Editor’s Note: This is a commentary on Chen C, Chen F, Mullan F. Teaching health centers: A new paradigm in graduate medical education. Acad Med. 2012;87:1752–1756.
In the years leading up to the passage of the Patient Protection and Affordable Care Act (ACA) of 2010, numerous policy makers had identified problems with the financing of graduate medical education (GME) in the United States, but there was no broad consensus on an acceptable approach to reform.1 Although the ACA provides a pathway to near-universal access to affordable health insurance, it leaves GME financing problems—as well as a variety of other, even more vexing health care system issues—to be addressed by future legislation. The ACA does, however, offer opportunities to experiment with potential solutions to various health care policy dilemmas. The Teaching Health Center Graduate Medical Education (THCGME) program can be seen in this context.2 In this issue of Academic Medicine, Chen and colleagues3 discuss the THCGME program and its implications for U.S. health care workforce policy and broader GME financing reform.
A Brief History of Federal GME Support
As Chen and colleagues3 note, the THCGME program offers a potential approach to addressing some of the issues of federal funding for GME. Understanding the policy context for these problems and possible solutions requires a brief review of the history of GME and its support in the United States.
The first half of the 20th century witnessed the rise of numerous medical specialties requiring diagnostic or therapeutic approaches that could not be provided in physicians’ consultation rooms or in patients’ homes. New medical school graduates wishing to learn these techniques therefore needed to spend time in the hospitals where these services were delivered. Hospital leaders saw advantages to accommodating these trainees, as did the specialized physicians, who were happy to have free, talented help. The costs of trainees’ housing and living stipends were built into the prices that private hospitals charged for their services (and into the budgets of hospitals supported by government or charitable organizations). The number of physicians in hospital-based training grew rapidly after World War II; by the 1960s, GME was a prominent feature of teaching hospitals, which built GME support into their operating costs. Accordingly, the original Medicare legislation explicitly recognized “that a part of the net cost of such activities (including stipends of trainees, as well as compensation of teachers and other costs) should be borne to an appropriate extent by the hospital insurance program.”4
In the mid-1980s, there was a shift in hospital payment from cost-based reimbursement (wherein teaching hospitals could recover their training expenses) to new approaches imple-mented by both public and private payers. Medicare adopted prospective hospital reimbursement through the diagnosis-related group (DRG) system.5 To accommodate continued Medicare support for GME, the legislation enacting Medicare DRGs authorized the Health Care Financing Administration (precursor to the Centers for Medicare and Medicaid Services) to pay teaching hospitals the “direct expenses” for GME occurring during the care of Medicare beneficiaries. These direct medical education (DME) payments addressed resident stipends, the salaries of requisite teaching faculty and staff, related fringe benefits, and educational facility costs. In addition, Congress acknowledged that teaching hospitals incurred indirect GME expenses as well. Indirect medical education (IME) adjustments to DRG payments compensated teaching hospitals for higher inpatient operating costs attributable to unmeasured patient complexity but not captured by the DRG and for other operating costs associated with being a teaching hospital. In the nearly 30 years since the introduction of Medicare DRG payments to hospitals, a variety of refinements have been made to this approach, but the basic structure of Medicare DME and IME funding has remained intact, resulting in more than $9.5 billion in annual payments to U.S. teaching hospitals.6
Not surprisingly, numerous problems have been identified related to using a system of GME financing that is tied more to past history than to current realities. In recent years, the most intense policy debate has been over the level of Medicare support for teaching hospitals’ IME costs. The Medicare Payment Advisory Commission6 (MedPAC) has long been a critic of what it considers Medicare’s flawed formula for determining such payments. Guided by MedPAC analyses (as well as the need to find Medicare cost savings), various deficit-reduction proposals have outlined substantial cuts in IME support. Another ongoing controversy has been the continued hospital orientation of the GME financing system. Primary care workforce advocates have noted that this approach cedes to individual teaching hospitals control of the resources that determine the numbers and specialty mix of physicians being trained and the educational venue for most GME in the United States. Medicare does not, however, require institutions receiving GME funding to be transparent or publicly accountable for the decisions they make at the institution level that affect physician workforce composition. However, at present there is no consistent source of funding that institutions can use for the GME costs that are not covered by Medicare. Policy makers have therefore been hard pressed to identify an alternative to teaching hospitals with the interest, infrastructure, and financial capacity to act as local sponsors of residency programs supported only by Medicare DME funds.
The THCGME Program: A Potential Model
Thus, the ACA funded the THCGME program to allow the Health Resources and Services Administration (HRSA) to support teaching health centers (THCs), defined as ambulatory care organizations that sponsor new or expanded primary care residency programs.2 In this legislation, the definition of primary care residency training goes beyond the focus of the HRSA Title VII primary care medicine training programs (family medicine, general internal medicine, general pediatrics7) to include geriatrics, obstetrics–gynecology, and psychiatry as well as dentistry (also a long-standing interest for HRSA Title VII programs7). As Chen et al3 note, THCGME funding comes with a range of reporting requirements that are quite different from Medicare’s. For example, each THC must produce an annual report that details the types of primary care resident training programs it offers, the number of approved training positions supported by THCGME funding, the number of residents who completed their residency training during the academic year, and other information deemed useful by HRSA (such as program graduates’ career choices and care of disadvantaged populations).
The THCGME and Medicare funding mechanisms have important differences as well. Although both base their direct expense payments on a “per resident” amount, the THCGME methodology does not adjust payments according to the facility payer mix. By statute, the Medicare program pays only the Medicare portion of GME expenses; however, the THCGME program supports the full direct training cost regardless of the source of financing for patient care. The indirect expense payment method of the THCGME program also differs fundamentally from that devised for Medicare. THCGME indirect expense payments are deter-mined by the secretary of Health and Human Services (HHS) and take into consideration the indirect training costs relative to supporting a primary care residency program in a qualified institution. Medicare IME support is determined through computations that would not be possible for the community-based ambulatory care entities that are typical THCs. One cannot, for example, estimate the intensity of teaching at a THC on the basis of a “residents-to-bed ratio.” Nor are there ready measures of the complexity of care (or related clinical teaching effort) at a THC comparable to the number of admissions and DRG severity weights relevant to Medicare IME payments. In addition, the scope of intended program support for the THCGME indirect expense payments may be distinct from that of Medicare IME. Medicare IME cost estimates consider not only productivity losses due to trainee learning (e.g., increased use of tests and ancillary services, longer operative times and hospitals stays) but also the greater severity of illness, higher concentration of high technology, increased standby capacity, and differences in the types of physicians and patients at teaching hospitals.4,8
There is an even more fundamental difference between THCGME and Medicare GME financing, however. As Medicare GME support is tied to payments for covered services to Medicare beneficiaries, it is an entitlement program that is not subject to the annual federal appropriations process. In contrast, after the first five years funded by the ACA, the THCGME program becomes subject to annual congressional appropriations like HRSA Title VII7 and the Children’s Hospitals GME program.9 If evaluations confirm the early promise of the THCGME innovations, additional congressional action will be required to sustain, let alone expand, the program.
Chen and colleagues3 suggest that the THCGME program is a “solution for funding GME to better meet the nation’s health care workforce needs.” The program certainly offers several interesting opportunities to gain information that might guide broader GME policy reform. The HHS secretary must develop a fundamentally new way to determine and allocate support for the indirect costs of medical education, which could inform the long-standing debate over Medicare IME payments. Furthermore, given that the THCGME program incorporates a diverse array of training programs—typical primary care specialties as well as clinical disciplines requiring more procedural training such as obstetrics–gynecology and dentistry—year-to-year and specialty-to-specialty differences in direct and indirect educational costs can be investigated. The diversity of residency programs with THCGME support may also allow a more complete analysis of how trainee-related productivity gains vary across venues, educational innovations, and sources of payment for clinical services. THCs’ detailed reporting will also provide data on the impact of assorted innovations on career choice and clinical care. All of this information could be useful in devising a more transparent and equitable system of support for the costs of GME.
THCGME program advocates must, however, be cautious in drawing any conclusions regarding how these lessons could guide broader Medicare GME policy. The biggest challenge is that Medicare GME payments must be tied to the care of Medicare beneficiaries. This fact is well known to the leaders of children’s hospitals, who care for few Medicare patients and, therefore, endure annual appropriation of funding for the Children’s Hospital GME program. Even if the next round of GME policy reform were to focus only on residency programs for adults (serving substantial numbers of Medicare beneficiaries), it is not within Medicare’s current scope to cover all the DME costs of this training. Furthermore, future evidence-based reform of IME reimbursements likely will make the gap in Medicare funding for DME a more pressing policy problem. Absent “all-payer financing” of GME, residency program sponsors will continue to lack sufficient Medicare funds to cover all training program costs and will necessarily fund resident positions in the clinical disciplines that serve their local, strategic institutional interests. If past institutional behavior under the Medicare GME position “cap” is any guide, organizational leaders will favor investments to add (or retain) specialty-focused resident and fellow positions over primary care positions.
Although, in theory, the THCGME program could serve as a platform for a new approach to funding the full costs of GME programs in workforce priority areas, the tremendous downward pressure on domestic discretionary spending makes this unlikely. Even if the billions of dollars required to support all training in the current THCGME disciplines could be achieved through congressional appropriations, would educators (and their trainees) be comfortable tying the future size, structure, and purpose of these critical GME programs to election cycles and the annual federal budget process? The struggles of the HRSA Title VII programs over the last 20 years do not provide confidence that a THCGME program supported through domestic discretionary spending offers a satisfactory long-term solution to GME financing.
Nonetheless, the HRSA Title VII programs have been critical to developing program innovations, faculty development, and resident training in underserved communities.10 Building on this tradition, the THCGME program provides an exciting new opportunity. The organizational infrastructures established, program leaders developed, data collected, and lessons learned can provide important resources to guide more fundamental reform of GME payment policy in the United States.
Other disclosures: None.
Ethical approval: Not applicable.
1. Iglehart JK. Medicare, graduate medical education, and new policy directions. N Engl J Med. 2008;359:643–650
3. Chen C, Chen F, Mullan F. Teaching health centers: A new paradigm in graduate medical education. Acad Med. 2012;87:1752–1756
4. Rich EC, Liebow M, Srinivasan M, et al. Medicare financing of graduate medical edu cation. J Gen Intern Med. 2002;17:283–292
6. Medicare Payment Advisory Commission. . Graduate medical education financing: Focusing on educational priorities. Report to Congress: Aligning Incentives in Medicare. June 2010 Washington, DC Medicare Payment Advisory Commission www.medpac.gov/documents/jun10_entirereport.pdf
. Accessed August 21, 2012
7. Advisory Committee on Training in Primary Care Medicine and Dentistry. Preparing Primary Care Practitioners to Meet America’s Future Healthcare Needs. The Critical Role of Title VII, Section 747. Fourth Annual Report to the Secretary of the U.S. Department of Health and Human Services and to Congress. 2004 Washington, DC U.S. Department of Health and Human Services
8. Zeidel ML, Kroboth F, McDermot S, et al. Estimating the cost to departments of medicine of training residents and fellows: A collaborative analysis. Am J Med. 2005;118:557–564
10. Reynolds PP. Title VII innovations in American medical and dental education: Responding to 21st century priorities for the health of the American public. Acad Med. 2008;83:1015–1020