% CI 0.8–44, p = 0.08) and former/current smokers (RR = 1.8, 95% CI 0.9–3.4, p = 0.08) had an increased risk of death. Substituting age for marital status in the model yielded only KPS as a significant factor. Some psychosocial factors may be related to patient and disease related characteristics which may yield selection bias in the decision to transplant or not. Further analyses of psychosocial factors in larger, more homogeneous patient populations may help clarify risk subgroups.