Background: The ranking of “America’s Best Hospitals” by the US News & World Report is a popular hospital-profiling system, but it is not known whether the 50 top hospitals for Neurology and Neurosurgery have lower 30-day risk-standardized mortality (RSMR) and readmission rates (RSRRs) after stroke compared with other hospitals, or whether outcomes are comparable among these 50 top hospitals.
Methods: The 50 top hospitals for adult Neurology and Neurosurgery were identified from the 2015-16 US News & World Report rankings. CMS Hospital Compare provided RSMRs and RSRRs within 30 days of ischemic stroke for hospitals with ≥25 stroke cases/year. We fit a mixed model to assess differences in RSMRs and RSRRs between top-ranked vs other hospitals, adjusted for hospital characteristics and weighted by hospital volume. Among the 50 top-ranked hospitals, we explored whether ranking order was associated with outcomes.
Results: Among 2718 hospitals, the 50 top-ranked vs other hospitals had lower mean 30-day mortality (RSMR 14.8% vs 15.3%, p<.01) but higher readmission (RSRR 14.5% vs 13.4%, p<.01). These patterns persisted in adjusted analyses; top hospitals had lower mortality by 0.98% (95% CI -1.36%, -0.60%) and higher readmission by 0.66% (0.41%, 0.92%). Among top hospitals, there was considerable variability in risk-standardized outcomes, ranging from 9.5-19.3% for mortality and 10.8-17.5% for readmission. RSMRs were lower than the national average for 14% of top hospitals while RSRRs were lower for only 2% and higher for 14% (figure). Hospital rank was not significantly associated with either mortality (RSMR 0.04% increase with each rank, 95% CI 0.0%, 0.08%) or readmission (RSRR -0.01%; 95% CI -0.04%, 0.02%) among the 50 top-ranked hospitals.
Conclusions: Admission to a top-ranked hospital for Neurology or Neurosurgery was associated with lower 30-day risk-standardized mortality but higher readmission. There was considerable heterogeneity among the top 50 hospitals.