Safety-net hospitals (SNHs) tend to be weaker in financial condition than other hospitals, leading to a concern about how the quality of care at these hospitals would compare to other hospitals.
To assess mortality performance of SNHs using all-payer databases and measures for a broad range of conditions and procedures.
Longitudinal analysis of hospitals from 2006 through 2011 with data from the Healthcare Cost and Utilization Project State Inpatient Databases, the American Hospital Association Annual Survey, and the Area Health Resources File.
A total of 1891 urban, nonfederal, general acute hospitals from 31 states.
SNHs were identified by the percentage of Medicaid and uninsured patients. Hospital mortality performance was measured by 2 composites covering 6 common medical conditions and 4 surgical procedures with risk adjustment for patient characteristics. Differences in each composite between SNHs and non-SNHs were estimated through generalized estimating equations to control for hospital factors and community resources.
Inpatient mortality rates declined over time for all hospitals. Small differences in risk-adjusted mortality rates between SNHs and non-SNHs were found only among teaching hospitals. After controlling for hospital factors, these differences were substantially reduced and remained significant only for surgical mortality rates. The small gap in surgical mortality rates diminished in later years.
SNHs appeared to perform equally well as other hospitals in medical and surgical mortality measures. Policymakers should continue to monitor the quality of care at SNHs and ensure that it would not decline under the current value-based purchasing program.