At a Glance
- Medicare has limited funding for graduate medical education (GME) to the number of residents a hospital trained in 1996.
- Hospitals may train dozens, even hundreds, of residents above the level that they trained in 1996, but they receive no GME payments for residents above the 1996 cap.
- The Centers for Medicare & Medicaid Services (CMS) closed the door to correcting the cap in a final rule published in December 2013.
- Unless Congress acts or a court invalidates CMS’s revised reopening regulation, GME funding is likely to remain fairly stagnant.
When it comes to managing Medicare payment, the Centers for Medicare & Medicaid Services (CMS) seems to have decided finality trumps accuracy as an operating principle.
Most providers, however, would likely choose accuracy over finality if finality means that an error will reduce future payments indefinitely. Last month, the Medicare program revised a regulation to tilt the balance decidedly toward finality by foreclosing all revisions to base-year “predicate facts” once the period to appeal or “reopen” those facts has passed, even when correcting a predicate fact would not alter the payment in a closed Medicare cost-reporting period. This revised regulation applies to any predicate fact, but it was a judicial decision granting hospitals the right to increase resident caps for graduate medical education (GME) that led the CMS to revise the regulation. CMS also may have been concerned that hospitals would seek to add residents to the cap based on statutory changes in the Affordable Care Act (ACA).
Medicare Payment Capped at 1996 Resident Levels
Broadly, CMS makes two types of payments for GME: direct GME (DGME) and indirect medical education (IME) payments. DGME payments cover the identifiable costs of training residents, such as salaries, fringe benefits, and the cost of administering resident training programs. IME payments cover harder-to-quantify costs associated with the teaching of residents and the education process, including:
- Increased demands placed upon staff participating in the education process
- Increased numbers of tests and procedures performed by residents as they learn their medical specialties
- The fact that teaching hospitals tend to attract sicker patients and provide more specialized services
Both payments are calculated based on the number of full-time equivalent residents (FTEs).
Congress capped the number of FTEs that a hospital may count at the number that the hospital trained during its most recent cost reporting period that ended on or before Dec. 31, 1996. These caps went into effect in the 1998 cost reporting year. Since then, however, many teaching hospitals have continued to expand their programs and to add additional programs. Now, 15 years later, these hospitals may train dozens, even hundreds, of residents above the level that they trained in 1996. These hospitals receive no payment from the Medicare program for residents above the 1996 cap.
This shortfall is plainly a problem for teaching hospitals struggling to support their programs. The shortage is particularly troubling in instances when the 1996 FTE cap is wrong because residents were not counted or certain types of activities were omitted from the count, either by the hospital itself or the hospital’s fiscal intermediary. Over the years, some hospitals have corrected errors and increased their 1996 FTE caps, either by appealing the 1996 cost report to the Provider Reimbursement Review Board (PRRB) or by requesting that the intermediary “reopen” the 1996 cost report to correct errors in the FTE count. The problem today is that the deadlines to appeal or to request reopening of the 1996 cost report have long passed. CMS’s policy has been that errors in the 1996 FTE count thus became frozen, permanently reducing the FTE cap of any hospital that noted such errors and denying the hospital funding for training residents that should have fallen within the FTE cap had it been correctly stated.
Opening the Door to Correcting the Cap
This situation changed when the D.C. Circuit issued its decision in Kaiser Foundation Hospitals v. Sebelius (708 F.3d 226 [D.C. Cir. 2013]). In 1996, the Kaiser hospitals failed to include in their DGME and IME FTE counts the time that residents spent training in certain clinics. When hospital leaders discovered this error, they requested that their intermediary add the time to the IME FTE count, but they did not request a similar addition to the DGME FTE count. Thus, based on the understated 1996 DGME FTE count, the Kaiser hospitals’ DGME FTE caps and payments were permanently depressed, even though everyone, including the Medicare fiscal intermediary, agreed that the 1996 FTE count was incorrect.
The Kaiser hospitals appealed later cost years (1999-2003) to the PRRB, arguing that the cap should be increased in these years, even though the deadline to appeal or reopen the 1996 cost report had passed. The U.S. Court of Appeals for the District of Columbia Circuit held that it is not necessary to reopen the 1996 cost report to change the cap in a later year. The court ruled that “the reopening regulation allows for modification of predicate facts in closed years provided the change will only impact the total reimbursement determination in open years.” The court explained that this interpretation of the reopening regulation is consistent with CMS’s past policy in instances where adjusting predicate facts would lower payments to hospitals. The Kaiser hospitals are now entitled to increase their DGME FTE caps in the cost years included in the appeal. This should also apply to future years.
CMS Reverses Kaiser and Closes the Door
In December 2013, CMS revised the reopening regulation to foreclose the revision of predicate facts more than three years after those facts are established or first used; the final rule was published in the Federal Register on Dec. 10, 2013. CMS summarized the change as follows:
Absent a specific statute, regulation, or other legal provision permitting reauditing, revising, or similar actions changing predicate facts: (1) A predicate fact is subject to change only through a timely appeal or reopening of the NPR for the fiscal period for which the predicate fact first arose or the fiscal period for which such fact was first determined by the intermediary; and/or (2) the application of the predicate fact subject to change through a timely appeal or reopening of the NPR for the fiscal period in which the fact was first used (or applied), by the intermediary to determine the provider’s reimbursement.
CMS asserted that the Circuit Court for the District of Columbia had misconstrued CMS’s longstanding interpretation of the reopening regulations. CMS intends to apply the revised regulation both prospectively and retroactively, claiming that retroactive application is necessary to comply with certain statutory provisions and to further the “public interest.”
DGME and IME Caps May Now Be Frozen
The Kaiser decision provided a sensible means for hospitals to correct errors in the caps and other types of predicate facts. Now, CMS has made this much more difficult, if not impossible. It might seem reasonable to some that CMS would foreclose cap revisions long after the base year that established the caps. After all, should hospitals be able to revisit their caps 15 or 20 years later by arguing over the details of its rotation schedules? A more difficult question, however, is whether statutory changes or judicial decisions about the legal standards for determining the caps should be incorporated into the caps.
A good example is the recent controversy over IME research. In 1996, fiscal intermediaries often excluded time that residents spent in “bench research” from the IME FTE count (although such time was generally included in the DGME FTE count). Because nearly all residency programs require a research component to the training, all teaching hospitals would have had residents who performed research in 1996. This time can add up quickly. For large teaching hospitals, research can easily total several dozen FTEs a year.
Prior to 2001, CMS’s policy on IME research was unclear. In 2001, CMS issued a new regulation stating, “The time spent by a resident in research that is not associated with the treatment or diagnosis of a particular patient is not countable” for IME. CMS claimed that this regulation reflected a “longstanding policy,” and CMS therefore applied it to cost years both before and after 2001 (Federal Register, May 4, 2001). Many hospitals disagreed that this 2001 rule was consistent with the prior regulation or CMS’s stated policies prior to 2001. Several federal district courts agreed that the research exclusion was a new policy that could not be applied retroactively.a The First Circuit Court of Appeals sided with CMS and held that the agency could reasonably interpret the pre-2001 IME regulation to exclude research.b Even so, other courts declined to follow the First Circuit.c
Then, in 2010, Congress enacted healthcare reform with the passage of the ACA. Section 5505 of the ACA addressed several aspects of the DGME and IME FTE counts, including research. Congress confirmed that research time may not be included in the IME FTE count from 2001 onward. Congress also specifically stated that this post-2001 statutory exclusion of research “shall not give rise to any inference as to how the law in effect prior to such date should be interpreted.”
Other provisions of section 5505, however, did seem to answer the question of how research should be treated prior to 2001. Section 5505 of the ACA also made other changes to the IME and DGME FTE count methodology that affect the 1996 cost year:
- All “non-patient care activities” that occur outside the hospital are added to the DGME payment calculation, but research is specifically excluded as of July 1, 2009.
- All “non-patient care activities” that occur in the hospital are added to the IME payment calculation, but research is not specifically excluded. This is retroactive to 1983.
Two hospitals—The University of Chicago Medical Center and Henry Ford Health System—argued that these provisions required CMS to add research as a non-patient care activity to the IME FTE count for periods prior to 2001. The Seventh Circuit Court of Appeals agreed, ruling that section 5505 of the ACA was a “clear, statutory answer” to the IME research dispute and required the Secretary to include research time in the IME FTE count prior to 2001 because “research activities are clearly a subset of non-patient care activities,” which section 5505(b) includes in the IME resident count.d This is confirmed by comparing the IME amendments in section 5505(b) with the DGME amendments at section 5505(a). Both address “non-patient care activities,” but subsection (a) specifically excludes research from non-patient care activities. The Seventh Circuit held that the DGME amendment in subsection (a) “compel[s] this understanding” that research is a subset of the non-patient care activities that subsection (b) includes in the IME resident count. Thus, the court said, “Congress spoke clearly when it retroactively allowed reimbursement for non-patient care activities starting in 1983.”
The Sixth Circuit disagreed. After the Seventh Circuit issued its decision, CMS issued regulations implementing section 5505 and specifically declined to include research prior to 2001. The Sixth Circuit held that section 5505 is ambiguous and that CMS’s regulation interpreting section 5505(b) to exclude research prior to 2001 “reasonably exercised the authority delegated to [CMS] under the Act....”e
As it stands now, hospitals located within the Seventh Circuit (i.e., Illinois, Indiana, and Wisconsin) should be able to include research in their IME FTE counts for periods before 2001.f The Sixth Circuit and the First Circuit have upheld CMS’s exclusion of research, but the First Circuit did so prior to the ACA’s section 5505.g Courts in the rest of the country have not addressed the issue. Most important, the D.C. Circuit has not interpreted section 5505 of the ACA, and hospitals nationwide can seek review in the D.C. Circuit. So even hospitals within the First and Sixth circuits could contest the exclusion of research in the D.C. Circuit.
New Rule Attempts to Halt Section 5505 Cap Adjustments
The problem, of course, is that the 1996 cap year is almost two decades past, and almost all of these Medicare cost reports are permanently closed and beyond review of a federal court. Prior to CMS’s change to the reopening regulation, hospitals could argue—relying on Kaiser and University of Chicago—that their IME FTE caps are understated and that this “predicate fact” should be fixed in current and future years by adding research time. CMS may have been concerned about this very possibility when it amended the reopening regulations. CMS claimed that “teaching hospitals have been allowed ample opportunity to correct their resident caps” and cited only court cases that concerned IME research or other non-patient care time.
Congress specifically forbade the reopening of settled cost reports to apply the section 5505 changes, with the exception of hospitals that had DGME or IME appeals pending when the ACA was enacted (section 5505 [d] of the ACA). It is clear, therefore, that Congress did not want the pre-2001 payments changed unless a hospital had a pending appeal. It is likely that only a small number of pre-2001 IME appeals were pending in 2010 when the ACA was enacted. It would be odd for Congress to weigh in and make extensive changes to the FTE count methodology, some retroactive to 1983, if those changes would affect almost no one.
Hospitals could reasonably conclude that Congress intended the section 5505 FTE changes to alter base-year FTE “predicate facts” to be applied in future cost reports, so long as the base year payment itself was not altered. Despite CMS’s protests that its longstanding interpretation of the reopening rules forbids changes to predicate facts after the time to appeal or reopen has passed, this view was certainly not universal. The Kaiser hospitals, the PRRB, the D.C. District Court, and the D.C. Circuit Court all agreed that predicate facts could be changed. If the Seventh Circuit is right that Congress intended to include research time in the IME FTE count for periods prior to 2001, Congress may have also intended to add this time to the caps prospectively. CMS’s amendment to the reopening rules appears, in part, designed to thwart any such revisions to the caps.
The Outlook for Teaching Hospitals
Teaching hospitals perpetually struggle to finance GME. The Medicare program is an important funding source for training the nation’s physicians, and that funding is not likely to support the training of the additional physicians that will be needed to treat an aging population and all the individuals that will enter the health insurance market due to healthcare reform. The FTE caps place a permanent damper on this training. Unless Congress acts or a court invalidates CMS’s revised reopening regulation, GME funding is likely to remain fairly stagnant.
Ronald S. Connelly, Esq., is principal, Powers Pyles Sutter & Verville PC, Washington, D.C. He represented the hospitals in University of Chicago v. Sebelius and Henry Ford Health System v. Sebelius.
a. Henry Ford Health System v. Sebelius, 680 F. Supp. 2d 799 (E.D. Mich. 2009); University of Chicago Medical Center v. Sebelius, 645 F. Supp. 2d 648 (N.D. Ill. 2009); University Medical Center Corporation v. Leavitt, No. 05-CV-495 TUCJMR, 2007 WL 891195 (D. Ariz. March 21, 2007); Rhode Island Hospital v. Leavitt, 501 F. Supp. 2d 283 (D.R.I. 2007), reversed, 548 F.3d 29 (1st Cir. 2008); Riverside Methodist Hospital v. Thompson, [2003-2 Transfer Binder] Medicare & Medicaid Guide (CCH) ¶ 301,341 (S.D. Ohio July 31, 2003).
b. Rhode Island Hospital v. Leavitt, 548 F.3d 29 (1st Cir. 2008).
c. University of Chicago Medical Center, 618 F.3d at 744; Henry Ford Health System v. Sebelius, 654 F.3d 660, 633 (6th Cir. 2011).
d. University of Chicago Medical Center, 618 F.3d at 744.
e. Henry Ford Health System v. Sebelius, 654 F.3d 660, 633 (6th Cir. 2011).
f. The Seventh Circuit encompasses Illinois, Indiana, and Wisconsin. A federal district court in Illinois, relying on University of Chicago, recently held that research time must be included in the IME FTE count. The Secretary of Health and Human Services appealed, and the case is currently pending before the Seventh Circuit.
g. The Sixth Circuit includes Michigan, Ohio, Kentucky, and Tennessee. The First Circuit covers Maine, Massachusetts, and Rhode Island.