Cancer Clinical Trials: Medicare Reimbursement Denials

Published by the Government Accountability Office on 1997-10-14.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

      United States
GAO   General Accounting Offhe
      Washington, D.C. 20648
      Health, Educathn and Human Services Division


      October 14, 1997

      The Honorable Connie Mack
      The Honorable John Rockefeller
      United States Senate

      Subject: Cancer Clinical Trials: Medicare Reimbursement Denials

      In order to determine whether new drugs are safe and effective, it is essential
      to conduct studies in which the drugs are given to patients and the results are
      monitored. These studies, known as “clinical trials,” can examine the effects of
      new (not yet approved) drugs or can test already approved drugs against new
      conditions. Aside from their contribution to our understanding of the best way
      to manage disease, clinical trials sometimes offer critically ill patients access to
      potentially helpful yet currently unavailable therapies.

      However, many costs are associated with clinical trials, and who should pay
      these costs is an issue of growing concern. The costs of clinical trials include
      those for the drugs and for the research (that is, the collection and analysis of
      data) and the costs of patient care (for example, the salaries of the physicians
      or the expenses involved in a hospital stay if one is necessary). Although there
      is agreement that the sponsors of the research should pay for the former costs,
      debate continues about whether health insurers should pay for the “routine”
      costs of care. Insurers argue that because they do not typically pay for
      “experimental” treatments, they should not pay any of the costs associated with
      clinical trials. Patient advocates argue that insurers should be required to pay
      for the routine care that patients receive while in a clinical trial because,
      although the drug may be experimental, the supportive care that accompanies
      it often is not. Another rationale given for why insurers should pay is that
      many of the costs of routine care would be incurred even if patients were
      treated with conventional therapy. (For example, the antinausea drug given to
      a cancer patient receiving an experimental form of chemotherapy would also
      be given to patients receiving standard chemotherapy.)

      It is currently HCFA’s policy that Medicare will not reimburse routine patient
      care costs for beneficiaries enrolled in a clinical trial. The Medicare Cancer

                                          GAOBIEHS-9%15R         Cancer Clinical Trials

Clinical Trial Coverage Act (S. 381), introduced in February 1997, would
establish a demonstration project requiring the Health Care Financing
Administration (HCFA) to cover routine patient costs for Medicare
beneficiaries who are enrolled in cancer clinical trials. The intent of the
legislation is to give Medicare beneficiaries who participate in the
demonstration project access to cancer clinical trials and eliminate any concern
they might have about reimbursement for the costs of patient care associated
with those trials.

In an effort to determine the potential effect of this legislation, you asked that
we attempt to quickly estimate the current rate at which Medicare carriers
deny reimbursement for routine patient care costs when beneficiaries are
enrolled in cancer clinical trials. (This rate is one essential component of any
effort to estimate the incremental costs to HCFA of requiring reimbursement
for beneficiaries enrolled in cancer clinical trials.) As agreed with your offices,
we focused on a subset of all trials that would be covered by the legislation
and contacted the physicians participating in those trials1 All physicians were
asked whether they had enrolled any Medicare beneficiaries in the trial and, if
so, whether any problems were experienced with reimbursement.

In summary, 186 physicians responded to our survey, approximately 55 percent
of those to whom surveys were sent. @era& 1,143 patients were admitted into
the trials we asked about and, of those , 21i were Medicare patients. Among
these patients, there were eight (all reported by the same physician) for whom
the Medicare carrier had not made reimbursement. Detailed data on seven of
those patients showed that, in each case, the services denied were
chemotherapy and drugs. Further, we learned that only one of the physicians
automatically excluded Medicare patients from clinicaIl trials because of
 concerns over reimbursement.

While our methods did not allow us to give a precise national estirnate of the
rate at which reimbursement is denied for Medicare beneficiaries enrolled in
cancer clinical trials, our results suggest it is relatively rare, given the
populations and time period of our review. A detailed description of our
methods is presented in the enclosure.

 ‘We examined only trials listed on the Physician’s Data Query system
 maintained by the National Cancer Institute because these were the only trials
 that we could identify in a timely manner. We have no way of knowing what
 percentage of all ongoing trials are represented by those listed in this database.

 2                                  GAOLHEHS-9%15R         Cancer Clinical Trials
In commenting on a draft of this report, HCFA staff knowledgeable about
Medicare reimbursement raised three issues. First, HCFA staff were concerned
that our questions and the way they were asked could be misinterpreted by
respondents in a way that would mask the extent to which claims are denied.
We discussed our questions and methods and pilot-tested the survey with trial
physicians and remain confident that the respondents understood our
questions. HCFA staff correctly noted that we did not ask respondents the
extent to which they might never submit a claim for reimbursement to
Medicare because of their expectation that it would be denied, per HCFA
policy. To do so would have expanded our work beyond its purpose of
examining only claims denials.

Second, HCFA staff commented that our response rate was inadequate, at just
slightly over half of physicians surveyed. We have no reason to believe that
physicians who had more difficulty with reimbursements would have been less
likely to respond to the survey. The uniformity of the results also makes the
findings worth reporting.

Finally, HCFA staff reported that, given our Endings, their actuaries have nearly
doubled their estimates of the extent to which Medicare mistakenly reimburses
claims for routine patient care costs2 Under HCFA’s current policy, any
reimbursement for care associated with a cancer clinical trial would be made
in error.

As anranged with your office, unless you publicly announce its contents earlier,
we plan no further distribution of this report until 30 days from the date of this
letter. At that time, we will send copies to interested parties and make copies

2HCFA initially estimated that 25 percent of the claims under Medicare Part B
and 10 percent of the claims under Medicare Part A were mistakenly
reimbursed. These estimates have been revised to 50 percent and 15 percent,

3                                  GAOLHEHS-9%15R         Cancer Clinical Trials
    B-278 145
    available to others on request. This letter was prepared by Carolyn Feis and
    George Silberman. Please call me at (202) 512-7119 or Dr. Feis at (202) 512-
    3864 if you or your staff have any questions .

    Sincerely yours,

    Bernice St&hardt
    Director, Health Services Quality
     and Public Health Issues


    4                                   GAO/HEHS-9%15R     Cancer Clinical Trials
ENCLOSURE                                                                    ENCLOSURE
                              SCOPE AND METHODOLOGY

As agreed with the requesters’ offices, we defined the scope of the study so as to allow
for data collection within a short time and with minimum respondent burden. Therefore,
we focused our attention on a subset of all trials. The criteria we used to define that
subset and their rationale are presented below.

First, because no database identifies all ongoing cancer clinical trials, we focused our
attention on trials listed in the Physicians Data Query (PDQ) database maintained by the
National Cancer Institute (NCI). NC1 and other National Institutes of Health (NM)
studies are automatically included in PDQ, as are others. Any study, however, may be
submitted for review. Approximately one-third of the active trials listed in PDQ are
submitted voluntarily.

Second, in order to ensure that Medicare patients were likely to be enrolled, we focused
on clinical trials of treatments for types of cancer that were most prevalent among the
elderly. We selected trials that enrolled patients with breast, colon, rectal, prostate, or
lung cancer.

Third, while the legislation would require Medicare carriers to reimburse patient care
costs independent of the phase of the trial, we limited our review to Phase IT and Phase
III trials. Phase I trials are studies in which a treatment is first tested in humans. The
purpose is largely to determine the safety of the treatment. Because of the risks to
patients from these untested modalities, Phase I trials involve few patients, and the
average enrollrnent per trial is very small3

Fourth, we selected March 1, 1996, through September 30, 1996, as our period of study.
Rapid changes in the U.S. health care system outdate information about reimbursement
decisions made years ago. Therefore, we were interested in patients treated recently.
However, because the ultimate decisions on reimbursement are often reached through
iterative questioning and appeals, we needed to allow sufficient time for the denial
decision to “mature.” In conversations with physicians who participate in trials, we
learned that the enrollment date was usually proximate to when treatment began and that
it could take as long as 6 months to resolve requests for reimbursement. Therefore, we
focused on patients enrolled between March 1 and September 30, 1996.

‘Phase II and IIl tria Is, in contrast, are aimed at determining the effectiveness of
treatments and threfore require larger numbers of patients in order to offer reasonable
chances for observing an effect, even if it is small.

5                                            GAOMEHS-9%15R          Cancer Clinical Trials
ENCLOSURE                                                                   ENCLOSURE
Finally, we excluded trials that provided solely inpatient treatment. This decision was
based on the reality that insurers rarely lmow what specific inpatient services are
provided. Reimbursement for inpatient services is determined by a diagnosis, not by the
specific treatment. Because the treatment modalities are not specified to the insurer,
insurers usually cannot discriminate between experimental and conventional therapy.

After identifying active trials investigating one or more of the five cancers we were
interested in, we surveyed the individuals who chaired the studies to confirm that a trial
had enrolled patients during the period we were interested in, to determine the number of
patients admitted to the trial, and to verify that outpatient services were provided.

Each trial could have multiple physicians listed as principal investigators. For the trials
that met our criteria, we determined the average number of patients per physician listed
as a principal investigator and selected those for which the average was greater than or
equal to two. We did this because the likelihood of obtaining relevant data was minimal
for trials with a smaller ratio. Therefore, we surveyed all principal investigators PDQ
listed as associated with the trials we included.

We sent surveys to 340 different physicians working on 49 different trials and included
one follow-up mailing. We completed our work between February and June 1997 in
accordance with generally accepted government auditing standards. Our findings are
based on surveys received from 186 doctors working on 39 difFerent trials, representing
response rates of 53 and 80 percent, respectively.



 6                                             GAOIEEHS-9%15R        Cancer Clinical Trials
Ordering    Information

The first copy of each GAO report and testimony is free.
Additional   copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order
made out to the Superintendent     of Documents, when
necessary. VISA and Mastercard      credit cards are accepted, also.
Orders for 100 or more copies to be mailed to a single address
are discounted 25 percent.

Orders by mail:

U.S. General Accounting      Office
P.O. Box 37050
Washington, DC 20013

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting  Office
Washington, DC

Orders may also be placed by calling (202) 512-6000
or by using fax number (202) 512-6061, or TDD (202)       512-2537.

Each day, GAO issues a list of newly available reports and
testimony.   To receive facsimile copies of the daily list or any
list from the past 30 days, please call (202) 512-6000 using a
touchtone phone. A recorded menu wiB provide information          on
how to obtain these lists.

For information on how to access GAO reports on the INTERNET,
send an e-mail message with Yinfo” in the body to:


or visit GAO’s World      Wide Web Home Page at:

United States
General Accounting Office
Washington, D.C. 20548-0001

Official Business
Penalty for Private Use $300
Address Correction   Reauested



                                       .. .._
                                        : ,.