Veterans Health Administration Hospitals Outperform Non-VHA Hospitals in Most Markets

Dr. Weeks and co-author Alan West, PhD, used the most current publicly available data to compare health outcomes for VA and non-VA hospitals within 121 local healthcare markets that included both a VA medical center and a non-VA hospital.

The Veterans Health Administration (VHA) is the largest integrated health care system in the United States, providing care at 1,243 health care facilities, including 172 VA Medical Centers and 1,062 outpatient sites. Many of the 9 million veterans enrolled in the VA healthcare program will, at some point, have to decide whether to seek care at a VA or non-VA facility. In a new study, researchers from The Dartmouth Institute for Health Policy and Clinical Practice and the White River Junction VA Medical Center in White River Junction, Vermont, used the most current publicly available data to compare health outcomes for VA and non-VA hospitals within 121 local healthcare markets that included both a VA medical center and a non-VA hospital.

In their findings, recently published in the Annals of Internal Medicine, Dartmouth Institute Professor William Weeks, MD, PhD, MBA, and Alan N. West, PhD, of the White River Junction VA Medical Center note that several recent studies using broad representative samples of VHA patients with representative samples not in the VHA system have found that outcomes at VA hospitals are at least as good as those in the private sector. Several circumstances they say could account for these findings: The VHA may provide better care than the private sector in every local area. Alternatively, non-VHA care may be better than VHA care in more local areas but by a small amount, whereas VHA care may be better than non-VHA care in fewer local areas but by a large amount in each area. The average across all patients and hospitals would favor the VHA in the former circumstance and might favor the VHA in the latter.

“We wanted to take a closer look at local healthcare markets and specific health conditions because if you’re a veteran deciding where to seek treatment what you’re really concerned with are the outcomes at your local VA,” Weeks says.

Weeks and West identified 15 outcome measures that were reported by VHA and non-VHA hospitals by using data from Hospital Compare, a Centers for Medicaid & Medicare Studies (CMS) website which provides information on how well hospitals provide recommended care to their patients. These measures included 30-day risk-adjusted mortality rates for four common diseases–acute myocardial infarction, COPD, heart failure, and pneumonia–plus 11 additional patient safety indicators. They used each hospital’s ZIP code to assign the hospital to one of 306 hospital referral regions–limiting their analyses to the 121 regions in which at least one VHA and one non-VHA hospital reported at least one of the measures. (The Dartmouth Atlas of Health Care defines these regions as distinct health care markets.) The researchers found that VA hospitals were likely to provide the best care in a local health care market and rarely provided the worst care in local markets.

“Our findings suggest that, despite some recent negative reports, the VA generally provides truly excellent care,” Weeks says. “If that is the case, outsourcing VA care to non-VA settings solely for patient convenience should be reconsidered.”

However, Weeks and West also raise the possibility that VA and non-VA hospitals may report data differently to Hospital Compare. If so, the authors recommend the VA and Centers for Medicare and Medicaid Services (CMS) take steps to adapt reporting methods to ensure fair comparisons by end users who are trying to make healthcare decisions.