Comment Letter to HCFA on
Fiscal Year 1999 Medicare Prospective Payment System Proposed
Rule
July 7, 1998
Nancy-Ann Min DeParle, Administrator
Health Care Financing Administration
Hubert H. Humphrey Building
Room 309-G
200 Independence Avenue, SW
Washington, DC 20201
Reference: HCFA-1003-P
Dear Administrator Min DeParle:
The Association of American Medical Colleges (AAMC or the
Association) welcomes this opportunity to comment on the Health
Care Financing Administration's (HCFA) proposed rule entitled
Medicare Program; Changes to the Hospital Inpatient Prospective
Payment Systems and Fiscal Year 1999 Rates. 63 Fed. Reg. 25,575
(May 8, 1999) (proposed rule). The AAMC represents all 125
accredited U.S. medical schools; approximately 400 major teaching
hospitals, including 75 Veterans Affairs medical centers;
86 professional and academic societies; and the nation's medical
students and residents.
Our comments focus primarily on the proposal to revise the
definition of the phrase "all or substantially all"
used in conjunction with the level of residents' costs that
a hospital must demonstrate that it incurs in order to receive
Medicare indirect medical education (IME) and direct graduate
medical education (DGME) reimbursements for residents training
in nonhospital sites. We also will comment briefly on direct
teaching payments for non-hospital providers, and the wage
index for hospital inpatient services used to adjust payments
under the inpatient hospital prospective payment system (PPS).
I. Proposal to Revise the Definition of "All or Substantially
All" Resident Training Costs
Under Medicare, hospitals are entitled to receive IME and
DGME payments associated with residents training in nonhospital
sites if they incur "all or substantially all" of
the residents' training costs. Currently, this criterion is
met if a hospital pays the residents' salaries and fringe
benefits.
The proposed rule would expand the definition of "all
or substantially all" to include "the portion of
costs of the teaching physicians' salaries and fringe benefits
that are [sic] related to the time spent in teaching and supervision
of residents." The new definition would also include,
where applicable, residents' travel and lodging expenses.
In addition to teaching hospitals, this new definition would
apply to entities seeking direct teaching payments as qualified
nonhospital providers, as defined by the proposed rule. The
proposed change is not mandated by the Balanced Budget Act
of 1997 nor any other law. The preamble to the proposed rule
states that HCFA is proposing this change because of an analysis
suggesting that resident salaries and benefits are slightly
less than half of hospitals' total graduate medical education
(GME) costs.
The AAMC believes that HCFA should retain the current definition
and withdraw this proposal. The current definition has proven
over time to be a straightforward mechanism that easily identifies
the entity that should receive teaching payments. The AAMC
believes the proposed definition is administratively infeasible,
has serious potential to impede the training of residents
in ambulatory sites, and is supported by a rationale that
does not outweigh the definition's corresponding deficiencies.
A. The current definition is logical, straightforward
and appropriate.
The current standard of requiring a hospital to document
that it pays residents' salaries and benefits in order to
be eligible for DGME and, since October 1, 1997, IME payments
when residents are in a non-hospital site, has proven over
time to be administratively straightforward. While additional
teaching costs are incurred in these sites, these costs are
difficult to isolate and quantify. By contrast, residents'
salaries and benefits are easy to identify and their administration
and record keeping can be monitored uniformly across the graduate
medical education community. In addition, resident salaries
and benefits are the key components of resident training costs.
In assuming responsibility for these costs, the teaching hospital
a'priori assumes responsibility for assuring that all residents
are provided appropriate educational environments, supervision,
and support for their training. Accordingly, the entity that
incurs these fundamental costs--whether it be the hospital
or a qualified nonhospital provider--is the entity that should
receive the teaching payments.
B. Implementing this proposal is not administratively
feasible.
In several places in the overall proposed rule's preamble,
HCFA rejects potential policy changes because they are deemed
to be not "administratively feasible." The AAMC
believes it is not administratively feasible to implement
a policy requiring hospitals to demonstrate that they have
incurred physician supervisory and teaching costs, as well
as resident travel and lodging costs, in order to receive
teaching payments for residents training in nonhospital sites,
many of which are community physicians' offices.
We recognize that there are costs associated with supervisory
and teaching responsibilities. However, unlike resident salaries
and fringe benefits, these costs are often difficult to identify
and quantify. They may be the responsibility of the hospital,
the nonhospital site, an individual physician practicing in
the community, an affiliated medical school, a combination
of these entities, or a separate entity altogether, depending
upon the specific arrangement. These costs may be covered
through related-party agreements or comprehensive shared services
arrangements. They may be compensated in a "lump sum"
that combines them with costs associated with training undergraduate
medical students. They may involve "in kind" remunerations,
such as providing continuing medical education or other benefits.
Finally, in some cases, some components of these costs may
be provided through voluntary contributions.
The proposed rule envisions a framework that requires some
type of identified and documented transaction that quantifies
these costs as well as identifies the entity bearing them,
even though these types of financial transactions may not
exist. It would be especially difficult to attempt to quantify
supervisory costs of community physicians in private practice.
Among other difficulties, there is no established methodology
for defining or quantifying supervisory costs. In addition,
even if supervisory and teaching costs could somehow be quantified,
they could vary depending on the specialty area and the year
of residency training. For example, residents in their first
year of training require more supervision than residents in
their third year, which would, necessarily, involve greater
costs. Keeping track of, and subsequently auditing, these
costs would require detailed information on residency training
years and specific specialties. A cost accounting infrastructure
capable of documenting these costs is virtually unimaginable.
At a minimum, current longstanding infrastructures would be
greatly compromised.
C. The proposal has the potential to seriously impede
resident training in nonhospital sites.
The Association believes that the proposed change could undermine
the current relationships between hospitals and nonhospital
providers and impede resident training in nonhospital settings.
Resident training occurring in community-based physician practice
offices is of particular concern. These offices currently
constitute a substantial source of ambulatory training sites
and their use will grow as curriculum and accreditation entities
add additional ambulatory care training to their requirements.
The administrative burden that would be imposed on physicians
under HCFA's proposal could seriously undermine the desire
of these community-based physicians to educate residents.
Many do not have the accounting and administrative infrastructures
that would be required by the proposed rule. As one example,
the proposed rule would require physicians to document the
precise number of hours they spend teaching or supervising.
These demands place an additional burden on physicians whose
current workloads are increasingly dominated by documentation
requirements. It is likely that such demands would cause physicians
to reconsider their decision to supervise and teach residents
in their offices and decide that the intrinsic rewards of
educating residents do not compensate for these additional
burdens.
The impediments arising from the new proposed standards are
even more disconcerting because the current standard has helped
to facilitate resident training in nonhospital sites. While
we are currently in the process of collecting data on the
precise extent to which residents are training in nonhospital
sites, there is strong anecdotal evidence that the number
of residents training in these sites is increasing. Given
HCFA's strong desire to encourage ambulatory residency training,
it is important that any changes be tested for the likelihood
that they will promote expanded ambulatory graduate medical
education. We believe the proposed change in the definition
does not pass this test.
D. The rationale for the proposal is insufficient to merit
a change in current policy.
The preamble to the proposed rule provides no policy rationale
for changing the current standard. Instead, it states that
a change is necessary because a HCFA analysis indicated that
resident salaries' and benefits were slightly less than half
of total hospital residency training costs. On a technical
level, we believe that because the proposed policy involves
nonhospital sites, it would be more appropriate to analyze
resident salaries and benefits as a share of overall training
costs at the nonhospital site, rather than overall hospital
costs. While these data are not yet available, we believe
it most likely would show that resident salaries and benefits
are a substantial component of overall training costs in these
sites. The preamble also recognizes that the components of
GME costs, including residents' salaries and supervisory costs,
would likely "constitute a different proportion of the
total GME costs in the nonhospital setting as compared with
the hospital setting." (63 Fed. Reg. at 25,597).
In addition, the costs HCFA is analyzing have no bearing
on teaching hospitals' direct GME payments, which are per
resident payment amounts based on hospital costs occurring
in 1984. As HCFA is aware, the level and types of costs that
were permitted to be recognized in 1984 varied across areas
and institutions. It is unclear whether all nonhospital supervisory
and teaching costs incurred by hospitals were recognized in
the 1984 base year amounts. Given that contemporary costs
have little relationship to the direct payments that hospitals
receive, it is equally unclear to what extent nonhospital
costs are reported on hospital cost reports. Consequently,
the data that HCFA analyzed may not reflect current environments.
We recognize that a further policy rationale for this change,
even though it was not articulated in the preamble, may be
the concern voiced by some non-hospital providers that they
are not being provided with sufficient resources to support
their activities in graduate medical education. The AAMC is
sympathetic to this concern but also recognizes that, at best,
the evidence for this concern is spotty and anecdotal. It
is also compromised, in many situations, by an inadequate
understanding of the Medicare methodology and proportional
payment system for direct graduate medical education and the
different methodology and rationale for indirect medical education
payments. It is our belief that a change in the definition
of ‘all or substantially all' of the costs will not resolve
these concerns and, as stated above, may seriously compromise
the existing and developing relationships between hospitals
and non-hospital graduate medical education sites. In the
end, these relationships must, and should be, voluntary and
it is up to the parties to define the appropriate parameters
of their relationships, including how costs beyond the resident
stipend and benefits should be accommodated.
E. The definition of "all or substantially all"
should not include residents' travel and lodging costs.
The proposed rule would define "all or substantially
all costs" to include "residents' travel and lodging
costs where applicable"(emphasis added). The preamble
provides no rationale for this change.
The Association believes these added criteria impose significant
additional reporting burdens with no offsetting benefits.
As discussed previously, the issue is not whether these costs
occur, but documenting a financial transaction that the costs
were incurred by the hospital, or qualified nonhospital site.
In addition, the phrase "where applicable" is vague,
requiring additional definitional language (related to distance,
means of travel, among others) if entities are to understand
their reporting obligations.
F. As an alternative to the proposed rule, HCFA could
initiate a demonstration project to learn more about supervisory
costs in nonhospital sites.
As discussed above, the AAMC believes that resident salaries
and benefits is a workable and fair standard for determining
which entity should receive teaching-related payments. If,
however, HCFA continues to believe that supervisory and teaching
costs should play a role in this determination, we recommend
that the Agency initiate a demonstration project to analyze
and test the merits of including these costs in the definition
of "all or substantially all" before making any
changes on a nationwide basis. In addition, a demonstration
project could shed some light on whether such a change would
encourage or discourage training in nonhospital sites.
While we recognize that HCFA will be collecting some of these
data because supervisory costs will be reimbursed under the
proposed methodology for making direct teaching payments to
Federally qualified health clinics and rural health clinics,
it is equally important to obtain and analyze data from other
nonhospital sites, such as physician offices. The AAMC would
be happy to work with HCFA to help structure a demonstration
project.
II. Teaching Payments to Nonhospital Providers
The AAMC believes that nonhospital entities should receive
direct Medicare payments for residency training costs so long
as they demonstrate they incur all or substantially all of
resident costs, defined as the residents' salaries and benefits--the
same criterion currently applied to hospitals. As discussed
above, we believe that reporting of additional costs,
such as supervisory and teaching costs, should not be a requirement
for these entities to receive direct teaching payments.
We believe that qualified nonhospital providers, including
Medicare+Choice plans, that meet the criteria for receiving
direct payments should receive payments associated with their
GME overhead costs. Over time, the data HCFA will have collected
on the direct costs of GME for these entities may necessitate
changes to the criteria and methodology contained in the proposed
rule. We also believe that HCFA should initiate studies to
address some of the methodological concerns and limitations
identified in discussion of the payment methodology for non-hospital
providers. The AAMC also would be happy to work with HCFA
on these studies.
III. Modifications to the PPS Hospital Wage Index
HCFA is proposing to include contract physician Part A costs
in the Medicare hospital wage index. The AAMC believes this
is an equitable policy for reflecting the wage costs that
hospitals face.
The preamble to the proposed rule also indicates that HCFA
will analyze the fiscal year 1996 wage data associated with
residents and certified registered nurse anesthetists to determine
whether it will propose any changes to the wage index for
fiscal year 2000.
HCFA historically has required complete and accurate data
before making any changes involving the wage index. The AAMC
commends this policy because the Medicare hospital wage index
is an important and substantial component of PPS hospital
payments. The AAMC believes that in making any future changes
to the wage index, HCFA's decision-making process should be
guided by its standard of relying on complete and accurate
data. To the extent any changes would result in significant
fluctuations in Medicare payments, the Association believes
these changes should be introduced over a period of time.
IV. Technical Comments Related to Proposed Regulatory
Language
When reviewing the proposed language for the regulations,
we identified the following technical inconsistencies.
A. 413.85(h)(2)(iii)(A)--reference to "paragraph
(f)(1)(ii) of this section"
This reference does not exist. To be consistent with allowable
direct GME costs as defined for RHCs and FQHCs, it appears
the appropriate reference should be 405.2468(f)(5)
B. 413.86(f)--several technical anomalies
"(2)" should be "(ii)"
"(3)" should be "(iii)"
"(4)" should be "(3)" and the phrase
"On or after July 1, 1987 and" appears to be a typographical
error that should be deleted.
C. 413.86(f)(4)(ii)(C)--cost reports for nonhospital sites
The reference to nonhospital sites reporting direct graduate
medical education costs on their cost reports should be clarified
to account for nonhospital sites, such as physicians' offices,
that do not have Medicare cost reports.
V. Conclusion
The AAMC commends HCFA's efforts to implement regulations
that encourage resident training in a variety of settings,
including ambulatory sites. We are greatly concerned, however,
that the proposal to revise the definition of "all or
substantially all" resident training costs has the potential
to undermine these intentions. We would welcome the opportunity
to work with HCFA on this issue.
If you have questions regarding our comments, please contact
Robert Dickler, Senior Vice President of the Association at
(202) 828-0491, rdickler@aamc.org,
or Karen Fisher, Assistant Vice President at (202) 862-6140,
kfisher@aamc.org.
Sincerely,
Jordan J. Cohen, M.D.
|
