Predicting Overall Hospital Mortality Rates

One strategy for improving the quality of care in hospitals in the United States is public reporting of hospital performance. Public reporting was instituted to (1) provide consumers with information they can use to determine where they want to seek care, (2) give feedback to healthcare providers about their own performance, and (3) provide payers with performance targets for financial incentives.

Public reporting of mortality rates has become particularly important because policy makers and consumers can readily interpret the outcomes and mortality rates represent meaningful end results of hospital care. Beginning in 2014, the Centers for Medicare & Medicaid Services will use mortality rates as a key measure in determining hospital payments through its Value-Based Purchasing program.

According to researchers, current public reporting and pay-for-performance programs focus on 3 conditions—acute myocardial infarction, congestive heart failure, and pneumonia—that account for only 13% of all hospitalizations for the elderly. “It is unclear whether high performance on these 3 conditions is an adequate measure for the overall performance of a hospital,” the researchers said.

The researchers recently conducted a study to determine whether mortality rates for publicly reported medical conditions are correlated with hospitals’ overall performance. Study results were reported online in JAMA Internal Medicine [doi:10.1001/jamainternmed.2013.7049].

The study was designed to answer 3 questions: (1) to what extent do mortality rates for high performance on the 3 reported conditions help consumers identify hospitals that are high performers overall or avoid hospitals that are low performers overall?; (2) what is the absolute mortality gradient associated with this effect?; and (3) how well do aggregate mortality rates for the 3 reported conditions compare with more traditional proxies for hospital quality (size and major teaching status) in identifying high-quality hospitals?

The study utilized data from the Medicare Provider Analysis and Review files to identify all elderly fee-for-service beneficiaries admitted to a nonfederal acute care hospital during 2008 or 2009. After applying exclusion criteria, the final sample of 2322 hospitals provided 90.3% of all acute care for elderly fee-for-service enrollees in the United States during the study period. The researchers compared performance at the 2322 hospitals on 30-day risk-adjusted mortality, aggregated across the 3 publicly reported conditions, with performance on a composite risk-adjusted mortality rate across 9 other common medical conditions, a composite mortality rate across 10 surgical conditions, and both composites combined.

The patient population selected, using the identified reported, medical, and surgical diagnoses, represented 6,670,859 admissions, accounting for 43.1% of all admissions and 57.0% of deaths among the study hospitals.

Overall hospital mortality was 3.6% lower at hospitals in the top quartile of performance on the publicly reported conditions compared with hospitals in the bottom quartile (9.4% vs 13.0%; P<.001). The differences in mortality rates were similarly large for the medical and surgical composites.

The top performing hospitals had 5 times greater odds of being in the top quartile on the overall combined composite risk-adjusted mortality rate (odds ratio [OR], 5.3; 95% confidence interval [CI], 4.3-6.5). Poor-performing hospitals on the publicly reported conditions had much lower odds of being in the best quartile of overall mortality.

Mortality rates for the index condition were predictive of medical (OR, 8.4; 95% CI, 6.8-10.3) and surgical (OR, 2.7; 95% CI, 2.2-3.3) performance when those groups were considered separately.

Large size (OR, 1.9; 95% CI, 1.5-2.4) and teaching status (OR, 2.4; 95% CI, 1.8-3.2) showed weaker relationships with overall hospital mortality rates.

In summary, the researchers stated, “Hospital performance on publicly reported conditions could potentially be used as a signal of overall hospital mortality rates.”