Mortality Among Neutropenic Cancer Patients Within the United States: The Association With Hospital Volume

Abstract

PURPOSE: Neutropenia is a serious complication of chemotherapy in patients with solid tumors. The influence of hospital volume on outcomes in patients with neutropenia has been little investigated. We hypothesized that large-volume hospitals would have reduced mortality rates for neutropenic patients compared with small-volume institutions. METHODS: We used the Nationwide Inpatient Sample database of the Healthcare Cost and Utilization Project, for the years 2007-2011. All adult inpatient episodes with a diagnosis of both neutropenia and solid-tumor malignancy were included. Hospital volume was defined as the number of neutropenic cancer episodes per institution per year. Mortality was defined as death during admission. A multilevel mixed-effects logistic regression model was applied. RESULTS: Twenty thousand three hundred and ten hospitalizations were included in the study, from 1,869 different institutions. Median age was 62 years. The overall inpatient mortality was 2.3%, and was dependent on age (age 50-59 years—1.6% and age 80-89 years—5.3%). The median number of neutropenic inpatient episodes in each institution per year was 14 (range, 1-168). Mortality was 3.3%, 2.7%, 2.2%, 2.2%, and 1.2% for each quintile of hospital volume (from lowest to highest volume, P < .001). Likewise, the proportion discharged home was 85.7%, 90.3%, 91.5%, 92.7%, and 95.4% ( P < .001). The association between hospital volume and mortality remained significant after adjustment for patient-level and hospital-level variables. DISCUSSION: Patients with neutropenia hospitalized in large-volume institutions have a substantially lower mortality compared with those hospitalized at low-volume institutions. Further study is required to validate our findings or overcome potential biases, understand mechanism, and investigate how smaller institutions can improve outcomes.