Characterizing Hospital Admissions after Proton Beam Therapy Using the National Inpatient Sample Database.

Abstract

To characterize the diagnoses, comorbidities, and length of stay in patients who received proton beam therapy during their hospital stay as part of their cancer treatment regimen. The National Inpatient Sample database was queried using ICD-10 codes to identify all hospitalizations with proton beam therapy between 2016 and 2018. Weighted frequencies for categorical variables and geometric mean with standard error for continuous variables were derived. Generalized linear models for clustered data in SAS v 9.4 (Cary, NC) were used to determine the association between length of stay with patient and hospital baseline characteristics. We studied 511 patients, representing a weighted estimate of 2,555 patients from the National Inpatient Sample from 2016 to 2018 who received proton beam therapy during hospitalization. The mean age was 40 years (StdErr = 3.0). The estimated average length of stay was 7.7 days (StdErr = 0.4). The cohort was 53.6% White, 15.3% Black, 10.4% Hispanic, 7.0% Asian or Pacific Islander, and 0.2% Native American. The most frequent diagnoses at admission were encounter for antineoplastic chemotherapy (10.0%), secondary malignant neoplasm of brain (9.2%), secondary malignant neoplasm of bone (9.0%), encounter for antineoplastic radiation therapy (6.1%), and neoplasm-related pain (3.3%). Large- and medium-sized hospitals were associated with significantly longer lengths of stay than small-sized hospitals (p<0.0001 and p = 0.0186, respectively). Compared to private investor-owned (proprietary) hospitals, nonfederal government-owned (public) hospitals are associated with prolonged length of stay (p = 0.0033). Hospital region and patient age, race, sex, comorbidity, and income quartile were not associated with a longer length of stay. Our findings demonstrate that hospital-level characteristics are more important than patient-level characteristics in predicting length of stay in patients undergoing proton beam therapy while hospitalized. Patients undergoing proton beam therapy while hospitalized at large- and medium-sized hospitals had longer lengths of stay than those at small-sized hospitals. Additionally, publicly owned hospitals were associated with prolonged length of stay compared to private hospitals. The underlying reason for this difference in length of stay is unclear but may suggest different management strategies, protocols, and discharge criteria for patients receiving proton beam therapy in larger and publicly-owned hospitals compared to smaller and privately-owned hospitals. Future investigations should explore the underlying causes for these discrepancies and identify potential strategies to shorten unnecessarily prolonged hospital stays in patients undergoing proton beam therapy.