Veterans, Medicare, and Health Maintenance Organizations

A s with all of health care, Medicare remains a topic for lively political debate. Medicare is a major health issue that is likely to occupy the agenda of the 105th Congress, as both Congress and the Clinton Administration struggle with the sensitive issue of Medicare reform. The Department of Veterans Affairs (VA), many of whose millions of veterans are eligible for Medicare, is understandably interested in Medicare reform. Under current legislation, veterans can use their Medicare benefits anywhere except in health care facilities run by the VA and the Department of Defense. Therein resides a continuing policy conundrum. Medicare-eligible veterans who use VA facilities will continue to do so and may prefer health care in VA hospitals rather than in alternative facilities. The VA has supported new legislation that, if enacted, would require the VA and the Department of Health and Human Services to evaluate a model project that provides Medicare reimbursement to the VA for health care services received by some Medicare-eligible veterans.

Enter managed care. Through capitated, risk contracting, managed care offers Medicare the potential of lower health care costs, at least for selected patients. Medicare patients have an opportunity to enroll in managed care plans just like the rest of us. What happens when a veteran enrolls in a Medicare managed care plan and then receives care in a VA Medical Center that is not a designated hospital in the managed care plan? This is an intriguing question and one that is likely to increase in frequency as more Medicare-eligible veterans are enrolled in managed care plans. Whom does the federal government pay? The managed care plan, the VA, or both?

This issue of JGIM contains a provocative article by Passman et al. about this scenario.¹ It describes veterans who are enrolled in Medicare, managed care plans in California but are hospitalized at a VA Medical Center. Passman and colleagues examined 1,000 nonpsychiatric admissions to their VA medical center and found that 20% of the patients “reported” having health insurance coverage. These authors also observed that 337 admissions involved elderly veterans who were eligible for Medicare. For 65% of these admissions, the veterans were enrolled in a Medicare- financed health maintenance organization (HMO). So who is Medicare paying ? Why, the HMO of course! Then who pays for the care these patients receive at the VA Medical Center? Why, the VA of course! Who provides the money for both these payments? Why, the taxpayer of course! As with other research that addresses controversial topics, this report raises more questions than it answers. To recognize this situation, the authors chose a question for the title of their paper: “Is the taxpayer paying twice?”

It is an important question, but the issue is not quite as simple as the question suggests, and the authors know it. The question could be posed another way: “Is this cost shifting? ” Cost shifting is a well-known strategy used by payers to limit their health care costs. Instead of assuming all costs of a patient needing expensive medical care, the cost for providing care is shifted to another payer, in some cases to a public-sector payer. Do the results in this article represent cost shifting for the Medicare- eligible veterans enrolled by the managed care plans? Did the HMO deny acute hospital care, perhaps appropriately, to a veteran who then sought care from the VA?

Still another question, equally appropriate, relates to the quality of care: “Are both the HMO and the VA providing appropriate care?” If these patients had readily accessible, fully satisfactory, continuity care from the VA, why would they need to sign up for Medicare benefits at all? Perhaps the managed care plans are providing more convenient access to outpatient care. On the other hand, perhaps the VA is providing more convenient, or less restricted, access to acute hospital care. If this were the case, then patients may be selecting what they perceive to be the best of both health care plans. If patients are not being served optimally by either plan, they may be seeking different types of care from both plans to get the care they think they need. In our current health care structure, these patients would be behaving in their own best interest.

Which of these questions are answered by the observations reported by Passman et al.? From their results we know that elderly patients who are veterans and were enrolled in Medicare- financed HMOs sought care in VA Medical Centers. The authors did not verify the health insurance status of the patients, but clearly some veterans were enrolled in HMOs because the VA recovered some billings from the HMOs. We also do not know the insurance status of the 80% of hospitalized patients who did not report their insurance coverage, so we do not know how big this problem is. The authors do observe correctly that the proportion of the elderly enrolled in Medicare HMOs is rising, so the problem probably is getting bigger. While the authors provide information on the diagnostic clusters associated with acute hospitalization, they do not report the actual reasons for hospitalization. Therefore, we do not know the patients’ levels of illness or disease severity. We do not know if the patients sought hospitalization from the HMO and were denied. We also do not know whether the indication for hospitalization was clear and defensible, or whether it was unnecessary and inappropriate but easier at the VA Medical Center than at the HMO. Some hospitalizations probably were necessary and appropriate, however, because the HMOs did reimburse the VA for some care, although for only a minority of care.

Though it cannot answer all our questions, this study does document the dual availability of health care options for elderly veterans and highlights the potential problems that can result. The time to answer these questions, or at least to collect additional data on a national scale, is clearly at hand. This additional research should achieve several goals. It should measure the real costs of providing care to veterans who receive their care in both Medicare-financed facilities and VA facilities. It should determine whether these results are generalizable to other parts of the country where managed care penetration is as extensive as it is in California. It should explain why these patients use VA health care when they have other options. For example, veterans might be motivated to use the VA system by familiarity, its national scope, or the array of benefits available to them. We also need to know to what extent veterans are attracted by the quality and comprehensiveness of health care available through a fully integrated health care system.

Until we get these answers, it is difficult to know how the VA Medical Centers and HMOs should behave when faced with these situations. Perhaps VA Medical Centers should seek prior approval before admitting a veteran who is an HMO enrollee. For eligible veterans who need care determined by a VA physician, however, the VA cannot (and perhaps should not) deny care despite the opinion of the HMO. Perhaps HMOs should consider the VA Medical Center a legitimate hospital for use by their enrolled patients, but HMOs may be unwilling to approve hospitalization unless their admission guidelines are followed. The VA’s responsibility is to their eligible veteran patients. When HMO providers accept capitation, they are responsible for all health care costs of their enrolled patients. There seems to be an impasse. Until it is resolved, we should not make elderly veterans choose between Medicare HMO enrollment and care in the VA system.

Is the taxpayer paying twice? Is cost shifting occurring? Are veterans being served well by these multiple health care options? Is it time for new federal health policy initiatives such as the model project proposed by the VA? I do not know. I do agree, however, that the results provided by Passman et al. suggest that it is time to find out. As I stated earlier, Medicare is a topic for lively political debate. At least we can agree that the taxpayer should only pay once and that patients should receive optimal health care commensurate with their needs.