National Academies Press: OpenBook

Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings (1989)

Chapter: Financing Graduate Medical Education in Primary Care: Options for Change

« Previous: Financing of Medical and Graduate Medical Education: Issues in Primary Care Education Support
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 197
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 198
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 199
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 200
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 201
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 202
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 203
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 204
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 205
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 206
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 207
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 208
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 209
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 210
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 211
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 212
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 213
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 214
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 215
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 216
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 217
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 218
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 219
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 220
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 221
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 222
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 223
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 224
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 225
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 226
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 227
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 228
Suggested Citation:"Financing Graduate Medical Education in Primary Care: Options for Change." Institute of Medicine. 1989. Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings. Washington, DC: The National Academies Press. doi: 10.17226/1487.
×
Page 229

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.

FINANCING GRADUATE MEDICAL EDUCATION IN PRIMARY CARE: OPTIONS FOR CHANGE Sandra C. Peinado, M.D. Fellow, General Internal Medicine and John M. Eisenberg, M.D., M.B.A. Sol Katz Professor of General Internal Medicine University of Pennsylancia School of Medicine Abstract To satisfy the nation's need for well-trained primary care physicians, graduate medical education in primary care requires adequate financial support. The current mechanisms of GME financing favor inpatient and procedural care, making the support of primary care programs difficult, since they are more oriented towards outpatient evaluation and management. The majority of graduate medical education funding comes from patient care reimbursement through Medicare Part A direct and indirect payments, and other third party payers. This scheme results in difficulties for primary care programs in resident and faculty compensation, as well as general difficulties for primary care program development. Criteria for evaluating proposals that aim to improve the financial support of primary care programs include financial, administrative, and educational implications of the options, as well as the views of interested stakeholders. The alternatives for sources of funds to support primary care GME include changes in existing Medicare payments, an increase in categorical GME funding, an increase in ambulatory payment, an increase in grants, commitments from future employers, and a redistribution of current funds. Alternatives for spending these funds to aid primary care programs include dividing the sources in three ways: on a per-resident basis, by competitive grants, or by incentives for primary care education. An analysis of the alternatives for changing GME financing is discussed in the paper. The most favorable options are summarized and recommendations for further analysis are made. The education of primary care physicians is of growing national concern. However, graduate medical education (GME) is principally funded through patient care reimbursement for inpatient and procedural care. While there has been a ~97

general shift from inpatient to outpatient sites for all medical training, primary care programs provide a greater proportion of care in ambulatory settings, and are less procedurally oriented than other specialties. Thus, residencies in family medicine, general internal medicine and general pediatrics are less likely to be adequately reimbursed by patient care funds. Alternatives for improved financing of primary care GME must be explored if the nation's need for an adequate supply of well-trained primary care physicians is to be fulfilled. Financial support is needed to assure both an adequate supply and a high level of quality in primary care programs. The current mechanisms for funding GME will be reviewed briefly here and the resulting difficulties encountered by primary care programs discussed. We delineate a set of criteria by which any alternative method of graduate medical education financing should be judged. The range of options for change is described and then analyzed in terms of the ability of each option to satisfy the proposed criteria. Current GME Financing Funds for residency programs are generated primarily from patient care reimbursement by third party payers. Additional funds derive from direct federal, as well as state and local support. The degree to which the financing of general internal medicine, general pediatrics, and family medicine programs differs, if at all, from the general pattern of GME financing is not well known. Payment for Patient Care Under the Prospective Payment System, Medicare Part A reimburses teaching hospitals for a proportion of the direct costs of medical education. These include residents' salaries and fringe benefits, administration and supervision costs of teaching physicians, education supplies, space costs, and associated overhead. Faculty salaries can be included if the teaching physician is salaried by or works under a written agreement with the hospital. Teaching hospitals are paid a set amount by Medicare for each resident, based on each hospital's 1984 costs, adjusted for inflation. Estimates of the magnitude of these payments in 1988 range from $975 million) to $1.4 billions. In addition to these direct cost payments, Medicare Part A pays teaching hospitals for the indirect costs associated with graduate medical education. Introduced by HCFA in 1980, these payments reimburse hospitals for the higher operating costs associated with resident-directed patient care. As described by 198

Judith Lave in a review of these payments, published in this volume, reasons for the higher costs of teaching hospitals may include: "the increased use of ancilIarsr services due to the residents' inexperience; the tendency in teaching hospitals to tog to make a more accurate diagnosis both for educational purposes and to satisfy the more academically minded physicians' need to know; the increased availability of state-of-the-art testing facilities and treatment technologies; the fact that very sick patients may be treated more aggressively and more innovatively; the decreased productivity of other employees such as nurses who have to break in the new residents and increased record keeping requirements."3 The indirect payment is based on a percentage of a hospital's ratio of residents to beds. For fiscal year 1989, PPS payments to teaching hospitals are increased by 7.7 percent for each 0.! increase in a hospital's resident-to-bed ratio). In 198S, these payments were estimated to be $2.02 billion dolIarsi. In addition to these payments to teaching hospitals under Medicare Part A, teaching physicians can bill for patient care under Medicare Part B. This includes patient care services provided while fulfilling supervisory duties. Medicaid has lagged behind Medicare in its reimbursement to teaching hospitals for the costs of GME. Of 37 states with Medicaid prospective payment systems, 23 include "some type of adjustment for teaching costs."2 Some states have threatened to withdraw their support of graduate medical education through Medicaid. Private insurers have traditionally paid teaching hospitals' higher charges, which have included the costs of GME. Specific funds are not targeted as a GME contribution by insurance companies, in large part because hospitals have not specified what portion of their charges could be attributed to resident and fellow training. With the advent of HMO's, PPO's, and other managed care organizations negotiating for lower hospital charges, however, a lower proportion of GME costs can be expected to be obtained from these sources4. Direct Federal Support The Veterans Administration currently supports 12% of residency positions in the U.S. In 1988 the VA invested $224 million in resident salaries and fringe benefits alone. As of 1986, this included 15% of all general internal medicine residents, only 1.~% of family medicine residents, and none in general pediatrics5. Title VIT of the Public Health Service Act enables the Bureau of Health Professions to help support approximately 10-15% of residencies in internal medicine and pediatrics. In 198S, 79 awards, totalling $13.S million, were 199

approved for general internal medicine and pediatric residencies. Grants to 184 programs in family medicine were approved for a total of $20 millions. Originally designed to help increase the overall number of physicians today, grants under Title VII are awarded to programs that aim to redress perceived geographic and specialty maldistribution of physicians7. This section of the Public Health Service Act also enables the Bureau of Health Professions to support faculty development in family medicine, general internal medicine and general pediatrics. In 198S, 29 family medicine programs received $4.5 million for this purposed. Direct State Support State support of GME is highly variable but generally stems from three sources other than Medicaid payment. State governments provided operating subsidies to teaching hospitals in 29 states as of 19872. Direct support for resident education was provided by 37 states, 17 of which targeted this support to family practice programs. State support to medical schools indirectly aids GME in its support of supervising physicians' salaries2. Issues For Primary Care GME Raised By Current Funding Scheme Primary care residencies spend most of their resources on personnel costs. The main components of their budget include the following: full and part-time faculty salaries, residents' salaries, administrative costs of academic units, operating costs of outpatient centers, and educational evaluation and/or quality controlled. These costs have been particularly difficult for primary care residencies to cover for several reasons. These include the facts that 1) the services pronded by primary care residents and faculty are generally paid at relatively low levels compared with inpatient and procedural services; 2) primaly care residents' contributions to the operation of hospital inpatient programs may be perceived by the hospital administration to be less essential than that of residents in other specialties; 3) primary care residencies include a number of rotations on other specialty services, usually in outpatient settings, where the residents' abilil~r to pronde service is limited; and 4) primary care programs generally include substantial amounts of formal education in topics such as prevention, medical decision-making, and doctor-patient communication, and this time in conferences and workshops may occur at the expense of service time. From the point of view of primary care residencies, the current scheme of financing graduate medical education principally through hospital payment is flawed. Physician payment also favors surgical and certain medical fields by paying them disproportionately for patient care, so that physicians' revenue from 200

patient care is less capable of cross-subsidizing educational programs. Other funds are insufficient for primary care program development. In the following paragraphs, the problems of the current GME reimbursement scheme are discussed in somewhat more detail in terms of how the arrangements result in difficulties in resident and faculty compensation, as well as general difficulties for primary care program development. Resident Compensation 1. Since the majority of patient care payment (and also the majority of GME financing) is obtained through hospital care, residents and their educational programs are disproportionately reliant on inpatient hospital services ii. 2. Residents are an important and apparently economical source of manpower for teaching hospitals. Several studies have shown that replacing residents with fully trained workers would cost hospitals more than residents' salaries 12. In shifting the training of residents, especially primary care residents, to the outpatient setting, hospitals will be forced to hire more clerical and technical support, nurses, and house physicians, and to develop better clinical information systems. Thus, unless hospitals are reimbursed for the time spent by residents in outpatient settings, they are likely to oppose the development of curricula that focus on those settings and deploy residents away from traditional inpatient settings. 3. Under Medicare Part A direct medical education payments, wide variation exists in per-resident costs among hospitals. The range of per resident amounts for the base year varies from below $10~000 to over $100~000 per residenti3. This differential is now fixed at 1984 levels and embedded into the formula used by Medicare7. This has almost certainly led to inequities among teaching hospitals in their ability to fund primary care programs and to support reform of residency education. 4. The Omnibus Budget Reconciliation Act of 1986 allows, under Medicare Part A direct GME payments, that residents' time in outpatient settings may be paid by hospitals if those hospitals incur "all or substantially all" of the cost of training. How this will be interpreted is important for the continued development of an adequate variety of outpatient teaching sites. While proposed rules have been published in the Federal Registerlll3, the final explication of those rules has not been established. 5. Although Medicare Part A direct GME payments pay for time spent by residents in outpatient sites, there is no requirement that these payments actually 20~

be allocated to support the teaching costs of those practices. Thus, faculty may report time spent in graduate medical education and the hospital may be paid for their efforts, but the hospital need not channel these funds directly into the teaching program. 6. Medicare Part A indirect medical education adjustment payments exclude the time spent by residents in outpatient sites other than hospital clinics from their calculations. 7. Most state Medicaid programs and private payers do not recognize teaching as a reimbursable cost in the outpatient setting. Faculty compensation 1. Outpatient payment for physician services is generally less than inpatient payments, allowing less for the support of residency programs. Medicare Part B reimburses 80% of an allowable physician charge for outpatient visits, and the customary charge for hospital-based teaching practices is set at 85% of the prevailing rate (in part to avoid it being even lower because of the lower profile of charges often rendered to the low-income clientele of these clinics). The population served by most teaching clinics is generally poor, resulting in low collection rates, and preventive services (almost all of which are outpatient) are infrequently covered by third party payersi4. For these reasons, outpatient sites are less able to support supervisory and other faculty costs. 2. Outpatient teaching is generally more facul~-intensive than inpatient teaching. Fewer residents and students can be involved in an ambulatory visit, and faculty must be present for the entire period of care. Patient visits are also longer, on average, in a residency practice, but are reimbursed under Medicare Part B and other payers at best as if in a private practice setting7. The fee differentials between evaluation/mana~ement ("cognitive") and procedural services ~. ~. . ~. . ~. compound the problem. Adoption ot a Resource based Relative Value Scale may partially compensate for this differential if it redresses the perceived inequity in payment levels for evaluation/management and procedural services. · ~ Be Faculty billing for outpatient visits may be paid at a rate that may provide more net income for hospital-based clinics than for non-hospital-based clinics (this results, in part, from the way in which payers assume overhead is paid by the hospital and thus pay a fee corrected for what overhead is assumed to have been). This formula may provide a disincentive to the development of community-based practices for resident training. 202

4. Primary care faculty may have a disproportionately difficult time developing their own teaching and research skills since they are often required to spend a large proportion of their time in patient care to generate practice income. Development of teaching and research skills is particularly needed at this time to improve the quality and efficiency of primary care residency programs. In addition, primaly care research tends to be poorly funded by federal agencies, thus making it more difficult for these faculty to gain support for their academic activities and to progress in academic careers. 5. Faculty time is reimbursable as supervisory time under Medicare Part A direct GME payments only if faculty are salaried employees of or have a written work agreement with the hospital. The fear of double dipping by faculty, that is billing under Part B while being reimbursed for time spent under Part A, has led to restrictive rules on faculty's ability to bill for services rendered by residents they are supervising. This presents a potential problem for reimbursement of clinic services under Part B. for the employment of part-time faculty, and for the partial compensation of voluntary teachersi5. General Concerns 1. Medicare Part A indirect medical education adjustments are based on an estimate that was doubled in the establishment of the Prospective Payment System to cover increased costs associated with case mix and market factors). This may cause an inequitable distribution of payments: hospitals with larger programs may be reimbursed relatively more since their increased costs of clinical care can be spread over more residents, though their costs may not be proportionately higher. The recent introduction of a declining rate of indirect adjustments addresses this issue, although the decrease in payment per resident is relatively small. On the other hand, if the indirect costs of medical education are disproportionately higher with larger programs (which tend to be located in major medical centers), smaller programs might benefit. Whether this relatively fixed rate of indirect payment per resident helps or hinders primary care programs is not known. 2. Support through Title VIT of the Public Health Service Act (for training grants administered by the Bureau of Health Professions) is small and suffers from year to year uncertainty. It also requires that at least 25% of a resident's time be spent in a continuity setting, which many programs find difficult to accomplish and unduly restrictive on their curriculums. The requirement that residents spend at least 25% of their time in a continuity practice has limited the 203

ability of programs to experiment and to utilize other ambulator care sites than continuity settings, such as specialty clinics and practices, home care, geriatric practices, and occupational health. 3. Internal Medicine and Pediatrics receive little state support. Family Medicine and Pediatrics receive little VA support. 4. The proportion of GME financing paid by HMO's and PPO's, which is unknown, may actually decline, as these managed care systems negotiate for smaller payments or switch to nonteaching hospitalst6. Criteria For Evaluating Alternative Proposals Any proposal for altering the current mechanism of financing GME with the purpose of improving primary care training should be judged from at least four viewpoints. As discussed here, these include 1) financial implications; 2) administrative requirements; 3) educational impact; and 4) views of interested stakeholders. (Table 1) Although we discuss these criteria individually, we recognize that to use them to evaluate any proposal, the complexity of the interrelationships between the criteria must be considered. Simply adding up a score of how well any one proposal meets the individual criteria may oversimplify its relative advantages or disadvantages. We do not expect any one proposal to satisfy all criteria, but by judging all proposals against the same criteria, we may find certain alternatives that make more sense than others. Financial Criteria 1. The proposal should be budget neutral, at least in terms of the federal budget. During a time of fiscal constraint for the U.S. government, it would be politically infeasible to raise total federal spending for GME substantially. Therefore, increases in funding for primary care graduate medical education would be likely to require decreases in funding in other federal expenditures, not necessarily but probably in other expenditures related to GME. The ideal of social budget neutrality also suggests that a reduction in other outlays will probably be required to identify additional funds for primary care education. It is likely that these reductions will be in the health care sector. While these reductions may occur in the funds available for hospital or physician 204

payment, they could also imply reduced profits or surpluses for third party payers or managed care systems, decreased state expenditures in other areas, lower salaries for residents or faculty, or decreased funds available for medical care services. Politically more difficult, but certainly possible, would be increased health insurance premiums, surcharges, or higher taxes. 2. AU those who benefits from GME In pr~marSr care should contribute to paying its costs. All of those entities which benefit from quality education of primary care physicians should contribute to its costs. These entitites include hospitals, patients, health maintenance organizations, physicians themselves, payers (e.g., HCFA, insurance companies), and society as a whole. 3. Funding should be preclictable. The financing of GME should not vary widely from year to year to such an extent that programs cannot plan for several years in advance. Program directors need to be exempt from year to year uncertainty in funding in order to develop stable, high quality residencies. Policy makers, in both professional and legislative organizations, also need time for continued development of financial policy and structures. 4. Funding should be sufficient. The support of GME in primary care should be enough to cover all reasonable costs. The definitions of "sufficient" and "reasonable" are clearly important. They would include clear delineation of what appropriate costs of residencies are, and should attempt to adjust those costs for inflation. The definition of "sufficient" should include some mechanism for adjusting to the shifting nature of GME. Curricula for primary care residencies will change, of necessity, and funding should be flexible enough both to adapt to those changes and to allow the decisions for curricular change to remain in the hands of qualified educators. The percentage of residents' time spent in an ambulatory setting is expected to increase in the next few years and "sufficient" funding options should be able to cover the cost of that increase. Another component of this criterion is a decision regarding the number of doctors this funding should support. For example, "sufficient" funding may be limited to that of all graduates of American medical schools, to all teaching hospital positions, or to all graduates who agree to fill federal health manpower needs. 205

Administrative Criteria 1. The implementation of any proposal should be administratively feasible. The initiation of any new funding scheme should be simple and not unduly expensive. If new administrative mechanisms are required, the design and development of these mechanisms could delay, complicate, or even prevent effective deployment of the funds to support primary care education. 2. The ongoing administration of any new funding scheme should be simple. Ideally, present administrative mechanisms should be simplified rather than made more complex. It is likely that administrative complexity would make the program difficult to monitor and unwieldy to payers, hospitals, and educators. 3. The ongoing administration of any new funding scheme should be · e nexpenslve. In order to maximize the portion of funds actually available for support of primary care education, the ongoing administration of the program should be designed to minimize the bureaucratic overhead required at all levels, from payer to educator. Educational Criteria I. Curmoular autonomy and flexibility should be maintained. The funding of GME in primary care should provide autonomy for educators in the choice and implementation of curricula, as well as flexibility for the development of innovative educational programs. 2. The growth and development of primary care curricular elements should be fostered. Change in GME financing should provide incentives for the development and strengthening of primary care in established programs. This applies not only to the creation of primary care elements within traditional internal medicine and pediatric programs, but also to the support of family medicine curricula and already established programs in primary care internal medicine and pediatrics. In internal medicine and pediatrics, emphasis should be placed on transforming 206

traditional programs to a primary care emphasis rather than adding still more residency positions in these disciplines. 3. High quality programs in non-p~mary care specialties should not be adversely affected. If any change in GME funding results in decreased funding available to non- primary care specialties, a mechanism should be adopted to prevent across-the- board-cuts to these programs. The better programs should not lose money, but marginal programs should be eliminated, especially in those specialties perceived to have more training positions than necessary for the nation's needs. However, for specialties now handsomely paid for physician services, loss of GME financing will probably have less impact on the viability of residency programs. 4. Incentives should favor high-quality primary care programs. No specific type of program is implied here. Rather, some recognition should be made of those programs that strive to achieve strong educational goals and are not simply sources of inexpensive manpower for their teaching hospitals. Stakeholders' Criteria Stakeholders are the entities that are most likely to be interested in and/or affected by change in GME financing. The categories are intended to be generic. We do not expect to be able to predict the response of any specific individual or individual organization (with the exception of HCFA) to any one option. 1. Society: U.S. citizens who benefit from an adequate supply of well-trained physicians, and tax payers whose dollars contribute to the training of physicians. 2. Federal Government: Congress, elected officials, and federal departments and agencies. 3. Health Care Financing Administration: considered separately because of its particular interest in and importance to the policies discussed. 4. State Government: state legislatures, elected officials, and agencies. 5. Private Payers: insurance companies, self-insured businesses, health maintenance organizations (HMOs), preferred provider organizations (PPOs), individual practice associations (IPAs), and employers who contribute to a health insurance plan. 207

6. Teaching Hospitals: hospitals that support programs in graduate medical education. 7. Non-Teaching Hospitals: graduate medical education. licensed Physicians and their representative professional hospitals that do not support programs in S. Physicians: ~ organizations (e.g., American Medical Association). 9. Primal Care Specialties: physicians practicing in internal medicine, pediatrics, and family medicine, as well as their professional organizations, specialty certifying boards, and residency review committees. 10. Non-Primary Care Specialties: physicians practicing all other medical and surgical specialties as well as their certifying boards, professional organizations, and residency review committees. 11. Primary Care Educators: faculty of residency and fellowship programs in primary care fields, and their professional organizations (e.g., Society of Teachers of Family Medicine, Ambulatory Pediatrics Association, Society of General Internal Medicine). 12. Non-Pr~mary Care Educators: faculty of residency and fellowship programs in all other medical and surgical specialties. 13. Primary Care Residents: physicians in primary care residency programs. 14. Non-Primary Care Residents: physicians in residency positions for other medical and surgical specialties. Neither educators, specialties, nor residents can be clivicled cleanly into all primary care or all not primary care clisciplines. One view would include only programs in general internal medicine, general pediatrics, and family medicine. A more moderate new, and the one used here, inclucles the first three years of all internal medicine and pediatric programs, and family medicine programs. A thirc] view might include, in addition, geriatrics fellowships, and residencies in obstetrics-gynecology, occupational medicine, and/or preventive · e mec Scone. 15. Patients: current and potential individual users of medical care services. 208

Source of Funds The question of how money is to be generated for the support of primary care graduate medical education may be considered separately from the question of how that money should be most appropriately spent. We will address both questions, but will not discuss the financial needs of residency programs in primary care. The relative merits of these proposals for identifying and spending funds for primary care GME are analyzed after the options are presented. Sources for funds to be targeted for primary care GME can be organized into SLY catergories, as described below. The various proposals for identifying and obtaining these funds are summarized in Table 2. In addition to these "extrinsic" sources of funding, it behooves primal care educators and specialty organizations to experiment with methods of improving the operating efficiency of their academic units and clinical sites. Change in Existing Medicare GME Payments Additional funding for graduate medical education in primary care could be obtained from cuts in existing programs which support GME. The only substantial programs, however, are Medicare Part A direct and indirect payments. One option (#I) would eliminate both direct and indirect payments and reallocate the money to favor ambulatory training. Alternatively, direct (#2) or indirect (#3) payments could be selectively eliminated. A reduction in direct (#4) or indirect (#5) payments should also be considered with the intention of redistributing the funds to primary care programs. Another proposal (#6) would limit Medicare direct and indirect payments to hospitals for the years required by physicians to gain their first certification. Adding incentives and disincentives to the Medicare Part A direct payment programs (#7) to favor primary care curricula would redistribute current funding. Increase in Categorical GME Funding Additional funding for GME in primary care could be obtained from new sources. These sources might involve an increase for categorical GME financing through such options as an increase in Medicaid reimbursement (#a), or a contribution from insurance companies and HMO's as part of their payment (#9~. The latter might be facilitated, for example, by federal regulations that require HMO's to pass on that portion of their reimbursement from Medicare that has been designated in the fee-for-service sector for the direct and indirect costs of graduate medical education (about 3% of the Medicare dollar). Medicaid funding 209

of GME could take place by requiring that Medicaid programs at least use a portion of federal matching dollars to support GME. Other sources for GME financing could include a new tax on health care providers or payers. A tax on providers might emerge as a sales tax on individual physicians' services (#10) or on hospitals, specifically those hospitals without a substantial commitment to medical education through support of residency programs (#all. Such a tax on services provided by non-teaching hospitals would recognize their dependence on teaching hospitals to train their future staff physicians and would partially offset the price advantage that non-teaching hospitals have in offering their services. A program that would more directly encourage primary care education would be one that levied a tax on hospitals without primely care programs. A tax on third party payments for health care services (#12) could be levied to establish a new primary care GME fundi7. In addition, a surcharge on physicians' licensing fees (#13) could generate funds for primary care educational. Increase in Ambulatory Payment An increase in the payment provided for ambulatory patient care would directly aid primary care programs. Adopting a Resource Based Relative Value Scale for Medicare Part B payments (#14) would redistribute funds from non- prima~y care to primary care specialty providers. This also might be achieved through the extension of outpatient insurance coverage by private and/or public sources (#15), from an increase in Medicare Part B payments through a teaching adjustment for ambulatory services (#16), or by allowing residents to bill for outpatient visits at a rate that would include the costs of supervision (#17~. Grant Support Another potential source of increased funding is through grants. These may originate from foundations (#id), an increase in Title VIT expenditures for residency support (#19), an increase in Title VI! faculty development grants for primary care (#20), an increase in Veterans Administration support (#21), or an increase in state grant support (#22~. State grant support could include a broadened commitment to primary care residency funding, including general internal medicine and pediatrics, and a renewed commitment to family medicine. 24 0

Commitment from Future Employer A fourth new source of funds might be provided by a commitment from future employers of physicians to support GME in primal care in return for a commitment from physicians to work for a specified period of time. For the purpose of evaluating the implications of this proposal, it is divided into support provided by HMO's (#23), by states (#24), by the Health Resources and Services Administration (#25), and by the National Health Service Corps (#26~. Redistribution of Funds In addition to changes in Medicare payments and generating various new sources of GME funding, a third source of financing for primary care would involve redistribution of the current amount dedicated to GME support (of course, some of the other proposals described here involve redistribution of current funds, such as reallocating the direct or indirect Medicare GME payments). These options include a redistribution of money from Title VII of the Public Health Service Act for primary care training grants (#27) among all primary care programs for the support of a specific aspect of primary care education (e.g., office supervision or behavioral science curriculum), or the encouragement of teaching hospitals and faculties to redistribute their clinical income to eliminate the discrepancy in primary care versus subspecialty clinical income (#281~9. Expenditure Alternatives The second question in considering the options for altering graduate medical education financing is how the money generated should be spent. There are three principal alternatives for dividing the money to benefit primary care programs (Table 3~. The first involves dividing the money equally among all primary care residencies on a per resident basis. Dividing the funds in this way may not account for the proportionately higher fixed costs of small programs, but it is hoped that residencies will aim for sizes that allow them to take advantage of economies of scale. The second alternative is to divide the money on the basis of competitive grants to primary care programs. This would involve a significant bureaucratic structure, either new or an expansion of an existing entity, to administer the application and disbursement process. A third alternative is to divide the money on the basis of incentives for primary care education (but to remain budget neutral). This option would also involve some administrative entity to establish and enforce criteria, and to disburse funds accordingly. These incentives might be based on the physician manpower recommendations of OBRA 'S6 or of the Council on Graduate Medical Education. This would involve 2 ~ ~

proportionately greater support of those specialties which are deemed undersupplied, and less support for those specialties which are well-supplied with · ~ P nyslclans. Within each of these alternative spending plans, funds could be divided among residencies to use as they deem most appropriate' or to be targeted for a specific purpose. The funds might be used 1) to support faculty salaries, specifically to decrease their dependence on clinical income and to encourage adequate teaching and research; 2) to support resident salaries, in order to make their training less dependent on hospital rotations; 3) as a supplement to the operating revenue of ambulatory teaching sites, in order to operate and improve existing sites and to aid the development of non-hospital based sites; 4) to support primely care curriculum development, either to improve overall quality or to develop specific areas (e.g. clinical economics, psychosocial aspects of medicine); 5) to help support the operation of the academic unit; 6) specifically to increase the percentage of time spent by residents in ambulatory sites; or 7) to foster cooperative teaching arrangements among the primary care specialties of pediatrics, internal medicine, and family medicine and with related disciplines. Analysis of Funding Options Each of these alternatives for raising funds for primary care GME and for spending on primary care GME has been considered in light of the criteria we have proposed. The options for sources of funds are analyzed first, followed by the options for spending. It is not our intent to suggest that compliance with each criterion and approval by each stakeholder is valued equally. The relative weights of each criterion and each stakeholder need further consideration by the IOM committee. In addition, the degree to which funding and spending mechanisms satisfy the criteria will vary, so that a full analysis of these options must weigh both the importance of the criterion or stakeholder and the degree to which it is satisfied or violated. In general, the resources needed to acquire this data exceed the scope of this report. The elimination or reduction in Medicare Part A direct and/or indirect payments to all teaching hospitals (options #~-5) in order to free money that could be redistributed for the benefit of primary care programs can be analyzed as a group. None of these options broaden the scope of payers of GME, nor do any of them provide that quality non-prima~y care programs will not be hurt. Non- primary care specialties, educators, and residents, as well as teaching hospitals, can be expected to object vociferiously to these options. They would have to develop new resources or cut into attending physician profits to support training in their respective fields. 2~2

The alternative of limiting Medicare support to physicians' first certification (option #6) would save money. It would not, however, necessarily provide any additional funds to foster the growth of quality primary care programs and it would limit non-primary care residencies. If used in combination with a spending plan that called for a reallocation of the savings toward improving primary care residencies, this alternative may be more favorable. It would be an extension of initiatives taken in OBRA '86, and we feel that these steps were appropriately extensive in this regard. The options of incorporating incentives and disincentives to Medicare direct and indirect payments (option #7) fulfill only a few criteria. These remedies would be budget neutral and the funding stream would be predictable. Primary care development could be encouraged by well constructed incentives and quality programs might be relatively well rewarded. The administration of such an option, however, might be difficult and expensive. However, certain measures could be instituted that would actually simplify Medicare payment of hospitals, such as including ambulatory care time in calculating the number of residents. Non-primary care residents, educators, and specialties can be expected to object. Teaching hospitals may oppose the measure, too, if they consider this an undue intrusion on GME, and a cause of significant administrative difficulties. Despite the fact that there would be substantial resistance to major shifts in Medicare direct and indirect payments, some changes are reasonable and could benefit primary care educators. First, time spent by residents in ambulatory care activities should be counted in the calculation of a teaching hospital's number of full time equivalent (FTE) residents for both direct and indirect payments. These shifts in the number of FTE residents should not cause an increase in the total direct or indirect payments but should be incorporated into a recalibration of the per- resident payments. Second, faculty effort in primary care education should be credited to the allowable cost of medical education that is included in the hospital's direct medical education payment. Hospitals should not be frozen at their 1984 levels, but should be allowed to adjust their reported costs for legitimate changes in the cost of operating their training programs. Third, an administratively simpler scheme would be to pay the same amount per resident to all hospitals, perhaps with different adjustments for geographic variation in the cost of operating programs, and perhaps with different payments for different specialties to reflect the cost of training or to incorporate incentives for the training of physicians in selected specialties. 2~3

Fourth, hospitals might be required to demonstrate that funds generated from direct medical education payments were actually expended on medical education. One reform would be for HCFA to pay these funds directly to the residency programs. It would make little sense to use Medicare's indirect medical education adjustment to pay for primary care education since these indirect payments were never intended to pay for GME. Instead, they were instituted to pay for the higher cost of care in teaching hospitals. Since residents' time in ambulatory settings was included in the initial calculation of the per-resident payments, this portion of residents' time should remain in the formula. The issue of providing adequate health care to the indigent should be considered as a separate topic, although better funding of care to the indigent would particularly benefit training programs located at inner city medical centers. The proposal that Medicaid programs conform with Medicare Part A direct cost reimbursement (option #in would foster growth and development of primaw care programs, since primary care residencies probably care for a disproportionate share of Medicaid patients. It would not specifically encourage quality primary care programs but it would also not hurt non-primary care programs. Because it is not intrinsically budget neutral, federal and state governments would probably oppose this option. However, if a major federal program that would relieve the present burden on Medicaid were instituted (e.g., mandatory employer-sponsored insurance, or a national program of long term care insurance), then the money made available to Medicaid for other expenditures could be partially allocated to enable states to share the federal burden of financing GME. A plan for encouraging voluntary contributions for primary care GME from insurance companies and HMO's fulfills few criteria. Its primary disadvantages are that the funding stream would be unpredictable and very likely insufficient. However, the option of mandatory payment for GME (option #9) by payers other than federal programs would more equitably distribute the responsibility of paying for ambulator GME. With the exception of insurers, stakeholders would generally favor such contributions. The option of imposing a sales tax on physician services (option #10) fulfills only a few criteria and would face significant stakeholder opposition. Both the initiation and ongoing administration of such a program would be difficult and costly. However, adoption of a popular spending plan that would fulfill the educational criteria might improve its relative advantage. Physicians (including practitioners, physician educators, and specialty organizations) and hospitals would 2~4

probably object strongly to such taxation. Patients, too, may object if they fear the cost wall be passed on to them. The alternative of imposing a tax on hospitals that do not substantially support GME (option #all fulfills the financial criteria, and ongoing administration of such a measure would be simple and relatively inexpensive. Initiation of this option, however, may be complicated and costly. Nonteaching hospitals would oppose the measure strongly and possibly pass on the cost to the health care consumer. Patients, too, may then object, but such a result might decrease the pricing advantage currently enjoyed by non-teaching hospitals. A tax or surch~r~n on third parlor payers (including Medicare, Medicaid, insurance companies, and HMO's) (option #12), would fulfill} all financial and administrative criteria. Those referring to educational goals, however, depend on how the funds would be spent. All educators, specialties, and hospitals would probably support the measure, since it would broaden and more equitably distribute the burden of GME costs. State governments, as well as insurance companies, could be expected to oppose such a surcharge or tax. Patients, too, might object if they were to perceive such a tax as resulting in higher prices for medical care. However, the federal government, and HCFA in particular, would likely welcome being better able to share the cost of GME with other payers. A tax or surcharge on physician licenses (option #13) satisfies all financial and administrative criteria except one; the initiation of such a program would require an organization to collect and disburse these funds, which might be expensive and unwieldly. Physicians would strongly object to this tax, although other stakeholders would generally be supportive. As with the other forms of taxation, this option's outlook might improve with a popular spending plan. Adoption of the Resource Based Relative Value Scale (RBRVS) (option #14) would provide primary care programs with more clinical income relative to other specialties. It would be predictable, administratively feasible, and as long as it does not coincide with an increased volume of service, budget neutral. Non- primary care specialties, educators, and residents are likely to oppose the measure, but recent advocacy by the American Medical Association and American Association of Retired Persons augers well. Congressional and administration support are likely to depend upon linkage with volume control mechanisms. In addition to its effect on funds available to primary care educators, an RBRVS may stimulate more interest among medical students and residents in primary care careers. 215

An extension of outpatient third party coverage to include more outpatient services (option #15), to increase the allowable charge payable in teaching settings, or to include more of the working poor, would aid almost all GME programs. However, quality primary care programs would not be particularly encouraged. Without offsetting decreases in funding for other services or populations, this solution is not neutral for the federal budget. Although it might be argued that increasing outpatient care, particularly preventive services, could save money due to lower inpatient costs, the evidence for this assertion is weak. Federal and state governments, as well as third party payers, would probably oppose such a measure as a way of enhancing primary care GME, but other stakeholders would be supportive. The option of adding a Medicare direct teaching adjustment for ambulatory services through Medicare Part B (option #16) would not be budget neutral unless the remaining payments were recalculated, nor does it broaden the scope of contributors to GME. The development of cost estimates for the adjustment may be difficult. However, such a payment could aid primary care programs, in particular if criteria are established regarding circumstances in which physicians' bills may be supplemented by a direct teaching adjustment. The option of allowing residents to bill third park payers for outpatient care (option #17) would not benefit primary care residencies exclusively, nor would it be budget neutral. If the billing costs were resource-based and if costs of supervision are included in the charges, this billing would probably be higher than that of attending physicians, especially for more junior trainees. The government and third party payers would be likely to oppose this option. Alternatively, foundations could be encouraged to support GME · - In primary care (option #18), perhaps through matching government-foundation programs. In this case, those who pay would not benefit directly, nor would all those who benefit from GME pay for it. The predictability of such grants could be increased by lengthening their time frame to a minimum of five or more years and by assuring a tapering period at both initiation and termination of the grant. Grants could encourage quality in primary care education by targeting the funding to programs that agree to use the money for predetermined purposes (e.g., faculty development, training site development, curricular innovation). . An increase in Title VIT funding and federal faculty development grants for primary care (options #19 and #20) each fulfills the criteria well with two significant negative effects. These options are not budget neutral unless money is found elsewhere to fund them. .. ~. ~. They do not extend the burden of GME payment to ace who keenest, except by passing on the costs through taxation. The federal government would be the only stakeholder expected to oppose the measure. 2~6

The option of increased VA support of primary care GME (option #21) would be budget neutral only if it were funded by a reallocation from other VA expenditures. While it would increase the scope of beneficiaries contributing to primary care GME, its focus would likely be narrow. VA funding of primary care education would principally aid internal medicine since few family medicine and pediatrics programs receive VA support. An increase in state grants to primary care residencies (option #22) fulfills the administrative criteria, since most states already have mechanisms in place for identifying and distributing such funds. As with all grants, the predictability of such funding could be aided by assurance of funding for a specified length of time, such as a minimum of five years. Educational objectives can be met as long as the funds are aimed at primary care programs for quality-enhancing purposes (e.g., funding of supervising physicians, curriculum development, development of new sites for training in underserved areas) and do not dictate curricula in ways that are idiosyncratic of the state legislature or administration. State tax payers and legislatures would likely object to an increase in their contribution to GME, especially in states where primary care physicians are practicing in adequate numbers and are adequately distributed. Funding primary care GME through commitments from future employers, such as HMO's, states, HRSA, or the National Health Service Corps (options #23- 26), involves an increase in overall expenditures for GME, but payers would benefit directly and more of those who benefit from GME would contribute to its costs. However, if this option requires a contract between the individual residents and future employers, the funding stream is not predictable for the training program. The administrative complexity of such a program would vary, depending on whether the future employer had a mechanism already in place for coordinating such efforts, such as the National Health Service Corps or military. Such an option, while increasing the funding of primary care GME, does not necessarily ensure an improvement in quality, although future employers would have a stake in assuring high quality and appropriate training to fulfill their organizations' needs. The stakeholders most likely to object strongly to this proposal are the primaly care residents themselves, who may view it as an inequitable solution whereby primary care residents are singled out from other residents and forced to take on the financial burden of their program. Primary care educators may agree with the residents and also dislike the option because of its unpredictability in financing their programs. Redistribution of money from Title VII of the Public Health Service Act among primary care residencies (option #27) fulfills almost all of the criteria. For only one criterion would this option be considered to have a negative effect; would not broaden the scope of payers. Not all of those who benefit from GME 2 ~ 7

would pay a portion of the costs. Stakeholders would generally favor such a proposal, with the important and probably vigorous exception of those programs who now rely on Title VI] for a significant portion of their funds. Since it is highly unlikely that a simple redistribution of Title VIT funds would provide for "sufficient" funding of primary care programs, regardless of how "sufficient" is defined, an increase in Title VI] funding along with a redistribution to primary care programs could have a saluto~r effect on primary care education with relatively less need for new funds than many other options. The alternative of encouraging academic health centers to devise plans whereby clinical income is redistributed from more clinically lucrative medical and surgical specialties to primary care departments (option #28) fulfills relatively few criteria. Administratively, it would be difficult for most hospitals to accomplish this task, since non-primar~r care physicians would strongly object. It would help support primely care programs in large teaching hospitals but not in smaller community hospitals with fewer physicians on a medical center practice plan. As a voluntary effort, or one instituted by medical center leadership on a local level, it is to be encouraged. Analysis of Spending Alternatives When the spending options described here are analyzed by the criteria discussed, three clusters emerge. The first theme includes eight options, described previously, that are all based on a division of funds on a per resident basis, distributed directly to primary care programs (options I.A - I.H in Table 31. This set of options fulfills the criteria of being predictable, and are administratively feasible. They also allow for curricular autonomy and encourage primary care growth and development. Whether or not they hurt non-prima~g care programs depends on the source of funds. While this increased funding for primary care does not ensure higher quality residencies, well conceived guidelines for how the money is to be spent would help, at least to even the level of training at an acceptable quality. The principal stakeholders who would object to these options would be non-primary care educators, specialties, and residents. The strength of their objections would depend on whether the money to card out these objectives is taken from the stakeholders' present funding sources or from new ones. The second cluster of options is less favorable when viewed against these criteria. These include the alternatives of dividing the funds on the basis of competitive grants for primary care programs to use as each program determines, or for faculty, residents, the ambulatory site, curriculum development, the 2~8

academic unit, increased resident ambulatory time, or for cooperative efforts between primary care residencies (options II.A - II.H in Table 3). These options fail in predictability, although this could be offset by setting the term of the grants as five years or more and by assuring a tapering period at the initiation and termination of the grant. They would also require significant start-up costs and a new or expanded administration to review grants, make site visits, dispense funds, and follow-up on their use. The ongoing costs of administration may also be significant. While the quality of some programs would certainly improve, not all primary care programs would receive grants and thus, a greater discrepancy in quality than currently exists could develop. Depending on the source of funds, both primary care and non-primary care educators, specialties, and residents may object to these alternatives. However, a more limited program of competitive grants, superimposed upon a program of basic payment for all programs, would be more popular and would likely stimulate improved training in primary care. The third cluster of spending options appears less feasible. These consist of those alternatives which Provide a system of incentives for Primary care education and corresponding disincentives for non-primary care education (options IlI.A- IlI.H in Table 3~. These fulfill few criteria, have significant administrative requirements, and would depend entirely on a set of criteria that would need development. In addition to educators, specialties, and residents (in both primary care and non-primary care), teaching hospitals could be expected to object to most of these alternatives as yet another set of regulations with which to comply and a potential loss of funds for their residencies. Summary The growth and development of graduate medical education in primary care has been hindered by current financing mechanisms. The alternatives discussed in this paper would attempt to rectify the situation. Criteria by which policy makers may judge funding alternatives have been proposed. These include financial' administrative, and educational implications of the options. The views of interested stakeholders also need to be considered, since many of the options would affect more than primary care educators and residents. We believe that no single remedy will be sufficient; instead, several solutions will be needed simultaneously. Judged against the criteria proposed here, our preferred options for raising money for primary care graduate medical education are as follows: 0 Adopt a Resource Based Relative Value Scale for payment of physicians and improve coverage of outpatient services. 2~9

o Include residents' primary/ambulatory care time in the calculation of resident FTEs for Medicare direct and indirect medical education payment, add incentives for primary care training in direct payments, and recalibrate payment per resident to maintain budget neutrality. 0 Increase state support through Medicaid participation in payment for GME and through grants for primary care education. O Require participation in payment for GME by other payers, including HMOs and private insurers, coupled with a surcharge or tax on revenues of non- teaching hospitals. O Increase and redistribute Title VIT funding for faculty development, curriculum design and other innovations. Encourage foundation support for similar purposes. Faculty development, in particular, should be allowed a separate funding stream. 0 Experiment with programs to commit residents to future employers, who in turn would support primary care GME. O Experiment with a direct medical education subsidy for outpatient payments to complement payment to hospitals to cover the costs of medical education. Consider an indirect adjustment to compensate for the higher cost of practice (e.g., overhead, more severely ill patients) in teaching settings. The spending options we judge best would involve division of the funds on a per resident basis to residencies in internal medicine, pediatrics, and family medicine for the development of primary care curricular elements through faculty stony site costs, curricular support, academic unit and primary care cooperative efforts, or to use as the individual residency chooses. This base funding would be coupled with competitive grant funding to stimulate innovation and faculty development. In addition, the appropriate and designated use of Medicare direct payments should be enforced by HCFA. support, resident support, ambul" costs, increased ambulator time, Recommendations For Further Analysis 1. A sensitivity analysis should be applied to all options to consider how the effects might change if the definition of primary care is changed to include only general internal medicine, general pediatrics, and family medicine; or if it were to be broadened to include geriatrics, occupational medicine, obstetrics-gynecology, and/or preventive medicine. 220

2. A calculation of the amount of funds being discussed in each case is beyond the scope of this paper. The lOM Committee may want to support a study to determine these amounts. 3. With most options, a mechanism for preserving quality non-primary care residencies needs to be adopted. 4. More options for both raising money and for spending for primary care may need to be considered than are presented here. One means of obtaining the most relevant options, especially for spending, might be to conduct a survey of primary care program directors to solicit their opinion on the components of their programs that are the most difficult to fund. 5. Since few options fulfill all criteria, and few fulfill the criterion that all who benefit from GME should pay for it, the IOM committee should consider combinations of two or more compatible options. 6. The criterion of "sufficient" funding needs to be defined. Most, if not all, of the options proposed may need an adjunct mechanism to fulfill the requirement of adaptation to the changing nature (and cost) of primary care education. 7. The lOM Committee to Study Strategies for Supporting Graduate Medical Education in Primary Care should investigate the costs of administering the current system of GME financing. The present requirement that teaching hospitals collect time and effort reports from faculty to estimate the cost of GME is Invalidated, time-consuming, and subject to interpretation and gamine. A more efficient, equitable system should be feasible. O S. Two methods of changing GME financing to benefit primary care training should be evaluated. The first would involve options that redistribute GME funds to bolster payment for ambulatory care education, and would thus support not only primary care programs but all specialty training programs with an important outpatient component. The second method would seek those options that are specifically targeted to aid primary care training. If the latter is preferred, the TOM Committee should evaluate each option for source and expenditure of funds in light of how well it specifically helps primary care programs. 22~

Table 1 Criteria for Evaluating Alternative Proposals for GME Financing Financial 1. The proposal should be neutral for the federal budget. 2. All those who benefit from GME in primary care should contribute to its costs. 3. Funding should be predictable. 4. Funding should be sufficient. Administrative, i. Implementation should be feasible. 2. Ongoing administration should be simple. 3. Ongoing administration should be inexpensive. Educational 1. Curricular autonomy and flexibility should be maintained. 2. Primary care curricular elements should be fostered and developed. 3. High-qualiJGy programs in non-primary care specialties should not be adversely affected. 4. Incentives should favor high-quality primary care programs. 222

Paw 2 Stakebolders 1. Society 2. Federal Oovernment 3. "C~ 4. State Covernment 5. Private Piers 6. Teaching hospitals 7. Non-teaching hospitals O ~ . O. scions O. Primal care specialties 10. Non-prima~ care specialties 11. Primp care educators 12. Non-prima~ care educators 13. Primal care residents 14. Non-prim care residents 15. Patients

Table 2 Alternatives for Sources of Funds Changes in Existing Medicare GME Pavments 1. Eliminate Part A direct and indirect payments 2. Eliminate Part A direct payments 3. Eliminate Part A indirect payments 4. Reduce Part A direct payments 5. Reduce Part A indirect payments 6. Limit Part A direct and indirect payments to first certification 7. Add incentives and disincentives to Part A direct payments Increase in Categorical GME funding S. Increase Medicaid 9. Mandate or encourage payments for GME payments by insurers and HMOs 10. Impose a tax on physician services If. Impose a tax on non-teaching hospital services 12. Impose a tax on third party payers 13. Impose a surcharge on physician licenses Increase in Ambulatory Payment 14. Adopt Resource Based Relative Value Scale 15. Extension of outpatient insurance coverage 224

Page 2 16. Adopt a Medicare Part B teaching adjustment to ambulatory sites 17. Allow resident billing including cost of supervision Increase in Grants 18. Foundations 19. Title VI] residency support 20. Title VIT faculty development grants 21. VA support 22. State grant support Commitment from Future Employer 23. HMO's 24. States 25. HRSA 26. National Health Service Corps Redistribution of Funds 27. Redistribute Title VIT money 28. Redistribute clinical income within teaching hospitals and faculties 225

Table 3 Alternatives for Expenditures I. Divide funds on a per-resident basis A. Use as each primary care residency chooses or, targeted for; Il. Divide funds on the basis of competitive Grants B. Primary care faculty C. Primary care residents D Primary care outpatient siteLs) ITI. Divide funds on the basis of incentives for primary care education of residents E. Primary care curriculum F. Primary care academic unit G. Increased ambulatory time for primary care residents H. Primary care cooperative efforts 226

Outs ~ appreciate the contr~ut~ns of James S. E~ricb, Edna P. Scb~, M.D~ and Herald T. Perky M.~.

REFERENCES 1. Anthony CR. Medicare support of medical education. Health Affairs 1988, Supplement: 158-62. 2. Watt dM. Discussion paper on the costs and financing of graduate medical education. The Subcommittee on Graduate Medical Education Programs and Financing of the Council on Graduate Medical Education. September 1987. 3. :Lave dR. The Medicare adjustment for the indirect costs of medical education: Historical development and current status. Association of American Medical Colleges January, i985. 4. Moore GT. HMOs and medical education: Fashioning a marriage. Health Affairs Spring 1986:147-53. 5. Gronwell dA. The Veterans Administration and graduate medical education. Health Affairs 198S, Supplement: 163-67. 6. Weaver DL. Director, Division of Medicine, Bureau of Health Professions. Personal communication, May 8, 1989. 7. Council on Graduate Medical Education. Report of the Subcommittee on Graduate Medical Education Programs and Financing, 1988. S. Delbanco TE, Calkins DR. The costs and financing of ambulatory medical education. Journal of General Internal Medicine 1988; 3(March/April, Supplement)534-43. 9. Pawlson KG, Watkins R. The costs of a family practice residency ambulatory care program. Journal of Family Practice 1979; 9:6, 1059-61. 10. Stern RS, et al. Graduate education in primary care. New England Journal of Medicine 1977; 297:12, 638-43. Il. Colwill dM. Financing graduate medical education in family medicine. Academic Medicine, in press. 228

Page 2 12. Eisenberg dM. Who finances graduate medical education? Presentation to the Graduate Medical Education Financing Subcommittee, Council on Graduate Medical Education, September 2, 1987. 13. Health Care Financing Administration. Medicare Program: Changes in Payment Policy for Direct Graduate Medical Education Costs. Federal Re~ster 1988 (September 211; 53:36589-36603. 14. Hadley J. Tique P. Financing graduate medical education: An update and suggestion for reform. Health Policy and Education 1982; 3:157-71. 15. Colwill JM. Dilemmas in Medicare reimbursement of teaching physicians in primary care residency programs. Family Medicine Teaching 1979; Summer:20-22. 16. Petersdorf RG. Current and future directions for hospital and physician reimbursement. Journal of the American Medical Association 1985; 253:17, 2543-48. 17. Petersdorf RG. A proposal for financing graduate medical education. New England Journal of Medicine 1985; 312:20, 1322-24. 18. Perkoff GT. Graduate medical education confronted. Journal of the American Medical Association 1988; 259:3, 402-4. 19. Perkoff GT. Teaching clinical medicine in the ambulatory setting. New England Journal of Medicine 1986; 314: 1, 27-31. 229

Next: Financing Primary Care Residency Training: Examples and Lessons from Successful Programs »
Primary Care Physicians: Financing Their Graduate Medical Education in Ambulatory Settings Get This Book
×
Buy Paperback | $53.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Because of changes in the health care system, the hospital has become less suitable as the primary focus of graduate medical education for primary care physicians. However, the current system of financing health care education and services makes it difficult to accomplish the needed shift to training in primary care ambulatory settings. This book suggests ways of lowering financial barriers to primary care training in ambulatory settings.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!