Supplemental Security Income: Additional Actions Needed to Reduce Program Vulnerability to Fraud and Abuse

Published by the Government Accountability Office on 1999-09-15.

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

                  United States General Accounting Office

GAO               Report to the Honorable Henry A.
                  Waxman, Ranking Minority Member,
                  Committee on Government Reform,
                  House of Representatives

September 1999
                  SECURITY INCOME
                  Additional Actions
                  Needed to Reduce
                  Program Vulnerability
                  to Fraud and Abuse

      United States
GAO   General Accounting Office
      Washington, D.C. 20548

      Health, Education, and
      Human Services Division


      September 15, 1999

      The Honorable Henry A. Waxman
      Ranking Minority Member
      Committee on Government Reform
      House of Representatives

      Dear Mr. Waxman:

      The Supplemental Security Income (SSI) program, administered by the
      Social Security Administration (SSA), is the nation’s largest cash assistance
      program. At the end of 1998, the SSI program was paying benefits to about
      5.2 million needy blind and disabled recipients and 1.3 million needy aged
      recipients. Program expenditures for the year totaled about $29 billion
      ($25 billion and $4 billion, respectively). Over the next 10 years, the
      combined federal cost alone for SSI and related Medicaid benefits is
      estimated at $122,000 per recipient.

      In the early 1990s, media reports and congressional hearings alleged that
      some SSI recipients may have improperly gained access to program
      benefits by feigning or exaggerating disabilities with the help of
      middlemen and medical providers. In 1995, we reported that some
      ineligible non-English-speaking applicants had obtained SSI benefits
      illegally by using middlemen, particularly interpreters, who had provided
      inaccurate translations or had coached applicants on how to appear
      disabled.1 As a result, we recommended that SSA develop a more
      aggressive and programwide strategy to obtain and share data about
      interpreters and middlemen. Similarly, some providers have submitted
      misleading diagnoses for SSI applicants, claiming mental impairments and
      other conditions that are difficult to verify, to help applicants obtain
      medical eligibility for SSI benefits.

      In light of these long-standing concerns, you asked us to (1) determine the
      extent to which SSI is vulnerable to individuals who obtain eligibility by
      feigning disabilities with the help of middlemen and medical providers;
      (2) describe SSA’s methods for preventing, detecting, and responding to
      this type of program fraud and abuse; and (3) identify additional strategies
      SSA could use to more effectively address this problem. Some of SSA’s
      actions discussed in this report were partially responsive to the
      recommendation in our earlier report.

      Supplemental Security Income: Disability Program Vulnerable to Applicant Fraud When Middlemen
      Are Used (GAO/HEHS-95-116, Aug. 31, 1995).

      Page 1                               GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                   To conduct our work, we met with a variety of personnel involved in the
                   administration of the SSI program, including claims representatives, claims
                   adjudicators, fraud investigators, administrative law judges (ALJ), and SSA
                   administrators. We also obtained and analyzed several databases to assess
                   program vulnerability and consider the value of potential changes in
                   administrative controls. We focused on six states with large SSI
                   populations (together, these states comprise about 40 percent of all SSI
                   recipients) or where SSA has experienced serious problems with disability
                   fraud and abuse.

                   To identify program vulnerability to fraud and abuse and possible ways to
                   enhance SSA prevention strategies, we enlisted the assistance of several
                   investigative organizations. To protect the confidentiality of their records,
                   these organizations provided information under special arrangements.
                   This information identified medical providers who had been investigated
                   or who were being investigated for fraudulent activities involving
                   Medicaid, Medicare, and the payment of private health insurance benefits.
                   We did not solicit information on the results of these investigations for
                   several reasons. In some cases, the outcome of the investigation was not
                   readily available because the case was still open, the organization lacked
                   the resources to provide a complete listing of the outcomes, or the charges
                   could not be substantiated. In the majority of cases, investigations do not
                   result in an admission of guilt or a conviction of fraud. An investigation
                   may be closed, for example, because a settlement is reached or the subject
                   agrees to make restitution in exchange for nonprosecution. However,
                   since investigations are not initiated on the basis of a simple complaint, we
                   included all investigated providers in our analysis.2 We use the term
                   “suspicious” to characterize medical providers or middlemen who had
                   been or were being investigated by these organizations at the time of our

                   Our work was done between October 1997 and May 1999 in accordance
                   with generally accepted government auditing standards. See appendix I for
                   additional information on our scope and methodology.

                   Although the number of people who have feigned injuries or illnesses to
Results in Brief   obtain SSI benefits is unknown, the SSI program is vulnerable to this type of
                   fraud and abuse. First, many SSI beneficiaries’ impairments are difficult to
                   objectively verify. From a sample file of beneficiaries—developed by SSA to

                    For example, one organization told us that before starting an investigation, it had to have a written
                   statement of facts supporting the position that false claims had been filed and the false claims did not
                   appear to be the result of an honest billing error or misinterpretation of requirements.

                   Page 2                                   GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

    research characteristics of the SSI population—we found that more than
    60 percent had such impairments, including psychoses, schizophrenia, and
    other mental disorders, as well as a range of physical disorders. In
    addition, providers who have been investigated for defrauding Medicaid,
    Medicare, or private insurance companies furnished at least some portion
    of the supporting medical evidence for more than 12,000 (6 percent) of the
    208,000 SSI disabled recipients in the six states we examined. Finally, over
    96 percent of the 158 officials and staff we interviewed said they believed
    that the practice of middlemen helping people improperly qualify for SSI
    benefits has continued.

    SSA has taken several actions, both on its own and in response to
    legislation, to reduce the program’s vulnerability to this and other forms of
    fraud. SSA has

•   established pilot fraud investigation teams in five states during 1998 to
    examine individual cases where significant fraud and abuse is suspected,
•   developed new policies and procedures to make it easier to deny claims or
    terminate benefits when program fraud or abuse is detected, and
•   strengthened its ability to handle its non-English-speaking clients.

    These steps have achieved positive results. For example, as of March 31,
    1999—just 6 to 14 months after they began their work—the pilot teams in
    five locations have provided information that contributed to cessations
    and denials of SSI benefits worth about $11 million. The overall
    effectiveness of SSA’s actions, however, has been limited by several factors.
    First, front-line staff largely rely on their experience and perceptions to
    identify suspicious claims; they lack other valuable information, such as
    the names of middlemen and medical providers suspected of fraudulent or
    abusive practices by other employees or organizations, that could help
    them judge a claim’s validity. In addition, SSA and Disability Determination
    Services (DDS) staff said that they do not always follow the new
    procedures because they believe the procedures conflict with agency work
    incentives that stress speed in processing claims and because they believe
    they are not adequately protected from legal liability that could arise if
    they were to follow claims denial procedures. They also question the
    agency’s commitment to fighting fraud, since they repeatedly see the same
    suspicious middlemen and medical providers involved in SSI cases, despite
    previous referrals for investigation.

    In our view, several additional types of actions could reduce SSI’s
    vulnerability to fraud and abuse by middlemen and medical providers. SSA

    Page 3                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

             could establish a national information system that identifies suspicious
             middlemen and medical providers. These type of data would help
             front-line staff, on whom SSA relies to fight program fraud and abuse, to
             better identify cases that warrant closer scrutiny. Further, SSA needs to
             (1) implement our recommendation from a previous report to reevaluate
             its work credit and incentive structure to encourage greater attention to
             fraud detection and (2) reexamine its policy regarding SSA-provided
             interpreters. The Congress may also wish to protect staff from legal
             liabilities that might arise from following new claims denial procedures.

             The SSI program, authorized under title XVI of the Social Security Act in
Background   1972, provides cash benefits to blind, disabled, and aged individuals whose
             income and resources are below certain specified levels. To qualify for
             benefits, blind and disabled individuals must meet medical and functional
             disability criteria as well as financial eligibility requirements.

             The benefit application process begins with initial interviews of applicants
             at any of SSA’s 1,298 field offices. During these interviews, SSA staff solicit
             information on applicants’ financial situation and the disability being
             claimed. Applicants can work directly with SSA staff or use middlemen
             who provide services, often for a fee, such as help in completing forms,
             interpreting for non-English-speaking individuals, and offering advice on
             how to navigate the application process. Interpreters supplied by SSA are
             also available to help non-English-speaking applicants through this

             The field offices forward the disability information gathered during the
             initial interviews to one of 54 state DDS offices, which are responsible for
             deciding if applicants meet the program’s criteria for disability. These
             offices develop evidence related to a claim by obtaining reports from the
             medical sources that an applicant has used to treat or diagnose the
             impairment. If necessary, the DDS office may require an applicant to have
             an SSA-paid medical (consultative) examination to evaluate and document
             the impairment further. At this stage, non-English-speaking applicants
             again may rely on either their own or SSA-supplied interpreters to help
             them answer questions raised by DDS staff during the adjudication process
             and by medical providers during required SSA medical exams.

             Individuals who are found eligible to receive SSI benefits are subject to
             periodic reevaluations of their financial status, known as
             redeterminations, and of their medical status, known as continuing

             Page 4                         GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

disability reviews (CDR). During a redetermination, the financial factors
related to the recipient’s eligibility—essentially earnings, assets, and
current living arrangements—are reviewed. A CDR is conducted to
determine whether a person is still medically and functionally unable to
work. Situations that can trigger a CDR include medical evidence
concluding that a condition is expected to improve, substantial earnings
reported to SSA that indicate a recipient is working, and medical
improvement reported to SSA by a vocational rehabilitation agency.

Individuals dissatisfied with SSA decisions to deny or terminate benefits
(whether for financial or medical reasons) can use SSA’s administrative
review process. First, a dissatisfied person may request a reconsideration
of the adverse decision. The reconsideration is an independent
examination, by a specially trained DDS staff member, of all evidence on
record plus any further evidence and information submitted by the
claimant or the claimant’s representative. If there is disagreement with the
reconsidered determination, a hearing before an ALJ from SSA’s Office of
Hearings and Appeals (OHA) may be requested. At the hearing level, OHA
personnel examine the evidence of record; the client or the client’s
representative may also introduce new evidence and new impairments.
Finally, if a disagreement remains with the OHA decision, persons may
request a review by SSA’s Appeals Council.

At each of these levels, the input of middlemen and medical providers can
be a factor. Middlemen and medical providers can have different motives
for assisting persons in obtaining SSI benefits. Some middlemen and
medical providers help individuals obtain SSI benefits because they want to
help persons who have backgrounds similar to their own or who need
financial assistance. Others are motivated by financial gain. Middlemen
often charge fees for their services contingent upon applicants becoming
eligible for program benefits. In most states, medical providers can bill
Medicaid for treating SSI recipients, and improper Medicaid billings have
been a long-standing problem.

Investigating possible fraudulent activity is the responsibility of SSA’s
Office of the Inspector General (OIG). In the past, SSA’s OIG has cooperated
in and reported on the results of investigations involving middlemen and
medical providers. For example, in December 1997, the OIG reported on an
extended family in Georgia that consisted of 181 members receiving SSI
benefits. DDS personnel performed CDRs on 151 of them and terminated
benefits to 88. The investigation disclosed that a psychiatrist who was
responsible for helping many of these individuals qualify for SSI benefits

Page 5                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                           then billed Medicaid for their treatment after they were awarded SSI

                           Our analysis of the SSI program indicates that SSI is inherently vulnerable to
SSI Is Inherently          people who, with the help of others, feign their impairments to obtain
Vulnerable to              benefits. Over 60 percent of SSI disability cases from an SSA statistical
Individuals Feigning       sample involved impairments that are difficult to objectively verify, and
                           thousands of SSI recipients in the six states we studied used suspicious
Disabilities With the      medical providers to gain access to the program. Middlemen also play a
Help of Providers and      significant role in SSI fraud and abuse, according to SSA officials and
                           front-line staff.
SSI Program Recipients’    OIG fraud investigators, SSA officials, and DDS staff told us that certain types
Impairments Are Often      of impairments that can be feigned are difficult to objectively verify. Some
Difficult to Objectively   specific impairments that they identified as falling into this category
                           include mental retardation, post-traumatic stress syndrome, and
Verify                     depression. Back impairments, unrelenting severe pain, and vision
                           problems that lack objective evidence, such as clearly documented
                           pathology or treatment history, are also potentially exaggerated or feigned

                           Our analysis of a sample file of SSI beneficiaries—which SSA developed to
                           research characteristics of the SSI population—shows that the majority of
                           disabled recipients had the types of impairments that SSA and DDS staff
                           considered susceptible to feigning.3 Specifically, we found that 64 percent
                           of disabled recipients in the April 1998 version of the sample file had
                           impairments susceptible to feigning. Table 1 shows the estimated number
                           of adults and children with impairments that SSA and DDS staff believe are
                           difficult to objectively verify within broad categories of impairments.

                            About 400,000 records (77.2 percent of the file) had information on SSI recipients’ disabilities,
                           representing over 4 million disabled SSI recipients.

                           Page 6                                    GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

Table 1: Estimated Number of SSI
Child and Adult Recipients in                                                                                                     Percent of
April 1998 Sampling by Category of                                                   Child            Adult             Total          total
Impairments Considered by SSA and        All SSI disabled recipients
DDS Staff to Be Vulnerable to Feigning   with identifiable impairments            799,730        3,251,580        4,051,310             100.0
                                         Recipients with mental impairments susceptible to feigning
                                         Psychoses and neuroses                   190,940          707,180          898,120               22.2
                                         Schizophrenia                               3,890         339,170          343,060                   8.5
                                         Mental retardation                       302,870          738,570        1,041,440               25.7
                                         Recipients with physical impairments susceptible to feigning
                                         Back disorders                                   •        136,490          136,490                   3.4
                                         Muscle, ligament, fascia
                                         disorders, sprains, and strains                  •          21,170           21,170                  0.5
                                         Epilepsy                                   11,950           30,870           42,820                  1.1
                                         Vision problems                                  •          44,930           44,930                  1.1
                                         Chronic pulmonary
                                         insufficiency                                    •          59,490           59,490                  1.5
                                         Total recipients with
                                         impairments susceptible to
                                         feigning                                 509,650        2,077,870        2,587,520               64.0
                                         Note: The data in the table represent persons who have impairments that are difficult to
                                         objectively verify. They do not suggest that individuals with these impairments are feigning them.
                                         Percentages have been rounded to the nearest 10th of a percentage point. The sampling errors
                                         for all but one of the numerical estimates in this table do not exceed plus or minus 6 percent of
                                         the estimate at the 95-percent confidence level.

Suspicious Medical                       From records maintained by SSA and other entities, we found that
Providers Are Assisting SSI              suspicious medical providers have helped individuals obtain or maintain
                                         SSI benefits and roughly estimated the program’s vulnerability to these
Applicants and Recipients
                                         types of activities. Using SSA records for SSI beneficiaries in the six states
                                         we studied, we identified 208,085 SSI recipients who—through a
                                         determination or a redetermination conducted between January 1, 1997,
                                         and June 30, 1998—were found eligible for SSI benefits on the basis of an
                                         impairment that was difficult to objectively verify. From government
                                         agencies that pay Medicare and Medicaid benefits and a private
                                         organization that supports health insurance companies, we obtained lists
                                         of suspicious medical providers and compared them with lists of providers
                                         used by these SSI recipients.

                                         Page 7                                  GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                                        Of the 208,085 SSI recipients who had impairments difficult to objectively
                                        verify, we found that 12,565 (about 6 percent) had used doctors identified
                                        as suspicious.4 (See table 2.)

Table 2: SSI Recipients With
Impairments Difficult to Objectively                                                      SSI recipients who
Verify Who Used Suspicious Medical                                  SSI recipients with         used medical
Providers to Support Their Disability                             impairments difficult providers suspected
Claim, by State                         State                      to objectively verify   of fraud or abuse                     Percentage
                                        California                                112,240                       7,028                      6.3
                                        Florida                                    28,764                       1,759                      6.1
                                        Georgia                                    12,969                         711                      5.5
                                        Louisiana                                    8,162                        551                      6.8
                                        Massachusetts                              15,668                       1,074                      6.9
                                        New York                                   30,282                       1,442                      4.8
                                        Total                                     208,085                      12,565                      6.0

                                        Of the suspicious providers identified by benefit-paying entities, we found
                                        that 1,447 assisted these SSI recipients in obtaining or maintaining benefits.
                                        Many assisted numerous SSI clients. For example, in California, 11
                                        providers had assisted from 100 to 300 SSI recipients with impairments
                                        difficult to objectively verify. We also found that one medical practice had
                                        submitted evidence for 632 recipients with such impairments. (See table

                                         The lists provided to us contained the names of hospitals, group practices, and individual medical
                                        providers suspected or convicted of fraudulent or abusive activity. In our analysis, we excluded
                                        hospitals as a suspect source of medical information because hospitals have many providers and we
                                        could not identify which providers were under investigation. If we had included the hospitals, the
                                        number of recipients with questionable medical sources would have risen from 12,565 to 34,153
                                        (16.4 percent of the recipients with impairments difficult to verify in the six states we analyzed).

                                        Page 8                                  GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

Table 3: Suspicious Medical Providers and SSI Recipients They Assisted Whose Impairments Were Difficult to Objectively
Verify, by State
                                                               Number of suspicious medical providers
                                                                                               100-     301-
Number of SSI recipients assisted                    1   2-5   6-10   11-25   26-50 51-100      300      500    500+     Total
California                                       279     314    85      83      22       10      11        1        •     805
Florida                                          120      88    26      14       8        1        •       •        1     258
Georgia                                             12    15    10       3       3        2        2       •        •      47
Louisiana                                           16    11     7      12       6        1        •       •        •      53
Massachusetts                                        8    12     4      12       7        1        3       •        •      47
New York                                         116      74    23      12       7        3        2       •        •     237
Total                                            551     514   155     136      53       18      18        1        1    1,447

                                           Although our analysis does not prove that any fraud or program abuse was
                                           committed in any of these cases, it shows that SSI recipients with
                                           impairments that are difficult to objectively verify have used evidence
                                           from medical providers who had been or were being investigated for
                                           fraudulent activities by other benefit-paying entities.

Front-Line Staff Believe                   We could not determine the extent to which middlemen participate in
Middlemen Continue to                      cases involving feignable impairments or identify which middlemen were
Help Persons Feign                         involved in a large number of cases because SSA does not routinely record
                                           the names and addresses of middlemen when a claim is filed. Therefore, to
Disabilities                               find out whether middlemen remain a significant source of potential fraud
                                           and abuse, we contacted 158 SSA, OIG, and DDS staff and managers in SSA’s
                                           Baltimore headquarters and in field offices in California, New York,
                                           Massachusetts, and Washington and asked them if they believed problems
                                           with middlemen continued. Of these, 96 percent (152) indicated that SSA
                                           remains vulnerable to middleman fraud.

                                           The following are examples of cases these staff cited.

                                       •   SSA and DDS staff in New York told us about a middleman whose clients are
                                           typically diagnosed as having severe mental conditions but continue to live
                                           at home and receive no treatment. The clients almost always have very
                                           low reported intelligence quotient scores and almost never have any
                                           historical medical records.
                                       •   In California, field office staff said some applicants are coached by
                                           middlemen on what to say and how to respond to questions before they
                                           come to the office. Staff in other offices told us that middlemen will use

                                           Page 9                         GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                            various aliases to mask their true identity or go to offices where they are
                            less known.
                        •   Field office staff in California said that middlemen are still active in SSI
                            cases, but are trying to hide their involvement. They said that when district
                            offices in Southern California hired bilingual staff to address concerns
                            about interpreter fraud, middlemen began taking their clients to other
                            district offices. Because SSA does not maintain a centralized database on
                            suspicious middlemen, its field staff cannot check whether a middleman
                            accompanying a claimant should be considered suspicious.
                        •   OIG investigators believe middlemen remain active because they have
                            observed middlemen waiting in cars while an applicant pursues a claim or
                            has a medical exam. OIG investigators further suspect that the middlemen
                            continue to prepare claims applications and to coach applicants on how to
                            act and respond to interview questions. In this regard, field staff pointed
                            out that suspicious claims applications are prepared using language that
                            mimics SSA policy manuals. They also said that suspicious applicants
                            always seem to know the “right” answer to SSA employee questions.

                            To reduce SSI’s vulnerability to fraud and abuse, SSA has undertaken
SSA’s Antifraud             several initiatives, some of which were required by legislation. SSA has
Initiatives Are             established pilot investigation teams in five states dedicated to examining
Valuable but                cases where fraud or abuse is suspected. It also has developed new
                            procedures that DDS staff handling claims must use when they encounter
Limitations                 suspicious disability claims and instituted new approaches for handling
Undermine                   claims of non-English-speaking individuals. While these initiatives are
                            useful steps in addressing potentially fraudulent cases, their effectiveness
Effectiveness               is limited by staff reluctance to routinely implement them. Staff perceive
                            that these actions conflict with other agency goals or are not convinced of
                            their effectiveness. Other staff believe that certain procedures expose
                            them to potential legal liability.

SSA Is Piloting Fraud       In 1998, SSA created as a pilot project five Cooperative Disability
Investigation Teams         Investigation (CDI) teams to investigate suspected cases of disability fraud
                            or abuse. The CDI teams are patterned after a fraud investigation unit
                            established in 1994 to respond to a large number of disability fraud and
                            abuse cases being identified in the Southern California area. Each CDI team
                            investigates cases referred through SSA’s OIG fraud hotline and by DDS and
                            SSA field office staff who have been instructed to refer all cases—both
                            applicants and recipients—in which they suspect disability fraud or abuse.

                            Page 10                       GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

The referrals can cover questionable situations, such as a recipient’s
failing to report work activity or feigning disabilities.

Each CDI team consists of four or five members and is headed by an agent
from SSA’s OIG. Other CDI team members typically include DDS examiners
and state law enforcement personnel, such as Medicaid fraud
investigators. SSA has placed these units in five cities that it believes have
serious disability fraud and abuse problems: Oakland, California; Chicago,
Illinois; Baton Rouge, Louisiana; Atlanta, Georgia; and Brooklyn, New
York. In cities that do not have a CDI team, SSA OIG offices continue to have
the responsibility to investigate fraud cases.

In conducting their investigations, CDI teams obtain information that helps
SSA decide whether applicants or recipients are truly qualified to receive
benefits. Although teams do not typically develop evidence for the
prosecution of criminal fraud, CDI investigations of individual applicants
and recipients may provide a basis for the OIG to conduct broader
investigations into the practices of medical providers and middlemen.

DDS staff continue processing the case even after referring it to the CDI
team. While the DDS office assesses the medical information, the CDI team
begins gathering evidence that either substantiates or contradicts
statements that applicants or recipients have made regarding matters such
as their income and how their disabilities limit their daily lives. The teams
typically do this by conducting undercover surveillance of the individual’s
daily activities and interviewing the individual’s neighbors, family, and
friends. Although the disability determination can be made before the CDI
team completes its investigation, if evidence is developed that affects the
determination, the DDS office may reopen the case.

The effectiveness of the CDI teams has been demonstrated. For example, in
1998, a state DDS office referred a case to a CDI team because the
applicant’s treating physician had a history of providing similar
information on multiple patients. The applicant alleged that headaches,
memory loss, weakness, asthma, and depression severely limited her
ability to carry out activities such as shopping and prevented her from
obtaining a driver’s license and learning English. The CDI investigation
disclosed that the applicant had a valid driver’s license, and during
surveillance, CDI staff observed the applicant grocery shopping. Staff also
approached the applicant with a question and discovered that she spoke
English. This information led to a denial of benefits in the case.

Page 11                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                          As of March 31, 1999, SSA’s OIG reported that the five CDI teams—which had
                          been operating for 6 to 14 months—had conducted 624 investigations that
                          contributed to the denial of benefits to 119 applicants and the cessation of
                          benefits to 58 recipients, according to SSA’s OIG. The OIG estimates total SSI
                          program savings from these claims denials and cessations of benefits
                          amounted to about $11 million.5 The original investigative team
                          established by SSA and DDS in Southern California has also had an effect on
                          the program. From November 1995 through March 1999, this team’s
                          investigations have resulted in the cessation of benefits in 42 cases and the
                          denial of benefits in 27 cases. According to the team, these investigations
                          have saved the SSI program an estimated $5.5 million. SSA is pleased with
                          these results and anticipates that similar teams will be placed in 12
                          additional locations by fiscal year 2003.

SSA Has Revised           In 1994, the Social Security Act was amended to require that evidence in
Procedures for Handling   eligibility determinations be disregarded “if there is reason to believe that
Suspicious Claims         fraud or similar fault was involved in the providing of such evidence.” SSA
                          issued implementing fraud or similar fault (FSF) procedures to the DDS
                          offices in April 1998. FSF implementing procedures for SSA field offices and
                          appellate adjudicators are still under development.

                          Under its implemented FSF procedures, DDS adjudicators must consider all
                          evidence in the case record before determining whether any specific
                          evidence should be disregarded. Supporting evidence should be
                          disregarded only if a preponderance of other evidence establishes a reason
                          to believe that fraud or similar fault was involved. Fraud or similar fault
                          involves knowingly making an incorrect or incomplete statement or
                          knowingly concealing material information. As is the case with the CDI
                          teams, the goal of the FSF procedures is to prevent individuals who are not
                          truly disabled from receiving benefits—not to develop sufficient evidence
                          to prosecute a person for fraud.

                          To help DDS staff identify high-risk cases, the FSF procedures first list
                          characteristics that have been commonly associated with fraudulent or
                          abusive cases in the past. The FSF procedures then recommend special
                          ways that high-risk cases should be handled and developed to determine
                          whether there is reason to believe fraud or similar fault was involved. The
                          special handling includes gathering additional evidence to determine
                          whether statements about the disabilities and functional limitations of

                           Because of the way CDI results were reported, it is probable that the actual number of cases
                          investigated, terminated, and denied is higher.

                          Page 12                                 GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                          clients are correct and complete and checking with appropriate staff to
                          see if there are any known problems with the person’s medical evidence
                          sources or any middlemen involved in the cases.

                          These new procedures require DDS staff to document a fraud or similar
                          fault finding and cite any evidence that is disregarded. After disregarding
                          any evidence, DDS staff should make the disability determination based on
                          the remaining evidence in the file. The procedures require that staff notify
                          SSA’s OIG of all cases where similar fault is suspected, alerting OIG to
                          suspicious middlemen or medical providers who may have been involved
                          in providing incomplete or incorrect statements. The OIG then has the
                          option of investigating these cases further to establish whether fraud

SSA Has Made Changes in   Program policy on the use of interpreters varies among the different
the Use of                components involved in making disability decisions. SSA has a general
Agency-Supplied           policy at its field offices of allowing non-English-speaking SSI applicants to
                          choose whether they want to use their own interpreters or an SSA-supplied
Interpreters              interpreter at the time a claim is filed. Interpreters provided by applicants
                          must now sign a form stating that they will accurately translate applicant
                          responses during the interview. However, if field staff suspect that an
                          applicant-supplied interpreter is not providing accurate information during
                          an SSA interview, they can stop the interview and reschedule it for a time
                          when an SSA-supplied interpreter is available. Failure to sign the form is
                          also grounds for SSA to stop and reschedule an interview with an
                          SSA-supplied interpreter.

                          During required consultative medical examinations, the DDS offices in most
                          states follow SSA’s field office policy of generally allowing applicants to
                          decide whether to use their own interpreter or one supplied by the DDS
                          office. Staff can also insist that the applicant or recipient use an
                          agency-supplied interpreter if they have suspicions about a case. One
                          state, however, requires all non-English-speaking applicants and recipients
                          to use DDS-supplied interpreters.

                          For cases denied by a DDS and then appealed, OHA requires that its ALJs use
                          a qualified interpreter. Interpreters have to be able to read, write, and
                          demonstrate fluency in the language of the claimant and in English. They
                          should have a basic familiarity with SSA terminology, agree to act in the
                          best interest of the claimant and the public at large, provide exact
                          translations, and comply with SSA disclosure and confidentiality

                          Page 13                       GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                             requirements. Sources considered as qualified include SSA and state
                             employees, consultative examination providers, family members, or
                             persons affiliated with churches and advocacy groups.

                             Because of both an increase in the number of non-English-speaking clients
                             and a heightened awareness of the problems associated with unscrupulous
                             interpreters, SSA has hired over 2,300 additional staff with bilingual
                             capabilities since 1993.6 However, SSA does not know how many bilingual
                             staff it has in total, nor has it determined how many it needs. SSA officials
                             told us that the agency has begun tracking claimant language preferences
                             so that it can target interpreter services more effectively. It is also placing
                             more emphasis on ensuring that adequate funds are available to pay for
                             non-English-speaking interpreter services where bilingual staff are not
                             available and providing specialized training for bilingual employees.

Staff Concerns Limit the     Each SSA initiative depends on its field and DDS staff first recognizing
Effectiveness of Antifraud   suspicious cases (which can be difficult) and then following the new
Initiatives                  procedures to refer the case for investigation by a CDI team, or use the new
                             FSF procedures, or arrange for an agency-supplied interpreter. However,
                             many of the staff whom we interviewed said they are reluctant to routinely
                             take these actions for several reasons. Some staff believe the new
                             procedures conflict with other agency goals, and some staff do not
                             perceive the procedures as being effective in preventing fraud and abuse.
                             In addition, some staff have concerns about their legal liability from
                             following the FSF procedures.

Concerns About Conflicting   Many DDS staff told us that they do not refer all suspicious cases to CDI
Agency Goals                 teams because such referrals require extra processing time. Specifically, in
                             cases where fraud or similar fault is suspected, staff must develop
                             evidence to support their suspicions; prepare referral forms that explain
                             the basis for their concern; and, to the extent possible, provide evidence
                             that supports their concern. Proposed referrals are then discussed with
                             DDS management, which decides whether to refer the case to a CDI team.

                             According to DDS staff, this extended processing time is inconsistent with
                             SSA’s goal to quickly and accurately process claims and post-entitlement
                             decisions, and SSA has not made allowances in its performance goals and
                             measures (work credits) for the additional time needed to identify and
                             handle suspicious cases. SSA continues to monitor processing times, and

                              The changes in SSA interpreter policy and the hiring of additional bilingual staff partially respond to
                             the recommendations that we made in our 1995 report.

                             Page 14                                   GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                                   staff believe that any delays in DDS decisions are viewed as negatively
                                   affecting performance. For example, one DDS examiner told us that when a
                                   case is held up, it has an adverse affect on an employee’s mean
                                   case-processing time. At another DDS office, staff said that case examiners
                                   do not always refer suspicious cases because they do not want their
                                   processing times to suffer.

                                   We heard similar comments about processing time concerns from staff in
                                   SSA field offices. For example, we were told that staff lack the time and
                                   resources to properly check claims. When they detect a possible problem
                                   during the interview and would like to follow up on those suspicions, they
                                   sometimes do not because they do not receive credit for the additional
                                   work. At another field office, a staff member said she believes the
                                   investigative teams are understaffed and she hates to let her processing
                                   time suffer by making referrals to them. At a third field office, we were
                                   told that the referrals were not an effective use of staff time.

                                   When we discussed processing times and the new FSF and agency
                                   interpreter procedures with front-line staff, concerns such as the following
                                   were raised:

                               •   DDS staff said that the new FSF procedures are more labor-intensive than
                                   those required for other claims. They can also require additional
                                   development of evidence. Further, they said that the guidelines on how to
                                   identify claims that might warrant special handling are so general that they
                                   could apply to most SSI claims.
                               •   SSA field office staff echoed these views. They told us that stopping
                                   interviews because of concerns over interpreters just extends the time
                                   needed to handle and close a claim. It takes time to establish another date
                                   when SSA can arrange for its own interpreter and for the applicant to
                                   appear at another interview. Consequently, the new policy can result in
                                   field staff missing processing time goals.

                                   According to an SSA official, the agency has developed a “culture” that
                                   values helping needy people and, within this culture, the prompt payment
                                   of benefits takes precedence over all other activities, including efforts to
                                   uncover fraud and abuse.

Concerns About Effectiveness       Both DDS and SSA field office staff perceive that SSA’s antifraud initiatives
of New Initiatives                 will have a limited effect on fraud and abuse, which adds to their
                                   reluctance to invest the time and effort required by these new initiatives.

                                   Page 15                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                                 Basically, staff believe that even if they deny a claim, applicants will
                                 ultimately be awarded benefits at the appeals level.

                                 One basis for this perception stems from the differences in procedures for
                                 ALJs and DDS offices. Because FSF procedures have not yet been issued for
                                 ALJs, they operate under adjudicative rules, which may cause them to
                                 reach a decision different from other SSA decision levels. According to OHA
                                 officials, by law ALJs must give controlling weight to the medical opinion
                                 provided by an applicant’s or recipient’s treating physician, provided the
                                 medical opinion is well supported by acceptable clinical and laboratory
                                 diagnostic techniques and is not inconsistent with other evidence. Under
                                 FSF procedures issued for DDS-level adjudicators, DDS staff may decide not
                                 to give controlling weight to the medical evidence from a treating
                                 physician when the DDS office has evidence that the physician has
                                 repeatedly provided identical diagnoses in other cases. Consequently, a
                                 denial determination under FSF procedures by DDS adjudicators may well
                                 be overturned by the ALJ when the judge does not have the necessary
                                 documentation about the reasons the DDS did not give controlling weight
                                 to the treating physician’s opinion.

                                 In addition, SSA and DDS staff told us that they are reluctant to refer all
                                 fraud or similar fault cases to SSA’s OIG for possible prosecution—although
                                 FSF procedures require them to do so—because they perceive that the OIG
                                 is not willing to investigate such cases. In the past, the OIG has devoted its
                                 limited resources to investigating fraud cases where large dollar amounts
                                 were involved or a conviction was likely. Furthermore, when fraud cases
                                 were referred to the OIG, there was no feedback on the outcome of the
                                 referrals. When staff continued to see the same middlemen and providers
                                 involved in other cases, they concluded that the OIG referrals were not a
                                 productive use of their time.

                                 The OIG is aware of these views and is developing systems to better inform
                                 field staff about the status of cases they have referred. There have also
                                 been staffing increases to improve its investigative capacity and efforts
                                 have been made to publish information about the outcome of fraud cases.

Concerns About Staff Liability   Finally, staff are concerned that they can be held liable for actions they
                                 take under the new procedures, which require them to place written
                                 statements in the files whenever they believe material information
                                 provided by applicants, medical providers, middlemen, or other third
                                 parties is misleading, inaccurate, or incomplete. Staff fear that if this type
                                 of statement becomes known, they could be sued and held liable for

                                 Page 16                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                            damages claimed by medical providers and interpreters alleging that DDS
                            staff impugned their reputations.

                            One state DDS has not yet implemented the new FSF procedures because of
                            these concerns. Although the principle of sovereign immunity generally
                            exempts states from liability suits based on actions taken by employees
                            performing their official duties, state laws and court decisions have
                            created some exceptions to that immunity. Officials in the state pointed
                            out that there is nothing to prevent providers, middlemen, or organizations
                            representing them from seeking to hold the state or its employees liable
                            under one of the exceptions. The state does not want to incur the time or
                            expense involved in defending itself and its employees or risk an adverse

                            SSA officials stated that the agency cannot guarantee that DDS employees
                            would be held harmless by a court. Such a guarantee would mean that the
                            government would have to defend any DDS employee, even if the employee
                            were negligent in making adverse statements about a medical provider or
                            other third party in a claims file. SSA officials also believe it is clear that the
                            guidelines for identifying suspicious claims are just that—guidelines—and
                            not mandates to apply the FSF procedures to each case meeting these

                            In our view, there are several additional actions SSA could take to help
Opportunities Exist to      reduce the SSI program’s vulnerability to fraud and abuse. Because SSA
Better Identify and         relies heavily on its front-line staff to detect suspicious claims and the
Track Suspicious            involvement of suspicious middlemen and medical providers, it is
                            important that resources and processes assist staff in their identification
Middlemen and               efforts and encourage them to use SSA’s new initiatives.
Medical Providers
Better Information Needed   Approaches that focus on obtaining and sharing information about
for Front-Line Staff        suspicious middlemen and medical providers programwide would likely
                            enhance SSA’s ability to identify cases where individuals may be obtaining
                            benefits by feigning disabilities. With this type of information, DDS
                            personnel and SSA’s field staff could better determine which claims should
                            receive increased scrutiny and target their investigations of current
                            beneficiaries to evaluate whether they should be removed from the
                            program. Such information could also help staff more readily identify
                            cases that meet certain profiles (suspicious middlemen and medical
                            providers), which should result in more effective referrals from DDS

                            Page 17                         GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

examiners and better use of CDI and OIG resources. The information could
also be used to identify those middlemen and providers who are involved
in multiple claims.

SSA  could use information it has to begin developing comprehensive
databases on suspicious middlemen. SSA and DDS staff could annotate the
database with the reasons for their suspicions about each identified
middleman. Because data on practicing middlemen are not readily
available, SSA would need to require that all third parties involved in claims
document their identity (for example, name, address, and social security
or driver’s license number). With these data, SSA could identify the cases in
which each middleman was involved, and SSA field and DDS staff could
check the database when handling claims and add new names to this
database as they became known. Thus, SSA, with its own data on
suspicious middlemen, could centralize and share this information
agencywide, as we suggested in our 1995 report. While SSA plans to
centralize information on suspicious middlemen within each DDS through
its new FSF procedures, this step may not be sufficient to address the
problem of middlemen operating among offices in more than one location.
Limiting the databases to specific geographic areas would likely reduce
their effectiveness as a tool to identify the involvement of suspicious
middlemen in SSI cases.

With databases that could be shared agencywide, the agency would be
better able to identify potential problem cases and unscrupulous
middlemen, regardless of the office being used. SSA could also require that
its own interpreters be used when an applicant uses a suspicious
middleman listed in the database, instead of requiring staff to rely on their
suspicions that an interpreter is providing inaccurate translations. To
facilitate the use of agency-supplied interpreters in these situations, SSA
could require that non-English-speaking claimants schedule an interview
at a field office where staff have the appropriate language capability. If this
is inconvenient for the client, SSA could schedule an interview at an office
of the applicant’s choosing and send an agency-supplied interpreter to that
office on the established appointment date.

SSA could supplement the middleman database with information on
suspicious medical providers identified by other entities (for example, the
Medicaid and Medicare programs and private insurance companies) to
identify cases for scrutiny. SSA’s past experience with investigating
disability fraud and abuse has shown that medical providers suspected or
convicted of Medicaid fraud have provided many SSI recipients with

Page 18                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

misleading medical evidence that helped them improperly obtain benefits.
Moreover, fraud investigators have told us that medical providers who try
to take advantage of one program often try to abuse or defraud other
programs as well.

Benefit-paying agencies typically maintain databases of suspicious
providers they have investigated for alleged fraudulent and abusive
activities. If SSA gathered and maintained this information, it could
determine through computer matching whether any SSI applicants or
recipients had used or were using these same providers. A match would
not prove that the applicant or recipient was actually feigning his or her
disability. However, it would alert DDS staff to the possibility of fraud or
abuse and highlight the case for more careful review either by them or by
a CDI team, if one is present at the DDS office. Establishing such a database
would require some changes in SSA recordkeeping practices. For example,
the agency would have to include in its electronic records the names of the
medical providers used by applicants and recipients to supply medical
evidence. Currently, only state DDS offices maintain provider names to
facilitate payment for medical evidence submitted on the behalf of

To ensure such comprehensive databases would be secure and the
information therein confidential, SSA would need to address widespread
weaknesses in controls over access to its systems, which we recently
reported on.7 These control weaknesses expose its computer systems to
external and internal intrusion, subjecting sensitive SSA information to
potential unauthorized access, modification, and disclosure. Although SSA
has developed and continues to pursue corrective actions to address these
problems, some organizations may not want to disclose data they maintain
on providers, fearing that improper handling would adversely affect their
own operations.

In addition, medical providers and middlemen may be concerned that their
reputations could be damaged if it becomes known that they had been
suspected of fraud or abuse and the suspicions may not have been
substantiated. There are ways to address these concerns. For example,
insurance laws in most states allow regulators to maintain databases of
suspicious medical providers and others suspected of defrauding
insurance companies. To encourage these companies to report the names
of suspicious providers and other parties in the claims they are evaluating,

 Information Security: Serious Weaknesses Place Critical Federal Operations and Assets at Risk
(GAO/AIMD-98-92, Sept. 23, 1998).

Page 19                                GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                          the laws guarantee that the companies cannot be sued by a suspicious
                          provider or other third party for maintaining or referring such data, as long
                          as the referral was made without malice or intent to harm. In addition to
                          these state-level databases, insurance companies provide the names of
                          suspicious individuals to the National Insurance Crime Bureau, a national
                          not-for-profit organization that maintains a central database for member
                          insurance companies to consult in their efforts to deter and prevent
                          insurance crimes. It is also used by law enforcement agencies in their
                          efforts to combat fraud.

                          Further, SSA is required by law to take certain steps to ensure the privacy
                          and security of data, whether that information was internally generated by
                          SSA or obtained from other agencies. These steps include traditional
                          safeguards such as developing a security plan, audit trails, automated
                          alerts to prevent inappropriate requests for personal information, personal
                          identification numbers and passwords, training, and periodic internal and
                          external evaluations of all privacy and security measures.

Encouraging Staff to      Fighting fraud and abuse will require changes in management approaches.
Pursue Suspicious Cases   SSA needs to demonstrate to its front-line staff that it is serious about
                          having them pursue questions about suspicious cases. Management
                          systems that emphasize timely processing of claims without recognizing
                          the additional time needed to develop evidence related to suspicious cases
                          are hindering SSA’s antifraud efforts.

                          Both the OIG and we have noted how staff perceive agency priorities. For
                          example, we concluded in a recent report that long-standing problems in
                          the SSI program are attributable to SSA’s ingrained organizational culture
                          that has historically placed a greater value on quickly processing and
                          paying SSI claims than on controlling program costs.8 We recommended
                          that SSA reevaluate its field office work-credit and incentive structure at all
                          levels of the agency and make appropriate revisions to encourage better
                          verification of recipient information and greater staff attention to fraud
                          prevention and detection. The OIG also noted that developing fraud cases
                          for referrals can require significant amounts of time and concluded that
                          SSA cannot simply measure claims processing by how many and how
                          quickly cases are processed because this approach creates a disincentive
                          to staff for developing fraud cases. It also suggested that incentives to

                           Supplemental Security Income: Action Needed on Long-Standing Problems Affecting Program
                          Integrity (GAO/HEHS-98-158, Sept. 14, 1998).

                          Page 20                              GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                      develop suspicious cases be provided and that adjustments to tracking
                      processing times be made.9

                      SSAtold us that giving special consideration when tracking staff claims
                      processing times in suspicious cases remains under review.

                      SSI and other benefit programs may be losing millions of dollars each year
Conclusions           because individuals improperly obtain benefits by feigning disabilities with
                      the help of medical providers and middlemen. Every individual who
                      obtains benefits in this manner will cost the federal government an
                      estimated $122,000 in SSI and Medicaid benefits over the next 10 years.
                      While SSA has made progress in addressing this problem since our 1995
                      report and its efforts have had positive results, detecting fraudulent and
                      abusive SSI cases remains difficult.

                      Because SSA relies heavily on its front-line staff to identify potential fraud
                      and abuse, it is important for staff to have the ability to detect suspicious
                      cases. Their detection abilities would be strengthened if they had
                      additional tools to meet this challenge. To the extent that information on
                      problem middlemen and medical providers can be developed, maintained,
                      and shared with staff, SSA’s fraud detection and prevention efforts will be
                      enhanced. In addition, by implementing our previous recommendation to
                      reevaluate its work-credit and incentive structure to encourage better
                      verification of recipient information and greater staff attention to fraud
                      prevention and detection, staff will be encouraged to use the new
                      procedures. Finally, we believe legislative action to address staff liability
                      concerns could enhance the use of established procedures to fight fraud.

                      We recommend that the Commissioner of Social Security take the
Recommendations       following actions:

                  •   Study the feasibility of obtaining information on suspicious medical
                      providers from federal, state, and private entities that face similar fraud
                      and abuse issues as SSA does in managing the SSI program.
                  •   Systematically track suspicious middlemen and medical providers
                      identified by SSA staff and outside agencies, and routinely share this
                      information throughout SSA. For example, SSA could electronically
                      maintain information on such medical providers and middlemen and on
                      the SSI applicants and recipients they serve. This information would help

                       SSA, Proceedings of the 2nd Annual Fraud Conference, Sept. 8-12, 1997.

                      Page 21                                GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

                         SSA (1) determine which claims should receive increased scrutiny to
                         prevent these applicants from improperly receiving benefits and (2) target
                         investigations of current beneficiaries to determine if they should be
                         removed from the program.
                     •   Reexamine SSA’s policy regarding SSA-provided interpreters for SSI
                         applicants with the aim of determining the extent to which it is followed
                         by field and DDS staff and its effectiveness, and whether the use of
                         SSA-provided interpreters should be required in situations which meet
                         certain profiles.

                         To address liability concerns related to maintaining lists of suspicious
Recommendation to        middlemen and medical providers and following FSF procedures, the
the Congress             Congress may wish to provide a limitation of the legal liability of state
                         employees who follow SSA policies that require them to identify and
                         document middlemen and medical providers suspected of providing
                         misleading, inaccurate, and incomplete evidence in disability claims.

                         We provided SSA a draft of this report for review and comment. In its
Agency Comments          written response, SSA agreed that more can be done to prevent fraud in the
and Our Evaluation       SSI program and endorsed our recommendation to reexamine its current
                         policy on the use of interpreters. However, the agency indicated that while
                         our other two recommendations have potential value, it wanted to explore
                         them further before committing to developing implementation strategies
                         for them. SSA also emphasized that its issuance of a plan to improve SSI
                         program management was evidence of its commitment to fight fraud and
                         noted that it has taken actions that can substantially reduce the potential
                         for such fraud.

                         Our views about several specific concerns raised in SSA’s letter follow.
                         SSA’s letter is reprinted as appendix III.

                     •   Regarding the finding that SSI remains vulnerable to middleman fraud, SSA
                         is concerned that our report relies almost exclusively on anecdotal
                         evidence. SSA said that while the middleman problem has not been
                         completely eradicated, it believes that it has taken actions that
                         substantially reduce the potential for middleman fraud and remains
                         committed to taking further action.

                         As our report states, SSA does not routinely record the names and
                         addresses of middlemen when a claim is filed. As a result, we could not

                         Page 22                       GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

    determine the extent of suspicious middleman involvement in SSI cases
    involving feignable impairments. As a substitute measure, we spoke with
    staff SSA relies on to identify potentially fraudulent cases (its field office
    employees and DDS staff) and SSA’s fraud investigators. Both said
    middleman fraud is a continuing problem.

•   SSA is also concerned that we may have overstated the extent of the
    problem with unscrupulous medical providers. It said that lists of
    suspicious providers may prove to be a valid indicator of the potential for
    fraud in a case. However, it also said that our inclusion of persons
    suspected of fraud rather than limiting the study to those convicted or
    otherwise sanctioned for fraud could overstate the problem. SSA noted that
    being investigated for fraud cannot and should not be equated with being
    convicted or sanctioned.

    Precisely measuring the SSI program’s vulnerability to fraud and abuse is
    difficult. By its nature, fraud is surreptitious and perpetrators are not
    always identified and prosecuted. Even if the rate is half what we
    measured, there is a problem that SSA needs to address. Some medical
    providers—an important component of the disability adjudication
    process—have been at least suspected of fraudulent activities by others.
    We believe SSA can improve staff ability to identify cases that deserve
    closer scrutiny by developing and maintaining lists of medical providers
    and middlemen whose past actions make their involvement in SSI cases

•   SSA emphasized that its October 1998 plan to improve SSI management
    addresses employee views that workload priorities overshadow antifraud
    activities. It said the plan makes it clear that SSA is pursuing initiatives
    designed to balance its program stewardship responsibilities with its
    public service responsibilities. Over time, it believes the plan activities will
    achieve this balance.

    We believe SSA’s issuance of a plan to improve SSI management is a positive
    step in its efforts to combat fraud and abuse in the program and that it has
    taken a number of actions to enhance program stewardship. However, the
    plan mentioned by SSA does not specify any initiatives that directly address
    employee perceptions that workload priorities overshadow antifraud
    activities. SSA needs to take some specific actions to overcome this
    widespread and deep-seated perception among its staff.

    Page 23                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

•   SSAcommented that the majority of the cost savings achieved by its five
    CDIteams are not necessarily related to fraud perpetuated against the
    program. It said that many if not most of the CDI team savings appear to
    involve instances of disability decisions being made incorrectly or without
    proper documentation rather than fraud.

    As our report notes, the purpose of the CDI teams was not to prove fraud;
    rather, it was to assist SSA and DDS staff in making benefit-related
    decisions. The report notes that the CDI teams believe that their
    investigative work contributed to denials and cessations of benefits—not
    that they contributed to prosecutions for fraudulent activity.

•   Finally, before pursuing two of our recommendations, SSA would like to
    have in-depth discussions about these approaches with its OIG staff and
    GAO. SSA said that implementing two of our three
    recommendations—tracking suspicious middlemen and medical providers
    SSA encounters and sharing this information with its staff; and studying the
    feasibility of obtaining information on suspicious medical providers from
    federal, state, and private entities to supplement this information—may be
    fruitful. However, SSA is concerned about the definition of suspicious
    medical providers or middlemen and the legal ramifications of tracking
    individuals who may not have been convicted or have not admitted guilt.
    Because the suspicious individuals in our study included people who had
    been or were being investigated as well as people who have been
    convicted or sanctioned, SSA states that this approach raises serious legal
    issues relative to the Privacy Act, the Freedom of Information Act,
    individual state and employee liability, and accessibility (security).
    Specifically, SSA notes that the Privacy Act requires that agencies maintain
    records that are accurate, complete, relevant, and timely as reasonably
    necessary to ensure fairness in any determinations made about the
    individual. Before establishing such a system of records, SSA would have to
    make the public aware of its plans by publishing a notice of its intended
    actions and allowing the public to comment. Once aware of the records
    system, the public could use the provisions of the Privacy Act to obtain
    records about themselves and the right to request correction of erroneous
    information in the records. If SSA inappropriately or incorrectly labels
    individuals as suspicious without the benefit of convictions or admissions
    of guilt, it could be vulnerable to legal challenges in civil actions brought
    by these individuals.

    We agree that SSA must comply with the Privacy Act and other relevant
    legislation and must act carefully and responsibly in characterizing

    Page 24                       GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

individuals as suspicious, particularly where that characterization could
lead to criminal prosecution, denial of benefits, or other adverse
consequences. Our definition of “suspicious” was reasonable for research
purposes but may not be appropriate for law enforcement purposes. In our
opinion, however, the Privacy Act is not an impediment to implementing
our recommendation to systematically track suspicious middlemen and
medical providers. SSA already maintains a system of records, the Program
Integrity Case Files, that contains the same kind of information and
complies with the Privacy Act. We believe this system of records—or a
similar one designed for this purpose—could be used to carry out our
recommendations as well. Information in the Program Integrity Case Files,
according to SSA’s published Privacy Act notice, includes the identity of
“persons suspected of violating Federal statutes affecting the
administration of programs under the responsibility of SSA.” We see no
reason why information about suspicious middlemen and providers in the
SSI program could not be maintained in the same fashion. Since SSA already
maintains such records, our recommendations create no new category of
risk of civil liability for incorrectly labeling individuals as suspicious.
Nevertheless, the intent of our recommendations is to provide SSA and DDS
staff with information, such as the involvement of suspicious middlemen
or providers in a case. This type of information will enable them to identify
potentially fraudulent cases for closer review.

We are providing copies of this report to the Honorable Kenneth S. Apfel,
Commissioner of Social Security. We will also send copies to other
interested parties on request. If you or your staff have any questions about
this report, please contact Barbara Bovbjerg, Associate Director, at
(202) 512-5491, or Rod Miller, Assistant Director, at (202) 512-7246. Other
major contributors to this report were Nancy Cosentino, Jill Yost, William
Staab, and Kevin Craddock.

Sincerely yours,

Cynthia M. Fagnoni
Director, Education, Workforce,
  and Income Security Issues

Page 25                       GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse

Letter                                                                                                1

Appendix I                                                                                           28
                       Interviews With SSA Managers and Staff                                        28
Scope and              Analysis of SSI Recipients in Susceptible Diagnostic Categories               28
Methodology            Analysis of Suspicious Medical Providers Involved in SSI Cases                29

Appendix II                                                                                          31

Considered by SSA to
Be Vulnerable to
Appendix III                                                                                         32

Comments From the
Social Security
Tables                 Table 1: Estimated Number of SSI Child and Adult Recipients in                 7
                         April 1998 Sampling by Category of Impairments Considered by
                         SSA and DDS Staff to Be Vulnerable to Feigning
                       Table 2: SSI Recipients With Impairments Difficult to Objectively              8
                         Verify Who Used Suspicious Medical Providers to Support Their
                         Disability Claim, by State
                       Table 3: Suspicious Medical Providers and SSI Recipients They                  9
                         Assisted Whose Impairments Were Difficult to Objectively Verify,
                         by State
                       Table I.1: Number of Interviews GAO Conducted With Individuals                28
                         to Ask About the Continued Existence of Middleman Fraud in the
                         SSI Program, by Organization
                       Table II.1: Impairments Considered Susceptible to Exaggeration                31
                         in SSI Claims

                       Page 26                      GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse


ALJ        administrative law judge
CDI        Cooperative Disability Investigation
CDR        continuing disability review
DDS        Disability Determination Services
FSF        fraud or similar fault
HCFA       Health Care Financing Administration
MFCU       Medicaid Fraud Control Unit
NICB       National Insurance Crime Bureau
OHA        Office of Hearings and Appeals
OIG        Office of Inspector General
SSA        Social Security Administration
SSI        Supplemental Security Income

Page 27                     GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
Appendix I

Scope and Methodology

                                      This appendix describes our approach for collecting and analyzing data
                                      and for interviewing officials in agencies coping with fraud and abuse in
                                      health insurance programs. Our work was directed at determining (1) the
                                      potential risk that recipients become eligible for SSI by feigning disabilities
                                      with the help of middlemen and medical providers; (2) how SSA prevents,
                                      detects, and responds to this type of program fraud and abuse; and
                                      (3) additional methods SSA could use to effectively address this problem.
                                      We did not, however, verify the accuracy of the automated data provided
                                      by SSA or the investigative organizations. We conducted our review from
                                      October 1997 to August 1999.

                                      To determine the beliefs of SSA managers, front-line staff, and various
Interviews With SSA                   fraud investigators about the continued existence of middleman fraud, we
Managers and Staff                    asked 158 individuals to discuss their opinions on and experiences with
                                      middleman fraud in the SSI program. These individuals were not randomly
                                      selected and were not in sufficient numbers to constitute a statistically
                                      valid sampling of the opinions of all individuals who work with the SSI
                                      program. Table I.1 shows the number of interviews we held, by

Table I.1: Number of Interviews GAO
Conducted With Individuals to Ask     Organization                                                 Number of interviews
About the Continued Existence of      SSA headquarters                                                                17
Middleman Fraud in the SSI Program,
                                      OHA headquarters                                                                 5
by Organization
                                      SSA regional offices                                                            10
                                      DDS offices                                                                     43
                                      SSA field offices                                                               43
                                      OHA regional offices                                                             7
                                      Investigators                                                                   33

                                      To learn which mental and physical disabilities are considered susceptible
Analysis of SSI                       to being feigned or exaggerated, we interviewed disability specialists at SSA
Recipients in                         headquarters in Baltimore, medical consultants and medical relations
Susceptible                           officers at DDS offices in seven states, and investigators who specialize in
                                      disability fraud. We also reviewed SSA’s Program Operations Manual,
Diagnostic Categories                 which lists impairments prevalent in claims involving fraud or similar
                                      fault. The specific categories we identified as susceptible to feigning are
                                      identified in appendix II.

                                      Page 28                        GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
                         Appendix I
                         Scope and Methodology

                         We then analyzed the distribution of diagnostic categories among
                         recipients in SSA’s Characteristic Extract Record, often referred to as the
                         “10-percent file,” to identify how many adults and children had mental or
                         physical disabilities that fell into the susceptible diagnostic categories.
                         Because over 20 percent of the records lack the diagnostic code which
                         would indicate the disability that qualified the recipient for SSI benefits,
                         our analysis reflects only those records in the 10-percent file that
                         contained the diagnostic code.

                         To determine the potential extent of SSI disability fraud and abuse by
Analysis of Suspicious   medical providers, we obtained records from SSA that identified SSI
Medical Providers        recipients whose disabilities were among those considered susceptible to
Involved in SSI Cases    being feigned or exaggerated. The records covered six states (California,
                         Florida, Georgia, Louisiana, Massachusetts, and New York). SSI recipients
                         in these states constitute about 40 percent of the total SSI population.

                         Using these recipient names and social security numbers, the DDS offices
                         for these six states created files containing records that identified both the
                         SSI recipients and the medical providers who had submitted evidence to
                         support their disability claims. (In many cases, the DDS record contained
                         only the name of a hospital, and it was not possible to identify the specific
                         doctor at the hospital who had been involved in a claim.) The names of
                         those medical providers were matched against lists of providers who had
                         been or were currently under investigation by agencies tasked with
                         investigating suspicious medical providers, the Health Care Financing
                         Administration (HCFA), the National Insurance Crime Bureau (NICB), and
                         the states’ Medicaid Fraud Control Units (MFCU). We did not verify the
                         accuracy of the data provided by these agencies.

HCFA and NICB Matches    We matched the name, address, and tax identification number of the
                         service providers in the DDS file against providers listed in the HCFA and
                         NICB files. These files contained identifying information for medical
                         providers who had been either suspected or convicted of defrauding or
                         abusing programs paying Medicare, Medicaid, and private health insurance
                         benefits. For those providers who appeared in both lists, we created a file
                         of the records for all SSI recipients who had obtained evidence from them.

MFCU Match               State regulations require state MFCUs to protect the privacy and
                         confidentiality of service providers investigated for possible fraudulent

                         Page 29                       GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
Appendix I
Scope and Methodology

activity. For this reason, we developed a protocol for this data match that
differed from those used with the HCFA and NICB data.

We created for each state MFCU a file in which we had assigned a control
number to each service provider identified in the DDS records. MFCUs
matched the name, address, and tax identification information in our file
against their databases of investigated service providers, then provided us
with a list of the control numbers associated with providers who appeared
in both files. Using the control numbers, we generated a file of SSI
recipients who had used medical evidence from these suspicious providers
to prove their disability.

Page 30                       GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
Appendix II

Impairments Considered by SSA to Be
Vulnerable to Exaggeration

                                     SSA uses a four-digit code to designate disabilities. The codes are based on
                                     the International Classification of Diseases, published by the Department
                                     of Health and Human Services. The diagnostic codes are divided into
                                     general areas, such as cardiovascular, musculoskeletal, and mental. To
                                     determine which of these disabilities were most likely to be feigned or
                                     exaggerated by a person applying for SSI disability benefits, we interviewed
                                     medical consultants and medical relations officers at DDS offices in seven
                                     states, disability specialists at SSA headquarters, and investigators who
                                     specialize in disability fraud. We also reviewed SSA’s Program Operations
                                     Manual, which lists impairments prevalent in claims involving fraud or
                                     similar fault. From these sources, we developed the following list of
                                     disabilities that were considered susceptible to being feigned or

Table II.1: Impairments Considered
Susceptible to Exaggeration in SSI   SSA disability code        Description
Claims                               Adult/childhood disabilities
                                     2900-2949                  Organic mental disorders
                                     2950-2959                  Schizophrenic disorders
                                     2960-2999                  Affective disorders
                                     3000-3009                  Anxiety disorders
                                     3010-3059                  Personality disorders
                                     3060-3169                  Somatoform disorders
                                     3170-3199                  Mental retardation
                                     3450-3459                  Epilepsy
                                     Adult-only disabilities
                                     3690-3699                  Blindness and low vision
                                     4960-4949                  Chronic pulmonary insufficiency
                                     7240-7249                  Disorders of the back (discogenic and degenerative)
                                     7280-7289                  Disorders of the muscle, ligament, and fascia
                                     8480-8489                  Sprains and strains (all types)

                                     Page 31                           GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
Appendix III

Comments From the Social Security

               Page 32   GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
Appendix III
Comments From the Social Security

Page 33                             GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
Appendix III
Comments From the Social Security

Page 34                             GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
           Appendix III
           Comments From the Social Security

(207024)   Page 35                             GAO/HEHS-99-151 SSI Vulnerability to Fraud and Abuse
Ordering Information

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

Orders by mail:

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

or visit:

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

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

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

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


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


United States                       Bulk Rate
General Accounting Office      Postage & Fees Paid
Washington, D.C. 20548-0001           GAO
                                 Permit No. G100
Official Business
Penalty for Private Use $300

Address Correction Requested