Health Affairs, 25, no. 1 (2006): 34-43 doi: 10.1377/hlthaff.25.1.34 © 2006 by Project HOPE	New Online   Senate Health Reform Bill   Rewarding Providers   Public Option Policy Brief   Health Reform & Abortion   Delivery System Reform

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Vladeck, B. C. Search for Related Content PubMed PubMed Citation Articles by Vladeck, B. C. Related Collections Hospitals Medicare Business Of Health Health Spending Consumer Issues

Mission Vs. Market

Paying For Hospitals’ Community Service

Abstract

 
U.S. hospitals incur costs of $25–$50 billion annually in providing "community service," primarily in the form of health professions education and standby costs. They also provide approximately $30 billion in uncompensated care. Historically, such "community service" costs have been subsidized explicitly by Medicare and implicitly in the prices paid by private payers. The sustainability of that system is highly uncertain. With a growing number of uninsured patients, allocating nonreimbursable costs to paying customers can create a "death spiral," in which fewer paying customers bear a larger proportion of such costs. The obvious solutions to this problem all have serious limitations.

The Community Service ‘Umbrella’

Top The Community Service... The Costs Of Community... The Joint Costs Problem Paying For Community Service Policy Options For Subsidizing... Concluding Observations NOTES

 
As economic entities, contemporary U.S. hospitals largely marshal a wide (and too rarely standardized) array of human and physical resources, under the direction of a physician who is likely not employed by the hospital, to treat identifiable patients with identifiable conditions and treatment patterns. For patients with insurance, payment to the hospital for provision of such services in such instances is generally a relatively straightforward—if often conflicted and controversial—process.

Outpatient and ED care. But that, of course, is not all that U.S. hospitals do. To begin with, they provide an increasing number and variety of services in a range of outpatient settings. They also generally operate emergency departments (EDs), whose principal clinical (and legally required) function is to figure out, if they can, what a patient’s problem is and then either treat it or send the patient to the right place, in or out of the hospital. Hospital EDs serve other social functions as well, including providing a place in which physicians, nursing homes, the police, human service agencies, and families can deposit people for whom they have run out of other options, and who they have no other place to put. For most outpatients and some ED patients, payments to the hospital comfortably follow the prevailing inpatient model; however, a higher share of outpatients, and a still higher share of ED patients, lack health insurance or are the financial responsibility of casualty and liability insurers, public agencies, or no one.

Services to the "nonsick." In response to exhortations that they demonstrate concern for the health of all members of the community, not just current patients, hospitals also increasingly provide various forms of health education, screening, and preventive services to people who are not actively ill. Sometimes these activities take place within hospitals’ physical facilities, and sometimes they do not. Rarely (perhaps more rarely than in the past), hospitals also participate in communitywide efforts to improve the health of the population through educational, civic, or economic development activities. Hospitals might receive some philanthropic subsidy for such activities, but frequently they are not paid directly at all. Post-2001 concerns about bioterrorism, combined with the recognition of the central role that hospitals play in community responses to disasters of any sort, have led to a greatly increased emphasis on hospitals’ disaster preparedness, often with limited financial support from a federal government that rarely covers hospitals’ costs for such preparation. Disasters are by definition infrequent, but most American communities would be unable to respond well to them at all without the central role of local hospitals.

Hospitals do tend to get paid for some of the patients they serve when simultaneously filling their roles in individual patient care and community public health by providing such services as trauma care, burn care, or intensive care for neonates. Such services tend to be characterized by both very high per case costs and very substantial standby costs, and while hospitals receive considerable per case revenues for such services from those who pay, in the aggregate, revenues from such services generally fail to equal expenses.

Health professions education. Many hospitals are also actively involved in the training of health professionals and paraprofessionals. Health professions education generally requires substantial direct clinical exposure for trainees, and the place where that exposure most commonly takes place is in hospitals (even as health professions educators increasingly lament that their activities are excessively hospital-based). Close to 1,500 hospitals—almost a third of the nation’s total—receive at least some Medicare payment in support of their programs in graduate medical education (GME), while even more provide sites for the clinical training of nurses, therapists, and other health care professionals. Medicare, of course, provides sizable subsidies for GME activities and limited subsidies for some nursing education; other payers—including Medicaid in many states—might also provide less explicit subsidies directly or indirectly. At the same time, it should be noted, hospitals often subsidize medical schools in compensation for the privilege of operating medical education programs.

Symbolic role. In many communities, hospitals also serve an additional purpose, at what might be called a more existential level: They are there. For many rural communities, the presence of a hospital is perceived as the only way to maintain an adequate standard of medical care. Without a hospital, the thinking goes, the community will never be able to attract a sufficient supply of physicians and other health professionals, let alone provide emergency and other services to its residents. Residents of poor inner-city communities often feel the same way, for the same reasons. Even when alternative avenues for the provision of medical care exist—or might be brought into existence—the local hospital plays an important symbolic role in the community’s self-image; like the high school, it demonstrates that the community has a certain presence and importance.¹ For larger communities, merely having a hospital might not be enough. The mayor of Las Vegas, for example, has suggested that a major academic medical center will be required for his city to achieve world-class status.²

The payment problem. Paying for all of the services that hospitals provide over and above general patient care poses a set of problems for any payment scheme, but those problems are magnified when the community contains more than a few people without the means to pay for needed hospital care themselves and when hospitals are expected to serve at least some of them. In discussions of the "social goods" provided by hospitals, uncompensated care is often lumped together with the other activities I have just discussed, and in many instances the subsidies supporting those other activities are similar in form and operation to subsidies for uncompensated care. But the distinction between the two categories of cost—and, one hopes, revenue—is vitally important, for reasons I hope to make clearer below.

Needless to say, the provision of community services of all sorts is not distributed randomly among hospitals, nor does it fall into a few neat patterns. Inner-city teaching hospitals generally provide active EDs, trauma and burn services, extensive health professions education, and other community education and public health services, but not all do so to the same extent, and many nonteaching hospitals perform similar functions in similar communities.³ Heterogeneity is the order of the day; some of it might have emerged from hospital leaders’ responsiveness to the specific needs and resources of their communities, but institutional idiosyncrasy and sheer randomness are also important explanatory variables. All of this further complicates, of course, the problem of designing payment strategies.

The Costs Of Community Service

Top The Community Service... The Costs Of Community... The Joint Costs Problem Paying For Community Service Policy Options For Subsidizing... Concluding Observations NOTES

 
I lump the activities described above under the term community service, conceding the inadequacies of that nomenclature but being unaware of a better one. I use the term here to include anything that hospitals do that is generally thought of as a good thing, but which payers have historically excluded from direct patient care costs. Hospital advocates often prefer the term public goods, but many of these services, although worthwhile and provided to the public, are certainly not public goods in the classic economic sense.

Data on the costs to hospitals of providing community services are extremely difficult to come by, not least because many hospitals have not been eager to make such information broadly available. But some rough estimates are possible. Including the costs of uncompensated care, total community service costs incurred by U.S. hospitals might amount to as much as $80–$95 billion a year, or a little more than 15 percent of the total economic activity of this $500 billion industry (Exhibit 1). Total uncompensated care is actually the easiest number to come by, although claims to any degree of precision would be misplaced; however, hospitals generally report an average of about 6 percent of their services as accounted for by the mysterious mélange of "charity care" and "bad debts," whose definitions have long included much art as well as science—and lots of arbitrary convention.

About 4,500 of the nation’s 5,500 general hospitals operate EDs, but only 1,500 of those meet self-defined criteria as trauma centers; slightly under 1,000 offer neonatal intensive care units (NICUs); and barely 200 have an organized burn-care service.⁵ Some of each of those services produce revenues equal to or greater than their costs. In support of the "existential" role of rural hospitals, the Medicare Payment Advisory Commission (MedPAC) recently estimated that the Critical Access Hospital program costs Medicare something on the order of $1.3 billion a year more than it would otherwise pay to participating hospitals for patient care services.⁶ The sums contributed by other payers to support such institutions are negligible. But there are other subsidies, from both Medicare and elsewhere, that support the "existential" function of many hospitals.

Within the context of a $500 billion industry, subsidizing somewhere less than 10 percent of the total costs devoted to other than direct patient care services ought to be a manageable problem, and indeed Medicare and Medicaid provide important subsidies for hospitals’ medical education activities and for the uncompensated care they provide. Private payers that pay hospitals in excess of direct patient care costs can also be said to be indirectly subsidizing community service activities and uncompensated care, although the total absence of correlation between an individual hospital’s ratio of costs to revenues and the extent of its community services suggests that any simple formulation does not accurately apply to private payers.

The Joint Costs Problem

Top The Community Service... The Costs Of Community... The Joint Costs Problem Paying For Community Service Policy Options For Subsidizing... Concluding Observations NOTES

 
The data and public policy problems associated with hospitals’ community service are both complicated by the fact that much of hospitals’ community service is not produced directly for itself at all. Most hospital-based health professions education takes place in the context of patient care services; the fixed costs of EDs or NICUs are incurred when they are full of patients as well as when they are empty; and the extra costs associated with having thinly used hospitals available even when they are not busy would be absorbed in patient care costs if there were enough patients. In other words, much of the costs of community service are joint costs, in the economic and accounting sense: Community service is produced in the process of providing revenue-generating services. In this sense, the problem is not what community services cost, but what third parties (and the few patients who pay directly) are willing to pay for under the rubric of "community service."

By definition, any allocation of joint costs into categories such as "patient care" and "non–patient care" is unavoidably arbitrary; there is no abstractly correct answer. So once again, if the provision of community service were uniformly distributed among all hospitals, it would be analogous to other overhead costs such as administration or medical record keeping and would appear relatively expensive in inefficient hospitals and relatively inexpensive in efficient ones. But the highly skewed distribution of community service activities creates a desire among payers and policy analysts to distinguish between those costs that result from (good) community service and those that are more directly attributable to (bad) inefficient resource use, a task that is empirically very difficult and conceptually almost impossible. That intellectual difficulty is not the primary cause of the policy muddle surrounding the financing of community service, but it does create a real obstacle to finding plausible solutions.

Paying For Community Service

Top The Community Service... The Costs Of Community... The Joint Costs Problem Paying For Community Service Policy Options For Subsidizing... Concluding Observations NOTES

 
Even in a world of cost-based reimbursement, paying for community service could be problematic, since payers have always sought to avoid bearing the costs associated with patients for whom they were not responsible. Thus, the exclusion of identifiable non–patient care costs was a principle of Blue Cross cost finding even before the advent of Medicare and Medicaid and was a basic precept of Medicare hospital payment from its inception until long after the implementation of the prospective payment system (PPS). In both instances, a relatively stable modus vivendi eventually emerged: Educational and standby costs were generally includable in the payment base, while a variety of more community-focused activities were not, although Medicare has long been very touchy about the costs associated with the maintenance of nurseries and NICUs, since few Medicare beneficiaries use such facilities. Hospitals presumably supported other forms of community service from other revenue sources.

Medicare and Medicaid DSH payments. For the biggest payer, Medicare, methods for supporting medical education and the existential costs of rural facilities have been the subject of enormous attention, reflecting the importance of those constituencies, although exactly how generous or niggardly Medicare’s subsidies for those activities might be is a highly controversial matter on which I avoid opining. Medicare’s disproportionate-share hospital (DSH) payments reflect a far more convoluted but highly illuminating approach. The formal rationale for Medicare DSH payments is not that hospitals need to recoup some revenue to support services to nonpaying patients, since doing so would violate the Medicare principle of paying only for its own beneficiaries. Nor are Medicare DSH payments supposed to compensate hospitals for their inability to collect certain revenues from Medicare patients—there is an entirely separate reimbursement process for Medicare bad debts. Rather, the argument has been made that it is more expensive to care for low-income Medicare beneficiaries than other beneficiaries. Thus, Medicare DSH payments are supposed to represent a refinement in the determination of patient care costs. I’m not sure how many analysts believe in the empirical validity of this rationale, but the principle continues to be important to policymakers.⁷

On the other hand, Medicaid DSH payments to hospitals must, by law, be associated with the costs of treating uninsured patients, although only uninsured in-patients and only those who are legally in the United States. The use by many states of Medicaid DSH mechanisms—especially in conjunction with provider taxes—as a way to increase the draw on federal dollars without additional state revenue contributions has obscured the payment side of the DSH equation, but concern about the sources of DSH funds should not prevent consideration of the importance of their uses. Medicaid DSH payments, for example, account for fully one-quarter of the total subsidies to safety-net hospitals for uncompensated care.⁸ And the controversy over intergovernmental transfers obscures the extent to which they have explicitly replaced direct state and local subsidies to public hospitals for care of the uninsured.

Private payers. Again, the extent to which hospitals have been able to recover community service costs from nongovernmental payers has been largely a function of the relative market power of payers and hospitals in specific markets at specific times and the relative generosity of the payers, as determined at least in part by the situational dynamics of the market for health insurance. But there’s an additional dimension to these very broad generalizations that is worthy of note: All other things being equal (which they occasionally are), the more affluent the market, the more generous the payers, but the less the demand for hospitals to supply nonremunerative community services. In other words, in communities in which people can afford to buy relatively generous health insurance, they are also more likely to be able to buy a range of other services that reduce the demands on hospitals to supply them. People who can get doctors to come to them don’t use emergency rooms.

Medicaid managed care. One particular manifestation of this phenomenon has had unexpectedly perverse effects in a number of communities in recent years. Efforts to "mainstream" Medicaid beneficiaries into middle-class systems of care, and, perhaps more importantly, to save money by enrolling them in managed care plans, have frequently had the effect of moving patients who continued to use safety-net hospitals from a payment system (fee-for-service Medicaid) that recognized medical education and DSH costs to a "mainstream" payment system that didn’t pay for those costs. Much of the savings attributed to Medicaid managed care might thus reflect a reduction in community service subsidies mislabeled or misrepresented as greater efficiencies. In response, safety-net providers have formed their own Medicaid managed care plans, thus restoring parts of the status quo ante with an additional bureaucratic layer and administrative costs.

State subsidy schemes. Reflecting the importance of GME in its health care system and its long tradition of relatively generous subsidization of health care for low-income people, New York State, when it made the transition from a regulated all-payer rate-setting system, established a series of mandatory payer surcharges to finance medical education costs, uncompensated care, and a variety of other forms of community service identified by the legislature as important. All private insurers now pay a surcharge of 8.95 percent on all hospital services for which the rates are otherwise presumably subject to negotiation, and an additional amount, which varies by region, for subsidies to health professions education.⁹ But as health care costs have continued to increase, and the number of privately insured residents has continued to fall while the number of uninsured residents has increased, the revenue generated by insurance surcharges has fallen increasingly short of the perceived subsidy needs. It has been supplemented first by the state’s proceeds from the tobacco litigation settlement and more recently from the proceeds of the conversion of Empire Blue Cross Blue Shield, but it is still not expected to be adequate in the years ahead.

New York has been second only to Maryland—which continues to operate an all-payer rate regulation scheme under a federal Medicare waiver—in its efforts to organize and finance statewide subsidies for community service, but the pressures that New York’s system is experiencing suggest that it simply might not be sustainable in an environment in which a growing proportion of the population has no health insurance. What has developed is a peculiar form of the insurance "death spiral": Increases in uncompensated care drive up the private insurance surcharges used to partially defray it, which are passed through into health insurance premiums, thus reducing the affordability of health insurance and increasing the number of uninsured people, in turn increasing the amount of uncompensated care that hospitals are expected to provide. It is getting harder and harder for states to find the patches to keep the system from imploding altogether.

Policy Options For Subsidizing Community Service

Top The Community Service... The Costs Of Community... The Joint Costs Problem Paying For Community Service Policy Options For Subsidizing... Concluding Observations NOTES

 
In the current U.S. health care environment, New York State’s teetering system represents a kind of middle way among the policy options for subsidizing community service. At one end of the spectrum is the system that prevails, de facto, in many parts of the United States: Subsidies for certain circumscribed components of community service are built into the Medicare inpatient payment system, and everyone else is on their own. As payments for patient care services come under increasing pressure, in a variety of forms, hospitals are increasingly hard pressed to sustain community-service activities. At the other end of the spectrum is direct public subsidization, independent of patient care reimbursement. This mechanism supports community service in a dwindling number of public hospitals and often suffers from the chronic underfinancing of public services that is prevalent in many U.S. communities.

For hospitals without direct public subsidy, where Medicare DSH and GME and Medicaid DSH payments are serving primarily to cover shortfalls in Medicare and Medicaid patient service payments, the difficulty of generating revenue to support community service is compounded by increasing political and public relations pressures around the issues of tax exemption. Crosssectionally, the magnitude of the implicit subsidies involved in the exemption of nonprofit hospitals from corporate income tax and, more important, local property taxes is increasingly identified as a potential "source" of community service financing. But that doesn’t offer much comfort to hospitals that have never paid taxes as they scramble to generate operating margins large enough to maintain creditworthiness, so they can borrow the money to purchase the information technology that everyone tells them they need to acquire and the equipment they need to compete successfully with the entrepreneurial efforts of their own physicians. Tax-exemption, expectations about community service, and access to capital are all connected in a circle that hospital managers are finding increasingly difficult to square.

Therefore, as long as the problem of the uninsured goes unaddressed and the major payers, public and private, find themselves increasingly constrained from engaging in too much gratuitous generosity, it will be increasingly difficult for hospitals to maintain historical levels of community service activity, let alone to expand such activity to meet growing public demands. One example of the potential effects is contained in a recent report from the U.S. Centers for Disease Control and Prevention (CDC): While the number of visits to U.S. EDs increased 26 percent between 1993 and 2003, the number of hospitals operating EDs fell 12 percent.¹⁰ Increased waiting times and ambulance diversions were an almost inevitable result.

Concluding Observations

Top The Community Service... The Costs Of Community... The Joint Costs Problem Paying For Community Service Policy Options For Subsidizing... Concluding Observations NOTES

 
Although much of what has been lumped together in this paper under the rubric of "community service" does not qualify as public goods in the strict, technical sense, hospital-provided community service appears to resemble more classic public goods in at least one important way: The more the financing of hospital care moves in the direction of a "perfect" market, the less and less funding for community service there will be. As long as hospitals retain strong pricing power in some markets, they will be able to extract sufficient surpluses from at least some payers to support valued non–patient care activity, but once that market power erodes, direct or indirect (New York–style) public subsidies will be the only alternative. The current condition of the public financing programs suggests that that is not a very promising long-term alternative, either, while few jurisdictions appear to have the political appetite for the kind of intrusive interventions in the health care system that have long been second nature in New York.

More generally, it is hard to see how adequate levels of community service can be sustained over time as long as the prevailing public philosophy insists that no one entity should be in charge of the health care system, or of the public’s health. To date, the continuing dynamism and creativity of U.S. hospitals, when combined with the ability of interests concerned with certain forms of community service to seek and find redress in public programs, and the weakness or obtuseness of many private payers, have supported many activities that are not sustainable in narrow economic terms. It’s not clear how long we’ll be able to continue to get away with that.

Editor's Notes

 
Bruce Vladeck (bruce.vladeck{at}ey.com) is senior health policy adviser, Health Sciences Advisory Services, at Ernst and Young LLP in New York City. He is a former administrator of HCFA (now the Centers for Medicare and Medicaid Services).

An earlier version of this paper was presented at the Conference on the Future of Hospital Financing, 15 July 2005, in Washington, D.C. The research assistance of Jackie Talbott is gratefully acknowledged, as is the editorial advice of Fredda Vladeck.

NOTES

Top The Community Service... The Costs Of Community... The Joint Costs Problem Paying For Community Service Policy Options For Subsidizing... Concluding Observations NOTES

 

1. B.C. Vladeck, "America’s Hospitals: What’s Right and What Could Be Better," Health Affairs 5, no. 2 (1986): 100–107.[Medline]
2. P. Harasim, "Medical Officials Examine Las Vegas," Las Vegas Review-Journal, 30 March 2005.
3. M. Regenstein and J. Huang, "Stresses to the Safety Net: The Public Hospital Perspective," June 2005, http://www.kff.org/medicaid/7329.cfm (accessed 20 October 2005).
4. Author’s estimate extrapolated from Medicare Cost Report data maintained by Ernst and Young LLP; see, for comparison, Medicare Payment Advisory Commission, A Data Book: Healthcare Spending and the Medicare Program (Washington: MedPAC, June 2005), 100.
5. American Hospital Association, Hospital Statistics (Chicago: Health Forum LLC, 2005).
6. MedPAC, Report to the Congress: Issues in a Modernized Medicare Program (Washington: MedPAC, June 2005), 159ff.
7. A thoughtful and dispassionate description of this issue can be found in Association of American Medical Colleges, "Medicare Disproportionate Share (DSH) Payments," http://www.aamc.org/advocacy/library/teachhosp/hosp0003.htm (accessed 20 October 2005).
8. Regenstein and Huang, "Stresses to the Safety Net," 13.
9. New York State Department of Health, "New York State Health Reform Act (HCRA)—Overview of Payor Obligations: Indigent Care and Health Care Initiatives Pools," September 2005, http://www.health.state.ny.us/nysdoh/hcra/overview.htm (accessed 20 October 2005).
10. L.F. McCaig and C.W. Burt, "National Hospital Ambulatory Medical Care Survey: 2003 Emergency Department Summary," Advance Data from Vital and Health Statistics no. 358 (26 May 2005), 2.

                      What's this?