AAMC Comment Letter on Medicare
Coverage of Clinical Trials
August 31, 2000
Clinical Trial Comments
Office of Clinical Standards and Quality
Health Care Financing Administration
7500 Security Boulevard, S3-02-01
Baltimore, MD 21244
The Association of American Medical College welcomes this
opportunity to comment on the Health Care Financing Administration's
proposed National Coverage Decision on Medicare Coverage
of Clinical Trials.
The AAMC represents approximately 400 major teaching hospitals;
all 125 accredited U.S. medical schools; 91 professional
and academic societies; and the nation's medical students
and residents.
The AAMC was very pleased when in June, President Clinton
directed the Secretary of HHS to "revise Medicare program
guidance to explicitly authorize payment for routine patient
care costs associated with clinical trials." Our comments
will focus on the definition of "routine costs of clinical
trials" and on the proposed "desirable characteristics
of a trial." In addition, we will raise a number of issues
that HCFA must address in order to pay providers accurately
for Medicare patients enrolled in clinical trials.
Definition of "Routine Costs of Clinical Trial"
The AAMC supports HCFA's proposal that among the items and
services excluded from coverage are those "provided by
the trial sponsor without charge." In consultation with
sponsors, principal investigators (PIs) and other knowledgeable
people, HCFA should make a general determination of which
costs have been borne by sponsors to this point and should
only pay for such costs in the future if sponsors can demonstrate
satisfactorily why these costs should no longer be paid by
them.
The AAMC is concerned about how it will be possible to identify
accurately items and services that are provided solely to
satisfy data collection needs (so called "protocol-induced
costs"). For instance, part of a protocol may be to give
a patient a monthly CT scan. If a patient develops a complication
and is given the scan for diagnosis and treatment of the complication,
rather than because it is called for in the protocol, how
will HCFA insure that the cost of this test that is now eligible
for payment, is paid? Other questions that must be answered
are:
- In order to bill Medicare, will every item and service
provided to a Medicare patient in a clinical trial have to
be coded as a routine cost, a cost related to a complication,
or a protocol-induced cost? If so, this will create an administrative
nightmare for hospitals and physicians and also may expose
them to claims of fraud if services are incorrectly coded
due to the imposition of an overly complicated system. Will
providers have to submit "no-pay" bills for those
items and services that are not covered
- HCFA needs to specify the responsibility of providers for
explaining the coverage benefits and limitations to beneficiaries.
Will providers be expected to develop an Explanation of Benefits
for Medicare beneficiaries enrolled in clinical trials?
- How
will hospitals and physicians know the point
at which investigational care becomes the standard of care?
Will they be expected to continually monitor the HCFA website
for both national coverage decisions and local medical review
policies? HCFA should establish a clear process by which
to determine on a national basis when experimental treatments
become accepted as the standard of care. The AAMC is willing
to work with HCFA to convene a group of knowledgeable individuals
to develop guidelines that address these very difficult issues.
Definition of "Clinical Trial"
The AAMC is pleased that HCFA has recognized that when a
clinical trial is not government sponsored, patients enrolled
in it should only be eligible to have items and services covered
if the trial meets certain standards or has, as HCFA terms
it, "desirable characteristics." The AAMC supports
the proposal that the Agency for Healthcare Research and Quality
(AHRQ) should spearhead a multi-agency effort to define those
characteristics. HCFA must ensure that whatever body it establishes
to develop the "desirable characteristics" list
does not duplicate any already existing bodies. AHRQ may wish
to consult with the Executive Committee of the Medicare Coverage
Advisory Committee.
We have the following additional comments about the HCFA
proposal:
- Payment for Medicare beneficiaries enrolled in government-sponsored
(i.e., "deemed") trials should begin as soon
as the NCD is finalized--retroactive to June 7--and should not
await the publication of a final list of "desirable
characteristics" for all other trials.
- HCFA should consider adding to the "desirable
characteristics"
list a requirement that the data from a clinical trial in
which Medicare beneficiaries are enrolled will be made available
to the public in accordance with standards protecting patient-specific
information and the requirements of scientific publication.
- The requirement that a desirable characteristic of
an acceptable trial be that the "trial is well-supported
by available scientific and medical information" could
be interpreted to mean that items and services associated
with "cutting-edge"
research will not be eligible for payment. Admittedly, HCFA
has a statutory obligation to provide beneficiaries with
treatment that is "safe and effective," and we
understand HCFA's intent in drafting the criterion as it
has. However, we suggest that better language would be that
the "trial
must have a sound scientific basis and medical rationale."
This language would ensure that arguments over semantics
would not be used to deny Medicare beneficiaries participation
in trials that provide them with the latest medical advances.
- Not included in the list of "desirable characteristics"
is that the trial be approved by an IRB or that there be
a written protocol (although HCFA's reference to a principal
investigator submitting a copy of the protocol to a registry
implies that a written protocol must exist). An explicit
requirement of a written protocol should be added, and
also the requirement for IRB approval.
- HCFA proposes that information on trials should be
kept in a Medicare clinical trials registry. The National
Library of Medicine, Food and Drug Administration,
and National Institutes of Health have just established
a clinical trials registry. It is suggested that the
information HCFA wishes to collect should reside in
that registry.
- While there is some appeal to the concept of a principal
investigator self-certifying that his/her clinical
trial meets the qualifying criteria, the consequences
of being wrong--for both the PI and the hospital--are
severe. If a hospital relies on a PI's self-certification
to bill Medicare, and it is subsequently decided that
there are problems with the self-certification, will
the hospital be subject to claims of violations of
the False Claims Act? What about the safety of Medicare
beneficiaries who are enrolled in trials that are
later found not to have the "desirable characteristics"?
HCFA should develop a quick-review system that will
confirm or deny the PI's self-certification.
- Principal investigators should be required to reveal
what financial interest, if any, they--or a member
of their immediate family or a business partner--have
in the item or service that is being tested.
- HCFA proposes excluding "trials that are designed
exclusively to test such things as toxicity levels
or basic disease biology."
This means excluding Phase I clinical trials. These trials
are of vital importance to cancer patients and those with
other life-threatening diseases. As a rule, patients are
enrolled in Phase I trials only if there is no standard
of treatment to combat their disease. It is possible that
this care could be covered under the doctrine of "compassionate
use,"
but it would be an easier and probably quicker process if
Phase I clinical trials were covered under the NCD. HCFA
could limit coverage of routine costs and costs related
to complications for patients enrolled in Phase I trials
to those patients who are certified by a physician to have
a life threatening disease for which there is no standard
of effective treatment.
- Category B devices should be covered under this NCD. The
coverage of Category B devices should not be left to the discretion
of individual carriers.
Impact on Medicare+Choice
The AAMC urges HCFA to consider that when a Medicare+Choice
enrollee goes out-of-network to participate in a clinical
trial, payment for items and services should be made directly
to the treating hospital or physician rather than to the M+C
organization. In this way, payment would go directly to the
person or entity that incurred the costs associated with the
Medicare beneficiary's enrollment in the clinical trial, and
it would not be necessary to redetermine the rates for M+C
organizations.
Other Issues
Since President Clinton's Memorandum stated that "HCFA
should inform all claims-processing contractors that Medicare
will immediately begin to reimburse routine patient
care costs and costs due to medical complications associated
with participation in clinical trials." (emphasis added),
it seems clear that once HCFA decides on the correct way in
which to implement this NCD, coverage should be retroactive
to June 7, 2000. As was noted previously, payment for clinical
trials with "deemed" status should not await the
resolution of issues that pertain only to other trials.
Finally, in the NCD HCFA says it will provide further clarification
about how to identify properly which costs are routine patient
care ones and which costs are protocol-induced. The AAMC
wishes to point out that without this essential information
it will be impossible to actually implement this NCD. Earlier
in the comments we discussed problems with putting costs
into certain categories since those categories may change
(e.g., a protocol-induced cost that unexpectedly becomes
a cost associated with a complication), depending on circumstances.
Before a final NCD is issued, HCFA must be certain as to
how it will identify costs that are eligible for payment
and must clearly communicate that information to all interested
parties. Accomplishing this goal may slow the process of
implementing the NCD, but if it is made retroactive, this
problem will be eased. Since so few Medicare beneficiaries
currently are enrolled in clinical trials, making this NCD
retroactive should not prove overly burdensome to carrier,
intermediaries, or providers.
If you wish to discuss these comments further, please contact
either Ivy Baer, J.D., M.P.H. at 202-828-0490 or Roger Meyer,
M.D., at 202-828-0567.
Sincerely yours,
Jordan J. Cohen, M.D.
cc: Nancy-Ann Min DeParle
|
