Objective: Patients undergoing allogeneic hematopoietic stem cell transplantation (HSCT) often demand a single estimate of their prognosis from their treating physician. Little is known about the validity of and the factors influencing these estimates.
Patients and methods: We explored physicians' prognostic estimates (six-point scale from very good to very poor) in a sample of 136 patients participating in a psycho-oncology study. Patients (mean age, 41 years) were followed-up for at least 2 years, and 72 patients died during the follow-up period.
Results: Physicians' estimates were clearly associated with overall survival (univariate Cox regression: hazard ratio=1.51, 95% confidence interval=1.24-1.82). In a multivariate Cox regression model, these estimates were independent predictors for survival in addition to previous treatment (former HSCT) and donor type (related vs. unrelated).
Discussion: Physicians on transplant units are able to give a concise prognostic estimate that corresponds to real outcomes. Patients often do not have a sufficient understanding of the risks of their treatment. Obtaining such condensed information instead of a complex prognostic estimate might be easier to understand. For the physician, it might be helpful to reflect his/her individual method of condensing information to be able to give a concise prognosis and to achieve informed consent with the patient. A systematic description of the extensive implicit and explicit experiences regarding objective and subjective data and its respective weight could help to improve the training of future hematologists and should be explored in further studies.