AAMC Comment Letter on Children's
Hospital Graduate Medical Education Payment Program (CHGME)
Federal Register Notice
July 19, 2000
F. Lawrence Clare, M.D.
Division of Medicine
Bureau of Health Professions
Health Resources and Services Administration
Room 9A-21
Parklawn Building
5600 Fishers Lane
Rockville, MD 20857
Re: Children's Hospitals Graduate Medical Education Payment
Program
Dear Dr. Clare:
The Association of American Medical Colleges (AAMC) welcomes
this opportunity to comment on the Health Resources and Services
Administration's (HRSA or the Agency) notice entitled "Children's
Hospitals Graduate Medical Education Payment Program: Proposed
Eligibility and Funding Criteria and List of Eligible Hospitals,"
65 Fed. Reg. 37985 (June 19, 2000). AAMC represents over 400
major teaching hospitals, including both free-standing children's
hospitals and general acute hospitals with large pediatric
divisions. The Association also represents all 125 accredited
U.S. medical schools; 86 professional and academic societies;
and the nation's medical students and residents.
The Children's Hospitals Graduate Medical Education (CHGME)
payment program was authorized by the Healthcare Research
and Quality Act of 1999 (Public Law 106-129) for two years,
federal fiscal years (FY) 2000 and 2001. Under the FY 2000
appropriations law, $40 million has been appropriated for
this program. The June 19 notice sets forth proposed eligibility
criteria, funding methodology, and performance measures for
the CHGME program.
I. The CHGME Payment Program Should Emulate Medicare Principles
and Methodologies
To the extent permitted by law, the CHGME program should
follow Medicare's principles and methodologies in distributing
payments to children's teaching hospitals.
For nearly 20 years, since 1983, the Medicare program has
made two distinctive payments to teaching hospitals. Direct
graduate medical education (DGME) payments cover Medicare's
share of the costs associated with physician graduate medical
education, including residents' stipends and benefits, faculty
supervisory costs, and allocated overhead costs. Indirect
medical education (IME) payments are intended to cover teaching
hospitals' higher patient care costs that are due to treating
a more complex and sicker patient population, providing an
advanced scope of services, and the inherent inefficiencies
associated with teaching residents. As with DGME payments,
IME payments reflect only Medicare's share of these higher
costs.
The Medicare methodologies are a time-tested means of distributing
monies for educating residents and compensating for the higher
patient care costs of teaching hospitals. The authorizing
legislation for the CHGME program recognized the value of
this history and methodology through numerous references to
the Medicare statute and regulations. Adopting Medicare principles
and methodologies would provide a firm foundation for the
CHGME program, as well as carry out the intent of the authorizing
legislation.
In addition, almost all children's hospitals receive some
level of Medicare DGME funding. Adopting Medicare principles
and methodologies for the CHGME program would reduce the confusion
and the administrative burden that would result from having
vastly disparate DGME payment programs.
A. Defining the Eligibility Criteria
The notice sets forth the criteria that HRSA proposes using
to determine which children's hospitals will be eligible to
apply for the CHGME program. The third criterion refers to
a hospital's participation in an "accredited GME"
program. We believe this criterion should be changed to refer
to "approved medical residency" programs, which
would align it with the Medicare criterion.
B. Determining Resident Counts in the CHGME Program
The June 19 notice implements the legislative mandate to
use the Medicare methodology for counting residents for DGME
payments, but is silent on the methodology for counting residents
for IME payments. The AAMC supports HRSA's proposal to follow
the legislative mandate for DGME payments and incorporate
Medicare's resident count methodology for IME payment purposes.
The June 19 notice also requests comments on whether the
CHGME program should require that participating children's
hospitals submit resident count data in the format specified
by the Medicare Intern and Resident Information System (IRIS)
system. While hospitals have encountered difficulties using
the current IRIS diskette, the discussion of which is beyond
the scope of this letter, the AAMC believes that it is the
best mechanism currently available for verifying resident
counts. Moreover, given its applicability in the Medicare
program, IRIS would help to ensure consistency in resident
counts across both GME programs. It also would make it possible
to continue to use IRIS to check for residents who have been
counted simultaneously by two or more hospitals, which is
always a possibility when residents rotate among hospital
sites.
Several of the examples contained under the Notice's section
on determining resident counts appear to be technical errors
or require clarification. As an overriding issue, it is important
to recognize that whether a teaching hospital is a "major
participating institution" should be irrelevant for the
CHGME program, as it is irrelevant for Medicare. A hospital
is entitled to DGME payments under Medicare so long as the
resident is in an approved medical residency program and is
either training in that hospital or is training in a nonhospital
site in which the hospital is incurring "all or substantially
all" of the training costs (see 42 C.F.R. § 413.86(f)(3)).
Example B states that a children's hospital had 24 residents
in 1996 and added two additional first residents in 1999.
The example concludes that the 1999 resident count is 25,
rather than 26, because the two residents accounted for "6
months each." However, the previous sentence stated that
the two residents spent all of their time at the children's
hospital. Despite this confusion, it appears that the example's
resultant count of 24 is correct since the resident count
is capped at the value in 1996, which was 24.
Example D indicates that an estimate of a children's hospital
1996 could be obtained by reducing the resident count in 1999
by 1.5 because during 1997 three residents spent 3 months
each at continuity clinics. Under Medicare rules, as discussed
above, if the hospital is incurring all or substantially all
of the training costs for the three residents in the continuity
clinic, those residents are permitted to be counted by the
hospital. In such a case, the hospital's 1999 resident count
would not be reduced.
C. The IME Methodology
The authorizing legislation gave HRSA significant discretion
in determining the IME payment methodology. The June 19 notice
does not provide a specific methodology, but indicates that
such a methodology will incorporate a resident count, a case
mix measure, a teaching intensity adjustment, and volume (65
Fed. Reg. at 37990).
As discussed above, the AAMC believes that the IME methodology
under the CHGME program should follow Medicare inpatient IME
policy. This could be accomplished by replicating Medicare's
IME formula with only those modifications, if any, that would
be necessary because of the fixed pool of IME funds available
under the CHGME program. With respect to case mix, we believe
it should be determined by treating children's hospitals'
discharges as if they were Medicare cases and assigning relative
weights based on the Medicare diagnosis-related group (DRG)
system. Teaching intensity should be measured according to
the ratio of residents to beds.
We also believe that the mechanism for distributing CHGME
IME payments should be tied to the case mix index and case
volume, and not resident counts. Resident counts affect the
level of teaching intensity, which is an important component
of an IME methodology. However, the purpose of IME payments
is to compensate for higher patient care costs, not resident
counts. Under Medicare, IME payments reflect a percentage
add-on to each Medicare case. While the fixed pool of IME
funds under the CHGME program may preclude replicating this
mechanism exactly, the IME methodology should be comparable
in concept.
II. HRSA Should Re-Evaluate its Proposed Performance Measures
According to the June 19 notice, the CHGME program is subject
to the Government Performance and Results Act of 1993 (GPRA).
The GPRA requires HRSA to prepare an annual performance plan
covering each program activity set forth in HRSA's budget.
As part of this process, HRSA asserts that to fulfill its
requirement to evaluate the effectiveness of the program,
it must obtain performance information from Children's hospitals
that receive funds under the CHGME. HRSA's proposed performance
goals are divided into two categories: "Eliminating Barriers
to Care," and "Improve Public Health and Health
Care Systems." The notice sets forth specific performance
goals under each of these categories and proposed data requirements.
The purpose of the CHGME program is to provide Federal funding
to children's hospitals that educate residents. The AAMC believes
strongly that the resident counts obtained by HRSA for purposes
of making DGME and IME payments are sufficient and appropriate,
in and of themselves, for HRSA to demonstrate the effectiveness
of the CHGME program.
If HRSA determines that additional performance measures must
be imposed, the AAMC believes that data requirements should
be tailored to reflect the purpose of the CHGME program. In
addition, where possible, information should be obtained from
existing data sources to minimize reporting and administrative
burdens on participating hospitals.
We believe that many of the proposed data requirements do
not reflect the purpose of the CHGME program and would be
difficult and burdensome to obtain. Specifically, under the
proposed goal of maintaining the number of FTE residents,
we believe total resident counts should be sufficient to determine
whether this goal is being met. Quantifying rotations of residents
across hospitals appears to be unrelated to this goal. Moreover,
given that Medicare does not factor sponsorship into its payment
determinations, requesting hospitals to distinguish resident
counts by sponsorship would be unduly burdensome and would
not provide relevant information.
We also believe that monitoring the proportion of uncompensated
care and Medicaid patients is beyond the scope of the CHGME
program. Indeed, many children's hospitals perform valuable
societal functions by providing care to these populations.
However, this mission is distinct from the purpose of the
CHGME program, which is to help finance the education of residents
and the additional patient care costs incurred by children's
hospitals that operate residency training programs.
Conclusion
The June 19 notice contains a number of additional issues
that will be addressed in comments submitted by the National
Association of Children's Hospitals. We also understand that
an opportunity for comment will be provided when HRSA determines
the methodology for an IME adjustment.
Thank you for this opportunity to present our views. We would
be happy to work with HRSA on the issues discussed above or
other topics that involve the academic health care community.
If you have questions concerning these comments, please feel
free to call Robert Dickler, Senior Vice President of the
Association, or Karen Fisher, Associate Vice President, both
of whom may be reached at (202) 828-0490.
Sincerely,
Jordan J. Cohen, M.D.
|
