Funding of graduate medical education, by whatever means, should be limited to three years or first eligibility for certifi cation in a general specialty. The three-year limit coincides with board eligibility in four primary care specialties—internal medicine, pediatrics, family practice and emergency medicine. The Task Force recognizes that other specialties require training beyond three years and that reasonable arguments can be made for funding them to completion under current requirements of eligibility for general certification (e.g., five years for surgery). Further training in subspecialties can be supported in other ways, such as professional fecs, faculty practice plans and scholarships. Funding of graduate medical education should not be limited The Task Force believes that three years is the minimum period of supervised practice that is needed to ensure that physicians are qualified before entering private practice. The Task Force does not favor proposals that would fund through health care revenues only the first year of training. That is a year in which residents are not yet licensed and cannot bill independently for their services. Under the one-year proposals, residents would charge for their services after the first year; teaching hospitals would neither pay nor be reimbursed for residents' salaries after licensure. Billing by residents early in their training has several difficulties. It would preclude billing by an attending physician if his or her services overlapped those of the residents. Also, the likelihood of residents earning enough to support their education is fading as more people join prepaid health groups and as more surplus specialists enter primary care. Funds for graduate medical education should support the The indirect cost of education comprises many expenses that arise solely because of teaching programs, such as more diagnostic procedures or the maintenance of a broader range of services. They should be supported through education funding. They should not be confused with the Medicare Prospective Payment System's "indirect medical education adjustment," which is a formula payment not necessarily associated with functions of graduate medical education. On the Issue of If money for graduate medical education is generated by means of a Lax, several options are available for distributing the funds. Aggregated at the Federal level, the funds could be disbursed as grants. However, if the government made much of a change in dollar amounts from year to year, it would gain control of graduate medical education; hospital administrative and education planning would be delayed until success in obtaining a grant was assured. • A block grant program could be implemented to transfer funds from the Federal government to the states. That would avoid the establishment of a large Federal bureaucracy to administer the grant program, but it would necessitate the creation of state-level offices of disburse ment and, more importantly, would give states discretion in allocating funds for graduate medical education. • If the funds initially were aggregated at the state level, state government would have virtually complete control of their disbursement. The difficulty with this, and to some extent with a block grant program, is the possibility that state control of education funds would mean more support for residency training in state and municipal hospitals than in private academic health centers. The following option for distribution of funds is the choice of the Task Force. A voucher system should be developed to distribute graduate The Task Force believes that a voucher system would bring to the distribution of graduate medical education funds a combination of desirable elements not fully available in the disbursement options described above. Medical school graduates would be entitled to receive vouchers with which to purchase their graduate education. They could choose from among accredited residencies offered by hospitals on the basis of what the hospital wished to provide to match its service needs, faculty, patient mix and other considerations. Vouchers could be financed by a combination of Federal, state and private sector funds in the relative proportions that they now support the direct costs of graduate education. The money value of a voucher would be held approximately equal to the cost of resident stipends, fringe benefits and other direct education costs. Disbursements of vouchers could be accomplished through any number of arrangements ranging from purely governmental to a voluntary private sector organization. Whatever system is put in place, the Task Force emphasizes that contributions by the private sector are crucial for preservation of graduate medical education. Although Medicare is the single largest payer of health services, the entire Federal govemment accounted for less than one-third of expenditures for personal health care in 1982. Most of the remainder of those expenses were paid by insurers, by businesses, and directly by patients-all in the private sector. On the Issue of Size The number of vouchers for first-year positions in graduate This recommendation, when coupled to the three-year-or-first-board eligibility recommendation stated earlier, would limit voucher support to little more than three times the number of graduates in the index year from schools accredited by the Liaison Committee on Medical Education and schools of osteopathic medicine. On the basis of 1983 graduates (15,885), the total number of vouchers offered would be 47,655 (not counting the extra-time slots for surgery and other first-board eligibility periods of more than three years). This is significantly fewer than the 59,176 occupied by United States medical graduates in 1983. Also in 1983, another 13,221 positions were held by foreign medical graduates (FMGs). In the proposed system, vouchers would not go to FMGs, but they would be free to seek other funding sources for training in the United States. The Task Force rec ognizes that exclusion of FMGs from a voucher system could pose detrimental effects for some public hospitals that have traditionally relied on FMGs. One solution would be the allocation of some vouchers to those hospitals most affected by the loss of FMGs so that the hospitals could recruit residents to meet their service needs. On the Issue of Program Content Principle #4 summed up the Task Force's strong opinion that funds for graduate medical education should no longer tie a resident exclusively to the teaching hospital or hospital-based services. The tertiary care training received in the high-technology hospital does not match the majority of cases seen in typical practice. There are many ways in which the discrepancy could be reduced. Teaching hospitals could establish satellite clinics in the communities, thereby giving residents more ambulatory care experience and perhaps enlarging the hospital referral base. Or hospitals could affiliate with group practices at non-hospital sites. Or residents could be encouraged to accept rotations in offices of physicians in private or group practice, as the University of Iowa does in tanuly practice programs. However, these and other possibilities of more appropriate clinical experiences for residents depend on separating funding for education from payment for hospital care. On the Issue of Academic health centers must become engaged in efforts to slow and ultimately reverse the trend toward specialty maldistribution in the United States. One reason is that an oversupply of specialists logically has some |