Graduate Medical Education That Meets the Nation's Health Needs (2014)

Appendix F

Illustrations of the Phase-In of the Committee’s Recommendations

This appendix provides three illustrations of the phase-in of the committee’s recommendations. See Appendix E for a description of the data and methods used here.

EXAMPLE OF A PHASED-IN ALLOCATION OF MEDICARE GME FUNDING TO THE OPERATIONAL AND TRANSFORMATION FUNDS

Aggregate funding levels in the Operational Fund will be reduced initially to 90 percent of current graduate medical education (GME) funding levels and transition to 70 percent by Year 5. Table F-1 illustrates how funds would be allocated between the Operational and Transformation Funds over the first 5 years of the transition. The illustration assumes that the base-year funding amount would equal the most recent estimates provided by the Centers for Medicare & Medicaid Services and presented in Chapter 3. One method for reducing the operational funding to generate the funding for the Transformation Fund would be to phase in a 50 percent reduction in indirect medical education (IME) operating payments to acute care hospitals. In the first year, a 14 percent IME reduction would be needed to fund the Transformation Fund. If the additional IME reduction were evenly phased in over Years 2-5, approximately an additional

TABLE F-1 Example of a Phased-In Allocation of GME Funding to Operational and Transformation Funds in Transition Years 1–5 ($ in Billions)

NOTE: Baseline amounts (column 1) reflect Medicare GME funding in 2012. Assumes that the funding for children’s hospitals and THCs would equal the same PRA as other training sites. Other specialty hospitals include psychiatric facilities, rehabilitation hospitals, other. CHGME = children’s hospital graduate medical education; DGME = direct graduate medical education; GME = graduate medical education; IME = indirect medical education; OF = Operational Fund; PRA = per-resident amount; TF = Transformation Fund.

9 percentage-point reduction would be made each year. For example, the Year 2 reduction would be 23 percent.¹

By Year 5, the funding formulas would be changed from hospital-specific amounts to a national combined per-resident amount (PRA). The separate direct graduate medical education (DGME) and IME funding streams would be changed to a combined PRA. The 50 percent weighting for residents beyond their initial residency program in the current DGME funding formula would be incorporated into the portion of the combined PRA attributable to DGME.

The combined PRA would be allocated initially on the basis of the number of Medicare-funded resident slots without regard to Medicare use rates. Ultimately, performance-based funding allocations would be implemented.

CALCULATING A COMBINED PER-RESIDENT AMOUNT

Table F-2 illustrates a general approach to determining the combined PRA. First, the average DGME payment per resident is calculated (exclusive of children’s hospitals). The PRA would be budget neutral to estimated aggregate DGME payments for the same set of hospitals after adjustment by the Medicare geographic adjustment factor (GAF). The resulting DGME per-resident amount was $37,300 before any adjustments for inflation.²

The amount for residents beyond their initial residency period would be 50 percent of this amount, or $18,650.

Next, we calculated an average GAF-adjusted IME payment per resident for general acute care hospitals that would be budget neutral to estimated IME payments if IME operating payments were reduced by 50 percent, consistent with the Medicare Payment Advisory Commission’s finding that the current levels are twice the amount empirically attributable to higher patient care costs (MedPAC, 2010). The resulting IME per-resident amount was $43,435.

The combined PRA, the sum of the IME and DGME component, or $80,735, would be applicable to residents in their initial residency period. The combined PRA for residents beyond their initial residency period would be $62,085 or 77 percent of the PRA for residents in their initial residency period. In other words, residents in subspecialty programs would count as 0.77 FTE if the 0.5 weighting were applied to the DGME portion of the

________________

¹ The reductions would be made only to the operating IME payment based on the Medicare Payment Advisory Commission’s findings. The capital adjustment is empirically derived, as are the IME payments to psychiatric and rehabilitation hospitals.

² This amount does not take into account the 6 percent differential between primary care and other residency programs that currently applies to hospital-specific PRAs but not to the national PRA applicable to new residency slots.

TABLE F-2 Illustration of Combined PRA Calculation, Before Inflation Adjustment

NOTE: DGME = direct graduate medical education; GAF = geographic adjustment factor; IME = indirect medical education; PPS = prospective payment system; PRA = per-resident amount.

SOURCE: IOM analysis of the 12/31/13 CMS Healthcare Cost Report Information System update.

composite rate and no weighting was applied to the IME portion. The committee suggests that the proposed GME Policy Council review this weighting scheme and also assess whether the combined PRA should vary for other types of residents, for example, residents in primary care, dentistry and podiatry, and rural training programs.³

________________

³ The GME Policy Council might also consider whether the geographic adjustment to the PRA should be revised to reflect specific GME cost components. See the Institute of Medicine report Geographic Adjustment in Medicare Payment. Phase I: Improving Accuracy for background and recommendations regarding the Medicare geographic price indexes (available at http://www.nap.edu/catalog.php?record_id=13138) (accessed April 23, 2014).

ILLUSTRATION OF THE IMPACT OF CHANGING TO A COMBINED PER-RESIDENT AMOUNT

Table F-3 illustrates the types of redistributions that will occur with the implementation of the combined PRA by type of hospital for the Prospective Payment System hospitals in our cost report analysis file. The percentage change in payment attributable to the 50 percent reduction in IME payments (–34 percent) is shown separately. It produces relatively minor differences in the impacts across hospital groups that reflect differing proportions of total GME payments attributable to IME. IME payments are on average a higher proportion of total GME payments in hospitals with a large number of Medicare discharges than hospitals with relatively fewer discharges. As a result, the IME reduction has a greater impact on GME funding for residents at the larger hospitals. The remaining changes are budget neutral in the aggregate.

Under current policy, the DGME counts and the IME counts are not the same because of differences in the rules for counting resident time. Moreover, because of the rolling average used in the current methodology, some hospitals are receiving funding for more residents than they are training. This policy was implemented when there was a projected surplus of physician supply and is no longer appropriate. Nevertheless, the illustration uses the resident counts to determine IME and DGME payments under current Medicare policies. The committee suggests that a single policy for counting residents (with appropriate weighting) should apply to the allocation of the combined PRA. Once the funding flows to the program sponsor, most issues that have complicated resident counts under current IME and DGME funding policies would be eliminated and the counting rules would be more straightforward.

REFERENCE

MedPAC (Medicare Payment Advisory Commission). 2010. Graduate medical education financing: Focusing on educational priorities. In Report to the Congress: Aligning Incentives in Medicare. Washington, DC: MedPAC. Pp. 103-126.

TABLE F-3 Illustration of Impacts of Changing to Combined PRA

NOTE: DGME = direct graduate medical education; IME = indirect medical education; PRA = per-resident amount.

This page intentionally left blank.