Comment Letter on Fiscal Year 2002 Medicare Prospective Payment System Proposed Rule
July 3, 2001
Thomas A. Scully, Administrator
Centers for Medicare & Medicaid Services
Room 443-G Hubert H. Humphrey Building
200 Independence Ave, SW
Washington, DC 20201
Attention: File Code HCFA-1158-P
Dear Administrator Scully:
The Association of American Medical Colleges (AAMC) welcomes
this opportunity to comment on the Centers for Medicare &
Medicaid Services (CMS or the Agency) proposed rule entitled
"Medicare Program; Changes to the Hospital Inpatient Prospective
Payment Systems (PPS) and Fiscal Year 2002 Rates," 66
Fed. Reg. 22646 (May 4, 2001). The AAMC represents over 400
major teaching hospitals and health systems; all 125 accredited
U.S. medical schools; 86 professional and academic societies;
and the nation's medical students and residents.
This letter addresses three issues: Changes to direct graduate
medical education (DGME) and indirect medical education (IME)
payment policies, Medicare payments for new technologies,
and collection of occupational mix data.
Medicare IME and DGME Payments
The proposed rule sets forth several clarifications and technical
methodology changes regarding the calculation of IME and DGME
payments. We very much appreciate the inclusion of examples
in the proposed rule preamble. Calculation of IME and DGME
payments has become much more complex since the introduction
of resident limits in the Balanced Budget Act of 1997 (BBA).
Examples provide important details that are sometimes difficult
to convey in descriptive text.
We also appreciate and support the proposal to permit temporary
upward adjustments for hospitals that take on and complete
the training of residents from residency programs that have
closed. To qualify for such an adjustment, the hospital in
which the program closure has occurred must agree to temporarily
reduce its resident limit until the residents complete their
training at the other hospital(s). We believe this approach
provides needed flexibility in those unfortunate situations
that require hospitals to close residency programs before
all of their residents can complete their training while,
at the same time, retaining the BBA's intent not to increase
the overall number of residents being trained.
We would like to provide additional views on two other areas
discussed in the proposed rule: DGME payment corridors, and
the counting of resident research time.
DGME Payment Corridors
We have no specific comments on the methodology for implementing
the DGME payment corridors; however, we believe some clarification
is needed regarding language contained in the preamble. Throughout
the preamble, there are references stating that the corridor
methodology is applicable for "cost reporting periods beginning
on or after October 1, 2000 and on or before September 30,
2005." We do not understand the basis for the September 30,
2005 end point. We believe that the per resident amount changes
authorized in the law were meant to be permanent in nature
and can find no basis to support the preamble reference. It
is important that CMS clarify this situation in the final
rule.
We also would very much appreciate if CMS would publish in
the final rule the CPI-U factors that must be used to update
the 1997 national average per resident amount to the current
payment period.
Counting of Resident Research Time for IME and DGME Payments
The proposed rule preamble includes a discussion about CMS'
policy regarding resident time spent performing research for
purposes of calculating the full-time equivalent (FTE) resident
count used for DGME and IME payments. For DGME, CMS' current
policy is that the time residents spend performing research
within the hospital complex as part of an approved residency
program may be counted for DGME payments. CMS does not plan
to make any modifications to its current DGME regulations
on this issue.
By contrast, HCFA states that for purposes of counting resident
time for IME payments, only research time that is associated
with delivering patient care is countable. HCFA proposes to
modify 42 C.F.R. §412.105(f)(1)(iii) to add a new subsection
(b) that states for IME purposes, "the time spent by a resident
in research that is not associated with the treatment or diagnosis
of a particular patient of the hospital is not countable."
We welcome the clarification regarding the counting of research
time for purposes of DGME payments. Our understanding is that
there are fiscal intermediaries that have excluded "bench"
research from the DGME calculations and the preamble clarification
will provide the necessary documentation to make sure this
time is included.
However, we have concerns, and believe additional information
is needed, regarding the rationale behind the policy to exclude
certain types of research for purposes of IME payments. On
a policy level, we believe more explanation is needed regarding
the origins of the IME adjustment and the relationship with
resident time.
The IME adjustment is intended to compensate for the higher
patient care costs that occur at teaching hospitals. Our understanding
of the development of the adjustment is that statistical analyses
showed that the use of an intern/resident-to-bed ratio (IRB)
was (and continues to be) the best proxy for the patient care
cost differences between teaching and non-teaching hospitals.
Given that the IRB is only a proxy, the relevance of a requirement
that residents themselves must be engaged in activities related
to patient care in order for their training time to be counted
in the IRB is unclear. It might also be informative to learn
which resident count was used when the initial IME statistical
analyses were computed in 1983. If all resident time spent
in approved programs was reflected in those analyses, it is
unclear why a change should be made.
Moreover, at a practical level, it seems that the proposed
regulation is unduly burdensome and introduces unnecessary
complexity into the IME calculations. First, CMS' position
requires hospitals to maintain different resident counts for
DGME and IME based on research activity or rotations. While
hospitals with distinct-part units must distinguish between
IME and DGME resident counts, that distinction is much more
administratively straightforward to maintain than a distinction
that depends on distinguishing between different types of
research activities. Second, the regulatory language itself
is ambiguous. For example, what is meant by the regulatory
language that says the research must be related "to a particular
patient?"
A number of residency programs require some type of research
activity. This is an important component of the educational
process. Medicare has consistently stated that to be counted
for purposes of DGME and IME payments, a resident must be
in an approved program. We believe this requirement has proven
to be workable and appropriate. We urge CMS not to make an
already complex methodology more complex and rescind its proposal
in the final rule.
Medicare Payments For New Technologies
The Medicare, Medicaid, and SCHIP Benefits Improvement and
Protection Act of 2000 (BIPA) mandated CMS to develop a process
to a) more rapidly incorporate new medical services and technologies
into the diagnosis-related groups (DRGs), and b) ensure adequate
payment for new medical services and technologies under the
Medicare inpatient PPS. We would like to comment on the regulatory
proposals to implement both of these mandates.
Incorporating New Medical Services and Technologies into
the Coding System
We support CMS' proposal to shorten the time frame for implementing
new ICD-9-CM procedure codes into the coding system. Such
changes will hasten the incorporation of new technologies
into the inpatient PPS, which will hasten adequate payments
for DRGs that include these new services.
Additional Payments for New Technologies
BIPA requires CMS to establish a mechanism to recognize the
costs of new services and technologies under the inpatient
PPS. BIPA also requires that any additional payments be budget
neutral; that is they must be financed by reducing the DRG
base standardized amount. The proposed rule sets forth a process
for complying with the BIPA requirements. The process involves
a) setting forth criteria for determining the technologies
that would qualify for additional payments, and b) calculating
the additional payment.
Identifying Applicable New Technologies
The proposed rule sets forth three criteria for a technology
to qualify for an additional payment. The technology must:
- Be a new, rather than existing, technology;
- Represent "an advance that substantially improves, relative
to technologies previously available, the diagnosis or treatment
of Medicare beneficiaries" (proposed 42 C.F.R. §412.87(b)(1));
and
- Result in standardized charges for a case that are at
least one standard deviation beyond the mean standardized
charge for all cases in the DRG to which the new technology
is assigned.
CMS plans to use a panel of "Federal clinical and other experts,
supplemented as appropriate with outside expertise" to determine
whether a technology meets the "substantial improvement" criterion.
(66 Fed. Reg. at 22694).
The AAMC agrees that the purpose of this provision is to
provide adequate payments for cutting-edge technologies that
enhance patients' quality of care. We have some reservation,
however, about the "substantial improvement" criterion because
of the ambiguity in its meaning and application. We urge CMS
to investigate other options, including criteria used by other
health care regulatory bodies, such as the Food and Drug Administration,
that might involve more objective criteria that could be used
to identify technologies that should qualify for additional
payments.
In the absence of any acceptable objective criteria, we support
the proposal to convene an expert panel-properly constructed
and with an appropriate charge-to identify the relevant new
technologies. However, our support is equivocal because the
proposed rule preamble provides little detail as to the composition
of the panel, its charge, or procedures. Nonetheless, we believe
that in certain cases, it will be fairly obvious to Federal
clinical experts that a new technology meets the eligibility
criteria. In other cases, the determination may require more
thorough review and expert input. In these latter cases, we
believe that the expertise of practicing clinicians from within
the provider community-particularly teaching hospitals-could
be valuable in ensuring appropriate decisions.
If experience demonstrates that the expert panel option is
unsatisfactory, we believe it would be imperative for CMS
to expeditiously propose, and/or implement on an interim final
basis, an alternative option.
We also have some concerns about the third criterion, which
requires that the per case charges be at least one standard
deviation beyond the mean standardized charge for all cases
in the DRG to which the new technology is assigned. It is
not altogether clear that "per case" costs is the appropriate
measure for determining new technology add-on payment eligibility,
or whether it is more appropriate that the costs of the technology
itself should be the locus of the determination.
Notwithstanding this concern, the proposed rule preamble
does not provide enough information to determine whether the
"one standard" deviation is feasible, let alone appropriate.
Standard deviations for per case charges can vary widely across
DRGs, particularly in high relative-weight DRGs (which are
likely candidates for the use of new technologies). In certain
cases, the standard deviation standard may be appropriate,
but in other cases, the standard deviation may be so large
that it will exclude eligibility for technologies that otherwise
merit inclusion. Before finalizing this criterion, CMS should
provide additional information about the range of DRG standard
deviations, particularly for those DRGs that involve the use
of technology.
Calculating the Additional Payment for New Technologies
The methodology proposed by CMS for complying with BIPA's
mandate to provide additional payments for new technologies
is to assign the new technology to the most appropriate DRG
and adjust payments for individual cases that involve a qualifying
technology when the costs of those cases exceed a threshold
amount. That is, hospitals would not receive a specific payment
associated with the technology, nor would there be an additional
payment for every case that involves the new technology. Rather
an additional payment would only be made in those situations
in which the costs of the entire case exceed the DRG amount
by a certain amount.
Under HCFA's proposed methodology, for cases that utilize
a new technology and that have costs that exceed the DRG payment
amount, the hospital would receive an additional payment equal
to one-half of the amount by which the costs of the case exceed
the DRG payment, up to a ceiling of 50 percent of the cost
of the new technology.
Teaching hospitals often are the sites where new technologies
are first introduced and used. These technologies are often
very costly when first introduced-costs that must be absorbed
by teaching hospitals because they are not recognized in the
DRG payment rates until several years after their introduction.
Consequently, providing adequate payments in these situations
is critical to ensure payment equity for the institutions
that shoulder this important responsibility.
We believe that the best solution involves incorporating
the costs of a new technology into the DRG rates as quickly
as possible. We recognize, however, that an interim solution
may be necessary until the data are available to permit this
incorporation.
We are concerned, however, that the proposed payment methodology
will not ensure payment equity for teaching hospitals that
use new technologies and, perhaps more importantly, may not
comply with the BIPA requirement that the additional payment
"adequately reflects the estimated average cost of the service
or technology." BIPA section 533. The proposed rule preamble
does not adequately explain how its proposal to limit additional
payments to 50 percent of a technology's costs complies with
the BIPA requirement.
If the criteria for identifying cutting-edge technologies
are developed and implemented appropriately, we believe that
only a limited number of technologies will qualify for additional
payments. For these limited technologies, we believe BIPA
requires that hospitals receive an additional payment that
much more closely approximates average costs than what has
been proposed.
Occupational Mix Adjustment
BIPA requires CMS to begin collecting data on the occupational
mix of short-term acute care hospital employees for purposes
of constructing an occupational mix adjustment to the wage
index that would affect hospitals' DRG payments beginning
in Federal fiscal year (FFY) 2005. Since the FFY 2005 wage
index will be based on wage data from hospitals' 2001 cost
reports, CMS is planning to conduct a special survey of hospitals
to obtain these data so that they coincide with hospitals'
2001 cost reports. The preamble to the proposed rule states
that more information about this survey will be provided in
the final rule.
A wage index that includes an occupational mix adjustment
essentially only recognizes differences across geographic
areas in terms of the price hospitals must pay for a particular
labor category. The fact that a hospital-such as a teaching
hospital-may have higher overall labor costs because its patient
population requires a larger quantity of highly skilled, higher
priced employees is not reflected in this type of wage index.
Typically, major teaching hospitals treat Medicare beneficiaries
with complex conditions, as evidenced by these hospitals'
higher case-mix indices (CMIs). These patients generally require
caregivers with high skill levels, which results in higher
labor costs for teaching hospitals.
It is generally acknowledged that the current DRG payment
system does not adequately recognize patient severity and
reimburse for the higher resource costs associated with complex
patients. However, under the current wage index methodology,
teaching hospitals could recoup some of these losses because
their higher employee skill mix was reflected through higher
wage indices in the areas where they are located. This will
no longer be the case if an occupational mix adjustment is
added to the wage index. Consequently, if this occurs, it
will be imperative that CMS refine the DRG system to ensure
that complex cases are adequately reimbursed. An example of
one such system is the "all patient refined diagnosis-related
groups (APR-DRGs)," although alternative DRG refinement systems
might also be appropriate.
In terms of CMS' specific proposal for collecting occupational
mix data, we would be happy to assist CMS in developing a
data collection methodology that would achieve the Agency's
goals without being unduly burdensome on providers. We are
particularly concerned about collecting hourly wages across
intervals. We also agree with the comments of the American
Hospital Association (AHA) that the job categories for which
CMS would collect wage data should reflect 100 percent of
hospital workers, not 60 percent, as reflected in the proposed
rule.
Conclusion
Thank you for this opportunity to present our views. We
would be happy to work with CMS on any of 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.
|
