Testimony on Comparing the Costs of the Veterans’ Health Care System With Private-Sector Costs

Testimony by Matthew Goldberg, Deputy Assistant Director of CBO’s National Security Division, before the Committee on Veterans’ Affairs’s Subcommittee on Health, U.S. House of Representatives.

CBO regularly examines issues related to veterans’ health care as well as other benefits that are provided by the Veterans Benefits Administration (VBA). Most recently, in December 2014, CBO released a report that compared the costs of health care provided directly at VHA facilities with the costs of private-sector care. My submitted statement today reprises that report, Comparing the Costs of the Veterans’ Health Care System with Private-Sector Costs. Although the structure of VHA and some published studies suggest that VHA care has been cheaper than care provided by the private sector, limited evidence and substantial uncertainty made it difficult for CBO to reach firm conclusions about those relative costs or about whether it would be cheaper to expand veterans’ access to health care in the future through VHA facilities or the private sector. Uncertainty about relative costs in the future is compounded by uncertainty about how VHA would structure contracts with private-sector providers.

CBO also produces budgetary baselines and cost estimates for legislative proposals that would modify veterans’ benefits. Among other measures, over the past eight months CBO has estimated the budgetary effects of the Veterans Access, Choice, and Accountability Act of 2014 (H.R. 3230, enacted as Public Law 113-146), including earlier versions of that legislation, and the Department of Veterans Affairs Expiring Authorities Act of 2014 (H.R. 5404, enacted as P.L. 113-175), which contained several amendments to P.L. 113-146.

In recent years, at the request of the Senate and House Committees on Veterans’ Affairs, CBO has reported on several related topics:

• Veterans’ disability compensation,¹
• VHA’s treatment of post-traumatic stress disorder and traumatic brain injury among recent combat veterans,² and
• The potential costs of providing health care to veterans of all eras.³

Among the many analytical challenges in conducting those studies are the problems CBO sometimes encounters in obtaining appropriate data from VHA or VBA. For instance, comparing health care costs in the VHA system and the private sector is difficult partly because the Department of Veterans Affairs (VA), which runs VHA, has provided limited data to the Congress and the public about its costs and operational performance.

Additional data, particularly if it was provided on a regular and systematic basis, could help inform policymakers about the efficiency and cost-effectiveness of VHA’s services. For example, the Department of Defense publishes an annual report to the Congress about its health care system, known as TRICARE (in response to a statutory requirement established in the National Defense Authorization Act for Fiscal Year 1996). The most recent of those reports contains more than 100 pages of operating statistics, including trends among beneficiaries and their demographics; funding by appropriation category; use and costs of inpatient, outpatient, and pharmacy services; beneficiaries’ cost sharing; and patients’ satisfaction with their care.⁴ A virtue of the annual, recurring nature of those reports is that each contains consistent trend data from the previous few years, and longer data series can be compiled by comparing past years’ volumes. A corresponding annual report on VHA—if one existed—would facilitate comparisons between VHA and the private sector. However, such comparisons would still be challenging, in part because private-sector data might also be incomplete, unavailable, or difficult to make comparable with VHA data.

The best study CBO could identify that compared the costs of health care directly provided by VHA with private-sector care was published in 2004, based on data from 1999. The authors of that study had access to detailed administrative data from six VHA medical centers and the clinical charts from the veterans treated at those centers in 1999.⁵ CBO could not replicate that study with more recent data, both because it had limited time and resources to perform its analysis and because, with few exceptions, VHA does not make either existing administrative data or clinical records (even with personal identifying information removed) available to researchers in other government agencies, universities, or elsewhere. Additional systemwide data from VHA would have facilitated the comparison of costs for care provided directly by VHA with the costs for care offered by the private sector. For example, it would be useful to know the average salaries, performance pay, and other elements of compensation that VHA provides for its physicians in various specialties and for its other clinicians; the number of patients its clinicians treat per unit of time (for example, in a typical week) and the length and intensity of those encounters; and the average prices it pays for pharmaceutical products—but VHA does not report that information publicly.

1. Congressional Budget Office, Veterans’ Disability Compensation: Trends and Policy Options (August 2014), www.cbo.gov/publication/45615.
2. Congressional Budget Office, The Veterans Health Administration’s Treatment of PTSD and Traumatic Brain Injury Among Recent Combat Veterans (February 2012), www.cbo.gov/publication/42969.
3. Congressional Budget Office, Potential Costs of Veterans’ Health Care (October 2010), www.cbo.gov/publication/21773; and Congressional Budget Office, Potential Growth Paths for Medical Spending by the Department of Veterans Affairs (July 2006), www.cbo.gov/publication/17962.
4. See Department of Defense, Evaluation of the TRICARE Program—Access, Cost, and Quality: Fiscal Year 2014 Report to Congress (March 2014), http://go.usa.gov/vEc4.
5. Gary N. Nugent and others, “Value for Taxpayers’ Dollars: What VA Care Would Cost at Medicare Prices,” Medical Care Research and Review, vol. 61, no. 4 (November 2004), pp. 495–508, http://dx.doi.org/10.1177/1077558704269795.