VA and Non-VA Hospitals Comparable for Heart Attack Care

Heart disease is the number one killer in the United States today, and a recent report in the New England Journal of Medicine suggests heart attack patients receive a comparable level of care in Veterans Affairs medical centers and non-VA hospitals.

In the largest clinical study to date comparing VA and non-VA care, a team led by Dr. Laura Petersen of the Houston VA Medical Center found that the post-heart attack death rate among Medicare patients at 1,530 non-VA hospitals equaled that among veteran patients at 81 VA medical centers. In all, the researchers studied the records of nearly 2,500 VA and 30,000 Medicare patients, all men age 65 and up. Collaborating with her on the study were researchers from Harvard Medical School.

"Comparing patients in VA and non-VA hospitals is difficult because of differences in the kinds of patients in the two settings," said Dr. Petersen, who is also an assistant professor at Baylor College of Medicine. "But using sophisticated statistical techniques, we were able to adjust for many of the important differences between the two groups."

The study, funded by VA Health Services Research and the Rockville, Maryland-based Agency for Healthcare Research and Quality, controlled for mortality risk factors such as other illnesses, severity of heart attack, age and race. The VA patients studied were slightly younger than the Medicare patients, and more likely to be African-American. They had significantly higher rates of diseases such as hypertension, stroke, lung disease, diabetes and dementia.

The research team used two separate statistical methods to control for patient differences, to achieve a level playing field between the VA and non-VA hospitals. One method used was logistic regression, a standard technique for determining how different variables are likely to affect an outcome, in this case mortality. The second method paired off VA patients with selected Medicare patients based on shared characteristics and then compared their outcomes, thus in effect creating a randomized, controlled trial within the main study. Both methods showed the same result, no significant difference in mortality rates between VA and Medicare patients. For health events such as heart attacks, short-term mortality is seen as a key measure of hospital quality.

According to study co-author Dr. Sharon-Lise T. Normand, associate professor of biostatistics in the Department of Health Care Policy at Harvard Medical School, the study was enabled in part by recent advances in statistics.

"The analytic methods needed to inform health policy leaders have evolved substantially over the past decade," said Dr. Normand. "VA patients and government oversight authorities can now compare and benchmark the VA system to other systems of care."

Dr. John M. Eisenberg, director of the Agency for Healthcare Research and Quality, said the study provides valuable information for American health care policymakers and consumers.

"This study, and similar research on health care outcomes, gives the nation a clearer picture of the quality of care that is delivered by different parts of the American health care system," said Dr. Eisenberg. "It shows how helpful it can be to evaluate outcomes using sophisticated analytic methods that consider differences in the patients and their severity of disease."

Dr. John R. Feussner, chief research and development officer for VA, said, "Quality of care is a major focus of VA research and medical care. VA is eager to compare health care delivery and patient outcomes between the non-VA sector and VA. As a publicly funded health system, VA health care can serve as a benchmark for other U.S. health systems, as shown in the current study."

---

©2006 Texas Medical Center

---

E-Mail: tmcinfo@texmedctr.tmc.edu
URL: http://www.tmc.edu/tmcnews/02_01_01/page_11.html