Association of Practice-Level Hospital Use With End-of-Life Outcomes, Readmission, and Weekend Hospitalization Among Medicare Beneficiaries With Cancer

To determine the relationships between hospital use of treating oncology practices and patient outcomes. Retrospective analysis of 397,646 Medicare beneficiaries who received anticancer therapy in 2012. Each beneficiary was associated with a practice; practices were ranked on the basis of risk-adjusted hospital use, that is, inpatient intensity. Outcomes included 30-day readmission, weekend admissions, intensive care unit stays in the last month of life, and hospice stay of ≥ 7 days. Outcomes were measured for each quartile of practice-level inpatient intensity. We fit multivariable logistic regression models to calculate adjusted odds ratios (ORs) for each outcome for each quartile of inpatient intensity. Total 30-day readmissions were 22.8% and 31.9% (OR, 1.45; 95% CI, 1.39 to 1.50) for patients in practices with the lowest versus highest quartiles of inpatient intensity, respectively; unplanned readmissions were 19.8% and 27.1% (OR, 1.36; 95% CI, 1.31 to 1.41), respectively. The proportion of admissions that occurred on weekends was similar across quartiles. Patients of practices in the highest quartiles of inpatient intensity had higher rates of death in an ICU stay in the last month of life (25.5% versus 18.0%; OR, 1.33; 95% CI, 1.19 to 1.49) and a lower rate of hospice stay of at least 7 days (50.9% to 42.5%; OR, 0.79; 95% CI, 0.74 to 0.86). Medical oncology practices that seek to reduce hospitalizations should consider focusing initially on processes related to end-of-life care and care transitions.