Risk Factors for 100-Day Readmission after Discharge in Allogeneic Hematopoietic Cell Transplant and Its Clinical Implications

Abstract

Introduction Hospital readmissions after hematopoietic cell transplant (HCT) carry significant morbidity and mortality in HCT recipients. Herein, we describe a single institution 100-day readmission rate, its risk factors and clinical implications using post-HCT cytoxan as graft vs. host disease (GVHD) prophylaxis. Methods Data on 125 consecutive allogeneic HCT (allo-HCT) patients at Levine Cancer Institute, Charlotte, NC, between March 2014 and May 2018 were reviewed for dates of readmission, reasons for admission, patient demographics, socio-economic status and clinical outcomes. The cohort was divided into 2 groups, readmission vs. no readmission, based on admission to the hospital within 100 days of discharge from index transplantation admission. Study variables were compared between the 2 groups via a descriptive statistical analysis. Fisher's exact test and two-sample t-test were used to calculate P-values for categorical and continuous variables, respectively. Univariate and multivariate logistic regression analysis was used to assess risk factors for readmission. Overall survival (OS) after the initial hospital discharge was evaluated between patients with and without 100-day readmission using the Kaplan-Meier method and log-rank test. The cumulative incidence of non-relapse mortality (NRM) was estimated in a competing risk setting with relapse related death as a competing event. Group comparison of incidences were determined by Gray's test. The effect of readmission on NRM was evaluated in a Cox regression model. Results 52 (41.6%) of 125 allo-HCT patients were readmitted after transplantation admission. Median time to readmission from hospital discharge was 22.5 days (range: 1-96) and median length of stay (LOS) was 12 days (range: 1-70). The most common causes for readmission were infectious complications (57.7%) and GVHD (13.5%). In univariate analysis for 100-day readmission (Table 1), variables such as primary insurance payer, graft cell source, infections and GVHD during index admission, LOS, marital status and ANC recovery did not affect readmission risk, however being non-Caucasian (n=33, 26.4%) had increased odds ratio (OR) of 2.05 (p=0.081). OS was also similar between the 2 groups (P = 0.189). The incidence of NRM for patients with and without readmission was 25.1% and 6.9%, respectively (P = 0.047). Multivariate Cox regression analysis identified 100-day readmission (HR 2.96, 95% CI 1.01-8.71, P = 0.043), and LOS (HR 1.48, 95% CI 1.05-2.10, P = 0.027) as significant risk factors for NRM (Table 2). Conclusions Although readmission after allo-HCT is associated with increased NRM, we could not identify any risk factor for readmission within 100 days of discharge at our institution. We observed increased OR for readmission amongst non-Caucasian race in univariate analyses, but it did not emerge as a significant variable in multivariate analysis.