Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
chapter 6 THE CONSEQUENCES CURRENT OF FINANCING METHODS FOR THE FUTURE ROLES AHCS OF The previous chapters have laid out an approach for how AHCs should adapt their roles to meet the public's needs in the 21st century. Although many responsibilities for accomplishing this fall to the leadership of the AHCs, financing policies should facilitate and reward those AHCs that undertake the transformation asked for in this report. AHCs are currently financed through a variety of sources that vary for each role. The academic functions of education and research are particu- larly dependent on public financing, whereas the patient care role is sup- ported through a combination of public and private funding. For most AHCs, the education and research roles are not believed to be self-support- ing but are subsidized from revenues derived from patient care. As patient care revenues have become constrained due to changes by both public and private payers, the funds available to subsidize these other activities are also constrained. This chapter reviews the current financing of education, research, and clinical care in AHCs; identifies behaviors brought about by current financ- ing methods; projects the consequences of continuing current financing methods; and identifies policy options that might harmonize the apparent discrepancy between society's future needs and current financing methods. The focus of this discussion is on AHCs' roles, not on the AHCs themselves. An assumption underlying the analysis in this chapter is that, regardless of mission, financial incentives affect behavior. Thus, depending on how fi- 92
CONSEQUENCES OF CURRENT FINANCING METHODS 93 nancing trends evolve, the manner in which the roles are funded and orga- nized could differ markedly from today's status quo. The committee finds the following: · Current approaches for financing the roles of AHCs will not sup- port the future directions in which the roles need to develop to meet the public's needs. · AHCs are heavily dependent upon public sources of funding, which are likely face problems in sustaining their levels of support in the future due to noncontrollable demographic shifts that affect program revenues and demands, especially in Medicare. · Policy makers will be faced with very difficult decisions if they are to ensure adequate levels of support for activities carried out by AHCs. But AHCs themselves will have to make hard decisions about what can be done within the level of resources available. · The prior chapters in this report have laid out an aggressive agenda for change and innovation. Financing for the AHCs' roles needs to support the process of change that is being asked of AHCs. In some cases, it may be possible to make adjustments within current methods; in other cases, more fundamental changes may be required. CURRENT STATUS OF FINANCING CLINICAL EDUCATION, RESEARCH, AND PATIENT CARE IN AHCS AHCs receive financial support for their roles from a variety of sources. The funding sources vary for each role. These funding patterns are only briefly reviewed here as they have been covered in great detail by other groups. Those interested in more extended discussions of the specific for- mulas for current payment methods should obtain reports produced other groups, such as The Commonwealth Task Force on Academic Health Cen- ters, the Council on Graduate Medical Education, or the Medicare Payment Advisory Commission. Support for clinical education is supported through a combination of cross-subsidies from patient care revenues and explicit funding, primarily from public sources, particularly for medical residents. Medicare pays hos- pitals for its share of residents' stipends, faculty salaries, and related ex- penses (called direct graduate medical education, or DGME, payments) plus an add-on to their inpatient diagnosis-related group (DRG) rates based on the number of residents per bed (called indirect medical education, or IME, payments). In 2000, Medicare provided almost $6 billion in IME payments, about $2 billion in DGME payments, and about $260 million for nursing and allied health programs (Boyle and Fisher, 2002; Medicare Payment Advisory Commission, 2001).
94 ACADEMIC HEALTH CENTERS Additional support for health professions training is provided by the Health Resources and Services Administration (HRSA) Bureau of Health Professions (about $400 million in 2002) and to individual scientists by the National Institutes of Health (about $650 million in 2002) (National Insti- tutes of Health, 2003). State Medicaid programs, non-Medicaid state ap- propriations, Veterans Administration, and Department of Defense also contribute public dollars to financing clinical education. While some of these dollars are devoted to non-physician training, by far the majority funds physician education. Medicare currently pays AHCs an average of approximately $65,000 per resident per year (Appendix A). The principal source of research dollars flowing to AHCs is the Na- tional Institutes of Health (NIH). AHCs received an average per institution in excess of $60 million in NIH funds in 2000, over $60,000 per faculty member (appendix A). The amount received per institution was somewhat variable, with the top 50 NIH recipients getting well in excess of $100 million and over $100,000 per faculty member (Appendix A). Total federal spending for biomedical research has been estimated at approximately $25 billion in 2000, with private foundation support contributing approxi- mately $8 billion to $10 billion, and private industry contributing as much as $55 billion to $60 billion (although only a fraction of the latter goes to AHCs) (Moses and Martin, 2001). While the budget of the NIH has in- creased rapidly over the past several years, the research expenditures by private organizations has increased even more rapidly such that the propor- tion of biomedical research financed by industry has been increasing. For AHCs, however, the NIH remains the principal funding source for their research enterprises. Over the 19902000 period, the amount of NIH fund- ing per AHC increased an average of 126 percent (Appendix A). AHCs do not differ markedly from other institutions in patient care revenue sources. Roughly, 30 percent of revenues come from Medicare, 20 percent from Medicaid, and the remainder primarily from private insur- ance. AHCs do differ, however, in their extent of uncompensated care, primarily for care to the poor. In 2000, uncompensated care was about 7 percent of their costs,1 about 3 percentage points higher than other large, urban nonteaching hospitals (Dobson et al., 2002). In part to compensate hospitals that serve a disproportionate share of the poor, Medicare and Medicaid pay a supplement to inpatient rates called the disproportionate share (DSH) adjustment. The Medicare adjustment, which is based on pro- portions of patients who are Medicaid and who are combined Medicare- Medicaid, is by far the larger of the two. In 2000, AHCs received an 1Defined as the sum of bad debt and charity care charges converted to costs by a hospital- specific ratio of costs to charges, minus the tax allowances for bad debt and charity care.
CONSEQUENCES OF CURRENT FINANCING METHODS 95 average of about $1500 per Medicare discharge in DSH supplemental pay- ments (Appendix A). It is important to recognize that most of the dollars flowing into AHCs are fungible; that is, they can be used for a variety of purposes regardless of the reasons they were paid (except for research grant dollars). GME and DSH dollars flow to the hospital's general operating revenues. Patient-care revenues obtained by AHC hospitals and faculty practice plans have his- torically been deployed both to cover necessary expenses and to finance discretionary spending related to institutional activities. Thus, the extent of support for education and research within an AHC is a function of external funding but is greatly affected by internal decisions about how funds are to be used (Kirch, 2002). OBSERVATIONS REGARDING CURRENT FINANCING METHODS AND AMOUNTS A number of general observations can be made about current support for the financing of the AHC roles. First, the financing of clinical education is Medicare-dependent. The sum of IME and DGME payments to AHCs far exceeds support for clinical education from other sources. As such, AHCs are dependent on a flow of Medicare inpatients to generate GME revenues. In particular, IME pay- ments are calculated as an add-on to Medicare inpatient DRG rates to compensate hospitals for the costs of patient care (for Medicare patients) associated with operating approved physician training programs, although evidence finds that the IME add-on exceeds the "empirical level" at which the costs of caring for Medicare patients in teaching hospitals exceed costs in nonteaching institutions of treating clinically similar patients (MedPAC, 2002; COGME, 2000). The private sector has also become less willing to subsidize education costs in its patient care payments to hospitals, in part, due to the growth of managed care in recent years and pressures to contain the costs of care. In the past, private payers routinely paid premiums to teaching hospitals on the order of 25 percent more than what would be paid for similar services in community hospitals. Now, according to some reports, managed care organizations are negotiating agreements with teaching hospitals with pre- miums no greater than 5 to 10 percent in order for the hospital to be included on a preferred provider list (Anderson et al., 1999; Committee on the Roles of Academic Health Centers in the 21st Century, 2002). Second, the public subsidy of graduate medical education is being ques- tioned. Historically, the achievement of a well-trained physician workforce has been seen as a justification for public subsidy of graduate medical education (Anderson et al., 2001; The Commonwealth Fund Task Force on
96 ACADEMIC HEALTH CENTERS Academic Health Centers, 1997a). Today, this "public good" rationale is not universally accepted (Newhouse and Wilensky, 2001; Gbadebo and Reinhardt, 2001). The fact that economists do not believe that medical education meets the textbook definition of a public good is probably less important than public dissatisfaction with the amount and distribution of these payments. The current financing method has encouraged teaching hospitals to employ more residents and arguably has contributed to physi- cian oversupply, maldistribution, and specialty imbalance (Young and Coffman, 1998; Institute of Medicine, 1996). For example, the inability of the graduate medical education system to produce more specialists in geri- atric medicine in the face of obvious demographic trends has caused some observers to question whether the public subsidy provides incentives to meet public needs. The amount of Medicare payments has varied enormously among hos- pitals in ways that are hard to relate to public objectives. Most importantly, the payment system does not permit any accountability for achievement of public goals regarding the size, composition, and location of the physician workforce, nor does it permit balancing subsidies to promote availability of nonphysician health care workers (Salsberg, 2001). Consequently, the DGME and IME adjustments are targets for cuts in the annual federal budget cycle (Matherlee, 2001). While cuts have not been implemented in every budget bill, the trend is clearly in the direction of reducing Medicare subsidy of physician education and is likely to remain so for the foreseeable future. Third, the financing of clinical education is mismatched with public needs. The current financing of clinical education that is so heavily reliant on Medicare DGME and IME payments is also oriented toward hospital inpatient and acute care and is primarily physician oriented. In contrast, the population medical needs, abetted both by demographic and medical tech- nology trends, is moving in the direction of home and community care (for living with chronic illness) and is being met through increasingly sophisti- cated services provided in outpatient and office-based settings. The popula- tion is not only aging but is also becoming more racially and culturally diverse. Training in interdisciplinary approaches to treatment, especially preferred for treating chronic illness and senior health problems, is not encouraged by the present financing system for education. To be sure, some changes in financing orientation have been made in recent years. Some funds have been made available to subsidize nursing education and, given the present shortage, more help is likely to be on the way. The DGME and IME payments have been structured so as not to discourage hospitals from deploying their residents in community settings, to encourage primary care residency training, and the number of residents qualifying for subsidy payments has been capped (Matherlee, 2001). These
CONSEQUENCES OF CURRENT FINANCING METHODS 97 changes are slight, however, in relation to the totality of the inpatient-based Medicare subsidy and the changing nature of society's needs. Fourth, the research role shares some of the stresses experienced by medical education. While research in AHCs is funded quite differently from clinical education, some of the threats to funding continuation appear simi- lar. Historically, for example, growth in both the education and research enterprises in AHCs was accepted as economically beneficial (Blumenthal and Meyer, 1996). Today, AHCs face continued incentives to grow these enterprises with no assurance that the added costs will be covered by direct payments or by surplus clinical revenues (Matherlee, 1995). This is in part due to the reduced willingness of private payers to subsidize the costs of both education and research. Research funding through the National Institutes of Health has grown substantially in recent years, but so has competition for NIH dollars. In addition, research funding by private organizations has grown as a propor- tion of the total, outpacing the increase in NIH funding. The mechanisms through which the public interest is served by these increases is a hodge- podge of disease advocacy, profit seeking, and investigator-initiated pursuit of discoveries in basic research. Although the public support for research subsidy remains strong, it is unlikely that the NIH budget will continue to increase at its recent pace (Korn, 2002), and the shift in funding toward the private sector implies a greater orientation toward discoveries and inven- tions with commercial potential (Matherlee, 2000; Moses and Martin, 2001). Fifth, the role of AHCs in translational and applied research is not supported by current financing methods. A number of organizations and individuals have advocated in favor of research institutions becoming more involved in research that converts basic discoveries into cost-effective medi- cal interventions (Institute of Medicine, 2002; Nathan, 1998). The "model" of basic research investigators competing for new and continuation grants from the NIH is not entirely compatible, however, with AHCs becoming more involved in clinical, health services and prevention research. The sources of clinical research funds (predominantly private), the methods of competing for research funding, and the disciplines required to conduct such research, are very different. Funding for research that does have a translational orientation, such as grants from the Agency for Healthcare Research and Quality (AHRQ), is minuscule compared to the levels of funding from NIH. Moreover, the costs associated with "vetting" new technologies before they diffuse into mainstream clinical practice are sel- dom supported by research funding or recognized by payment systems. Sixth, surplus revenues from clinical services in AHCs are shrinking. As noted above, private payers' willingness to subsidize non-patient-care costs in AHCs is diminishing as pressure is placed on private health plans to
98 ACADEMIC HEALTH CENTERS control spending. Private payers are looking elsewhere for routine services and seeking arrangements with "centers of excellence" for specialized ser- vices. Pressure is also exerted on Medicare and state Medicaid programs to control the growth of program spending. At the same time, medical tech- nology advances are pushing services into the outpatient setting where (at least for Medicare) operating margins are smaller than for inpatient ser- vices. As a result of these trends, surplus funds available in AHCs to subsi- dize their roles are shrinking and are likely to shrink further. Seventh, the AHC safety net role is being stretched. Many of the costs of caring for the uninsured poor are borne by AHCs and other safety net hospitals. GME and DSH payments help to defray these costs, but they are not well targeted. Medicare DSH payments, for example, are based on the hospital's Medicaid caseload. This mechanism creates an incentive for hospitals to accept Medicaid inpatients (since these patients will increase the add-on to Medicare inpatient payments) but a disincentive to accept patients with no insurance. In addition, hospitals in states with relatively generous Medicaid eligibility requirements have higher DSH payments and less uncompensated care, other things being equal, whereas hospitals in states with stringent Medicaid eligibility have lower DSH payments but more uncompensated care (The Commonwealth Fund Task Force on Aca- demic Health Centers, 1997b). Thus DSH payments are not well targeted to cover the costs of uncompensated care to the uninsured poor. The methodology also encourages the provision of hospital services rather than the development of care models that can better meet the needs of this population. The persistence of the large segment of uninsured population in the U.S. combined with pressure on state budgets to control spending suggest that the problem of uncompensated care is unlikely to dissipate. The in- creased unwillingness of private payers to voluntarily subsidize higher costs in AHCs also contributes to the problem. The public relies on some AHCs and other hospitals to provide safety net services to the uninsured poor. Except for the relatively few public hospitals whose budgets are funded by state and local governments, there are few payment mechanisms to directly compensate most hospitals for the costs of this public service. FUTURE CONSEQUENCES OF CONTINUING CURRENT FINANCING METHODS This section assesses the consequences of a "straight-line" continuation of current financing methods in light of trends in population needs, technol- ogy advancement, and cost pressures discussed earlier in this report. The future contemplated here is sufficiently distant that it is beyond effective consideration by our current policy decision-making apparatus, but suffi-
CONSEQUENCES OF CURRENT FINANCING METHODS 99 ciently near that prudent organizations would incorporate consideration of it into long-range planning. The discussion begins with alternative sce- narios describing the future of public and private health care financing, and continues with possible consequences for the AHC roles. In terms of public financing for health care, health care spending as a proportion of Gross Domestic Product (GDP) will continue to increase, with the result that the future federal policy environment is likely to be dominated by the need to limit the growth of Medicare spending. The leading edge of the population segment known as the "baby boomers" (people born between 1946 and 1964) will begin to age into Medicare eligibility in 2011. The effect on Medicare spending will be slight at first (most of the young baby-boomer seniors will be relatively healthy), but eventually the health problems of this population segment, combined with its sheer size, will have a profound effect on Medicare spending. Under current law, Medicare spending is projected to double from historical levels to 4.5 percent of GDP in 2030, about the time that the Medicare Health Insurance trust fund is projected to become insolvent (Social Security Ad- ministration and Medicare Boards of Trustees, 2002). The potential conse- quences of this trend for taxes, the national debt, and/or the "crowding out" effect on nonentitlement spending will place enormous pressure on the federal government to limit Medicare spending. These possible consequences could have mixed effects on health care provider organizations. On one hand, there is likely to be substantial money in the system as the proportion of GDP devoted to health care continues to grow. On the other hand, providers who rely on Medicare and other federal programs for funding are likely to be affected by efforts to control federal spending. Of course, many developments on the national policy scene could affect this scenario. Efforts to restructure Medicare, such as proposals con- templated by the National Bipartisan Commission on the Future of Medi- care in 1999, may be resurrected and implemented. It is also possible that a national system of health financing, such as that advanced in the first Clinton term, could be enacted. While such changes would alter the picture considerably, the failure of the federal government to create such reforms in the past suggests the likelihood of their being enacted in the near future as unlikely. In the private sector, efforts to control health care spending increases will likely create a new breed of informed consumers with financial incen- tives to purchase cost-effective products and services. As noted earlier in this report, a number of trends appear to be leading to consumers of the future having more direct responsibility for their health care purchases. On one hand, the explosion in medical technology advances is creating new diagnostic and treatment options. Direct-to-consumer advertising and the Internet will continue to provide consumers opportunities to be informed
100 ACADEMIC HEALTH CENTERS (and sometimes misinformed) about the strengths and weaknesses of the options before them. On the other hand, the introduction of new forms of cost-sharing in insured plans, combined with the persistence of a large segment of the population which is uninsured or underinsured, suggests that consumers will have a direct financial incentive to seek alternatives that minimize out-of-pocket costs. Cost pressures facing payers are likely to result in expansions of such concepts as three-tiered drug copayments and preferred provider arrangements that contain costs by shifting part of the responsibility to consumers. As noted before, the same concept is being applied to hospitals in some areas. As in the case of the governmental programs, the future presents the potential for mixed consequences for providers. More money will flow into the system as new technologies are brought on line and consumers continue to demand all that technology has to offer. In addition, it is at least possible that some of these new technologies will be cost-decreasing. Also, reforms in the tort system may alter how health plans make coverage decisions if it is possible to substitute clinically equivalent but less costly alternatives without threat of malpractice proceedings and awards. Despite these possi- bilities, it still seems likely that the consumer of the future will seek to have their perceived health needs met in settings and by providers where their out-of-pocket costs are minimized. A continuation of current methods of financing, in light of future trends, presents difficulties for AHCs to continue to fulfill clinical education, re- search, and patient care roles as they presently do. Many of the trends affecting health care financing in the future are well under way. In every case, the stresses experienced by AHCs in the face of such trends are likely to intensify in the absence of financing change. In the case of clinical education, a continuation of current financing methods will exacerbate the growing imbalance between the acute, inpa- tient, physician orientation of financing and the chronic, outpatient, multi- disciplinary nature of patient needs. Continued variability in GME pay- ments, coupled with the lack of accountability to public health care workforce goals, will continue to undermine the public-good rationale for Medicare subsidy of medical education costs. Eventually, these subsidies may fall to a point where hiring residents will appear financially unattrac- tive to many AHCs. In the case of research, a slowdown in growth of NIH and other pub- licly funded research, coupled with shrinking patient care surpluses, will exacerbate the problem of unreimbursed overhead costs. Much new re- search, especially that funded within the private sector, will be conducted outside of AHCs in settings consistent with trends toward outpatient care for chronic illness. AHCs may be faced with incentives to downsize their
CONSEQUENCES OF CURRENT FINANCING METHODS 101 research enterprises, set clearer priorities, and/or seek new commercial de- velopment opportunities. In the case of patient care, the ability of many AHCs to generate excess revenues to support mission activities will diminish and possibly disappear altogether. Safety net institutions, in particular, will be under extreme pres- sure to continue to provide uncompensated care to the uninsured poor. The ratio of routine to specialized services will likely lessen as payers will seek less expensive services in alternative settings. AHCs are likely to respond by seeking new opportunities to leverage their role as developers of new tech- nology into enhanced revenues by demanding payment for their most unique services, in essence, exploiting their monopoly niche in the subset of special- ized services that are uniquely provided by AHCs. However, as the patient care setting becomes more specialized, the patient mix is less representative of the general population, potentially affecting the education role. POLICY OPTIONS What financing policy options might be pursued to reshape the educa- tion, research, and patient care roles and ensure that they are fulfilled?2 Below we examine options for changing financing in each of the three areas. Incremental changes are those that can be accomplished with current financing structures remaining in place. Fundamental changes are those that would discard the current structures and replace them with new fi- nancing structures. In some cases the discussion below refers to specific changes in federal financing policy; in others it refers to actions that AHCs may take in response to a changing financing environment. Financing the Education Role Incremental Change In principle, there is no reason that current GME funding under Medi- care could not be more targeted to perceived health care workforce needs, as several states have done (Matherlee, 2001). DGME payments could be 2Numerous proposals for reform have been advanced in the literature but are not reviewed here. See, for example, Medicare Payment Advisory Commission, Report to the Congress: Rethinking Medicare's Payment Policies for Graduate Medical Education and Teaching Hos- pitals, Washington, DC: MedPAC, 1999. Anderson, G.F., G. Greenberg, and B. Wynn. "Graduate Medical Education: The Policy Debate" Annu Rev Public Health 22:35-47 (2001). Aaron, Henry (ed.), The Future of Academic Medical Centers, Washington, DC: Brookings Institution Press, 2001. Reports of The Commonwealth Fund Task Force on Academic Health Centers (www.cmwf.org).
102 ACADEMIC HEALTH CENTERS restructured to encourage teaching institutions to train more geriatricians and other medical specialties in relatively short supply, and more nurses and other allied health professions. DGME could also be added onto other prospective payments, such as outpatient care or skilled nursing facility payments. Similarly, IME payments could be restructured to encourage teaching hospitals to train needed specialties and professions in the settings where they are most appropriately deployed. The add-on to inpatient DRG payments based on residents per bed could be replaced with other measures used as the basis for payment, that more directly track to society's health workforce needs. For example, training support for gerontologists could be linked to visits by older patients. The advantage of this approach would be to make educational subsidy payments more consistent with technology and population trends, provid- ing incentives to bring training into the community where more health care services are being delivered. Keeping the Medicare add-on structure in place would maintain the linkage with "automatic" entitlement funding. The major disadvantage of this approach is it perpetuates the incongru- ity of trying to accomplish national health workforce objectives through Medicare reimbursement. It would also be more complex, requiring Con- gress and the Executive Branch to work together to establish objectives and implement them through changes to Medicare payment formulas, not a straightforward process. Finally, by maintaining the link with an entitle- ment program, this approach would be vulnerable to future budget cuts designed to control the growth of Medicare spending. Fundamental Change One major structural change would be to remove health workforce subsidization from Medicare and replace it with a separate program de- signed to formulate public goals and fund them directly. The program could be a separate entitlement that is linked to the achievement of national objectives or a separately authorized and appropriated "line item" in the federal budget, similar to the National Institutes of Health. The advantage of a separate program would be to have national health workforce objectives supported by a national financing system. Once a national plan is formulated, institutions could be rewarded in proportion to their contribution to the achievement of national objectives. Additionally, the policy could be broadened to include all health professions receiving training in diverse settings, and adjusted periodically as perceptions of workforce needs change. A disadvantage of this approach is that it would require some level of workforce planning, which may not fit American tastes. In addition, clini-
CONSEQUENCES OF CURRENT FINANCING METHODS 103 cal education would have to compete with other budget priorities in the appropriations process, with uncertain outcome regarding the level of fund- ing. This could be mediated with multiyear funding streams to provide some level of stability for the programs, yet permit periodic adjustments (similar to the funding of large, multiyear research studies). The uncertainty of year-to-year funding levels would be less of a problem if a new entitle- ment program were created. In this case, however, the beneficiaries of the program, and what they are entitled to, would need to be established. Whether appropriation or entitlement, creating such a fund would only address half the question; how the money is distributed, to whom, and for what purposes would also need to be addressed. Financing the Research Role Incremental Change One approach to shrinking surplus patient care revenues and poten- tially reduced growth in government research sponsorship is to "broaden the base" of research undertakings to include more clinical research and to seek more opportunities for commercial support. NIH sponsorship may remain the mainstay of investigator-initiated research in AHCs, but it may also be supplemented with more clinical research, including that sponsored by private companies, to help cover the fixed costs of the research enter- prise and secure new sources of revenue. AHCs might also place greater emphasis on making their technology transfer activities into profit-making enterprises by developing more of their discoveries into health care prod- ucts, retaining property rights, and eventually adding to revenues through licensing and royalty fees. The advantage of adopting this approach, which may already be under way in some institutions, is to reduce reliance on government-sponsored, basic science research funding sources and bring the research enterprise closer to the settings where health care is delivered. Doing so may help existing technologies to be utilized more effectively and earn a reputation for advancing technologies with the greatest potential for clinical applica- tion. These might include, of course, applications in genetics and proteomics and other leading-edge advances in modern medicine. Disadvantages of this approach include potential dilution of the seren- dipitous discovery characterized by investigator-initiated research, and it would require more complex partnerships between scientific disciplines, including the social sciences, not often found in great numbers in AHCs. In addition, exploiting more private sector opportunities implies less freedom to independently pursue faculty research interests in AHCs.
104 ACADEMIC HEALTH CENTERS Fundamental Change Fundamental change in research could come from a shift in how priori- ties are set and by whom. As the availability of both old and new technolo- gies grows and health care costs continue to rise, there will be increasing pressure on research institutions to provide more information on "what works, what doesn't, and at what cost" to aid in technology adoption and spending decisions. Payers, consumers, and regulators will have stronger input into the research agenda, in addition to the scientific community, with the potential to create a "sea change" in research priority setting within both public and private sectors. The Food and Drug Administration (FDA) and Center for Medicare & Medicaid Services (CMS) can expect to be pressured to include cost-effectiveness considerations in product ap- proval and coverage decisions, respectively, and research enterprises (in- cluding NIH) will be pressured to place clinical research, including research on costs and benefits of new technology, on par with basic research. Im- proved coordination among funders, especially the federal funding agen- cies, could significantly affect priority setting for health-related research. Congressional action may be required to amend necessary statutes. A change in research orientation of this magnitude would require AHCs and other research institutions to alter the mix of skills of personnel com- peting for, and performing, research to include more clinicians of various types and more social scientists. The resulting change in research focus would enable providers and payers to make more informed decisions re- garding the deployment of medical technology and, possibly, to slow the pace at which new technology contributes to health care spending inflation. A disadvantage for AHCs would be a reorganization of how research gets done. AHCs would have to aggregate the skills and expertise to conduct clinical, health services, and prevention research. If they are unable to do this efficiently, the result could be a high proportion of such research being performed outside AHCs. More generally, a change in research priorities of this magnitude may slow the pace of discovery of new biomedical inven- tions and lose or delay their corresponding health benefits. Financing the Patient Care Role Incremental Change As surplus revenues from insured services diminish, one incremental change that would provide greater support to safety net institutions would be to retarget Medicare and Medicaid DSH payments to institutions for uncompensated care to the uninsured and underinsured poor. Such pay- ments could also be made to be less inpatient-focused by adding to reim-
CONSEQUENCES OF CURRENT FINANCING METHODS 105 bursement in different settings through formulas that directly measure un- compensated care rather than relying on proxy measures. The advantage of this approach is that it would better target care subsidies to the greatest need. By doing so, it would remove the bias inher- ent in using Medicaid services as a proxy and help preserve institutions most relied upon for care to the poor. Disadvantages include continued reliance on entitlement programs to accomplish public policy concerning nonentitlement objectives, continued vulnerability to Medicare and Medic- aid budget cuts, and potential difficulty in devising direct measures of uncompensated care that are unambiguously calculated and not subject to gaming. Fundamental Change If insurance programs, including Medicare, will become increasingly unwilling to subsidize the AHCs' roles, including uncompensated care, one rather dramatic solution is to enable all population segments to pay their own way, at least for some socially determined minimum standard of care. From a public policy standpoint, this means eliminating the uninsured either by providing them with insurance (or mandating insurance and pro- viding subsidies based on financial and medical need) or by directly com- pensating providers for care to the uninsured (in essence, government is the insurer). Advantages of this approach are to reduce unmet need and protect the institutions under pressure to find ways to fund uncompensated care. At- tempts to widely broaden health insurance coverage have been tried, unsuc- cessfully, going back decades into the post-World War II, pre-Medicare era and, most recently, in the first Clinton term. One grave risk is that before we are able to get to a national solution to the uninsured, cost pressures may force some institutions to close and present serious access problems to the uninsured. So, a disadvantage of the nationalized approach to the unin- sured is that it may require a dismantling of the safety net before we are able to get there politically. Also, its costs may be prohibitive. IMPLICATIONS FOR THE FUTURE AHCs can expect to face continued financial pressures in the future. Medicare will face continued stresses as the post-World War II "baby- boom" generation ages into the program and increases the volume of ser- vices that will be paid by it. At the same time, pressure is growing on Medicare to revise its benefits and include coverage for prescription medi- cations. As demands for care increase and greater benefits are desired, the
106 ACADEMIC HEALTH CENTERS aging of the population is resulting in fewer workers paying into Social Security (relative to the size of the population over age 65) and, therefore, reduced funding for Medicare. The combination of these forces sounds an alarm for AHCs that rely on Medicare as a major source of funding for their activities, particularly for medical education that is funded through an entitlement. Much of what is written in this chapter suggests that financing changes, through either natural evolution or public policy measures, will make it increasingly difficult for AHCs to cross-subsidize their roles by taking sur- plus revenues from insured patient services and applying them to shortfalls in education, research, and uncompensated care. It has been pointed out that because many of the costs of performing disparate roles are joint costs, it is impossible to determine what it costs to perform any single role with precision (Lewin, 2002; Vladeck, 2002). Nevertheless, it appears that fi- nancing trends are headed in the direction of making the roles performed by AHCs become "tubs on their own bottom." The advantage of this trend, especially in gauging public subsidies, is transparency. If we "see" what it costs to pay for clinical education, for example, we are better able to gauge the necessity for, and returns to, public subsidy, and also to set goals and require accountability. On the other hand, if education and research, in particular, have to compete on a level playing field for public dollars at full cost, there may be fewer dollars available to pay for these roles than there would be under a system that relies heavily on cross-subsidization. The end result for some AHCs may be tough decisions that result in resizing their education and research enterprises. Making all or some of these decisions will require that AHCs have better information on the resources and results of each of their roles toward improving health. Mission-based management is one approach for under- standing the contribution of the various activities, but many schools have not budgeted systematically for their different activities performed (The Commonwealth Fund Task Force on Academic Health Centers, 2002). The Fund Flow Project of the University Healthcare Consortium is another approach to understanding the financial interdependencies between the component organizations of the AHC. AHCs will need to have systems in place to improve their understanding of the costs of each role, as well as the magnitude and direction of cross-subsidies in order to make the hard deci- sions with which they will be faced. Financing policies will also need to change to support the innova- tive and collaborative approaches that AHCs are being asked to under- take in their roles. Research funders can do this by funding the collabo- rative, interdisciplinary approaches that can facilitate the development of knowledge across a whole continuum of research that includes not only discovery but also translation into care and evaluation. Payers for
CONSEQUENCES OF CURRENT FINANCING METHODS 107 health services should be willing to support experimentation in both the delivery and the financing of care in order to develop processes of care that are safe, effective, and efficient, and that are designed to improve the health of patients and populations. Special attention should be given to improving care for underserved populations or to condi- tions that exhibit disparities in outcomes. The committee believes that among all the AHC roles, the greatest pressure for change will be in the education role. Chapter 3 described the need for significant reform in the content and methods of clinical educa- tion, and this chapter described the current and projected pressures on its financing. Whereas funding for research has been expanding, support for education has been contracting. The education reforms that AHCs are being asked to undertake are of sufficient magnitude that organizations will undoubtedly face costs in retooling their programs. Funding for education needs to support needed reforms and should be made available immediately for this purpose. Either new funds should be provided or existing resources be redirected. If existing resources are redi- rected, the options are to redirect a portion of DGME or a portion of IME. As noted before, DGME covers Medicare's share of the hospital's expenses for resident stipends, faculty salaries, and related expenses. IME payments are provided as an add-on to the hospital's DRG payments to support the additional costs of caring for Medicare patients that are attributable to teaching activities. For example, extra tests may be performed as students learn, staffing levels may be higher because of the demands placed on other hospital staff, or patients at teaching hospitals may be sicker than is ac- counted for in the case mix index. These increase the costs of treating Medicare patients in teaching hospitals. Analyses by the Medicare Payment Advisory Commission (MedPAC) found that Medicare's IME adjustment formula is about twice the calcu- lated estimate of the relationship between teaching intensity and the in- creased costs of patient care (MedPAC, 2002). It is has been recognized for some time that the actual adjustment for IME is higher than the statistical estimates. Congress made an explicit decision to approximately double the IME factor at the time that the prospective payment system (PPS) was being put in place for hospitals because they were concerned about its impact on academic health centers. When Congress originally passed the PPS provi- sions in 1983, the amount provided for IME was 11.6 percent for each 10 percent increase in a hospital's ratio of residents to beds. Since then, the rate has been reduced multiple times to reach a level of 5.5 percent for 2003 and beyond (MedPAC, 2001; Matherlee, 2001). Analyses have found that the effect of medical education on patient care costs has decreased, due to combined improvements in the Medicare case-mix index, the DRG patient classification system, and the relative efficiency of major teaching hospitals (Lave, 2001).
108 ACADEMIC HEALTH CENTERS For 2003, MedPAC estimates that about 2.5 percentage points of the 5.5 percent IME add-on, or about $2.5 billion, are in excess of the current cost relationship (MedPAC, 2002). The Council on Graduate Medical Edu- cation has recommended that IME payments be set at no more than the analytically justified level for teaching activities (COGME, 2000). At its January 2003 meeting, MedPAC identified the need to continue to study possibilities for better targeting of IME funds (MedPAC, 2003). Rather than reducing the level of IME, the amount that exceeds the relationship between teaching and Medicare patient care costs could be redirected to explicitly support reforms in education. This would bring the IME factor in line with estimate of the increased costs of patient care associated with teaching activities, as it is intended to do, and provide a means to jump start the process of education reform. As noted in chapter 3, educational reform will entail much more than just changing curricula, but will also include changes in the care settings in which education takes place, so the use of IME funds for educational reform is also consistent with its intent. Although immediate reforms in individual educational programs can be supported, the committee believes that long-term, more broadly based changes in the financing of education will also be required. This chapter has already described the problems and gaps in current financing methods for clinical education and its ability to respond to future needs of the popula- tion. Proposals have been put forward to create a fund, with contributions from all payers, to support medical education (COGME, 2000; The Com- monwealth Task Force on Academic Health Centers, 1997a). Although such proposals have offered guidance on alternatives for creating a pool of funds, less consideration has been given to how such funds should be disbursed. For example, should there be a relationship to broader workforce goals, are there desired attributes that training programs must demonstrate (e.g., use of nonhospital or interdisciplinary approaches), is there a relation- ship between the training of physician and nonphysician clinicians, who should be the recipient of funds, or what might be mechanisms of account- ability. Alternative approaches might also be explored. For example, the can- cer centers program of the National Cancer Institute requires different responsibilities from three levels of cancer centers and provides different amounts of core support (although the percentage of support is equal).3 3Three types of centers are recognized, with varying scope of activities. Comprehensive cancer centers conduct basic, clinical, and prevention/behavioral/population-based research, and perform outreach and education to the health professionals and people in the community
CONSEQUENCES OF CURRENT FINANCING METHODS 109 The corollary in clinical education would be the presence of three types of education programs, such as those in AHCs, major teaching hospitals, and other teaching hospitals, each with varying scopes of education programs. Additional work would be needed to explore whether and how such a model would have applicability to clinical education, but it illustrates the possibility of alternative approaches. If AHCs are being asked to innovate across all of their roles to meet the changing needs and demands of the future, financing policies will also need to support, encourage, and facilitate such innovation. served. Clinical cancer centers conduct at least clinical research and may do other research as well. Cancer centers conduct research in a narrowly defined area, such as population re- search. Core funding, set at 20 percent of the NCI-supported research program at the institu- tion, supports infrastructure and developmental work. Centers obtain program funds through the competitive NIH grant process. Centers are evaluated every five years. (The Cancer Cen- ters Branch of the National Cancer Institute, www3.cancer.gov and personal communication, Brian Kimes and Linda Weiss, September 27, 2002).