AAMC Letter to CMS Administrator
Scully Regarding Medicare Resident Limits
January 25, 2002
Thomas Scully, Administrator
Centers for Medicare & Medicaid Services (CMS)
200 Independence Avenue, SW
Washington, DC 20201
Dear Administrator Scully:
On behalf of the Association of American Medical Colleges,
I write to share with you our concerns about the current Medicare
"resident limit" restrictions and to ask that CMS propose
several changes to the resident limit regulations. We believe
regulatory changes, while not eliminating the primary problem
of the resident limits themselves, will provide some temporary
relief until the more fundamental issue can be addressed by
the Congress. At the end of our letter, we also ask that CMS
modify its regulations regarding Medicare reimbursement associated
with residents educated in nonhospital sites-an issue of continuing
concern and confusion for our members.
The resident limits were required by the Balanced Budget
Act of 1997 (BBA). In general, the law mandated that, for
purposes of Medicare indirect medical education (IME) and
direct graduate medical education (DGME) reimbursement, a
hospital's number of allopathic and osteopathic residents
may not exceed the number reported on the hospital's most
recent cost report that ended on or before December 31, 1996
(42 U.S.C. 1395ww(h)(4)(F)). Dental and podiatry residents
are excluded from the resident limit provision.
I. The Resident Limits Are Having a Chilling Effect on
the Educational Missions of Teaching Hospitals
High quality residency education is fundamental to ensuring
a physician workforce that many consider the best in the world.
Medicare has been an important source of financial support
for residency programs by reimbursing the program's share
of the costs of residency education at teaching hospitals.1
The provision in the BBA to essentially "freeze" the number
of residents that are associated with Medicare reimbursement
has imposed significant difficulties on teaching hospitals
and medical schools that sponsor and conduct graduate medical
education programs. As time passes, the policy is beginning
to impede the continued development of the educational mission
at many of our institutions. We believe that, after being
in place for over four years, it is time-at a minimum-to substantially
modify the resident limit policy.
Residency education is a dynamic process. Teaching hospitals
and medical schools routinely examine their graduate medical
education programs and incorporate changes as a result of
numerous factors, including advances in medicine and changing
needs for future physicians. These efforts increasingly are
being hampered by the imposition of resident limits.
Staff of the Accreditation Council for Graduate Medical Education
(ACGME) have noted that the limits are impeding even preliminary
discussions about the establishment of new residency programs.
For example, there is a pressing need and desire to increase
residency training programs that involve preventive medicine
and public health yet, according to discussions with ACGME
staff, the resident limits are having a chilling effect on
these planning activities.
Staff of the Accreditation Council for Graduate Medical Education
(ACGME) have noted that the limits are impeding even preliminary
discussions about the establishment of new residency programs.
For example, there is a pressing need and desire to increase
residency training programs that involve preventive medicine
and public health yet, according to discussions with ACGME
staff, the resident limits are having a chilling effect on
these planning activities.
In some parts of the country there is a need for more emergency
room physicians, yet the ability to start emergency room residency
programs is being thwarted by the resident limits. There also
are emerging specialties that are in the process of being
recognized by the ACGME for accreditation. The resident limits
are barriers to teaching hospitals' abilities to incorporate
these "cutting edge" specialty residency programs without
diminishing other programs. As the population ages, additional
residency programs will also be needed to ensure a pipeline
of specialists in areas such as cardiology, rheumatogy and
oncology.
There are other problems associated with the resident limits,
including the fact that the BBA provision is based on a snapshot
of activity, essentially "freezing" the status of residency
education at a random point in time-1996. A number of institutions
were undergoing transitions during that time such that 1996
was an anomaly in terms of residency counts. For these institutions,
the resident limits can have profound effects on their educational
missions.
The resident limits have now been in place for over four
years. In other areas, decisions to impose a "freeze" are
temporary in nature. In health care, and in Medicare in particular,
we are unaware of policies that have not factored in the need
for modifications after a certain period of time. In sum,
we believe it is time to reconsider the resident limits policy.
II. Resident Limit Regulatory Changes Are Needed and Are
Appropriate
The BBA gave the Secretary authority to implement exceptions
to the resident limits through the regulatory process. This
authority was explicit with regard to new residency programs
established on or after January 1, 1995 and when hospitals
of the same affiliated group (as defined by the Secretary)
wish to have the resident limits apply on an aggregate basis.
We believe that CMS has not fully utilized this authority
to recognize new residency programs.
We also believe that several of the regulations that have
been implemented to date are not necessary to comply with
the statute and, in some cases, may actually contravene statutory
intent. We urge that CMS modify these regulations to help
ensure that the resident limit statutory provision is implemented
in a way that is fair and reasonable.
A. CMS Should Permit Resident Limit Adjustments to Reflect
New Residency Programs
The BBA gave the Secretary the authority to permit adjustments
to hospitals' resident limits for new programs. CMS implemented
this provision by adjusting the resident limits for rural
hospitals to reflect any new programs they establish. A very
limited adjustment also has been permitted for urban hospitals
that participate in rural training track residency programs.
We believe that CMS should utilize its statutory authority
to also allow adjustments to resident limits for urban teaching
hospitals that establish new programs.
The Conference Agreement accompanying the BBA noted that
the conferees were "concerned" about the impact of resident
limits on the establishment of new programs.2
Specifically, the conference agreement stated:
The Conferees understand that there are a sizeable number
of hospitals that elect to initiate [programs after January
1, 1995] (as well as terminate such programs) over any period
of time, and the Conferees are concerned that within the
principles of the cap that there is proper flexibility to
respond to such changing needs, including the period of
time such programs would be permitted to receive an increase
in payments before a cap was applied.
- From BBA Conference Report 105-217 at S-203
We recognize that the Secretary's authority is not unlimited.
The Conferees also noted in their report that "the Secretary's
flexibility is limited by the conference agreement that the
aggregate number of FTE residents should not increase over
current levels." CMS has not published any data indicating
that the current number of residents exceeds the resident
count that existed when the BBA was passed. Preliminary analyses
of Medicare data by the AAMC suggests that the current national
number of residents is less than it was when the BBA was passed.
Consequently, we believe there is ample opportunity for CMS
to consider expanding its current exceptions requirement for
new programs while adhering to the Conferees intent.
As discussed above, there are needs for additional residency
programs. We understand that CMS might not want to implement
an open-ended policy, but we think criteria could be developed
that would give CMS the option to permit limited expansions
for new programs. It also is important to remember that this
suggestion is limited to new programs, not expansions of existing
programs. In view of this, we believe there is little potential
for significant increases in resident numbers.
B. CMS Should Consider Other Regulatory Changes Relating
to Resident Limits
We also urge CMS to reexamine its current regulations relating
to resident limits and implement the following modifications.
1. Resident Limit Adjustments and Hospital Closures
Through regulation, CMS permits temporary upward adjustments
to the resident limits of hospitals that take on and complete
the training of residents from hospitals that are closing.
We appreciate CMS' recognition of this problem and the efforts
to minimize disruptions in residents' education. However,
we urge CMS to consider permitting permanent resident limit
adjustments in these situations.
Teaching hospital closures can disrupt the balance of educational
programs within a geographic area. The loss of a residency
program that is the only one of its kind within a community
can have important detrimental effects on the likelihood of
these types of physician specialists practicing in that area.
In addition, the closing teaching hospital may serve as an
important rotation site. If the sponsor hospital is at its
resident limit and cannot absorb these rotated resident slots
on a permanent basis, the residency program ends up with fewer
slots-a result that could impact the quality of the program.
Because it involves a hospital closure, allowing permanent
adjustments in these situations would not increase overall
resident counts. We urge CMS to consider implementing this
regulatory change.
2. Incorporating Weighted Resident Counts in the Resident
Limit Regulations
Medicare DGME payments are based on an weighted
count of residents, whereby residents in their initial residency
periods are counted as 1.0 FTE and residents that undergo
additional training in subspecialties are counted as 0.5 FTE.
By contrast, the DGME resident limit is based on an unweighted
resident count. Under the current DGME resident limit
regulations, if a hospital's current year unweighted resident
count exceeds its resident limit, its current year weighted
resident count will be reduced in the same proportion that
the number of unweighted residents exceeds the resident limit,
even if the weighted count of residents does not exceed the
resident limit. We believe this policy contravenes legislative
intent and should be withdrawn.
The resident limit provision applies only to unweighted resident
counts. Nowhere in the BBA is it mandated that the weighted
count of residents must be adjusted to coincide with the resident
limit. If Congress wanted to address this issue, they would
have done so in legislation-which they did not. The effect
of incorporating weighted counts into the resident limit regulations
essentially dictates to hospitals the number and types of
residents that get reimbursed, since hospitals with residents
in subspecialties ("weighted" residents) are most affected
by CMS' policy. Congress expressly rejected this type of workforce
tampering when they "emphatically" noted in the Conference
Report accompanying the BBA that workforce matters "should
remain within each facility, which is best able to respond
to clinical needs and opportunities." (BBA Conference Report
at S-203). Accordingly, we believe CMS should reconsider its
position on this issue and withdraw its current policy.
3. Resident Limit Affiliations with New Urban Teaching
Hospitals
Under current CMS policy, urban nonteaching hospitals that
decide to become teaching hospitals are not permitted to enter
into resident limit affiliation agreements with other teaching
hospitals. According to the May 12, 1998 Federal Register,
CMS is concerned that, without the prohibition, existing teaching
hospitals might encourage nonteaching hospitals to start residency
programs and then essentially move the program to their own
institution through the use of a resident limit affiliation
agreement (see 63 Fed. Reg. at 26333). We believe the Agency's
concern is purely speculative and the prohibition should be
eliminated.
Establishing new residency programs is not a trivial endeavor.
In addition to the effort and time associated with complying
with Medicare requirements, the procedure entails an intense,
time-consuming process associated with stringent accreditation
rules. The likelihood that a nonteaching hospital would undergo
these efforts to be a conduit so that an established teaching
hospital could circumvent the resident limit rules seems extremely
unlikely. A more likely scenario is one in which a new teaching
hospital with a fledgling graduate medical education program
enters into rotation agreements with another teaching hospital.
We believe these two hospitals should be permitted to enter
into aggregation agreements to reflect the rotation schedules.
4. Educating Residents From VA Hospitals
The current regulations permit a temporary adjustment to
the resident limit of a hospital that takes on and completes
the training of residents from a Department of Veterans Affairs
(VA) hospital but only under very specific conditions, including
that the resident transfer occurred between January 1, 1997
and July 31, 1998 and an accreditation issue was involved.
We believe this regulation correctly recognizes the difficulties
that can arise when there are issues with resident education
at VA institutions. However, we believe it is too narrowly
construed. The VA system is restructuring many of its facilities,
which may involve residency program closures. We think that
in these situations, temporary resident limit adjustments
should be provided to non-VA teaching hospitals that take
on and complete the education of residents affected by these
decisions. More broadly, we believe that consideration should
be given to making these adjustments permanent since the total
number of resident slots is not increasing but would only
be redistributed.
5. Impact of Program Closures and the Intern/Resident-to-Bed
Ratio (IRB) Limit
On a more technical issue, we believe that the regulations
concerning residency program closures and the limit on the
IRB ratio should be changed so that teaching hospitals that
take on displaced residents are not financially penalized.
In the August 1, 2001 Federal Register, CMS recognized the
need for a temporary adjustment to resident limits for hospitals
that take on and complete the training of residents that come
from residency programs that are closing. The Agency also
recognized that these residents should not be reflected in
the calculation of the three-year rolling average. However,
while acknowledging the issue, CMS decided to defer a change
in the regulations that would essentially exclude counting
the displaced residents in the calculation of the IRB limit
(see 66 Fed. Reg. at 39901).
We believe changing the regulations to exclude displaced
residents in the calculation of the IRB limit is appropriate
and would be consistent with CMS' other decisions in this
area. The discussion in the preamble states that while hospitals
that take on these residents would be disadvantaged in the
first year, they would have the benefit of a higher IRB limit
after the last year of training the displaced residents. If
we understand CMS' position correctly, all hospitals would
be disadvantaged in the first year, while only hospitals that
are over the resident limits will receive the "benefit" in
the last year.
6. Deadline for Submitting Resident Limit Affiliation
Agreements
Currently, resident limit affiliation agreements must specify
adjustments based on a 12-month basis, from July 1 to June
30 of each year. It appears that this date was chosen because
a number of residency training programs follow a July 1 to
June 30 educational year (see 63 Fed. Reg. at 26338).
We believe that the requirement should be changed so that
hospitals may execute resident limit affiliation agreements
at any time during the year. While many residency programs
follow a July 1 to June 30 academic cycle, a number do not.
In addition, for reimbursement purposes, the more important
time period relates to a hospital's cost reporting period.
Regardless of the date it is executed, the resident count
set forth in the agreement must be reconciled with the hospital's
cost reporting period. Establishing a uniform July 1 date
requirement does not address this situation because many hospitals
have different cost reporting periods. Consequently, we believe
the permitting hospitals execute agreements at any time would
reduce hospitals' administrative burdens without imposing
much, if any, additional hardship on Medicare program administration.
III. CMS Should Modify Its Regulations on Resident Rotations
at Nonhospital Sites
Medicare makes IME and DGME payments associated with residents
educated in nonhospital sites so long as the teaching hospital
incurs "all or substantially all" of the costs of the residency
training in those sites (and the resident limit is not exceeded).
Prior to January 1, 1999 this requirement was met if the teaching
hospital paid the residents' stipends and benefits during
the residents' educational time at the nonhospital site. CMS
since expanded the requirement such that the "all or substantially
all" criterion now includes, where applicable, the costs of
teaching physician supervisory costs at the nonhospital site.
The AAMC strongly supports ambulatory training in nonhospital
sites. However, we are concerned that requiring hospitals
to demonstrate that they are incurring teaching physician
supervisory costs in order to receive Medicare teaching reimbursements
may result in fewer residents training in these sites. Moreover,
there continues to be confusion regarding CMS' policy on physicians
who volunteer to supervise residents.
In the FY 2000 PPS Final Rule (July 30, 1999), CMS stated
that "[h]ospitals may receive payment for the costs of training
physicians in the nonhospital site even though the hospital
might not be incurring any costs for supervisory physician
activities." (64 Fed. Reg. at 41518). Medicare Program Memorandum
A-98-44 (December, 1998) also recognizes that physicians may
volunteer their time spent in supervisory activities.
Despite these pronouncements, communications with CMS staff
suggest that there continues to be ambiguity on CMS' volunteer
physician policy. The AAMC respectfully requests that CMS
explicitly state that, so long as the other criteria are met,
hospitals may receive Medicare payments for residents training
in nonhospital if the written agreement, which is signed by
both the hospital and nonhospital site, indicate that the
supervisory physician and associated site has agreed to volunteer
his/her time supervising activities.
Thank you very much for considering our requests. If have
any questions about our views, please feel free to contact
Robert Dickler, Senior Vice President or Karen Fisher, Associate
Vice President in the AAMC's Division of Health Care Affairs,
both of whom may be reached at 202-828-0490. We appreciate
and value the collegial relationship with the CMS staff who
oversee Medicare payments to teaching hospitals and we look
forward to working with you and them in the future.
Sincerely,
Jordan J. Cohen, M.D.
| CC: |
Tom Gustafson, CMS
Tzvi Hefter, CMS
Robert Dickler, AAMC
Karen Fisher, AAMC |
1. Medicare also pays its share of
the higher patient care costs at teaching hospitals through
the IME payment adjustment. [Back]
2. "New" refers both to programs
in existing specialties that a hospital currently does not
have as well as programs in newly emerging specialties that
are recognized by the ACGME, the American Board of Medical
Specialties, or other accreditation bodies. [Back]
|
