Prolonged Hospitalization for Hematopoietic Cell Transplantation: Characteristics, Risk Factors and Associations with Patient-Reported and Clinical Outcomes.

Patients hospitalized for hematopoietic cell transplantation (HCT) may experience prolonged length of stay (PLOS). However, the associations between PLOS and patient-reported outcomes (PROs) during and after HCT hospitalization is unknown. We aimed to evaluate the associations of pre-HCT demographic and disease characteristics and PROs with PLOS, as well as the associations between PLOS and trajectory of PROs and risk of rehospitalization in the first year post-HCT. We conducted a secondary analysis of data from adult patients with hematologic malignancies undergoing HCT who were enrolled in a prospective observational study or one of two randomized clinical trials evaluating integrated specialty palliative care during HCT hospitalization. PLOS was defined as ≥30 continuous days for allogeneic HCT and ≥21 continuous days for autologous HCT. Quality of life (QOL; Functional Assessment of Cancer Therapy Bone Marrow Transplant), symptom burden (Edmonton Symptom Assessment Scale), anxiety and depression symptoms (Hospital Anxiety and Depression Scale and Patient Health Questionnaire-9), and posttraumatic stress symptoms (PCL) were measured at time of admission (ie, prior to HCT), 2 weeks, and 3 and 6 months post-HCT. Multivariate logistic regression was used to assess the association between pre-HCT PROs and PLOS adjusting for relevant covariates. Linear mixed-effects models were used to characterize the trajectory of PROs by PLOS during and after HCT. Cox proportional hazards regression was used to evaluate differences between LOS groups in time to readmission or death in the first year post-HCT. A total of 606 patients (mean age = 55.7 years [18.3 to 78.0 years]; 56.6% male; 81.5% White; 53.1% allogeneic HCT) were included. Patients with PLOS were younger (mean 53.3 versus 56.6 years, P = .004), in complete remission prior to HCT (52.8% versus 46.3%, P = .02), diagnosed with acute leukemia (34.2% versus 26.1%, P < .001), and underwent allogeneic HCT (62.1% versus 49.9%, P < .0001). In multivariate analyses, worse pre-HCT QOL (OR 0.99, P = .003), symptom burden (OR 1.02, P = .01), and depressive symptoms (OR 1.07, P = .01) were associated with higher risk of PLOS. Patients with PLOS reported worse QOL at 2 weeks (∆ = -12.3, P < .0001), 3 months (∆ = -6.9, P = .002), and 6 months post-HCT (∆ =-4.8, P = .02) compared to those without PLOS. Patients with PLOS reported greater symptom burden at 2 weeks (∆ = 10.2, P < .0001) and 3 months (∆ = 3.9, P = .04), but not 6 months post-HCT (∆ = 0.5, P = .79). Patients with PLOS reported higher depression burden at 2 weeks (∆ = 2.5, P < .0001) and 3 months (∆ = 1.1, P = .03), but not 6 months post-HCT (∆ = 0.6, P = .19). Patients with PLOS experienced shorter time to death or re-admission in the first year post-HCT (median 221 days versus not reached, HR 1.7; CI 1.3 to 2.2, P < .001). Pre-HCT PROs including QOL, symptom burden, and depressive symptoms were associated with PLOS. Moreover, patients with PLOS go on to experience worse QOL, symptom burden, and depressive symptoms up to 6 months post-HCT and are at an increased risk of mortality and greater healthcare utilization. Patients with PLOS may have unique needs compared to the usual HCT population and may benefit from augmented supportive care during and after HCT.