Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website
FORUM - Translating research into quality health care for Veterans

» Back to Table of Contents


Care Quality in VA versus Medicare

Key Points


  • VA is shifting from a system that provides direct care to Veterans to a system that pays for and coordinates care for Veterans. The quality implications of this shift merit further research.

  • This article discusses the lead author discusses research on end-of-life care quality for a cohort of Veterans who died from cancer. Investigators found that VA-reliant Veterans receive higher-quality end-of-life care than Medicare-reliant Veterans.

In response to concerns over poor Veteran access to VA healthcare, Congress passed the Veterans Access, Choice and Accountability (Choice) Act in 2014. The Choice Act allowed VA-enrolled Veterans to bypass VA entirely and obtain healthcare from the private sector if:

1) the Veteran had to wait more than 30 days for an appointment;

2) the Veteran lived more than 40 miles from the closest VA; or

3) the Veteran faced geographic hardship in accessing VA care.

The Choice Act allocated $10 billion over three years for Veterans to access private sector care.

In June 2018, Congress passed the VA Maintaining Internal Systems and Strengthening Integrated Outside Networks (MISSION) Act. The MISSION Act further expanded the ability of Veterans to seek care in the private sector, including waiving the 30-day and 40-mile criteria for accessing non-VA care and allowing Veterans who used VA in the past two years to access walk-in community clinics. While not yet funded, analysts estimate the MISSION Act will cost $52 billion over five years. To put this figure in context, Congress appropriated $68 billion to the Veterans Health Administration in 2017. With the passage of these acts, an increasing number of Veterans will receive care in the private sector instead of from VA.

As VA shifts from a system that directly provides most of its care to a system that also pays for and coordinates care, the question remains as to the quality implications of such an approach. One way this question can be answered is by evaluating care practices in VA versus Medicare. Medicare providers practice in the private sector; thus, as VA shifts to purchasing care, many Veterans will see these providers in the community, either through VA-paid mechanisms or due to their status as Medicare beneficiaries. More than 90 percent of Veterans 65 and older are enrolled in Medicare. In this study, we leveraged data from this dually-enrolled population to understand differences in care quality in VA versus Medicare.

Using data from FY 2010-2014, we studied end-of-life care quality for a cohort of Veterans who died from cancer. The American Society of Clinical Oncology (ASCO) recommends a reduction in medically-intensive service in the last weeks of life for cancer patients; the National Academy of Medicine notes such care is at odds with the focus on quality-of-life that should be the priority at this stage of illness. We focused on cancer for two reasons. First, while prognosticating death is never easy, the pattern of functional decline prior to death for persons dying of cancer is much stronger and more consistent than is the case with many other common causes of mortality. It should thus be clearer to providers that intensive care for these patients would be burdensome. Second, there are well-established quality metrics created and supported by oncology specialty societies pertaining to care provided in the weeks prior to death. Thus, there is both agreement within the oncology community about what constitutes intensive care as well as indications from oncology societies that they believe near-term death can be anticipated for patients dying of cancer.

Our study evaluated a cohort of 87,251 Veterans aged 66 or older who died from cancer and were continuously enrolled in fee-for-service Medicare for one year prior to death. The Veterans in our cohort could have opted to receive care through VA, Medicare, or both. We allocated Veterans to a system (VA or Medicare) based on where they received the majority of their medical/surgical care in the six months prior to the last thirty days of life (the period for which outcomes were evaluated).

Using ASCO and National Quality Forum (NQF) metrics, we evaluated quality of care as the proportion of patients who experienced the following: two or more emergency department  (ED) visits, chemotherapy, a hospital admission, an ICU stay, and death in the hospital. We also evaluated the number of days spent in the hospital. Poor-quality care was indicated by higher proportions of patients with these experiences. Care was evaluated using VA, fee-basis, and Medicare administrative data. Fee-basis care was allocated to VA, as in the pre-Choice Act time frame of our analysis, non-emergency fee-basis care had to be authorized by VA before being provided. In addition to adjusting for comorbidities that could affect receipt of intensive services, we adjusted for variables previously shown to influence Veterans’ reliance on Medicare versus VA: enrollment priority, service-connected disability, distance from VA, race, age, and rurality.

In adjusted analyses, we found Medicare-reliant Veterans were significantly more likely to receive poor-quality, high-intensity care than were VA-reliant Veterans. In their last month of life, Medicare-reliant Veterans were more likely to have the following: chemotherapy, a hospital admission, admission to the ICU, more days spent in the hospital, and death in the hospital. However, these Veterans were significantly less likely than VA-reliant patients to have multiple ED visits in the last month of life.

Our work indicates VA-reliant Veterans receive higher-quality end-of-life care than Medicare-reliant Veterans. This begs the question as to why. There are major organizational and financial dissimilarities between VA and Medicare. Financially, VA is a non-revenue generating system with salaried providers. In traditional Medicare, on the other hand, providers are paid more when they provide more services. Thus, Medicare providers have financial incentives to provide more care, even at the end-of-life, that VA providers do not face. Organizationally, VA is an integrated system that largely delivers care. Medicare is simply a payer of services and is a reimbursement mechanism for a diverse and non-integrated set of providers nationwide. VA has strong operational support for palliative care services, which may help circumvent unnecessary medical treatment at the end-of-life.

To investigate this, we evaluated the relationship between palliative care and medically intensive care using VA data only (palliative care data are not available in Medicare datasets). We examined whether facilities with high levels of palliative care penetration had higher-quality, less-intensive end-of-life care. Our models found no significant relationship between palliative care and end-of-life cancer quality metrics. Thus, higher-intensity end-of-life care may be driven by financial incentives, which are present in fee-for-service Medicare but not in VA’s integrated system.

Our results have important implications for the future of VA care. As congressional funding shifts VA into being less of a direct provider and more of a purchaser of care, Veterans facing end of life may experience more aggressive care than accepted quality indicators would recommend. Our study also adds to the substantial body of literature showing that across multiple metrics and health conditions, VA provides care that is of similar or higher quality than that provided by non-VA providers.

Our work indicates that care coordination will be increasingly important in order to avoid poor-quality care. Coordination efforts will be required of both VA and Medicare providers; in our cohort, the majority of Veterans received services from both systems rather than relying on one system or another for their care. To avoid putting VA-reliant Veterans at risk of receiving lower-quality care in the private sector, VA should continue to develop formal coordination and quality monitoring programs to guard against purchasing overly intensive end-of-life care. VA should also work with the Centers for Medicare & Medicaid Services (CMS) to ensure CMS is aware of the need for their providers to also be informed of the care Veterans are receiving in VA. As Veterans receive more care through the private sector, it is imperative that providers in both systems be made aware of the care received in as close to real-time as possible, to avoid overuse of intensive services, poly-pharmacy problems, and duplication of services.

  1. Teno JM, et al. “Dying Trajectory in the Last Year of Life: Does Cancer Trajectory Fit Other Diseases?” Journal of Palliative Medicine 2001; 4(4):457-64.
  2. Gidwani-Marszowski R, et al. “The Quality of End-of-Life Care is Higher in VA than in Medicare: An Evaluation of a Dually-Eligible Population,” Health Affairs 2018; 37(1).
  3. Trivedi AN, et al. “Systematic Review: Comparison of the Quality of Medical Care in Veterans Affairs and non-Veter­ans Affairs Settings,” Medical Care 2011; 49(1):76-88.

Previous | Next



Questions about the HSR&D website? Email the Web Team.

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.