Under the Affordable Care Act, health care reimbursement will increasingly be linked to quality and costs. In this environment, teaching hospitals will be closely scrutinized, as their care is often more expensive. Furthermore, although they serve vital roles in education, research, management of complex diseases, and care of vulnerable populations, debate continues as to whether teaching hospitals deliver better outcomes for common conditions.
To determine the association between risk-standardized mortality and teaching intensity for 3 common conditions.
Using CMS models, 30-day risk-standardized mortality rates were compared among US hospitals classified as Council of Teaching Hospital (COTH) members, non-COTH teaching hospitals, or nonteaching hospitals. These analyses were repeated using ratios of interns and residents to beds to classify teaching intensity.
The study cohort included Medicare fee-for-service beneficiaries aged 66 years or older hospitalized in acute care hospitals during 2009–2010 for acute myocardial infarction (N=342,145), heart failure (N=647,081), or pneumonia (N=598,366).
The 30-day risk-standardized mortality rates for each condition, stratified by teaching intensity.
For each diagnosis, compared with nonteaching hospitals there was a 10% relative reduction in the adjusted odds of mortality for patients admitted to COTH hospitals and a 6%–7% relative reduction for patients admitted to non-COTH teaching hospitals. These findings were insensitive to the method of classifying teaching intensity and only partially explained by higher teaching hospital volumes.
Health care reimbursement strategies designed to increase value should consider not only the costs but also the superior clinical outcomes at teaching hospitals for certain common conditions.
Supplemental Digital Content is available in the text.
*Massachusetts General Hospital, Boston, MA
†Dartmouth Hitchcock Medical Center, Lebanon, NH
‡Harvard Medical School and Harvard School of Public Health, Boston, MA
Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website, www.lww-medicalcare.com.
D.M.S., X.L., G.S.M., and D.F.T. are all employed by Academic Medical Centers. The other author declares no conflict of interest.
Reprints: David M. Shahian, MD, Center for Quality and Safety and Department of Surgery, Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114. Email: firstname.lastname@example.org.