The study, published in the Journal of the American Medical Association, compared Medicare mortality rates for pneumonia and the two heart conditions for high- and low-performing hospitals with those for 19 medical and surgical conditions, such as stroke, renal failure, coronary artery bypass grafting and colon resection.
It found a composite of the medical and surgical mortality rates was 3.6% lower, after adjusting for risk, among hospitals in the top performance quartile on the trio of Medicare mortality rates when compared with the worst-performing hospitals.
Medicare first reported the cardiac mortality measures six years ago, and pneumonia followed a year later. Performance soon will be tied to Medicare payment. Starting in 2014, CMS will use mortality rates as measures in the Patient Protection and Affordable Care Act's value-based purchasing initiative, wrote Jha and co-authors Dr. Marta McCrum, Dr. Karen Joynt, E. John Orav and Dr. Atul Gawande.
Jha said he entered the analysis skeptical that the work would find a strong correlation. The results came as a surprise. “I expected there to be a very, very small effect,” he said. “This is why we do research.”
The correlation suggests there may be factors at work across an entire hospital—not just one department, such as cardiology—that improve quality outcomes, he said. “Leadership and culture probably matter a lot,” he said.
Further collection and public reporting on quality measures now could focus on expanding outcomes measures other than mortality, since a limited number of mortality rates provide a snapshot of overall mortality rates, he said.
This would benefit patients in several ways. Patients deciding where to go for care would not be overwhelmed by dozens of mortality statistics. And they would benefit from public reporting on additional quality measures such as infection rates and medication errors.
