Medicare Outpatient Payments: Rates for Certain Radioactive	 
Sources Used in Brachytherapy Could Be Set Prospectively	 
(24-JUL-06, GAO-06-635).					 
                                                                 
Generally, in paying for hospital outpatient procedures, Medicare
makes prospectively set payments that are intended to cover the  
costs of all items and services delivered with the procedure.	 
Medicare pays separately for some technologies that are too new  
to be represented in the claims data used to set rates. It also  
pays separately for certain technologies that are not new, such  
as radioactive sources used in brachytherapy, a cancer treatment.
The Medicare Prescription Drug, Improvement, and Modernization	 
Act of 2003 required separate payment for the radioactive	 
sources. It also directed GAO to make recommendations regarding  
future payment. GAO examined (1) how Medicare determines payment 
amounts for technologies that are not new but are separately paid
and (2) how payment amounts for iodine, palladium, and iridium	 
sources used in brachytherapy could be determined.		 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-06-635 					        
    ACCNO:   A57379						        
  TITLE:     Medicare Outpatient Payments: Rates for Certain	      
Radioactive Sources Used in Brachytherapy Could Be Set		 
Prospectively							 
     DATE:   07/24/2006 
  SUBJECT:   Cost analysis					 
	     Health care costs					 
	     Hospitals						 
	     Medical services rates				 
	     Medical technology 				 
	     Medicare						 
	     Prices and pricing 				 
	     Prospective payments				 
	     Medical supplies					 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-06-635

     

     * Results in Brief
     * Background
          * Radioactive Sources Used in Brachytherapy
          * Evolution of Medicare Payment for Outpatient Services
          * OPPS Payment for Radioactive Sources
     * When Paying Separately for Technologies That Are Not New, CM
          * When Paying Separately for Technologies That Are Not New, CM
          * Certain Technologies That Are Not New and Are Not Suitable f
     * When Paying Separately for Iodine and Palladium, CMS Could S
          * CMS Could Set Prospective Payment Rates for Iodine and Palla
          * Suitable Methodology for Determining Separate Payment Amount
     * Conclusions
     * Recommendations for Executive Action
     * Agency and External Reviewer Comments and Our Evaluation
          * CMS Comments and Our Evaluation
          * Manufacturer and Provider Comments and Our Evaluation
     * Sample Design
     * Data Collection and Data Processing
     * Estimates of Mean and Median Purchase Prices for Iodine and
          * Weighting the Hospital Sample
          * Mean Purchase Price Using Volume and Hospital Weights
          * Median Purchase Price Using Volume and Hospital Weights
          * Coefficients of Variation for Mean Purchase Price
     * GAO Contact
     * Acknowledgments
     * GAO's Mission
     * Obtaining Copies of GAO Reports and Testimony
          * Order by Mail or Phone
     * To Report Fraud, Waste, and Abuse in Federal Programs
     * Congressional Relations
     * Public Affairs

Report to Congressional Committees

United States Government Accountability Office

GAO

July 2006

MEDICARE OUTPATIENT PAYMENTS

Rates for Certain Radioactive Sources Used in Brachytherapy Could Be Set
Prospectively

GAO-06-635

Contents

Letter 1

Results in Brief 3
Background 5
When Paying Separately for Technologies That Are Not New, CMS's General
Practice Is to Set a Rate Based on an Average Cost across Hospitals 8
When Paying Separately for Iodine and Palladium, CMS Could Set Prospective
Rates, but Suitable Payment Methodology for Iridium Is Unclear 10
Conclusions 13
Recommendations for Executive Action 14
Agency and External Reviewer Comments and Our Evaluation 14
Appendix I GAO Survey of Hospital Purchase Prices for Iodine, Palladium,
and Iridium Sources Used in Brachytherapy 19
Appendix II Comments from the Centers for Medicare & Medicaid Services 26
Appendix III GAO Contact and Staff Acknowledgments 28

Table

Table 1: Reported Iodine and Palladium Purchase Prices, July 2003-June
2004 11

Abbreviations

ABS American Brachytherapy Society ACCC Association of Community Cancer
Centers ACRO American College of Radiation Oncology APC ambulatory payment
classification ASP average sales price ASTRO American Society for
Therapeutic Radiation and Oncology CAB Coalition for the Advancement of
Brachytherapy CMS Centers for Medicare & Medicaid Services MMA Medicare
Prescription Drug, Improvement, and Modernization Act of 2003 OPPS
outpatient prospective payment system

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

United States Government Accountability Office

Washington, DC 20548

July 24, 2006

Congressional Committees

Under Medicare's hospital outpatient prospective payment system (OPPS),
hospitals are paid a fixed, predetermined-that is, prospectively
set-amount for each procedure they provide to Medicare beneficiaries.1
Hospitals are expected to use this prospective payment to cover the costs
of items and services, such as anesthesia and medical supplies, associated
with the procedure. In creating one payment bundle for items and services
associated with a procedure, Medicare provides hospitals with an incentive
to operate efficiently, as they retain the difference if the payment
exceeds the cost the hospital incurs in performing the procedure. Although
bundled payment is a fundamental principle of the OPPS, Medicare pays
separately for certain high-cost technologies because bundling them into a
payment with their associated procedures could financially disadvantage
hospitals even if they operate efficiently.2 Some technologies are paid
separately because they are new and their costs are not represented in the
historical data used to set bundled payments for procedures. However,
certain other technologies that are not new and have historical claims
have also been designated for separate payment either by Congress or by
the agency that administers Medicare, the Centers for Medicare & Medicaid
Services (CMS) in the Department of Health and Human Services.

Brachytherapy is an example of a procedure involving a technology that is
not new and is separately paid. During the procedure, radioactive
materials, called sources, are implanted in or near a cancerous tumor. The
three radioactive sources most commonly used in this treatment are
iodine-125 and palladium-103, which provide a prolonged, low dose of
radioactivity, and iridium-192, which provides a brief, high dose of
radioactivity.3 In 2002, these three sources were billed on 98 percent of
the claims for radioactive sources associated with brachytherapy. Medicare
pays separately for these, as well as other radioactive sources associated
with brachytherapy,4 at each hospital's cost.5 According to our estimates,
payments in 2004 for iodine, palladium, and iridium sources represented
less than one-half of 1 percent of the $15.9 billion in OPPS spending.

1For purposes of this report, "procedure" can refer to a service that
constitutes a clinical course of action, such as an outpatient surgery; a
medical test; or another service, such as an office visit.

2In this report, we use "technologies" to refer to certain products that
are used in outpatient procedures. These products include drugs; devices;
biologicals, which are derived from living sources, including humans,
animals, or microorganisms; and radiopharmaceuticals, which are
radioactive chemical agents provided orally, injected, or provided through
other means for diagnostic or therapeutic purposes.

The Medicare Prescription Drug, Improvement, and Modernization Act of 2003
(MMA) required that all radioactive sources used in brachytherapy be paid
separately rather than bundled into payment for their associated
procedures.6 The MMA specified that these separate payments be made at
each hospital's cost through December 31, 2006. While the MMA required
separate payment after this date as well, it did not specify a methodology
for determining the separate payment amounts. Rather, it directed us to
conduct a study and make recommendations regarding future payment for
radioactive sources. As discussed with the committees of jurisdiction,
this report examines (1) how CMS determines payment amounts for
technologies that are not new but are separately paid and (2) how payment
amounts for iodine, palladium, and iridium sources used in brachytherapy
could be determined.

To examine how CMS determines payment amounts for technologies that are
not new but are separately paid, we reviewed federal law and regulation
pertaining to the OPPS. We also interviewed officials at CMS. To examine
how payment amounts for iodine, palladium, and iridium sources used in
brachytherapy could be determined, we conducted a survey of purchase
prices paid by 121 hospitals from July 1, 2003, through June 30, 2004.7
These hospitals were selected to be representative of all hospitals
purchasing these sources in 2002, the most recent year from which data
could be used to construct a sample.8 We assessed the reliability of the
data we received from these hospitals. After excluding questionable data,
we determined that the remaining data were suitable for our purposes. Our
final results represented data from 62 hospitals, or slightly more than 50
percent of the hospitals in our sample. Our results can be generalized to
the larger population of hospitals providing these sources in the
outpatient setting that met our sampling criteria. (See app. I for more
information on our hospital survey.) We also interviewed representatives
from a trade association of radioactive source manufacturers, six
radioactive source manufacturers, three associations representing
physicians and other health professionals involved in brachytherapy, an
association of cancer hospitals, and seven individual hospitals. We
conducted a site visit to a hospital that provides brachytherapy. We also
reviewed federal law and regulation pertaining to the OPPS and interviewed
officials at CMS. We did our work in accordance with generally accepted
government auditing standards from June 2004 through July 2006.

3For the remainder of this report, we refer to iodine-125 as "iodine,"
palladium-103 as "palladium," and iridium-192 as "iridium." While
iridium-192 can also be provided in low-dose form, this method of
treatment is rare. Therefore, we refer to iridium in its high-dose form,
unless otherwise specified.

4Medicare pays for 12 radioactive sources used in brachytherapy: high- and
low-activity iodine, high- and low-activity palladium, gold-198, low-dose
iridium, high-dose iridium, yttrium-90, cesium-131, liquid iodine-125,
ytterbium-169, and linear palladium-103.

5Unless otherwise specified, paying "at each hospital's cost" refers to a
particular methodology CMS uses to estimate a hospital's cost of providing
a technology. This methodology relies on the charge a hospital identifies
on its claim for reimbursement, which CMS converts to cost using the ratio
of aggregate costs and charges from the hospital's most recent cost
report. An alternative method of paying at each hospital's cost relies on
the costs reported by the hospital on its most recent cost report.

6Pub. L. No. 108-173, S: 621(b), 117 Stat. 2066, 2310 (to be codified at
42 U.S.C. S: 1395l(t)(2)(H) and (16)(C)).

                                Results in Brief

When paying separately for technologies that are not new, CMS's general
practice is to set prospective rates based on an average-that is, the mean
or median-unit cost of the technologies across hospitals. For example, CMS
currently pays separate prospective rates for certain high-cost drugs and
biologicals9 based on the mean estimated per-unit acquisition cost, as
derived by CMS from data provided by drug manufacturers. A prospective
rate, even for technologies that are separately paid, is desirable because
basing a rate on an average encourages those hospitals that provide the
technology to minimize their acquisition costs. If CMS determines that a
technology's unit cost varies substantially and unpredictably, or that
reasonably accurate data on which to base an average unit cost are not
available, CMS pays for the technology at each hospital's cost. When the
cost of a technology varies substantially and unpredictably, a prospective
rate based on a historical average may not adequately pay hospitals even
if they operate efficiently. One example of such a technology is corneal
transplant tissue. After analyzing data submitted by hospitals and other
stakeholders, CMS determined that the fees eye banks charge hospitals for
corneal transplant tissue vary substantially and unpredictably over time
and across eye banks in a given year. The amount of the fee charged by an
eye bank depends heavily on the level of charitable donations it receives,
which it uses to subsidize the cost of providing the tissue to hospitals.
As a result of the variation in fees hospitals pay, CMS pays for the
tissue at each hospital's cost.

7Specifically, we asked hospitals to report the prices they paid for
sources upon receiving the product. These prices are net of discounts, but
they do not reflect rebates from manufacturers, which are not commonly
provided for brachytherapy sources, nor any costs hospitals may incur in
storing and handling the radioactive sources.

8These hospitals were Medicare providers as of July 2004.

9For the remainder of this report, we use "drugs" to refer to both drugs
and biologicals.

CMS could set prospective payment rates for iodine and palladium due to
the general stability in their unit cost and the availability of
reasonably accurate data. According to interviews we conducted with
hospital and manufacturer officials, iodine and palladium have an
identifiable unit cost, the price per source. When we surveyed hospitals
on their purchase prices, we found that the prices do not vary
substantially or unpredictably. Furthermore, we determined that a
reasonably accurate source of data, historical OPPS claims, is available
for setting prospective rates for iodine and palladium. We were unable to
identify a methodology CMS could use to determine future payment amounts
for iridium. In contrast to iodine and palladium, where multiple sources
are permanently implanted in one patient, a single iridium source is
temporarily implanted. Because an iridium source can be implanted in
multiple patients over its 3-month life span,10 and each patient can
receive multiple treatments with the source, the appropriate unit cost of
an iridium source is the average cost of all treatments administered
across all patients. Although we surveyed hospitals on the per-treatment
costs of iridium, we did not receive sufficient data to estimate an
average cost across hospitals. However, hospital claims data are available
to CMS for estimating an average per-treatment cost across hospitals that
have used iridium. Using these data, CMS would be able to evaluate whether
the range of cost comprising the average is substantial and whether cost
varies unpredictably over time. Such an analysis would help CMS identify a
suitable methodology for determining a separate payment amount.

10We use "life span" to refer to the period of time iridium is
sufficiently radioactive to be used for high-dose brachytherapy.

In this report, we make recommendations to the Secretary of Health and
Human Services regarding payment for iodine, palladium, and iridium
sources. Specifically, we recommend that the Secretary direct the
Administrator of CMS to (1) set prospective payment rates for iodine and
palladium sources, with each rate based on the source's mean or median
cost across hospitals estimated from OPPS claims data, and (2) use claims
data to evaluate the unit cost of iridium, so that a suitable, separate
payment methodology can be determined. In response, CMS stated that it
will take GAO's recommendations into consideration.

                                   Background

Iodine, palladium, and iridium are the radioactive sources most commonly
used in brachytherapy. The brachytherapy procedure is typically performed
in the outpatient setting where, under the OPPS, costs associated with a
procedure are generally bundled in order to promote hospital efficiency.
However, since the OPPS was implemented in 2000, an increasing number of
technologies have been paid separately. Except in 2003, the one year in
which iodine and palladium used to treat prostate cancer and iridium were
bundled into payment for brachytherapy procedures, all radioactive sources
used in brachytherapy have been paid separately.

Radioactive Sources Used in Brachytherapy

Radioactive sources are used in brachytherapy to treat a variety of types
of cancers. The most prevalent brachytherapy procedure is low-dose
brachytherapy with iodine or palladium, which is typically provided for
early-stage prostate cancer. During this procedure, approximately 20 to
200 tiny iodine or palladium sources are implanted in the prostate,
deliver radiation over a period of months, and then remain permanently in
the body. Generally, the choice between iodine and palladium is determined
by the aggressiveness of the tumor, and the number of sources by the size
of the prostate.11

In recent years, utilization of the high-dose brachytherapy procedure,
which typically uses iridium, has grown. Iridium can be used to treat a
variety of advanced-stage cancers-most commonly gynecological cancers. In
high-dose brachytherapy, a single, highly radioactive iridium source is
implanted in the tumorous area for a brief period-a matter of minutes or
hours-and then withdrawn. Depending on a patient's clinical needs, the
patient may receive one or more such treatments, also known as fractions,
with the same source over the course of several days. Because an iridium
source emits sufficient radiation for 3 months, the same source can be
used to treat multiple patients.

11Although iodine and palladium both emit relatively low levels of
radiation, palladium emits radiation at a higher rate, making it generally
appropriate for more aggressive tumors.

Evolution of Medicare Payment for Outpatient Services

The payment methodology for outpatient services has varied in the degree
to which it relies on bundled payments to promote hospital efficiency.
Prior to OPPS implementation in 2000, payment for outpatient items and
services was not bundled; rather, hospitals were paid under a complex
array of cost-based reimbursement methods and fee schedules. Generally,
neither of these payment methodologies provides a strong incentive to
furnish services efficiently. Under a cost-based methodology, each
hospital is paid its cost based on information it reports to CMS. Under a
fee schedule methodology, all hospitals receive a prospectively determined
rate for each item and service they provide, but little incentive exists
for them to provide only the necessary items and services.

Under the Balanced Budget Act of 1997, CMS was required to implement the
OPPS, which was designed to streamline the historically complex system of
payment for outpatient care and better promote hospital efficiency.12 CMS
assigns each outpatient procedure to one of approximately 850 ambulatory
payment classification (APC) groups. Each APC group includes procedures
that share cost and clinical similarities and has one payment rate for all
procedures in the group.13 To set an APC rate, CMS uses historical claims
to calculate a median cost across a group's procedures that includes the
costs of the associated bundled services and supplies, which are known as
"packaged" costs. A median, rather than a mean, gives less weight to
extreme values. That median cost is then converted into a numeric weight,
which determines the payment hospitals receive for all procedures assigned
to the APC. Because the OPPS provides a single payment to cover the
average total cost of a procedure, the incentive for each hospital to
efficiently provide the necessary items and services associated with that
procedure is greater than when the hospital is paid its cost or a separate
fee schedule payment for each item and service used in the procedure.

12Pub. L. No. 105-33, S: 4523, 111 Stat. 251, 445-50.

13For example, APC 396, "Bone Imaging," includes the following procedures:
"bone imaging, limited area"; "bone imaging, multiple areas"; and "bone
imaging, whole body."

Although bundling is a fundamental principle of the OPPS, the number of
technologies that are paid separately from their associated procedures has
increased since the implementation of the payment system.14 Beginning in
2000, the first year of the OPPS, CMS was required to make temporary,
separate payments-referred to as "transitional pass-through payments"-for
technologies that it determines to meet specified criteria for being new
and high cost.15 These payments supplement the bundled payments for
outpatient procedures associated with the technologies, and are designed
to compensate hospitals for the additional cost. A new technology is
eligible for pass-through payments for 2 to 3 years, after which time the
technology is no longer considered new and CMS can include the technology
in the payment bundle for the associated procedure. Over time, other
high-cost technologies that are not new-mainly certain drugs and
radiopharmaceuticals-have also been designated for separate payment either
by Congress or by CMS.

OPPS Payment for Radioactive Sources

The payment methodology for radioactive sources associated with
brachytherapy has changed several times since the inception of the OPPS.
CMS was required to make separate pass-through payments for all
radioactive sources associated with brachytherapy beginning in 2000. In
2003, these technologies were no longer eligible for pass-through
payments. Because they are considered devices by Medicare, and devices are
typically bundled into payment for their associated procedures, CMS
bundled iodine and palladium into the payment bundle for the low-dose
brachytherapy procedure for prostate cancer, and iridium into the payment
bundle for the high-dose brachytherapy procedure, regardless of cancer
type. For iodine and palladium sources provided for conditions other than
prostate cancer, CMS continued to pay separately. Instead of paying
separately for these radioactive sources at each hospital's cost, CMS set
prospective rates for 2003 based on the median cost of each source across
hospitals. The MMA mandated that all brachytherapy sources be paid
separately after 2003 and specified that from January 1, 2004, through
December 31, 2006, separate payments for the sources be at each hospital's
cost. The MMA did not specify a methodology for paying separately after
this date.

14See Medicare Payment Advisory Commission, Report to the Congress:
Selected Medicare Issues (Washington, D.C.: June 2000), and Barbara O.
Wynn, Medicare Payment for Hospital Outpatient Services: A Historical
Review of Policy Options, a working paper prepared for the Medicare
Payment Advisory Commission by RAND Health, June 2005.

15The Medicare, Medicaid, and SCHIP Balanced Budget Refinement Act of
1999, Pub. L. No. 106-13, app. F, S: 201(b), 113 Stat. 1501,
1501A-337-1501A-339.

When Paying Separately for Technologies That Are Not New, CMS's General Practice
           Is to Set a Rate Based on an Average Cost across Hospitals

When paying separately for technologies that are not new, CMS's general
practice is to set a prospective rate for all hospitals, based on an
average unit cost across hospitals. However, certain technologies may vary
in cost substantially and unpredictably or there may not be reasonably
accurate data on which to base an average cost across hospitals. In either
case, CMS pays for these technologies at each hospital's cost.

When Paying Separately for Technologies That Are Not New, CMS's General Practice
Is to Set Prospective Rates

Although CMS does not use published criteria to determine payment amounts
for separately paid technologies that are not new, we found that its
general practice is to pay prospectively based on the average historical
cost of each technology across hospitals. A prospective rate, even for
technologies that are separately paid, is desirable because basing a rate
on an average encourages those hospitals that provide the technology to
minimize their acquisition costs.16

To set prospective rates for these separately paid technologies, CMS
currently uses two sources of historical data: manufacturer data and OPPS
claims. For example, CMS pays for certain high-cost drugs prospectively
based on average per-unit acquisition cost.17 To calculate hospital
acquisition cost, CMS relies on per-unit average sales price (ASP) data,
which manufacturers are required to submit to CMS and are used in making
payments for physician-administered drugs.18 CMS also uses ASP data to pay
a per-unit rate for particular orphan drugs, which are drugs used to treat
patients with rare conditions and are typically high in cost. For drugs
where CMS does not have ASP data, CMS pays based on the mean cost
calculated from OPPS claims.

16For example, if two manufacturers sell the same technology, and there is
not a discernable difference in quality between the two products, then the
hospital would have the incentive to purchase the technology from the
manufacturer offering the lower purchase price.

17These include certain drugs known as specified covered outpatient drugs
and other drugs with per-day costs of $50 or more.

Certain Technologies That Are Not New and Are Not Suitable for Prospective
Payment Are Paid at Cost

When a technology's unit cost varies substantially and unpredictably, or
when reasonably accurate cost data are not available, CMS pays for the
technology at each hospital's cost. If the cost varies substantially and
unpredictably, a prospective rate based on a historical average may not
adequately pay hospitals even if they operate efficiently. CMS pays each
hospital's cost, for example, for corneal transplant tissue and certain
vaccines, including those for flu and pneumonia.19 CMS uses this
methodology for corneal transplant tissue because, after analyzing data
submitted by hospitals and other stakeholders, the agency determined that
the fees eye banks charge hospitals for this tissue can vary substantially
and unpredictably over time and across eye banks in a given year. The
amount of the fee charged by an eye bank depends heavily on the level of
charitable donations it receives, which it uses to subsidize the cost of
providing the tissue. The cost to hospitals of providing vaccines also
varies substantially and unpredictably due to instability in the nation's
vaccine supply.

In other cases, CMS makes cost-based payments for technologies when it
determines that reasonably accurate historical data on unit cost are not
available. For example, the MMA mandated separate payment for certain
radiopharmaceuticals. As we discussed in our 2006 report on OPPS payment
for certain drugs and radiopharmaceuticals,20 differences among hospitals
in how these technologies are purchased make it difficult for CMS to set a
prospective rate based on an average cost across hospitals. As a result,
payment for these radiopharmaceuticals is based on each hospital's cost.

18The MMA required manufacturers to report quarterly ASPs for drugs sold,
with certain exceptions, to all purchasers. MMA S: 303(i)(4)(B)(iii), 117
Stat. 2254. An ASP must be net of volume discounts, prompt-pay discounts,
cash discounts, free goods that are contingent on any purchase
requirement, chargebacks, and all rebates but those owed to Medicaid. MMA
S: 303(c), 117 Stat. 2240-41.

19CMS pays hospitals at cost for corneal transplant tissue somewhat
differently than it pays hospitals at cost for other separately paid
technologies. Specifically, CMS instructs hospitals to record the
acquisition cost on the claim instead of a charge and pays them for this
amount. CMS later conducts a reconciliation of these payments with the
costs indicated on the hospital's annual cost report to ensure that the
payments were based on reasonable costs.

When Paying Separately for Iodine and Palladium, CMS Could Set Prospective
         Rates, but Suitable Payment Methodology for Iridium Is Unclear

Based on our analysis, the absence of wide variability in the unit costs
of iodine and palladium and the availability of reasonably accurate
historical data makes these radioactive sources suitable for prospective
payment rates. We were unable to establish a unit cost for iridium and, as
a result, could not identify a suitable payment methodology. CMS has OPPS
claims data from hospitals that provided iridium, and would be able to use
these data to calculate an average unit cost across hospitals and to
identify which methodology is suitable for determining a separate payment
amount.

CMS Could Set Prospective Payment Rates for Iodine and Palladium

Our analysis suggests that CMS would be able to develop prospective rates
for iodine and palladium beginning in 2007. Based on interviews we
conducted with hospital and manufacturer officials, and the results of our
hospital survey, we determined that iodine and palladium have identifiable
unit costs and that these costs do not appear to vary substantially and
unpredictably across hospital purchases at a given point in time or from
year to year. Both hospitals and manufacturers told us that hospitals
generally purchase iodine and palladium sources at a per-source price,
making the calculation of a unit cost straightforward. According to our
survey of 121 hospitals on the prices they paid during 1
year-specifically, from July 2003 through June 2004-the range of iodine
and palladium prices is not wide.21 This is indicated by the relative
level of precision-technically, the coefficient of variation-achieved for
our estimated mean price.22 (See table 1.) We also note that iodine and
palladium are not subject to the same supply and demand conditions as
corneal transplant tissue and flu and pneumonia vaccines-conditions that
lead to substantial and unpredictable cost variation from year to year.

20GAO, Medicare Hospital Pharmaceuticals: Survey Shows Price Variation and
Highlights Data Collection Lessons and Outpatient Rate-Setting Challenges
for CMS, GAO-06-372 (Washington, D.C.: Apr. 28, 2006).

21Our survey requested per-source purchase prices from hospitals. These
prices do not reflect storage and handling costs associated with the
radioactive sources. Prior to April 1, 2004, CMS had not articulated a
policy specifically on reimbursement for these costs. Effective on that
date, CMS provided several avenues for hospitals to identify the costs
associated with the storage and handling of radioactive sources, so that
these costs might be recognized in the payment system.

Table 1: Reported Iodine and Palladium Purchase Prices, July 2003-June
2004

                        Number of                                             
              Number of hospitals   Estimated                                 
Type of     reported reporting  mean price    Coefficient of variation for
source    purchasesa purchases per sourceb              the mean estimatec
Iodine         1,926        52      $29.54                           1.59% 
Palladium        941        40      $45.35                           0.68% 

Source: GAO survey of purchase prices from July 2003 through June 2004.

aA reported purchase refers to an individual hospital's purchase of a
given quantity of the radioactive source at a particular price on a
specific date.

bThe estimated mean price per source is weighted according to the
methodology described in app. I.

cThe coefficient of variation measures the magnitude of dispersion around
the mean. In statistical terms, a coefficient of variation below 10
percent is considered to be low. (See Morris H. Hansen, William N.
Hurwitz, and William G. Madow, Sample Survey Methods and Theory (New York:
John Wiley & Sons, 1953), 124,129-130.)

Although CMS uses ASP data to set a prospective rate for certain high-cost
drugs, CMS currently does not have ASP data for radioactive sources used
in brachytherapy. However, we found that OPPS claims provide a reasonably
accurate source of data for setting a prospective rate for iodine and
palladium sources. To determine if claims could be used as a reasonable
data source, we compared the payment rates for 2003 and the proposed
payment rates for 2004,23 which were based on median costs calculated from
historical claims, with the median of the per-source purchase prices
reported directly to us by hospitals. Although the payment rates applied
only to sources used in non-prostate brachytherapy, CMS officials told us
that they were calculated using prostate and non-prostate brachytherapy
claims with iodine and palladium sources. We found that for iodine the
prospectively set rate for 2003 and proposed rate for 2004 were $31.33 and
$36.35, respectively, and the median of reported purchase prices was
$25.37.24 For palladium, the prospectively set rate for 2003 and proposed
rate for 2004 were $43.96 and $44.00, respectively, and the median
reported purchase price was $45.46.

22To the extent that variation exists across either palladium or iodine
prices, it could be attributed to differential pricing by specific source
characteristics, such as radioactivity level or the configuration in which
they are purchased-that is, whether they are stranded together for
insertion or are individual, "loose" sources. We did not receive enough
data from hospitals to reliably identify any price differences by source
characteristic. However, we instructed hospitals to report all their
purchases during the survey period. Therefore, any price variation due to
source characteristic should be reflected in our data. Regarding activity
level, we note that the MMA required CMS to establish payments that
account for the radioactive intensity of sources. As a result, in 2005,
CMS established separate billing codes for high- and low-activity iodine
and palladium sources. CMS is therefore expected to have the data
available to set separate rates for high- and low-activity iodine and
palladium in 2007.

23These rates were proposed for 2004; however, they were not implemented
due to the MMA requirement to pay for the sources based on each hospital's
cost.

Since 2004, when CMS was required to pay separately for all iodine and
palladium sources, the agency has been accumulating claims data that
include separate charges for these sources. As a result, CMS will have
data from 2005 for the 2007 payment year.25 These data could be used to
set prospective payment rates, either based on a mean-as is currently done
with certain high-cost drugs-or based on a median-which CMS used to set
the 2003 and proposed 2004 rates for iodine and palladium sources.

Suitable Methodology for Determining Separate Payment Amount for Iridium Is
Unclear

Due to the reusable nature of the iridium source, identifying its unit
cost is not as straightforward as identifying the unit cost of iodine and
palladium. Over the course of its 3-month life span, an iridium source can
be temporarily implanted in multiple patients and each of those patients
can receive about 1 to 10 such treatments with the same source. Therefore,
the appropriate unit cost of an iridium source is the per-treatment
cost-the average cost of all treatments administered across all patients
over a 3-month period. When hospitals purchase an iridium source, they may
not know the exact number of patients they will treat or the number of
treatments each of those patients will receive. Therefore, hospitals must
bill Medicare based on projections of their unit cost, and will only be
able to identify their actual unit cost retrospectively.

We asked hospitals to provide the per-treatment cost of iridium sources
they purchased over a previous 12-month period in order to identify a unit
cost. However, we did not receive enough data to identify the
per-treatment cost. Of 121 total hospitals surveyed, 19 responded with
data on iridium, and the majority of these 19 hospitals did not provide
data we could use to estimate the cost per treatment. Specifically, 11
either did not provide the number of treatments, reported a questionable
source price, or both. Eight hospitals reported a source price and the
number of treatments from which a unit cost could be calculated. However,
among these 8 hospitals there were inconsistencies in the data provided.
Some hospitals reported the total price of their iridium contracts,26
while other hospitals isolated the price of the radioactive source within
their contracts and reported that price. Because we could not establish a
unit cost, we could not assess if the unit cost of iridium varies
substantially and unpredictably over time.

24The median reported price is weighted according to the methodology
described in app. I.

25All payment rates for a given year are based on claims for services
provided 2 years prior.

Although we could not identify an average per-treatment cost from our
survey data, CMS has OPPS claims data from hospitals that provided
iridium. Using these data, CMS would be able to evaluate whether the range
of costs comprising the average is substantial and whether the cost varied
unpredictably. Such an analysis would help CMS identify a suitable
methodology for determining a separate payment amount.

                                  Conclusions

Under the OPPS, an increasing number of technologies have been designated
for separate payment, either by Congress or by CMS. Pursuant to the MMA,
radioactive sources used in brachytherapy, including iodine, palladium,
and iridium, are among those technologies. Based on our analysis, CMS can
pay separately for iodine and palladium sources using prospective rates
because the unit cost of the sources does not vary substantially and
unpredictably. In addition, CMS has data available to identify reliable
average costs across hospitals to set prospective payment rates beginning
in 2007. Paying prospectively in this manner would help encourage hospital
efficiency. However, we were not able to identify a suitable methodology
for determining a separate payment amount for iridium sources because we
did not receive sufficient information from hospitals to estimate an
average per-treatment cost across hospitals. In order to identify a
suitable methodology for determining a separate payment amount, CMS would
be able to use OPPS claims data to evaluate whether the range of costs
comprising the average is substantial and whether the average
per-treatment cost varies unpredictably over time.

26Most hospitals purchase the iridium source as part of an annual contract
that covers the cost of four sources-one for each quarter-and the cost of
maintaining the sources.

                      Recommendations for Executive Action

In order to promote the efficient delivery of radioactive sources
associated with outpatient brachytherapy, we recommend that the Secretary
of Health and Human Services direct the Administrator of CMS to take the
following two actions:

           o  Set prospective payment rates for iodine and palladium sources
           with each rate based on the source's average-that is, the mean or
           median-unit cost across hospitals estimated from OPPS claims data.
           o  Use claims data to evaluate the unit cost of iridium so that a
           suitable, separate payment methodology can be determined.

Agency and External Reviewer Comments and Our Evaluation           
			  
			  We received written comments on a draft of this report from CMS
           (see app. II). We also received oral comments from individuals at
           five organizations representing manufacturers of radioactive
           sources used in brachytherapy and providers of brachytherapy.
           These included the Coalition for the Advancement of Brachytherapy,
           which represents manufacturers of radioactive sources; the
           Association of Community Cancer Centers (ACCC), which represents
           hospitals that provide cancer treatment; and three organizations
           representing physicians and others involved in providing
           brachytherapy: the American College of Radiation Oncology (ACRO),
           the American Brachytherapy Society (ABS), and the American Society
           for Therapeutic Radiation and Oncology (ASTRO). We also received
           technical comments from CMS and the external reviewers, which we
           incorporated as appropriate.

CMS Comments and Our Evaluation           
			  
			  In reviewing our draft report, CMS stated that it appreciated our
           analysis and will consider our recommendations on iodine,
           palladium, and iridium as it develops payment policy for 2007. CMS
           also noted that we did not make recommendations on payment for
           other radioactive sources associated with brachytherapy that may
           be separately payable in 2007.

           As stated in our draft report, we examined how payment amounts for
           iodine, palladium, and iridium could be determined. In 2002, these
           three sources were billed on 98 percent of the claims for
           radioactive sources associated with brachytherapy. Medicare pays
           for seven other radioactive sources used in
           brachytherapy-gold-198, low-dose iridium, yttrium-90, cesium-131,
           liquid iodine-125, ytterbium-169, and linear palladium-102. We did
           not examine how payment for those sources could be determined
           because sufficient data on those sources were not available in the
           2002 claims used to construct the sample of hospitals for our
           survey. Medicare did not pay for cesium-131, ytterbium-169, and
           linear palladium-102 in 2002, and gold-198, low-dose iridium,
           liquid iodine-125, and yttrium-90 together appeared on 2 percent
           of the approximately 22,000 claims for radioactive sources in that
           year. Although we did not examine how payment amounts could be
           determined for these seven sources, the analytical framework we
           used may apply to them as well.

Manufacturer and Provider Comments and Our Evaluation           
			  
			  Comments from external reviewers representing manufacturers of
           radioactive sources and providers of brachytherapy centered on
           three different areas: our recommendation to pay prospectively for
           iodine and palladium sources; our recommendation that CMS evaluate
           the unit cost of iridium; and payment for radioactive sources
           other than iodine, palladium, and iridium.

           Representatives from CAB disagreed with our recommendation to set
           prospective rates for iodine and palladium using OPPS claims data.
           They asserted that price variation due to the range of available
           iodine and palladium products makes it inappropriate to pay for
           sources prospectively based on averages. In their opinion, our
           finding that the unit costs of iodine and palladium sources are
           generally stable was compromised by limitations in our hospital
           survey-specifically, our exclusion of outlier data and the absence
           of source configuration information in many of the surveys we
           received from hospitals. ACCC stated that OPPS claims data are
           flawed and that prospective rates may be appropriate but only when
           a more accurate data source is available. They also noted, as did
           ACRO representatives, that costs incurred by hospitals for storing
           and handling radioactive sources were not represented in our
           survey results. Representatives from ASTRO, ABS, and ACRO agreed
           with our recommendation that payment can be based on an average.
           ACRO representatives cautioned that the data used to set the
           payment must be representative of different types of hospitals,
           and ABS representatives suggested that the data should reflect the
           increased use of stranded sources, which they stated are more
           costly but considered clinically advantageous by many physicians.

           Regarding our recommendation that CMS use OPPS claims data to
           evaluate the unit cost of iridium in order to determine a suitable
           separate payment methodology, representatives from CAB said the
           report accurately conveys the difficulties of identifying a
           per-unit cost for iridium. However, they disagreed with our
           recommendation because they said it would not be possible for CMS
           to fully evaluate a unit cost using OPPS claims data, which they
           asserted to be flawed. They stated that the cost of iridium varies
           substantially and unpredictably and would not be appropriate for
           prospective payment based on an average. Representatives from
           ASTRO, ABS, and ACRO agreed with our recommendation, although they
           expressed confidence that the unit cost of iridium would be found
           to vary substantially and unpredictably and would therefore be
           inappropriate for prospective payment based on an average cost
           calculated across hospitals.

           Finally, other comments focused on payment for radioactive sources
           other than iodine, palladium, and iridium. Representatives of
           ASTRO and CAB noted that we did not specifically address payment
           for the other radioactive sources used in brachytherapy-gold-198,
           low-dose iridium, yttrium-90, cesium-131, liquid iodine-125,
           ytterbium-169, and linear palladium-102-and ASTRO asked whether we
           would be making recommendations on payment for these other
           radioactive sources.

           Concerning the comments that variation in source price makes it
           inappropriate to pay prospectively for sources, as noted in the
           draft report, we based our finding on the low coefficient of
           variation we calculated from surveys received from our
           representative sample of hospitals. We do not believe that our
           exclusion of outlier data masked the true degree of price
           variation. We used standard statistical trimming principles, which
           resulted in the exclusion of only 2 percent of reported purchases
           of iodine and none of the reported purchases of palladium.
           Although many of the responding hospitals did not indicate on the
           survey the configuration of the sources purchased, we instructed
           hospitals to list prices for all sources purchased during the
           survey period. Therefore, the variation we calculated from
           hospital responses can be expected to reflect the range of
           products purchased by hospitals at the time. Representatives from
           ACRO and ABS stated that they believed the average prices
           presented in the draft report were consistent with prices for the
           types of sources-loose, low-activity sources-commonly used during
           the survey period. If costlier stranded sources have become more
           frequently used since the survey period of July 1, 2003 through
           June 30, 2004, as stated by representatives of ACRO and ABS, the
           use of those sources would be captured in OPPS claims data from
           subsequent years and reflected in future prospectively set rates.
           Regarding the concerns about basing prospectively set rates for
           iodine and palladium on OPPS claims data, as noted in the draft
           report, we based our recommendation on our comparison of average
           purchase prices for those sources from our hospital survey with
           CMS payment rates for 2003 and proposed payment rates for 2004,
           which CMS derived from OPPS claims data. Concerning the comments
           about the cost of storing and handling radioactive sources, CMS
           has provided guidance to hospitals on how they can receive
           reimbursement for those costs.

           With respect to our recommendation on payment for iridium, as
           noted in the draft report, we are recommending that CMS use its
           claims data to evaluate whether the range of costs comprising the
           average for a given year is substantial across hospitals and
           whether this average unit cost varied unpredictably over time.
           Consistent with its general practice for paying separately for
           technologies that are not new, CMS could pay for iridium at each
           hospital's cost if OPPS claims did not prove to be a reasonable
           source of data or if CMS determined that the unit cost varies
           substantially and unpredictably over time.

           As we noted in our response to comments received from CMS, we
           limited our examination of payment for radioactive sources to
           iodine, palladium, and iridium because sufficient data on the
           other sources were unavailable in the 2002 claims used to
           construct the sample of hospitals for our survey, and these three
           sources were billed on 98 percent of the claims for radioactive
           sources associated with brachytherapy.

           We are sending a copy of this report to the Administrator of CMS.
           We will also provide copies to others on request. The report is
           available at no charge on GAO's Web site at http://www.gao.gov .

           If you or your staffs have any questions, please contact me at
           (202) 512-7119 or [email protected] . Contact points for our
           Offices of Congressional Relations and Public Affairs may be found
           on the last page of this report. GAO staff who made major
           contributions to this report are listed in appendix III.

           A. Bruce Steinwald Director, Health Care

           List of Committees

           The Honorable Charles E. Grassley Chairman The Honorable Max
           Baucus Ranking Minority Member Committee on Finance United States
           Senate

           The Honorable Joe L. Barton Chairman The Honorable John D. Dingell
           Ranking Minority Member Committee on Energy and Commerce House of
           Representatives

           The Honorable William M. Thomas Chairman The Honorable Charles B.
           Rangel Ranking Minority Member Committee on Ways and Means House
           of Representatives

           The Honorable Nathan Deal Chairman The Honorable Sherrod Brown
           Ranking Minority Member Subcommittee on Health Committee on Energy
           and Commerce House of Representatives

           The Honorable Nancy Johnson Chairman The Honorable Pete Stark
           Ranking Minority Member Subcommittee on Health Committee on Ways
           and Means House of Representatives

           This appendix summarizes the sample design, methods for collecting
           and processing the data, and methods for estimating mean and
           median purchase prices for iodine and palladium sources used in
           brachytherapy.1 Though we were not able to estimate mean and
           median purchase prices for iridium, this appendix also includes a
           discussion of the data we received.

           We developed a random sample of hospitals to survey for the
           purchase prices of iodine, palladium, and iridium sources used in
           brachytherapy. The sample frame consisted of 949 hospitals that
           (1) had charged Medicare for radioactive sources during 2002, the
           most recent year for which usable data were available;2 (2) were
           still Medicare providers on July 1, 2004; and (3) were a subset of
           sample hospitals drawn for a survey we conducted of hospital
           outpatient drug prices.3 The sampling frame contained 98 percent
           of the 968 hospitals that submitted Medicare claims for the three
           brachytherapy sources in 2002. We drew a sample of 121 hospitals
           from the sample frame, on the basis of an expected response rate
           of 50 percent. Our results can be generalized to the larger
           population of hospitals providing iodine and palladium in the
           outpatient setting and meeting the above criteria.

           To improve the precision of our estimates of mean and median
           purchase price, we stratified the sample of hospitals. The
           objective was to obtain a sample of hospitals that mirrored the
           distribution of hospitals billing Medicare for these sources.
           Because we did not have a measure of purchase price of radioactive
           sources at the time we selected the sample, we used total hospital
           outpatient drug charges to Medicare as a proxy for purchase price
           variation. We used a regression model to identify stratification
           factors (such as teaching hospital status) that would maximize the
           difference in mean purchase price (as proxied by Medicare drug
           charges) among strata. We grouped hospitals into major teaching
           hospital, nonmajor teaching hospital, urban nonteaching hospital,
           and rural nonteaching hospital strata. We placed small hospitals
           in a separate stratum to ensure that hospitals with no or minimal
           charges for drugs during the first 6 months of 2003 were
           appropriately represented.

           In our sample design, we defined a major teaching hospital as a
           hospital for which the ratio of residents to the average daily
           number of patients was at least 1 to 4 and a nonmajor teaching
           hospital as one having a ratio of residents to patients of less
           than 1 to 4. We defined a hospital as urban if it was located in a
           county considered a metropolitan statistical area (as defined by
           the Office of Management and Budget) and rural if it was located
           in a county not considered a metropolitan statistical area. We
           defined a small hospital as a hospital with total Medicare drug
           charges of less than $10,000 during the first 6 months of 2003.

           To develop our survey of hospital purchase prices for radioactive
           sources, we interviewed representatives from the Coalition for the
           Advancement of Brachytherapy (CAB). CAB reports that it represents
           manufacturers of 90 percent of all brachytherapy sources and 100
           percent of high-dose rate brachytherapy sources in the United
           States. We also interviewed representatives of the American
           Brachytherapy Society, the American College of Radiation Oncology,
           the American Society for Therapeutic Radiology and Oncology, and
           the Association of Community Cancer Centers. We also interviewed
           representatives from six radioactive source manufacturers and
           seven hospitals and officials at the Centers for Medicare &
           Medicaid Services. In developing the survey, we obtained
           information from these associations and individual hospitals and
           pilot tested the survey with 5 hospitals prior to sending it to
           the entire sample of 121 hospitals. As a result, we clarified
           certain protocols and procedures but did not substantially change
           the survey instrument.

           The survey instrument was five pages long with one page for each
           radioactive source, one page for rebate data, and one page
           defining the terms in the previous pages. We collected data by
           reported purchase-that is, the purchase of a given quantity of a
           radioactive source at a particular price on a specific date. For
           iodine and palladium sources, we asked hospitals to provide the
           name of the manufacturer; the number of sources; the price per
           source; and certain characteristics of the sources purchased, such
           as radioactivity level. For iridium, we asked hospitals to provide
           the name of the manufacturer, the number of treatments delivered,4
           the source price, and the rebate eligibility. We also asked
           hospitals to report information on any rebates they received for
           these purchases.

           We contracted with Westat to administer the survey. Westat began
           data collection on September 27, 2004. Key components of the data
           collection protocol were

           o  a first mailing to the chief executive officer or chief
           financial officer of each hospital explaining the survey, followed
           by a telephone call to identify the main point of contact;
           o  a second mailing to the main contact outlining the data that
           were needed and describing the options for submitting the data;
           o  a follow-up telephone call to facilitate the main contact's
           understanding of the data collection, provide technical assistance
           as needed, and obtain some basic information about the hospital;
           and
           o  telephone calls at regular intervals to remind the hospitals to
           submit their data and to provide assistance as needed.

           Hospitals could submit data in one of three ways: by uploading
           electronic files through the study Web site, by sending an e-mail
           to the study address with data attached, or by sending electronic
           media or paper submissions through the mail. When our contractor
           received a brachytherapy survey from a hospital, it forwarded the
           survey to us for processing and analysis.

           Of the 121 hospitals surveyed, 62 hospitals submitted usable data,
           resulting in an overall response rate of 51 percent. We considered
           iodine and palladium data usable if we were able to identify the
           price per source and the number of sources purchased. We
           considered iridium data usable if we were able to identify the
           price per source and the number of fractions provided with the
           source. Of the 62 hospitals, 52 hospitals submitted usable data
           for iodine and 40 hospitals submitted usable data for palladium,
           with some providing data for both radioactive sources. Sixty-five
           percent of hospitals providing data for iodine and 63 percent of
           hospitals providing data for palladium were teaching hospitals.

           Our data were not sufficient to measure overall price differences
           by radioactivity level and other characteristics across each of
           the two types of sources. Specifically, hospitals did not indicate
           activity level for 37 percent of their reported purchases of
           iodine and 47 percent of their reported purchases of palladium.
           They did not indicate source configuration for 43 percent of their
           reported purchases of iodine and 51 percent of their reported
           purchases of palladium.5 Although we did not receive enough data
           from hospitals to reliably identify any price differences by
           source characteristic, we instructed hospitals to report all their
           purchases during the survey period. Therefore, any price variation
           due to source characteristic should be reflected in our data.

           We applied statistical trimming rules to eliminate outliers in the
           data. Accordingly, 2 percent of the reported purchases of iodine
           were trimmed, and none of the reported purchases of palladium were
           trimmed. The resulting data allowed us to calculate the mean and
           median price per source for iodine and palladium.

           Few hospitals reported receiving rebates. This is consistent with
           information we received from hospitals during interviews-that
           manufacturer rebates were not commonly provided for radioactive
           sources. Therefore, we did not factor rebates into our mean and
           median purchase prices.

           We determined that there were insufficient data to estimate the
           price of iridium. Of the 19 hospitals submitting iridium data, 11
           either did not provide number of treatments, reported a
           questionable iridium source price, or both. Eight hospitals
           reported an iridium source price and the number of treatments from
           which a unit cost could be calculated. However, among these 8
           hospitals there were inconsistencies in the data provided. Some
           hospitals reported the total price of their iridium contracts,
           which includes the cost of maintaining the iridium source, while
           other hospitals isolated the price of the iridium source within
           the contracts and reported that price.

           Estimates of Mean and Median Purchase Prices for Iodine and Palladium Sources
			  
			  This section describes the rationale and method for weighting the
           hospital sample, calculating mean purchase price, calculating
           median purchase price, and calculating the associated coefficients
           of variation-or standard error reflecting sample design and
           weights.

           Weighting the Hospital Sample
			  
			  To estimate hospitals' mean and median purchase prices for iodine
           and palladium sources, the sample hospitals' purchase price data
           were weighted to make them representative of the sample frame of
           hospitals from which the sample was drawn. The less likely that a
           hospital was sampled, the larger its weight. For example, if each
           hospital had a 1 in 10 probability of being sampled, its sample
           weight was 10. That is, each hospital in the sample represents 10
           hospitals in the sample frame. Consequently, if 5 hospitals in a
           sample bought a particular radioactive source, and the sample
           weight was 10, we estimate that 50 hospitals in the frame bought
           that radioactive source. In this report, we refer to sample
           weights as "hospital weights." Our sample was stratified, so all
           hospitals in a particular stratum (for example, major teaching
           hospitals) had the same weight. Since in our sample the
           probability of a hospital's being selected varied by stratum,
           hospitals in different strata had different weights.

           We calculated the hospital weight as

           Wjh =

           where

           o  Wjh denotes the hospital weight for the jth radioactive source
           in the hth stratum;
           o  Njh denotes the sample frame (the total number of hospitals)
           that according to Medicare outpatient claims, billed for the jth
           radioactive source in the hth stratum; and
           o  Rjh denotes the total number of hospitals in the hth stratum
           that purchased the jth radioactive source, according to their
           survey submissions.

           This weight recognizes that not all hospitals responded to our
           survey, since the weight's denominator is Rjh-the number of
           hospitals that responded to the survey and indicated that they
           bought the jth radioactive source.

           Mean Purchase Price Using Volume and Hospital Weights
			  
			  To summarize hospitals' purchase prices for iodine and palladium
           sources-reflecting purchases made, in many cases, at different
           prices and in different quantities-we calculated a mean purchase
           price for each radioactive source. This mean purchase price for a
           particular radioactive source is, in effect, a weighted mean. To
           reflect the differences among hospitals in purchase prices and
           purchase volumes, we used both the hospital weights and purchase
           volume as weighting variables in estimating the mean purchase
           price. All calculations were done at the individual purchase level
           but reflect the hospital and purchase volume weighting variables.

           The mean purchase price is estimated from our sample data, based
           on the following equation:

           Yj = (Sh Si y*jhi) / (ShSi x*jhi)

           where

           o  Nh represents the total number of hospitals in the hth stratum;
           o  nh represents the size of the sample of hospitals in the hth
           stratum;
           o  y*jhi = Sk yjhik, which represents the total dollar amount for
           the jth radioactive source listed on the kth invoice for the ith
           hospital in the hth stratum; and
           o  x*jhi = Sk xjhik, which represents the total number of units
           for the jth radioactive source listed on the kth invoice for the
           ith hospital in the hth stratum.

           The equation estimates the mean purchase price of a radioactive
           source as the ratio of the total amount purchased in dollars to
           the total number of units purchased.

           In addition to the mean purchase price, we calculated the
           estimated median of each radioactive source's purchase price. To
           calculate this median, we first applied volume and hospital
           weights to each hospital's purchases of a given radioactive
           source; we then ranked the weighted hospitals' purchase prices
           from lowest to highest and selected the midpoint of these prices.
           More precisely, the estimated median-based on the population
           cumulative density function F for hospital purchase prices-is
           given by

           X0.5 = inf {yjhik : F(yjhik) >= 0.5},

           where

           o  X0.5 denotes the median estimate of hospital purchase price for
           a particular radioactive source;
           o  yjhik denotes the unit purchase price of the jth radioactive
           source listed in the kth invoice record submitted in our survey by
           the ith hospital in the hth stratum;
           o  F, the cumulative density function, is the probability that the
           variable yjhik takes on a value greater than or equal to a
           particular value (in this case, 0.5);
           o  inf {a : b} refers to the minimum value of a, which satisfies
           the condition specified in b (in this case b is the condition that
           F(yjhik) >= 0.5); and
           o  the estimated population cumulative density function, F, is
           defined as

           F(x) = {Sh Si Sk I(yjhik =< x)} / Sh  Si Sk}.

           Coefficients of Variation for Mean Purchase Price
			  
			  To assess the precision of our estimates of the mean purchase
           price, we calculated coefficients of variation for the estimated
           mean purchase price. We also used the coefficients of variations
           as an indicator of price variability across hospitals. We
           estimated the mean purchase prices, median purchase prices, and
           the coefficients of variation for the means using specialized
           software for survey data analysis-SUDAAN(R).6
         
Appendix I: GAO Survey of Hospital Purchase Prices for Iodine,
Palladium, and Iridium Sources Used in Brachytherapy

                                 Sample Design

1Radioactive sources commonly used in brachytherapy include iodine and
palladium, which provide a prolonged, low dose of radioactivity, and
iridium, which provides a brief, high dose of radioactivity.

2Although 2003 data were available at the time the sample was constructed,
there was neither a separate billing code for iridium sources nor separate
billing codes for iodine and palladium sources used in prostate
brachytherapy.

3GAO, Medicare: Drug Purchase Prices for CMS Consideration in Hospital
Outpatient Rate-Setting, GAO-05-733R (Washington, D.C.: June 30, 2005).

                      Data Collection and Data Processing

4The survey asked for number of fractions, which refers to the number of
individual treatments provided.

5Of the iodine and palladium purchases that contained information on
activity level, about 90 percent were identified as low activity. Of the
iodine and palladium purchases that contained information on source
configuration, 86 percent of the iodine purchases and 95 percent of the
palladium purchases were identified as loose.

 Estimates of Mean and Median Purchase Prices for Iodine and Palladium Sources

Weighting the Hospital Sample

Mean Purchase Price Using Volume and Hospital Weights

Median Purchase Price Using Volume and Hospital Weights

Coefficients of Variation for Mean Purchase Price

6B.V. Shah, B.B. Barnwell, and G.S. Bieler, SUDAAN: User's Manual Release
7.5, vols. 1 and 2 (Research Triangle Park, N.C.: Research Triangle
Institute, 1997). SUDAAN (R) is a registered trademark of the Research
Triangle Institute.

Appendix II: Comments from the Centers for Medicare & Medicaid Services

Appendix III: GAO Appendix III: GAO Contact and Staff Acknowledgments

                                  GAO Contact
											 
 A. Bruce Steinwald, (202) 512-7119 or [email protected]

                                Acknowledgments

           In addition to the contact above, Maria Martino, Assistant
           Director; Shamonda Braithwaite; Melanie Anne Egorin; Hannah Fein;
           Nora Hoban; Dae Park; Dan Ries; Anna Theisen-Olson; Yorick F.
           Uzes; and Craig Winslow made contributions to this report.											 
										  

(290354)

          GAOï¿½s Mission

           The Government Accountability Office, the audit, evaluation and
           investigative arm of Congress, exists to support Congress in
           meeting its constitutional responsibilities and to help improve
           the performance and accountability of the federal government for
           the American people. GAO examines the use of public funds;
           evaluates federal programs and policies; and provides analyses,
           recommendations, and other assistance to help Congress make
           informed oversight, policy, and funding decisions. GAO's
           commitment to good government is reflected in its core values of
           accountability, integrity, and reliability.

           Obtaining Copies of GAO Reports and Testimony
			  
			  The fastest and easiest way to obtain copies of GAO documents at
           no cost is through GAO's Web site ( www.gao.gov ). Each weekday,
           GAO posts newly released reports, testimony, and correspondence on
           its Web site. To have GAO e-mail you a list of newly posted
           products every afternoon, go to www.gao.gov and select "Subscribe
           to Updates."

           Order by Mail or Phone
			  
			  The first copy of each printed report is free. Additional copies
           are $2 each. A check or money order should be made out to the
           Superintendent of Documents. GAO also accepts VISA and Mastercard.
           Orders for 100 or more copies mailed to a single address are
           discounted 25 percent. Orders should be sent to:

           U.S. Government Accountability Office 441 G Street NW, Room LM
           Washington, D.C. 20548

           To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax:
           (202) 512-6061

           To Report Fraud, Waste, and Abuse in Federal Programs
			  
			  Contact:

           Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail:
           [email protected] Automated answering system: (800) 424-5454 or
           (202) 512-7470

           Congressional Relations
			  
			  Gloria Jarmon, Managing Director, [email protected] (202) 512-4400
           U.S. Government Accountability Office, 441 G Street NW, Room 7125
           Washington, D.C. 20548

           Public Affairs
			  
			  Paul Anderson, Managing Director, [email protected] (202)
           512-4800 U.S. Government Accountability Office, 441 G Street NW,
           Room 7149 Washington, D.C. 20548

www.gao.gov/cgi-bin/getrpt? GAO-06-635 .

To view the full product, including the scope

and methodology, click on the link above.

For more information, contact A. Bruce Steinwald at (202) 512-7119 or
[email protected].

Highlights of GAO-06-635 , a report to congressional committees

July 2006

MEDICARE OUTPATIENT PAYMENTS

Rates for Certain Radioactive Sources Used in Brachytherapy Could Be Set
Prospectively

Generally, in paying for hospital outpatient procedures, Medicare makes
prospectively set payments that are intended to cover the costs of all
items and services delivered with the procedure. Medicare pays separately
for some technologies that are too new to be represented in the claims
data used to set rates. It also pays separately for certain technologies
that are not new, such as radioactive sources used in brachytherapy, a
cancer treatment. The Medicare Prescription Drug, Improvement, and
Modernization Act of 2003 required separate payment for the radioactive
sources. It also directed GAO to make recommendations regarding future
payment. GAO examined (1) how Medicare determines payment amounts for
technologies that are not new but are separately paid and (2) how payment
amounts for iodine, palladium, and iridium sources used in brachytherapy
could be determined.

What GAO Recommends

GAO recommends that Medicare

(1) in paying separately for iodine and palladium, use outpatient claims
to set prospective rates, and (2) use claims data to evaluate the unit
cost of iridium, so that a suitable separate payment methodology can be
determined. In response, CMS stated that it will take GAO's
recommendations into consideration.

In paying separately for technologies that are not new, the Centers for
Medicare & Medicaid Services (CMS) generally sets prospective rates based
on the average unit cost of the technologies across hospitals. For
example, CMS currently pays separate prospective rates for certain
high-cost drugs based on the mean per-unit acquisition cost, as derived by
CMS from data provided by drug manufacturers. A prospective rate is
desirable because basing a rate on an average encourages those hospitals
that provide the technology to minimize their acquisition costs. However,
when CMS determines that the unit cost of a technology designated for
separate payment varies substantially and unpredictably over time, or that
reasonably accurate data are not available, it pays each hospital its cost
for the technology. For example, CMS pays each hospital its cost for
corneal transplant tissue, because it determined that the fees eye banks
charge hospitals vary substantially and unpredictably.

GAO's analysis suggests that CMS could set prospective payment rates for
iodine and palladium because their unit costs are generally stable and CMS
can base the payments on reasonably accurate data. According to interviews
GAO conducted with hospitals and manufacturers, iodine and palladium have
an identifiable unit cost that does not vary unpredictably over time. In
addition, the results of GAO's survey of hospital purchase prices suggest
that the unit cost of iodine and palladium does not vary substantially.
Furthermore, GAO found that Medicare claims would be a reasonably accurate
source of data for setting prospective rates for these sources. GAO was
not able to determine a suitable methodology for paying separately for
iridium. In contrast with iodine and palladium, which are permanently
implanted in patients, iridium is reused across multiple patients, making
its unit cost more difficult to determine. Although GAO surveyed hospitals
on the unit cost of iridium, it did not receive sufficient data to
identify and evaluate an average unit cost across hospitals. However, CMS
has outpatient claims data from all hospitals that have used iridium. In
order to identify a suitable methodology for determining a separate
payment amount, CMS would be able to use these data to establish an
average cost and evaluate whether the cost varies substantially and
unpredictably.
*** End of document. ***