Medicare Billing: Commercial System Will Allow HCFA to Save Money, Combat
Fraud and Abuse (Testimony, 05/19/98, GAO/T-AIMD-98-166).

More than three years after GAO recommended that Medicare acquire
commercial software to detect inappropriate billings--which could save
hundreds of millions of dollars each year--the Health Care Financing
Administration (HCFA) has tested the software and plans to install it.
Incorrect billings, fraudulent and otherwise, cost Medicare about $1.7
billion in improper payments in 1997. This report analyzes HCFA's
progress in testing and acquiring a commercial system for identifying
inappropriate Medicare bills, the consequences of HCFA's initial
management decisions, and its current plans for immediate
implementation.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  T-AIMD-98-166
     TITLE:  Medicare Billing: Commercial System Will Allow HCFA to Save 
             Money, Combat Fraud and Abuse
      DATE:  05/19/98
   SUBJECT:  Fraud
             Audits
             Commercial products
             Health care programs
             Program abuses
             Claims processing
             Erroneous payments
             Medical expense claims
             Contract administration
             Management information systems
IDENTIFIER:  Medicare Program
             Civilian Health and Medical Program of the Department of 
             Veterans Affairs
             VA TRICARE Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed 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.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Before the Subcommittee on Oversight and Investigations, Committee on
Commerce,
House of Representatives

For Release on Delivery
Expected at
2 p.m.
Tuesday,
May 19, 1998

MEDICARE BILLING - COMMERCIAL
SYSTEM WILL ALLOW HCFA TO SAVE
MONEY, COMBAT FRAUD AND ABUSE

Statement of Joel C.  Willemssen
Director, Civil Agencies Information Systems
Accounting and Information Management Division

GAO/T-AIMD-98-166

GAO/AIMD-98-166T


(511251)


Abbreviations
=============================================================== ABBREV

  CCI -
  CHAMPVA -
  HCFA -
  HHS -
  OMB -
  TRICARE -

============================================================ Chapter 0

Mr.  Chairman and Members of the Subcommittee: 

We are pleased to join you today in examining the actions of the
Health Care Financing Administration (HCFA)--an agency of the
Department of Health and Human Services (HHS)--in assessing the
benefits of commercial claims-auditing software for nationwide
implementation with its Medicare processing systems.  Such software
can be a critical tool in helping HCFA address fraud and abuse in the
Medicare program.  Fraud and abuse within Medicare is pervasive;
accordingly, we have designated the program a high-risk area for the
federal government.\1 According to HHS' Office of Inspector General,
incorrect coding by physicians cost Medicare about $1.7 billion in
improper payments during fiscal year 1997, an increase of about $630
million from fiscal year 1996. 

Commercial systems to detect inappropriate coding/billing have been
available for several years.  As early as 1991 commercial firms
marketed specialized auditing systems that identified inappropriately
coded claims.  Both the HHS Inspector General and we have noted the
potential value of such systems.  In 1991, the Inspector General
reported that commercially available claims-auditing systems had the
potential to save $12 million annually at one Medicare processing
site alone.\2 Similarly, in 1995 we reported that such systems could
save Medicare about $600 million annually if implemented on a
nationwide basis.\3

Instead of acquiring available commercial software, however, HCFA
initially chose to develop its own system.  In 1991, HCFA directed
its Medicare insurance carriers to begin developing edits to be
included in its claims-auditing systems.  In 1994, it awarded a
contract for further development of these edits, which it called the
correct coding initiative--a system HCFA now owns and operates. 
According to HCFA, these edits helped Medicare save about $217
million in 1996 by successfully identifying inappropriate claims. 
Now, 3 years after our recommendation, HCFA has tested a commercial
system and found that it could indeed save substantially higher
sums--in this case, about $465 million annually
in addition to the savings resulting from the coding initiative. 
Consequently, HCFA now plans to acquire this commercial
claims-editing capability as soon as possible. 

In a report being released today, we analyzed HCFA's progress in
testing and acquiring a commercial system for identifying
inappropriate Medicare bills.\4 My statement today will discuss how
HCFA tested this commercial system, its initial management decisions
and their consequences, and its current plans for immediate
implementation. 


--------------------
\1 High-Risk Series:  Medicare (GAO/HR-97-10, February 1997). 

\2 Manipulation of Procedure Codes by Physicians to Maximize
Reimbursement, Office of Inspector General, Department of Health and
Human Services, CIN:  A-03-91-00019, August 30, 1991. 

\3 Medicare Claims:  Commercial Technology Could Save Billions Lost
to Billing Abuse (GAO/AIMD-95-135, May 5, 1995). 

\4 Medicare Billing:  Commercial System Could Save Hundreds of
Millions Annually (GAO/AIMD-98-91, April 15, 1998). 


   HCFA'S TEST METHODOLOGY
---------------------------------------------------------- Chapter 0:1

HCFA used a test methodology that was comparable with processes
followed by other public insurers who have successfully tested and
implemented such commercial systems.  Other public insurers--such as
the military's TRICARE, Veterans Affairs' CHAMPVA, and the Kansas and
Mississippi Medicaid offices--each used four key steps to test their
claims-auditing systems prior to implementation.  Specifically, they
(1) performed detailed comparisons of their payment policies with
systems' edits to determine where conflicts existed, (2) modified the
commercial systems' edits to comply with their payment policies, (3)
integrated the systems into their claims payment systems, and (4)
conducted operational tests to ensure that the integrated systems
processed claims properly.  This is a comprehensive approach that
requires significant time to complete.  For example, TRICARE took
about 18 months for two sites and allowed about 2 years for its
remaining nine sites. 

HCFA's approach was similar.  From contract award on September 30,
1996, through its conclusion 15 months later at the end of December
1997, both HCFA and contractor staff made significant progress in
integrating the test commercial system and evaluating its potential
for Medicare use nationwide.  HCFA used both a policy evaluation team
and a technical team to concentrate separately on these aspects of
the test. 

A detailed comparison of the commercial system's payment policies
with those of Medicare identified conflicting edits--inconsistencies
that in some cases would increase and in others decrease the amount
of the Medicare payments.  For example, the commercial system would
pay for the higher cost procedure of those deemed mutually exclusive,
while Medicare dictates paying for the lower cost procedure.  (A
mutually exclusive procedure would be, for instance, the same
patient's receiving both an open and a closed treatment for a
fracture.) Conversely, the commercial claims-auditing system would
deny certain payments for assistant surgeons, while Medicare allows
them.  These and all other identified conflicts were provided to the
vendor, who modified the system's edits to make them consistent with
HCFA policy. 

The technical team carried out three critical tasks.  First, it
developed the design specifications and related computer code
necessary for integrating the commercial system into the Medicare
claims-processing software.  Second, it integrated the
claims-auditing system into the system that processes Medicare part B
claims.\5 Finally, the team conducted numerous tests of the
integrated system to determine its effect both on processing speed
and accuracy.  HCFA management was kept apprised of the status of the
test through regular progress reports and frequent contact with the
project management team. 

HCFA found that the edits in this commercial system could save
Medicare up to $465 million annually by identifying inappropriate
claims.  Specifically, HCFA's analysis showed that the system's
mutually exclusive and incidental procedure edits\6 would save about
$205 million, and the diagnosis-to-procedure edits\7 could save about
$260 million.  HCFA's analysis was based on a national sample of paid
claims already processed by Medicare part B and audited for
inappropriate coding with HCFA's internal software.  We reviewed the
reports of HCFA's estimated savings, but did not independently verify
the national sample from which these savings were derived.  However,
the magnitude of savings--$682 million, including the savings derived
from HCFA's internal software, which HCFA reported at $217 million
for 1996--is in line with our 1995 estimate that about $600 million
in annual savings are possible.\8

On November 25, 1997, HCFA officials notified the Administrator of
the successful test of the commercial system.  This was a far
different conclusion than the one reported by HCFA 2 months earlier,
while testing was ongoing.  At a September 29, 1997, hearing before
this subcommittee, a senior HCFA official stated that the agency was
testing the commercial system as a stand-alone system against
Medicare's claims-processing system.  He testified that "for the
month of August, our system, the CCI system [correct coding
initiative] achieves savings of $422,000 more than the [commercial]
system would have achieved if that would have been what we were
using.  We were outperforming a [commercial] product." However, as we
testified at that same hearing,\9 the test needed to compare the
commercial system as a supplement to the existing one, rather than as
a replacement.  Before HCFA completed its test it did compare the
commercial system as a supplement.  This comparison showed that
commercial systems offer the potential for substantial Medicare
savings. 


--------------------
\5 Medicare part B claims are those submitted by providers, such as
physicians, laboratories, and outpatient clinics; part A covers
hospitals, home health agencies, and other in-patient-facility care. 

\6 An incidental procedure is one that is clinically integral to and
covered by the primary procedure, such as control of intraoperative
bleeding with a tonsillectomy. 

\7 Diagnosis-to-procedure edits compare bills for procedures that are
unexpected for a given diagnosis, such as a corneal transplant with a
diagnosis of pneumonia. 

\8 As with any claims editing, some of the denied items will likely
be appealed and paid.  The estimates are not adjusted for this. 

\9 Medicare Automated Systems:  Weaknesses in Managing Information
Technology Hinder Fight Against Fraud and Abuse (GAO/T-AIMD-97-176,
September 29, 1997). 


   MANAGEMENT DECISIONS COULD HAVE
   COST MONTHS AND HUNDREDS OF
   MILLIONS OF DOLLARS
---------------------------------------------------------- Chapter 0:2

Despite the successful outcome of the test, two early management
decisions, if left unchanged, would have significantly delayed
national implementation of claims-auditing software in the Medicare
program.  First, the use of the test system was limited to its single
Iowa location, thereby requiring another contract for nationwide
implementation.  Second, HCFA's initial plan following the test was
to proceed with developing its own edits, rather than to acquire
those available through commercial systems.  This plan would not only
have required additional time before implementation, but could well
have resulted in a system less comprehensive in its capacity to flag
suspect claims than what is available commercially.  I would now like
to provide some details surrounding both of these decisions. 


      LIMITED TEST CONTRACT
      PRECLUDED SPEEDY NATIONWIDE
      IMPLEMENTATION
-------------------------------------------------------- Chapter 0:2.1

HCFA's contract limited the use of the test system to its Iowa site
and did not include a provision for implementation throughout the
Medicare program if the test proved successful.  As a result,
additional time will now be needed to award another contract to
implement the test system's claims-auditing software or any other
approach nationwide.  According to a HCFA contracting official, it
could take as much as a year to award another contract using "full
and open" competition--the method normally used for such
implementation.  This entails preparing for and issuing a request for
proposals, evaluating the resulting bids, and awarding the contract. 
HCFA's estimated savings of up to $465 million per year demonstrates
the costs associated with delays in implementing such payment
controls nationwide. 

Along with additional time and lost savings from the lack of early
nationwide implementation, awarding a new contract could result in
additional expense to either develop new edits or for substantial
rework to adapt the new system's edits to HCFA's payment policies if
a contractor other than the one performing the original test wins the
competition.  If another contractor were to become involved, much of
the work HCFA performed during the test period would have to be
redone.  Specifically, another company's claims-auditing edits would
have to be evaluated for potential conflict with agency payment
policy. 

Other options were open to HCFA from the beginning.  For example,
HCFA could have followed the approach used by TRICARE, whose contract
provided for a phased, 3-year implementation at its 11 processing
sites following successful testing.  According to HCFA's
Administrator, the agency is doing what it can to avoid any delays
resulting from the limited test contract.  The Administrator said
HCFA is evaluating legal options to determine if other contracting
avenues are available, options that would allow expedited national
implementation of commercial claims-auditing software. 


      INITIAL HCFA PLAN TO DEVELOP
      OWN EDITS EXPENSIVE AND
      INEFFECTIVE
-------------------------------------------------------- Chapter 0:2.2

In reporting the test results, HCFA representatives recommended that
the HCFA Administrator award a contract to develop HCFA-owned
claims-auditing edits to supplement the correct coding initiative,
rather than acquire these edits commercially.  They provided the
following rationale:  First, this approach could cost substantially
less than commercial edits because HCFA would have the option of
changing contractors for edit updates, it would not have to pay
annual licensing fees, and the developmental cost would be much less
than purchasing the capability commercially.  Second, according to
HCFA officials, this approach would result in HCFA-owned
claims-auditing edits, which are in the public domain and
consequently allow HCFA to disclose its policies and coding
combinations to providers, as it currently does with the correct
coding initiative edits.  Officials also explained that if a
commercial vendor bid, won, and agreed to allow its claims-auditing
edits to enter the public domain, HCFA would allow the vendor to
start with its existing edits, which should shorten development time. 

We found serious flaws in this approach--in terms of cost, overall
effectiveness, and underlying assumptions.  First, upgrading the
edits by moving from the initial contract developer to one unfamiliar
with them would not be easy or inexpensive; it is a major task,
facilitated by a thorough clinical knowledge of the existing edits. 
Second, the annual licensing fees that HCFA would avoid with its own
edits would be offset to some degree by the need to pay a contractor
with the clinical expertise to keep the edits current.  Third, while
the commercial software could cost more than developing HCFA-owned
edits, this increased cost has already been more than justified by
HCFA's test results demonstrating that commercial edits provide
significantly more Medicare savings.  Finally, the cost of delay is
significant:  HCFA has realized no savings from such commercial
software over the past 6 years. 

Moreover, we found that HCFA's plan to fully disclose its edits to
the medical community is not required by federal law and is not
followed by other public insurers; it could also result in limiting
the number of potential contractors with an interest in bidding.  In
May 1995 HHS' Office of General Counsel informed HCFA that no federal
law or regulation precludes it from protecting the proprietary nature
of the edits and the related computer logic used in commercial
claims-auditing systems.  Further, HCFA's Deputy Director of the
Provider Purchasing and Administration Group stated that the agency
has no explicit Medicare policy requiring it to disclose to providers
the specific edits used to audit their claims.  Rather than
disclosing the edits, other public insurers, such as CHAMPVA and
TRICARE, notified providers that they were implementing the system,
and supplied examples of categories of edits that would be used to
check for such disparities as mutually exclusive claims. 

Finally, while it is true that development time would likely be
shortened if a commercial claims-auditing vendor were awarded the
contract and used its existing edits as a starting point, it is
doubtful that such vendors would bid on the contract if resulting
edits were to be in the public domain.  This response was confirmed
to us by an executive of a company that has already developed a
claims-auditing system; he said he would not enter into such a
contractual agreement if HCFA insisted on making the edits public
because this would result in the loss of the proprietary rights to
his company's claims-auditing edits. 

HCFA's plan to develop its own edits was also inconsistent with
Office of Management and Budget (OMB) policy in acquiring information
resources.\10 HCFA has not demonstrated the cost-effectiveness of its
plan to develop edits internally.  In fact, a prime example showing
otherwise is HCFA's own estimate that every year it delays
implementing claims-auditing edits of the caliber of those used in
the commercial test system in Iowa, about $465 million in savings
could be lost. 

Developing comprehensive HCFA-owned claims-auditing edits could take
years, during which time hundreds of millions of dollars could be
lost annually due to incorrectly coded claims.  To illustrate:  HCFA
began developing its database of edits in 1991 and has continued to
improve it over the past 6 years.  While HCFA reported that its
correct coding initiative identified $217 million in savings in 1996
(in the mutually exclusive and incidental procedure categories), this
database did not identify an additional $205 million in those
categories identified by the test edits, nor does it address the
diagnosis-to-procedure category, where the test edits identified an
additional $260 million in possible savings.  HCFA has no assurance
that its own edits would be as effective as those available
commercially. 


--------------------
\10 OMB Circular A-130, 8b(5)(b) states that in procuring information
resources, agencies shall "acquire off-the-shelf software from
commercial sources, unless the cost-effectiveness of developing
custom software to meet mission needs is clear and has been
documented."


   CURRENT PLANS WILL EXPEDITE
   IMPLEMENTATION OF COMMERCIAL
   EDITS
---------------------------------------------------------- Chapter 0:3

This past March, after considering our findings and other factors,
the HCFA Administrator said that the agency's plans had changed.  She
said that HCFA plans to begin immediately to acquire and implement
commercial claims-auditing software in as expedited a manner as
possible. 

We are encouraged that after a slow start, HCFA now plans to move
quickly to take advantage of the comprehensive claims-auditing
capability that is available, and we are looking forward to seeing
HCFA's milestones for expeditiously implementing this capability. 
Typically, such milestones would include dates for awarding a
contract for the commercial claims-auditing edits, initiating and
completing implementation at the first Medicare site, and
implementing the edits at the remaining Medicare processing sites. 


-------------------------------------------------------- Chapter 0:3.1

Mr.  Chairman, this concludes my statement.  I would be happy to
respond to any questions that you or other members of the
Subcommittee may have at this time. 

*** End of document. ***