Two new studies of standard quality metrics at Veterans Affairs medical centers show that the system has made substantial improvements in quality, in some cases providing substantially better care than is available in private insurance plans. But for all its improvements, the VA hasn't been able to close a gap between the health outcomes of white and black patients.
As recently as the 1990s, the Veterans Affairs health care system had a subpar reputation for quality, but two new studies of standard quality metrics, both led by Amal Trivedi, assistant professor of community health at Brown University and a physician at the Providence VA Medical Center, show that the system that cares for more than 5 million patients has improved markedly in the last decade.
In one study, published March 18 in the journal Medical Care, Trivedi found that the VA's care for seniors is consistently better than what is available through private Medicare Advantage plans, but another analysis published in the April issue of Health Affairs shows that while care has improved for both black and white patients, racial disparities persist in health outcomes.
Trivedi, who receives some of his funding from the VA, said each peer-reviewed analysis was based on widely accepted quality indicators in the health care industry. The measures indicate whether care providers followed recommended processes given the patient's condition and whether patients experienced favorable outcomes. For example, for patients with diabetes, quality care requires both testing and controlling their cholesterol. In his studies, Trivedi and his co-authors analyzed millions of documented cases of care delivery.
The race gap
For the analysis published in Health Affairs, Trivedi studied 10 quality indicators in the records of more than 1.1 million veterans. He found that during the last decade, VA doctors became significantly more likely to provide better care to members of each racial group, both for process (ordering appropriate tests) and outcomes (obtaining improved results). But because care for black veterans did not increase in quality faster than it did for white veterans, the disparity that existed before has remained stubbornly in place.
"The disparities that we saw were for outcomes measures, or getting the right result, as opposed to the provider doing the correct thing," said Trivedi, who teaches in Brown's Department of Community Health and sees patients as a hospitalist at the Providence VA Medical Center.
For example, in 2009, 63 percent of black veterans with diabetes had controlled cholesterol compared to 71 percent of white veterans, a disparity of 8 percentage points. In 2005, the disparity was 9 percentage points.
While the disparity didn't change, the quality of care did. Among patients with diabetes, only 52 percent of black veterans and only 61 percent of white veterans had controlled cholesterol in 2005. Between 2005 and 2009, whites and blacks saw 10 and 11 percentage-point jumps in better outcomes respectively.
Trivedi and co-authors Regina Grebla, Steven Wright, and Donna Washington said that the disparity is not found in just a few VA medical centers, but persists widely throughout the system.
He said the study can't account for why the gap has remained even though the VA has improved quality overall, but the data present the VA with the opportunity to recognize the disparity and focus on it.
"I think it's important for all health systems, not just the VA, to track performance for vulnerable groups," he said.
Other research, he said, has shown that racial disparities on similar measures is 1.3 to 2 times higher in the Medicare system than at the VA.
Advantage for seniors
Overall, the VA trumps private Medicare plans for patients 65 and older, according to the analysis Trivedi published in Medical Care last month. With co-author Grebla, Trivedi looked at comparable quality indicators in 293,000 VA records from 142 VA medical centers and more than 5.7 million from 305 Medicare Advantage plans between 2000 and 2007.
"On these indicators the VA outperformed private sector Medicare Advantage plans by a wide and increasing margin," Trivedi said. "Quality is going to be a function of incentives and capabilities, and in the VA the incentives may be better aligned for providing high-quality care."
In the study's first year, Trivedi said, the VA system scored higher in 10 of 11 quality measures, missing the mark only on providing eye exams to diabetic patients, by 2.6 percentage points.
After the first year, the VA's quality improved faster than the private plans in a majority of indicators. By the final year in the study, Trivedi said, the VA surpassed Medicare Advantage providers in all 12 measures he studied.
The VA's higher quality margin in 2007 ranged from 4.3 percentage points in the case of testing LDL cholesterol among patients with coronary artery disease to 30.8 percentage points in providing colorectal cancer screening.
In all 12 measures, the VA also had lower quality disparities than private plans between patients living in areas with the highest and lowest incomes and education levels. For example, for diabetic patients, the quality difference in controlling blood sugar between patients living in the richest and poorest areas was 0.6 percentage points in the VA system, compared to 8.1 percentage points in Medicare Advantage plans.
"The VA holds clinicians and managers accountable for quality, and emphasizes primary care and health information technology," Trivedi said. "These practices can also be used by private-sector systems to improve care."
In addition to the VA, the Robert Wood Johnson Foundation funded the studies published in Medical Care and Health Affairs.
- Amal N. Trivedi, Regina C. Grebla. Quality and Equity of Care in the Veterans Affairs Health-Care System and in Medicare Advantage Health Plans. Medical Care, 2011; 1 DOI: 10.1097/MLR.0b013e31820fb0f6
- A. N. Trivedi, R. C. Grebla, S. M. Wright, D. L. Washington. Despite Improved Quality Of Care In The Veterans Affairs Health System, Racial Disparity Persists For Important Clinical Outcomes. Health Affairs, 2011; 30 (4): 707 DOI: 10.1377/hlthaff.2011.0074
Cite This Page: