Hospital quality of care in the U.S. territories

Background The U.S. territories are home to four and a half million residents, yet health care quality in the territories is often absent from national discussions regarding the elimination of health care disparities in hospital quality. We sought to compare hospitals in the U.S. territories with hospitals in the U.S. states regarding hospital structural characteristics and several quality measures including recommended processes of care and mortality outcomes for patients discharged after acute myocardial infarction (AMI), heart failure (HF), and pneumonia (PNA).

Methods We conducted a cross-sectional study to compare the U.S. states versus the U.S. territories on hospital quality, defined as reported hospital performance on recognized process and outcome measures for Medicare patients with AMI, HF, and PNA. The sampling frame included all short-term acute and critical access nonfederal hospitals in the U.S. states and the U.S. territories that submitted at least one inpatient Medicare claim to The Centers for Medicare & Medicaid Services (CMS) for discharges between July 1, 2005 and June 30, 2008 for Medicare fee-for-service beneficiaries with a principal discharge diagnosis of AMI, HF, or PNA. Using data from Medicare claims, Hospital Quality Alliance (HQA), and the 2007 American Hospital Association (AHA) Annual Survey, we compared the mean reported performance for hospitals in the U.S. states versus the mean reported performance for hospitals in the U.S. territories on each quality measure and estimated multivariable linear regression models to adjust for differences in hospital structural characteristics.

Results The final analysis included 4,416 hospitals in the U.S. states and 50 hospitals in the U.S. territories. The mean hospital 30-day RSMR in the U.S. territories was significantly higher than in the U.S. states for AMI (18.1 vs. 16.6, p <0.001), HF (12.1 vs. 11.2, p <0.001), and PNA (14.5 vs. 11.6). Compared with the states, hospitals in the territories were more likely to have hospital characteristics associated with poorer performance (P-values <.001). Hospitals in the U.S. territories also reported significantly lower performance on the majority of process measures assessed (P-values <.001). Statistically significant differences in 30-day RSMR persisted in multivariable analysis after adjusting for process measures and hospital structural characteristics.

Conclusion Collectively, hospitals in the U.S. territories have significantly higher 30-day mortality rates for patients discharged after AMI, HF, and PNA than hospitals in the U.S. states. Re-examining existing federal policy disparities in Medicaid and Medicare payment that disadvantage the U.S. territories may be important next step.