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 3 OPTIONS AND RECOMMENDATIONS FOR FINANCING GRADUATE MEDICAL EDUCATION FOR PRIMARY CARE PHYSICIANS IN AMBULATORY CARE SETTINGS* Numerous individuals and committees have recommended ways to overcome barriers to financing GME in primary care and GME in ambulatory settings. Recently reported strategies include the expansion of Medicare GME payments to all ambulatory and outpatient settings (Council on Graduate Medical Education, 1988), and the expansion of state capitation grants to include primary care internal medicine and general pediatrics programs (New York State Council on Graduate Medical Education, 1988). Eisenberg (1989) offered a number of options including increased Medicare payment for ambulatory care services, Medicare teaching adjustments for Part B payments, and contributions from future employers of physicians in exchange for practice commitments. Bentley et al. (1989) made several suggestions, including that practice plan revenues could help finance ambulatory care education, and that the public sector contribute through targeted grants and appropriations that provide economic incentives for training in ambulatory settings. Proposals such as these have identified several potential sources of funds, with varying implications for such factors as hospital revenues, the costs of services, state and federal budgets, and physician incomes. The committee approached the task of developing policy opi;ions for financing GME for primary care physicians in ambulatory settings first, by identifying existing barriers; second, by constructing a set of criteria against which to evaluate options; and finally by developing the committee's recommendations. This chapter will briefly review the existing barriers, and describe criteria against which proposals for altering the mechanisms for funding GME in primary care can be evaluated, and finally lay out the committee's recommendations. *Parts of this chapter are based on a paper commissioned by the committee: Financing Graduate Medical Education In Primarv Care: Options for Change by Appendix Sandra C. Peinado and John M. Eisenberg. This paper can be found B. 51
BalTiers Costs of Training Certain characteristics of training in ambulatory sites account for the high costs of training in these settings. These characteristics include: the one-on-one nature of the interactions between residents and faculty; the loss of income experienced by host physicians for which compensation may be required; and the need for additional space to accommodate residents. In addition, primary care teaching clinics are often the site of services for medically indigent individuals. The provision of uncompensated care may be regarded as a cost or as a reduction in revenue. The Reimbursement Playing Field is Not Level Payment for non-primary care services and inpatient care is often greater than for primary care services and outpatient services. Differences in reimbursement include: incomplete recognition by Medicare of residents' time in outpatient settings; non-recognition by many third-party payers of teaching as a reimbursable cost in outpatient settings; lower payments by third-party payers for services provided by primary care physicians, outpatient physician services, and non-procedural services; and frequent lack of coverage of preventive services. In addition, community-based outpatient clinics are at a disadvantage compared with hospital-based clinics, because at the latter sites payers sometimes pay for an overhead assumed to be contributed by the hospital to its clinics. Academic Barriers There is widespread recognition that medicine's academic values and reward systems are not oriented toward the primary care specialties and outpatient services. In addition, the demands of clinical service, which may be higher for the less well reimbursed primary care faculty, make it difficult to develop the research and teaching skills needed to improve the quakier and efficiency of primary care practice. Moreover, federal funds for primary care research are very limited, and few academic institutions either recognize or reward the development of the curricula and special teaching techniques that are needed for effective outpatient care training. The low esteem in which primary care and clinic work is held is 52
transmitted to residents, and the lack of role models of esteemed, successful, practicing/teaching primary care academicians is detrimental to education. Finally, faculty support is not sufficient to help programs develop adequately trained faculty in areas such as decision analysis and quality assurance. These areas must be emphasized to ensure that they are included in the training of · · - prlmalg care p ~yslclans. Criteria for Evaluating Policy Options Criteria developed by the committee reflect several concerns; some concerns have to do with political and fiscal realities--policy should be made with an understanding of the constraints under which decision makers would try to implement change; some concerns are based on values held by members of the committee that were felt to be of such importance that policy options should be judged in terms of whether these values could be sustained under the proposed change. The committee reviewed many more criteria than are listed here (most will be found in the paper by Peinado and Eisenberg in Appendix B). While many of the criteria have merit, they were considered to be either not of sufficient importance that they should be a significant part of the committee's deliberations, or they were too complex to be useful in the limited time available to the committee. An example of the latter is the notion that all those who benefit from expenditures on GME in primary care should bear the costs. While this proposal has evident appeal, the question of who benefits has been the subject of much discussion, and numerous beneficiaries have been posed. To revisit this debate and make conclusions that would be operationally helpful for the purposes of this study would be a large undertaking from which little of immediate use would result. Budget Constraints The committee was convinced that since major new expenditures for GME ~, the federal government are highly unlikely, only recommendations that are budget neutral will be seriously considered by policy-makers. Thus, rather than risk having recommendations rejected because they violate today's political realities, the committee accepted the need to constrain its recommendations to ways of reallocating resources. Thus if a recommendation is made for significantly increased federal spending on primary care GME, a case for reallocating resources from other areas must also be made. Although the committee accepted this limitation on its recommendations, members were concerned that it should not be interpreted as indicating a belief 53
that the problem of shifting the focus of primary care GME to ambulatory settings is either unimportant or of little public benefit. It is therefore emphasized that, while recognizing today's political realities, the urgent need to initiate actions that will accelerate the needed change in primary care GME should not be underestimated. Since state budgets are also under pressure, it is unlikely that increases in state contributions to GME are likely to occur unless the case can be made that the state will benefit. For instance, some states may benefit from a change in physician specialty mix toward a greater proportion of primary care physicians. State (and local) budgets may be a source of support for primary care ambulatory training if mutually beneficial ground can be found, such as arrangements between the primary care ambulatory training sites and the state or locality to provide care for medically needy populations. The concept of health budget neutralizer suggests that, unless a source of additional health care funds has been identified, proposals for change should be evaluated in terms of resource redistribution within the health care system. Thus, for example, increases in funding for primary care GME may cause reductions in hospital or physician income or some services may be reduced. The impact of changes on the health system must be evaluated, and particularly valuable institutions and services should be protected. These include such institutions as teaching hospitals that make a substantial contribution to access to care for medically needy populations. The ability of some of these hospitals to sustain the level of care, or even to survive, might be threatened by the withdrawal of residents from inpatient care or reductions in some sources of revenues. Matching the Needs of Primary Care Residency Programs Residency programs seeking to expand their primary care ambulatory training can be greatly assisted by grant money at start-up and when major program changes are being made. However, in general, to enable directors to engage in relatively long term planning, programs need predictable financing that will not vain greatly from year to year. The predictability and stability of the funding stream should therefore be a criterion by which a proposal is evaluated. Changes that encourage the support of primary care should not include such narrowly prescriptive language that the autonomy of educators is unduly constrained and the development of innovative approaches to training is hampered. 54
Economic incentives aimed at achieving desired goals should be incorporated into proposals. These may include incentives that encourage the following: an increase in the number of primary care residencies; a refocusing on primary care and ambulatory care activities in traditional programs; the development of curricula that will prepare primary care physicians in skills and settings that match the requirements of practice; efficient operations of training sites; and physicians to select careers in primary care. Do No Harm to Valued Institutions and Activities Frequently, alterations in the status quo have unintended side effects. This is particularly likely to occur when making budget neutral changes that, by definition, reallocate resources. The committee identified some areas that are specially vulnerable when GME funding is shifted, and that are of sufficient value that their preservation should be a criterion against which changes are evaluated. In altering GME funding it is important to ensure that high-quality programs are not adversely affected. Programs that are of less high quality are not the subject of this concern, and programs training physicians to practice specialties in which supply is substantially in excess of the nation's needs are also of less concern. The committee recognizes that defining and identifying high- quality programs is a difficult task--that criteria for excellence such as pass-rates on examinations are not perfect--but believes that moderate reductions in GME financing for some specialties, as recommended by the committee, are not likely to impair the viability of good programs. Some teaching hospitals today are in precarious financial health. Many of these hospitals also provide substantial amounts of uncompensated care. Alterations in the funding of GME could exacerbate the already serious financial problems of these hospitals. First, reductions in Medicare direct or indirect GME payments would have important adverse effects on the operating margins of some hospitals. Second, any reduction of residents' inpatient service time that would result from successful financing of primary care outpatient training would also require that hospitals employ substitute labor. The cost to the hospital of this change may be significant. Either of these two effects might undermine the ability of some hospitals to continue to provide uncompensated care, or even in the long run to survive. However, primary care training in ambulatory sites, which also provide substantial amounts of uncompensated care, can help lighten the load for their local hospitals by providing timely preventive care. In addition, there are examples of arrangements between local or state agencies and primary 55
care training sites whereby an agency has funded care for medically indigent people at the ambulatory site and thus help support primary care training. The committee's concern that changes not undermine existing arrangements for the care of indigent people, and the committee's interest in furthering the interaction between primaly care GME and care for indigent people, derives from an interest in maintaining or increasing the provision of care for indigent people. This position does not imply that the committee believes that those unable to pay for care should receive a different quality of service than those able to pay. Furthermore, ambulatory sites of GME do, and should continue to, attract paying patients. Finally, policies that reduce hospital margins run the risk of backfiring in terms of support of primary care GME. If hospitals begin to reduce residency support the most likely candidates for elimination are the primary care residency programs rather than the specialties that earn more revenue. Administration Proposals to increase the support of primary care GME should not be so administratively complex or costly as to overwhelm administrators and divert the funds intended for education into administration. Conclusions and Recommendations The committee developed its recommendations mindful of four major considerations. First, the charge to the committee was to improve the quality of primary care graduate medical education by developing financing mechanisms both to increase the amount of time that primary care residents spend in ambulatory settings, and to increase the number of training sites that closely resemble practice conditions likely to be experienced by the physician in future practice. Second, that because of the time and resource constraints under which this study was conducted, the committee found it feasible to approach change only in an incremental manner. The committee was concerned with developing recommendations that would immediately begin to move the policy process in appropriate directions, and acknowledges that there are complex issues that this committee did not address. Third, that proposed changes should foster the attainment of two secondary goals: (~) expand primary care physician manpower, (2) sustain or enhance access to care for medically indigent people. Fourth, that recommendations be developed, as far as possible, in accordance with the criteria set at the beginning of this chapter. 56
The committee considered the options presented in the paper by Peinado and Eisenberg (see Appendix B), as well as the other papers and discussion at the workshop. Because federal budget constraints were perceived to be real and expenditure restraint to be socially desirable, the task for the committee was to identify areas of funding in which resources could be reallocated in ways that would support primary care residencies in ambulatory settings and encourage increases in the supply of primary care physicians. The committee reviewed the major sources of support for GME to assess the extent to which resource redistribution was feasible without endangering other socially desirable goals of paramount importance. The committee also considered whether some new sources of funds could be found. Believing that support of GME is the responsibility of the private as well as the public sector, such sources as contributions from commercial insurance, a tax on all hospitals or non-teaching hospitals, and an extension of outpatient insurance were considered, but were found not to be practical. For instance, voluntary, explicit payment of education costs from commercial insurance is unlikely to occur. If such a payment were made mandatory, or were captured through a new tax, the cost would be passed on by insurers, many of whom claim that health insurance is already a low margin business. The customer to whom the cost would be passed is most often the employer, and further premium increases would be likely to hasten the move to self insurance, thus defeating the purpose of the initiative. A tax on hospitals was thought to be undesirable because some states were already employing this method to create a pool of money for care for the medically indigent. In addition, such an increase in hospital costs would not be budget neutral, and would likely be passed on to the consumer. Expansion of outpatient coverage would not be budget neutral either for the federal budget or for overall health care spending. In view of the wide variety of primary care teaching programs and of existing and potential sites for ambulatory training, the committee believes that no single approach to overcoming financial barriers will solve the financing problems of all primary care programs. Rather, the committee sought to develop a group of recommendations that together would have a significant positive impact on the ability to establish high quality, appropriate primary care GME programs. Academic I~eadersh~p The influence of academic leadership in helping training programs to develop in innovative and useful ways should not be underestimated. Committed leaders have had significant success in overcoming financial barriers and establishing 57
arrangements for training primary care residents in ambulatory settings. These leaders also have an important function in developing the professional values and mores that will encourage young physicians to enter primary care specialties with an understanding of their place in the health care system, and with the respect that will sustain them throughout a career in primary care. The committee encourages deans and faculty members to emphasize the importance of primary care ambulatory training, and: urges the implementation of academic systems that reward those who provide role models for future generations of primary care physicians and devote time to developing curricula and [caching shills needed to make training in ambulatory settings a useful and positive experience. Efficient Use of Training Resources The committee was convinced that the efficiency with which outpatient training sites are operated makes a significant difference in the financial health of the training program. Such factors as patient volume and flow, the use of non- physician health personnel, and the effectiveness of bill collection are worthy of managers' attention. The committee believes that budgeting and planning for primary care ambuZatory training sites should take into account the need to develop effective clinic management. In addition, to fAe extent that economies of scale can be achieved by the joint use across specially lines of facilities arm other resources, these cost savings should be sough! and interapeciaZ!y barriers towered. Physician Payment Reform Reimbursement for patient care services is potentially the most powerful financial policy instrument available for influencing physician's career decisions and the medical education system ....Reimbursement ... affects all health care providers on a continuing basis (Sloan, 1980, p.57~. Much of the care provided by primal care physicians for their patients is paid for at a relatively lower price than the services provided by other specialists. Many have suggested that some services, for example surgical and diagnostic services, are relatively overpaid while such services as cognitive and preventive care are relatively underpaid. In order to correct these inequities, as well as to change the financial incentives that may result in patients receiving inappropriate care, a fee scale has been developed that is based on the costs of resources (including time) used to provide a unit of service. This fee scale, the Resource ~8
Based Relative Value Scale, has been proposed as the basis for Medicare physician payment by a commission created to advise Congress on the reform of physician payment methods. The commission, also charged with restraining the rate of Medicare cost increases, developed a Medicare budget neutral fee schedule under which internal medicine and family practice physicians are expected to increase their income from Medicare, while some specialties would experience a decrease. However, volume control mechanisms may be needed to sustain budget neutrality. In addition, confining Resource Based Relative Value Scale payment to federal programs affects only a portion of physicians' incomes. There is a possibility that total private sector physician service costs will increase if primary care charges rise to the level paid by Medicare, and procedure charges continue at the current level or even rise to make up for lost Medicare income. Thus, to achieve health care budget neutrality and to bring pediatricians into the fee schedule, all payers should adopt a resource based relative value scale payment system. This will also have the effect of redistributing more than only the Medicare portion of physician Income. The committee supports the proposal that Medicare adopt a resource based relative value scale method of payment for physicians, and recommends that all payers adopt such a payment scale. The implementation of a physician payment system using a resource based relative value scale will have several effects. The financing of primary care GME in ambulatory settings will be facilitated by an increase in patient care revenues from sponsored patients, and the improved earnings ability of primary care faculty will increase the ability of faculty practice plans to support teaching physicians. In addition, as the earnings of primary care physicians increase and the differential between the primary care specialties and other specialties decreases, the economic incentives that deter some physicians from entering primary will diminish. To the extent that economic considerations are a determinant of specialty choice, diminishing negative incentives will increase the number of primary care physicians and help rectify the imbalance in the physician labor force between primary care and non-primary care specialists. Finally, the adjustment of fees will signal an appreciation of the importance of primary care services and should enhance the standing of those specialties. Medicare Direct Graduate Medical Education Payment Medicare payment for the direct costs of medical education is based on the number of full-time equivalent residents in a hospital, multiplied by a hospital- specific amount per resident. However, a proposed rule would introduce a weighting factor that diminishes the payment for a resident who has passed the 59
initial residency period (the period of board eligibility plus one year, not to exceed five years). Thus, for example, internal medicine subspecialties receive full funding for four years--three years prerequisite for the general internal medicine program plus one additional year. The second year of subspecialty training has a reduced weight factor because it exceeds the initial board eligibility plus one year. For surgical specialties, where five or more years are required for board eligibility, only the first five years receive the full weighting (Federal Register, 1988). Thus, Medicare's direct GME payment will not only recognize the direct costs of training, but by incorporating financial incentives--reduced payments for subspecialties that are generally oversupplied--it will become a tool of physician manpower policy that has the potential of altering the composition of the physician workforce. The committee suggests building on this precedent by shaping further incentives in the Medicare direct GME payment. The committee recommends an adjustment to the Medicare payment for the direct costs of GME that would create an incentive to establish residencies in primary care and place those residents in primary care ambulatory settings. The mechanism shouM be a differential in the full-time equivalent calculation between primary care residents and other residents. Residents in genera Z internal medicine, genera' pediatrics, and family medicine shouH receive a higher weighting factor than other residents. Primary care residents who spend 2{i percent or more of their time in a primary care ambulator setting (not including specialty clinics) would receive a larger weighting factor. The committee was unable to calculate the weighting factors that should be applied in order to achieve its goal of creating a sufficient incentive to have an impact on the numbers and sites of primary care residencies. Although some data on hospitals' direct per-resident costs exist, there is no available itemization by specialty. Moreover, 1984 data indicate a wide range ($7,500 to $~87,500) in hospital per-resident costs, much of which is not yet explained (Council on Graduate Medical Education, 1988). The committee supports the recommendation of the Council on Graduate Medical Education that a study of the variation in per-resident direct costs be carried out expeditiously. The findings of this study, and further analyses of data pertaining to the Medicare direct GME payment, should provide a basis for the development of incentive weighting factors. However, if such studies cannot be accomplished with alacrity, weighting factors should be introduced and adjusted as suggested by monitoring of the outcomes. The committee believes that the incentives in their recommendation both will make the provision of primary care residencies more attractive to hospitals and generate revenues needed for the development of quality training programs in 60
community practice sites. In addition, the committee believes that it is reasonable to offer extra support to needed specialties while directing small amounts of resources away from those that are better financed and for which the supply is considered to be more than adequate. The committee considered the implications of making incentive payments directly to department chairmen or others, but concluded that the hospitals must accept responsibility for the entire residency program. In so doing the hospital would respond to the financial incentives by making the shifts to true primary care residencies that the committee seeks, and utilize the additional revenues that accrue to them by helping the development of appropriate training sites. If necessary the additional revenues could, in part, support personnel needed to replace resident time lost to outpatient training. Medicare Indirect Graduate Medical Education Adjustment In its review of the major sources of GME funding, the committee directed its attention to the Medicare indirect medical education adjustment as a potential source of funds that might be reallocated to support primary care education. However, for several reasons the committee decided not to suggest radical changes in this item. For example, the revenue from the indirect adjustment is of major significance to some hospitals that provide large amounts of uncompensated care. However, since Medicare's indirect GME payment is a recognition of the costs of education, it is appropriate that hospitals use some of this revenue to support the primary care ambulatory care services that are an essential cost of training primary care physicians. The committee urges hospitals to commit a portion of the revenue from the Medicare indirect GME adjustment to direct financing of services at community-based ambulatory sites used for training primary care physicians. The Medicare indirect medical education adjustment was originally developed to compensate for a number of factors that increase the costs of teaching hospitals but are not directly attributable to the support of residents or faculty. Such costs derive from a variety of factors such as the test-ordering behavior of residents. Evidence suggests that outpatient sites of residency training experience costs associated with teaching activities similar to those recognized by the Medicare indirect medical education teaching adjustment. The committee recommends that Medicare include in the calculation of the indirect medical education adjustment time spent by primary care residents in all primary care ambulatory settings. This recommendation, which is not budget neutral, would extend the Medicare indirect medical education adjustment to all primary care outpatient 61
sites, making additional money available to help support the costs of training in those sites. State and Local Roles The committee recognized the importance of state funding of GME generally, and in funding state schools and family practice programs in particular. State Medicaid programs also can play an important role in enabling primary care ambulatory training to remain financially functional. The committee encourages Medicaid programs that cio no! now support GME to follow Medicare GME payment policies. However, the committee appreciates that states and localities face budget pressures that are often no less severe than those faced by the federal government. In particular, the committee applauds the expansion of Medicaid to increase the coverage of pregnant women and children, and believes that financing such services is a high priority. By some estimates, the nation now faces an insufficient supply of primary care physicians, which already is reflected in some states. Case studies indicate that state GME capitation payments contribute appreciably to the ability of the funded primary care residencies to support primary care ambulatory residencies. The committee recommends that states assess their need for primary care physicians, bearing in mind the special roles of these physicians. States that determine that an increased supply of primary care physicians would benefit their citizens, and states that find a potential shortage of primary care practitioners, should increase their financial support of GME arm widen their support; to include genera! internal medicine and genera! pediatrics as we!! as family practice. To encourage involvement of state and local governments, training programs should take the initiative in finding ways to exploit the natural affinity of primary care training and services to medically needy populations. Case studies offer numerous examples of ways in which committed leaders have negotiated arrangements mutually beneficial to ambulatory training sites and to states or localities seeking cost-effective care for specific populations. Such arrangements can offer a stream of new patient care revenues needed to make feasible the support of residents in an ambulatory clinic. In trying to encourage state and local support of primary care GME, programs must be ready to demonstrate the ways in which support of specific GME components or services will be beneficial to the relevant governments. Program directors should become aware of local needs, such as health care for the homeless, which their ambulatory clinics may be uniquely able to fulfill. The committee recommends that primary care GME programs assume the responsibility of informing [egisiators and agencies 62
of ways in which primary care ambulatory GME could provide services that would benefit needy populations as well as the education programs. The programs shouH also make efforts to endure continued support by maintaining contact zaith the relevant agencies and legislators through such means as nezasletters. Grants The committee looked at the uses of short terms funds that can be obtained from government and private grants. Bearing in mind that residency programs cannot make needed long-range plans in the face of insecure funding, grants should not be regarded as protracted operating support. Rather, grants are important catalysts in the initial development of ambulatory sites, in supporting innovative educational arrangements, in enabling creative financial arrangements to be developed, and in helping develop the faculty needed to initiate a quality program. The committee recommends that the funds available through Nile VII of the Public Health Service Act be targeted to the development of innovative model programs and demonstration sites from which others can learn of nezo Keys of arranging and supporting quality primary care ambulatory training programs. In addition, these grant programs can continue to play a role in faculty development in the early years of programs. Private fourufations, both local aru] national, interested in medical education and the provision of health services, should add their support to such activities, thus multiplying the impact of the limited federal grants furu]s that are available. In conclusion, believing that quick action is needed to ensure the future supply of appropriately trained primary care physicians, the committee has developed the foregoing recommendations for ways of improving the ability of educational programs and health care providers to support GME for primary care physicians in ambulatory settings. These recommendations are intended to motivate several different entities to act decisively and expeditiously; entities to whom the committee addresses recommendations include federal, state, and local governments, hospitals, and private foundations. This dispersion of responsibility for making needed changes reflects the committee's belief that GME is of benefit to, and is correctly the concern of, numerous participants in the health care system. The committee's recommendations are not addressed to all those whose influence could appropriately be brought to bear on the problem, nor do the recommendations cover all possible solutions. Rather, the recommendations are intended as first, immediate, steps in a direction that the committee believes must be pursued. 63
REFERENCES Bentley, James D., Knapp, Richard M. and Petersdorf, Robert G. 1989. Education in Ambulatory Care - Financing is One Piece of the Puzzle. Special Article. New England Journal of Medicine. 320~23~: 1531-1534. Council on Graduate Medical Education. 1988. First Report of the Council. U.S. Department of Health and Human Services, Public Health Service, Health Resources and Services Administration, Bureau of Health Professions, Division of Medicine. Rockville, Md. Eisenberg, John M. 1989. How Can We Pay for Graduate Medical Education in Ambulatory Care? Special Article. New England Journal of Medicine. 320~231: 1525-1531. Federal Register. 1988. Medicare Program; Changes in Payment Policy for Direct Graduate Medical Education Costs. Proposed Rules. Wednesday, September 21. 53(183):36589-36608. New York State Council on Graduate Medical Education. 1988. First Annual Report, 1988. Albany, New York: New York State Council on Graduate Medical Education. Sloan, Frank A. 1980. Patient Care Reimbursement: Implications for Medical Education and Physician Distribution, in Medical Education Financing. Ed. Jack Hadley. New York: Prodist. 64
