Supplemental Security Income: SSA Is Taking Steps to Review Recipients'
Disability Status (Letter Report, 10/30/96, GAO/HEHS-97-17).
Pursuant to a congressional request, GAO reviewed the Social Security
Administration's (SSA) strategy for conducting legally required
continuing disability reviews (CDR) on Supplemental Security Income
(SSI) recipients, focusing on: (1) SSA plans to conduct legally required
SSI CDR in fiscal years 1996 through 1998; (2) the resources committed
to meeting this requirement; (3) how SSA selects recipients for SSI CDR;
(4) the potential benefits of conducting CDR on the SSI population; and
(5) potential options for improving the CDR process.
GAO found that: (1) SSA planned to conduct required CDR on about 118,000
SSI recipients in fiscal year (FY) 1996; (2) SSA also planned to conduct
an additional 100,000 CDR on SSI recipients that were not legally
required; (3) as of June 1996, SSA had completed about 60 percent of the
required CDR; (4) other competing priorities may make it difficult for
SSA to conduct all required SSI CDR after FY 1996; (5) in FY 1996, SSA
limited its selection for CDR to those recipients for whom medical
improvement is either expected or possible; (6) SSA estimates that
conducting CDR will result in removing only about 5 percent of SSI
recipients from the rolls, but without CDR, the number of ineligible
recipients will likely increase over time; (7) SSA estimates that
conducting CDR on SSI adult recipients for whom medical improvement is
expected or possible results in about $3 in federal program savings for
every $1 spent conducting CDR; and (8) SSA needs to establish less rigid
requirements for determining who should be scheduled for CDR, ensure
that contact is made with all SSI recipients, and develop a legislative
proposal to obtain the authority needed to extend this new process to
all recipients.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: HEHS-97-17
TITLE: Supplemental Security Income: SSA Is Taking Steps to Review
Recipients' Disability Status
DATE: 10/30/96
SUBJECT: Eligibility determinations
Eligibility criteria
Medical examinations
Cost control
Cost effectiveness analysis
Disability benefits
Social security benefits
Federal social security programs
Persons with disabilities
Vocational rehabilitation
IDENTIFIER: Medicaid Program
Old Age and Survivors Insurance Trust Fund
Social Security Disability Insurance Program
Supplemental Security Income Program
Supplemental Security Income Record Description Database
Social Security Disability Insurance Trust Fund
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Report to the Chairman, Subcommittee on Oversight, and the Chairman,
Subcommittee on Human Resources, Committee on Ways and Means, House
of Representatives
October 1996
SUPPLEMENTAL SECURITY INCOME - SSA
IS TAKING STEPS TO REVIEW
RECIPIENTS' DISABILITY STATUS
GAO/HEHS-97-17
SSI Continuing Disability Reviews
(106806)
Abbreviations
=============================================================== ABBREV
AFDC - Aid to Families With Dependent Children
CDR - continuing disability review
DDS - disability determination service
DI - Disability Insurance
HCFA - Health Care Financing Administration
IFA - individualized functional assessment
MIE - medical improvement expected
MINE - medical improvement not expected
MIP - medical improvement possible
OD - Office of Disability
SGA - substantial gainful activity
SSA - Social Security Administration
SSI - Supplemental Security Income
SSIRD - Supplemental Security Income Record Description
VR - vocational rehabilitation
Letter
=============================================================== LETTER
B-270337
October 30, 1996
The Honorable E. Clay Shaw
Chairman, Subcommittee on Human Resources
The Honorable Nancy L. Johnson
Chairman, Subcommittee on Oversight
Committee on Ways and Means
House of Representatives
In recent years, the Congress has been concerned that ineligible
individuals may be receiving disability benefits under the
Supplemental Security Income (SSI) program, which provides benefits
to low-income disabled and blind\1 individuals. These concerns stem,
in part, from frequent allegations of fraud, waste, and abuse in the
program. For example, we recently confirmed allegations that some
ineligible non-English-speaking applicants were illegally obtaining
benefits through the use of middlemen, who provide translation and/or
other services for a fee to help individuals apply for benefits.\2
Concerns about the program also stem from sharp increases in the
number of SSI recipients in recent years. From 1990 to 1996, the
population of disabled SSI recipients increased by over 40 percent,
from about 3.4 million to nearly 4.8 million.
To help ensure that only eligible individuals are receiving SSI
benefits, the Social Security Independence and Program Improvements
Act of 1994 required that the Social Security Administration (SSA)
review the disability status of a minimum number of SSI recipients.
In August 1996, the Congress enacted the Personal Responsibility and
Work Opportunity Reconciliation Act of 1996, which amended the Social
Security Act to require SSA to conduct continuing disability reviews
(CDR) within the first year of life on all low-birth-weight babies
receiving SSI benefits and at least once every 3 years on all other
SSI recipients under age 18 for whom medical improvement is likely.
The 1996 legislation repealed the requirement that SSA conduct CDRs
on one-third of SSI recipients attaining age 18 in each of fiscal
years 1997 and 1998 and replaced it with the requirement that,
starting on the date of the legislation's enactment, SSA redetermine
medical eligibility on the basis of adult criteria for all recipients
within 1 year of their 18th birthdays.
To assist in your oversight of the SSI program, you asked us to
examine SSA's strategy for conducting the legally required CDRs on
SSI recipients. In response, we addressed the following questions:
-- What steps is SSA taking to conduct legally required SSI CDRs in
fiscal years 1996 through 1998;
-- what resources has SSA committed to meeting this requirement and
what additional resources, if any, are needed;
-- how does SSA select recipients for SSI CDRs;
-- what are the potential benefits of conducting CDRs on the SSI
population; and
-- what potential options exist for improving the CDR process?
In companion reports, we discussed SSA's general strategy for
conducting CDRs, which are also required under the Disability
Insurance (DI) program.\3 The DI program provides disability benefits
to individuals who have enough work experience to be insured under
Social Security.
To address these questions, we (1) interviewed federal and state
officials; (2) reviewed relevant laws, regulations, policies,
procedures, and reports; and (3) reviewed SSA data on CDR work plans,
budgets, costs, savings, and previously conducted CDRs. Furthermore,
we reviewed the process SSA uses in selecting recipients for CDRs and
the composition of the formulas used to estimate the likelihood of
benefit termination for that process. We estimated savings from
benefit terminations after CDRs using assumptions provided by SSA;
the Health Care Financing Administration (HCFA), which administers
the Medicaid program; and the Congressional Budget Office. In
addition, to characterize the population of SSI recipients currently
due or overdue for CDRs, we analyzed SSA data from the Supplemental
Security Income Record Description (SSIRD) for a 15-percent sample of
the SSI disabled population due or overdue for CDRs on the basis of
SSA's current system for scheduling CDRs. We used SSIRD data as
provided and did not evaluate its accuracy or the validity of SSA
formulas used in the selection process. (App. I contains more
details on our scope and methodology.)
--------------------
\1 In this report, the term "disabled" includes recipients classified
as either blind or disabled under the SSI program. People over the
age of 65 who are not disabled also receive SSI if they meet the same
resource and income requirements that apply to the disabled. Because
this report is concerned with CDRs of disabled recipients, we did not
include the aged in our analyses.
\2 Supplemental Security Income: Disability Program Vulnerable to
Applicant Fraud When Middlemen Are Used (GAO/HEHS-95-116, Aug. 31,
1995).
\3 Social Security Disability: Improvements Needed to Continuing
Disability Review Process (GAO/HEHS-97-1, Oct. 16, 1996) and Social
Security Disability: Alternatives Would Boost Cost-Effectiveness of
Continuing Disability Reviews (GAO/HEHS-97-2, Oct. 16, 1996).
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
For fiscal year 1996, SSA planned to conduct required CDRs on about
118,000 SSI recipients--one-third of SSI recipients attaining 18
years of age and 100,000 recipients in other age groups. SSA also
planned to conduct other CDRs that were not legally required on over
100,000 additional SSI recipients. As of June 1996, SSA was on
schedule, having completed about 60 percent of the required CDRs.
SSA is currently revising its SSI CDR plans to meet the new
legislative requirements enacted in August 1996. Although the
Congress and SSA took steps to provide over $4 billion for CDRs in
fiscal years 1996 through 2002, legally required DI CDRs, which
include a large backlog, and other competing priorities may make it
difficult for SSA to conduct all required SSI CDRs after fiscal year
1996.
In fiscal year 1996, SSA limited its selection for CDRs to a portion
of SSI recipients it considers cost-effective to review. In general,
these include recipients for whom medical improvement is either
expected or possible, who make up about one-half of all SSI
recipients due or overdue for CDRs. Although SSA estimates that
conducting CDRs will result in removing only about 5 percent of SSI
recipients from the rolls, without CDRs, the number of ineligible
recipients would likely increase over time. Moreover, SSA estimates
that conducting CDRs on SSI adult recipients for whom medical
improvement is expected or possible results in about $3 in federal
program savings for every $1 spent conducting CDRs.
In our companion reports, we identified two options that would enable
SSA to further enhance its CDR process. First, we recommended that
SSA (1) establish less rigid requirements, to the extent it has the
authority, on who is scheduled for CDRs and how often; (2) ensure
that it makes contact with all recipients; and (3) develop a
legislative proposal to obtain the authority needed to extend this
new process to all recipients. These actions would give SSA greater
flexibility than the current CDR scheduling system to concentrate its
efforts on recipients with the greatest potential for medical
improvement and subsequent benefit termination. Second, we
recommended that, to more efficiently use CDR resources and
strengthen its return-to-work initiatives, SSA should test the use of
its contacts during CDRs to also determine recipients' vocational
rehabilitation service needs and provide them with the support and
assistance they need to enter or reenter the workforce.
BACKGROUND
------------------------------------------------------------ Letter :2
The SSI program provides financial assistance to disabled people
whose income and resources are below specified amounts. As of
January 1996, about 2.6 million adults and 1.1 million children were
receiving SSI disability benefits and an additional 1.1 million
adults were receiving both SSI and DI benefits.\4 Most SSI recipients
qualify for Medicaid coverage, and 48 states and the District of
Columbia also supplement federal SSI payments with state SSI
benefits. In 1995, SSI disability recipients received a total of
about $21 billion in federal SSI benefits and $2.6 billion in SSI
state supplements.
Reviewing recipients' disability status, especially those most likely
to improve, is an important component of good program management.
Even though the SSI program was created to provide benefits to people
who are severely disabled or terminally ill,\5
some people do improve through treatment, surgery, or the passing of
time. Amidst concerns about fraud, waste, and abuse in the SSI
program, the Congress passed the Social Security Independence and
Program Improvements Act of 1994, which required SSA to conduct CDRs
on one-third of SSI recipients attaining age 18 and another 100,000
recipients in each of fiscal years 1996 through 1998.
The 1996 amendments to the Social Security Act required that SSA
conduct CDRs on all low-birth-weight babies within their first year
of life and at least once every 3 years on all children under age 18
whose conditions are likely to improve. The amendments single out
low-birth-weight babies because historically a relatively high
percentage of these babies, about 40 percent, have had their benefits
terminated after CDRs. The 1996 amendments replaced the requirement
that SSA conduct CDRs on one-third of recipients attaining age 18
with the requirement that SSA redetermine disability eligibility
using adult criteria for all recipients attaining age 18.\6 These
redeterminations differ from CDRs in that SSA bases decisions for
disability eligibility redeterminations on whether recipients meet
eligibility requirements; for CDRs, SSA bases eligibility decisions
on whether recipients' impairments have improved since the last
determination. Since these disability eligibility redeterminations
can be counted as CDRs on SSI recipients, this report examines SSA's
plans to conduct both of these types of reviews. In addition, this
report focuses on CDRs of SSI-only recipients because provisions in
the 1994 and 1996 laws apply only to recipients who are receiving
disability benefits solely under the SSI program, and not under both
the SSI and DI programs. Legislation has required CDRs of DI
beneficiaries, including those also receiving SSI benefits, since
1980.
The number of SSI CDRs required by the 1994 and 1996 legislation
represent large increases over the number conducted in previous
years. In fact, the number of CDRs required in fiscal year 1996
alone exceeds the total of all SSI CDRs conducted in fiscal years
1991 through 1995 (see table II.1). According to SSA, it conducted
few CDRs of SSI recipients in those years because the agency had
limited resources and no legal requirement existed. However, because
SSA had the authority to conduct SSI CDRs, SSA continued to schedule
SSI recipients for CDRs and, as a result, about 1.9 million SSI
adults and children are now due or overdue for CDRs. Tables II.2
through II.4 present selected characteristics, including age,
impairment, and length of time receiving benefits, for the SSI
population who were due or overdue for CDRs in fiscal year 1996.
SSA administers the SSI program with the help of state agencies,
called disability determination services (DDS). DDSs make disability
determinations for SSA, process initial applications, assess
recipients' potential for medical improvement, and set due dates for
and conduct CDRs. DDSs determine when recipients will be due for
CDRs on the basis of their potential for medical improvement. On the
basis of recipients' impairments and ages, DDS officials classify
individuals into one of three categories: medical improvement
expected (MIE), medical improvement possible (MIP), or medical
improvement not expected (MINE). Individuals are then scheduled for
CDRs at 6- to 18-month intervals if classified as MIE, at least once
every 3 years if classified as MIP, and once every 5 to 7 years if
classified as MINE.
In recent years, given limited resources for conducting CDRs and the
large backlog of 2.4 million DI CDRs due or overdue, SSA developed
new processes in an effort to conduct CDRs in a more cost-effective
manner. SSA developed a mailer CDR process to obtain self-reported
information on current medical conditions, treatments received, and
work activities as a low-cost alternative to full medical CDRs. The
full medical CDR process is labor-intensive and generally involves
(1) 1 of 1,300 SSA field offices that determines whether disabled
recipients continue to meet the financial eligibility requirement
regarding income and resources and (2) 1 of 54 state DDSs that
determines whether recipients continue to be disabled, which
frequently involves medical exams by at least one doctor. The
average cost of a full medical CDR is about $1,000, while the average
cost of the mailer CDR is between about $25 and $50.
In addition, on the basis of the outcomes of previously conducted
CDRs on DI beneficiaries, SSA developed statistical formulas to
estimate the likelihood of benefit termination as a result of CDRs
using recipient characteristics, such as age, impairment, length of
time on disability rolls, and previous CDR activity. SSA sends
mailer CDRs to a portion of individuals who it estimates, on the
basis of its formulas, have the lowest likelihood of benefit
termination. Cases selected for mailer CDRs are later sent to DDSs
for full medical CDRs only if responses to the mail questionnaire and
information used in the formulas to estimate the likelihood of
benefit termination warrant a more comprehensive review.
--------------------
\4 SSI recipients can also qualify for disability benefits under the
DI program if they have enough work experience to be insured under
Social Security.
\5 The Social Security Act defines disability as the inability to
engage in substantial gainful activity (SGA) by reason of any
medically determinable physical or mental impairment expected to last
for a continuous period of at least 12 months or result in death.
SGA is defined as earnings of more than $960 a month for blind
individuals and $500 a month for other disabled individuals.
Children qualify for benefits if they have impairments of comparable
severity.
\6 For most recipients, SSA starts basing disability decisions on
adult rather than child criteria at age 18.
SSA PLANS TO CONDUCT REQUIRED
SSI CDRS AND TEST THE USE OF
THE MAILER CDR
------------------------------------------------------------ Letter :3
For fiscal year 1996, SSA planned to conduct full medical CDRs for
the legally required SSI CDRs and to test the mailer CDR process on
over 100,000 additional SSI recipients. Table 1 presents, for fiscal
year 1996, the number of CDRs SSA specified in planning documents and
the number SSA had initiated and DDSs had completed as of June 1996.
As of June, DDSs had completed about 60 percent of the required
reviews in each category.
Table 1
Number of SSI CDRs Planned in FY 1996
and Initiated and Completed as of June
1996
Completed/
Initiate returned CDRs as
Category Planned CDRs d CDRs of June 1996
---------------- ------------ -------- ----------------
Legally required CDRs
----------------------------------------------------------
Full medical-- Between 18,000\a 10,500
18-year-olds 16,000 and
18,000
Full medical-- 100,000 109,800\ 64,000
other age a
groups
Additional CDRs
----------------------------------------------------------
Mailer 119,400 107,900\ 53,000
b
==========================================================
Total CDRs Between 235,700 127,500
235,400 and
237,400
----------------------------------------------------------
Note: Because of updating, the numbers reported in this table differ
somewhat from planned CDRs contained in SSA's 7-year plan for fiscal
years 1996 through 2002.
\a Full medical CDRs sent to SSA field offices.
\b Mailer CDRs sent to SSI recipients. SSA revised its initial plans
and sent out mailer CDRs to 107,900 SSI recipients in fiscal year
1996.
SSA is currently modifying its CDR plan for fiscal years 1997 through
2002, which were developed before the enactment of new SSI CDR
requirements under the Personal Responsibility and Work Opportunity
Reconciliation Act of 1996. Under the new requirements, SSA must
conduct at least 150,000 SSI CDRs in each of fiscal years 1997 and
1998, which includes disability eligibility redeterminations on
18-year-olds.\7 Prior to the new requirements, SSA's CDR plan
specified more SSI CDRs in fiscal year 1997, but fewer than 150,000
SSI CDRs in fiscal year 1998. Estimates of the minimum number of SSI
CDRs that will be required in later years were not available at the
time our fieldwork was completed. However, SSA's plan called for
dramatic increases in the number of SSI CDRs in fiscal years 1999
through 2002, ranging from 367,000 to 625,000 per year.
--------------------
\7 SSA has determined that newly required CDRs on low-birth-weight
babies and children under age 18 whose impairments are likely to
improve may count toward the 100,000 CDRs required under the Social
Security Independence and Program Improvements Act of 1994.
RESOURCES APPEAR SUFFICIENT FOR
REQUIRED SSI CDRS, BUT
COMPETING PRIORITIES POSE
CHALLENGES
------------------------------------------------------------ Letter :4
SSA plans to use CDR funds to conduct the legally required SSI CDRs
and disability eligibility redeterminations in fiscal years 1996
through 2002. In fiscal year 1996, SSA for the first time set aside
regular administrative funds for CDRs, and the Congress took steps to
increase funding for CDRs. SSA set aside $200 million for both SSI
and DI CDRs and plans to continue that level of funding at least
through fiscal year 2002. In addition, the Contract With America
Advancement Act of 1996 established a new funding mechanism\8 for
CDRs and authorized up to an additional $2.7 billion for SSI and DI
CDRs through fiscal year 2002. This was about $1 billion less than
the amount SSA had requested from the Congress and believed would be
sufficient, along with regular administrative funds, to conduct CDRs
on all SSI recipients who were due or overdue for review and all
required DI CDRs through fiscal year 2002. The 1996 amendments to
the Social Security Act subsequently authorized an additional $250
million for SSI CDRs and disability eligibility redeterminations\9 in
fiscal years 1997 and 1998. Combined, regular administrative funding
for CDRs and the new budget authority could total over $4 billion in
fiscal years 1996 through 2002.
Competing priorities, such as conducting legally required DI CDRs,
which include an enormous backlog of reviews, may pose challenges to
conducting all required SSI CDRs in fiscal years 1997 through 2002.
Furthermore, the same DDS staff who conduct SSI CDRs also conduct DI
CDRs and process initial applications and other reviews of disability
eligibility required by law. Although SSA has estimated that funds
are sufficient to conduct all required SSI and DI\10 CDRs in fiscal
years 1996 through 2002, in our companion reports we question whether
CDR funds are sufficient to meet those CDR goals. To the extent that
CDR funds are not sufficient to conduct all required SSI and DI
reviews, SSI CDRs may be scaled back, since SSA generally considers
them to be less cost-effective than DI CDRs. Only SSI CDRs were
scaled back when SSA received less CDR funding for fiscal years 1996
through 2002 than it had requested from the Congress.
Also, according to SSA, the ability to conduct CDRs is always
vulnerable to unexpected increases in initial applications and
disability eligibility redeterminations. In fact, both of the 1996
laws require work in these areas that may compete with CDRs for DDS
staff. First, SSA currently estimates that, because the Contract
With America Advancement Act eliminated drug and alcohol abuse as a
basis for receiving disability benefits, benefits will be terminated
for some of the 196,000 SSI recipients and DI beneficiaries whose
primary impairments were drug abuse and/or alcoholism. SSA expects
many of those terminated to reapply on the basis of other
impairments. Second, in fiscal years 1997 and 1998, the 1996
amendments to the Social Security Act require SSA to redetermine the
disability eligibility of between 300,000 and 400,000 children
currently receiving SSI benefits. Although these disability
eligibility redeterminations can count toward the required 100,000
SSI CDRs in those years, the law gives them precedence over required
CDRs on other children. SSA is currently evaluating the impact of
this and other required work on its ability to conduct CDRs.
--------------------
\8 The legislation authorizes funding for 7 years for CDRs from the
Old-Age and Survivors Insurance Trust Fund and the Federal Disability
Insurance Trust Fund.
\9 In addition to CDRs and disability eligibility redeterminations on
18-year-old recipients, these funds can also be used for disability
eligibility redeterminations of between 300,000 and 400,000 child
recipients who qualified for SSI under individualized functional
assessments (IFA). The 1996 amendments discontinued IFAs and
required SSA within 1 year to redetermine the disability eligibility
of all recipients who had qualified as a result of IFAs. Prior to
the 1996 amendments, children with severe impairments who did not
qualify for disability benefits based solely on their impairments
received IFAs to determine whether they would qualify on the basis of
their impairments and their inabilities to engage in age-appropriate
activities.
\10 SSA plans to conduct required CDRs on all DI worker beneficiaries
under age 59 but currently excludes from its plan required CDRs on
three categories of DI beneficiaries: workers aged 59 and over, and
disabled widows and widowers and disabled adult children of disabled,
retired, or deceased DI worker beneficiaries.
SSA USES DIFFERENT APPROACHES
TO SELECT ADULT, CHILD, AND
18-YEAR-OLD RECIPIENTS FOR SSI
CDRS
------------------------------------------------------------ Letter :5
In fiscal year 1996, SSA only conducted SSI CDRs on a portion of
recipients it considered to be cost-effective to review. In general,
these included MIEs or MIPs, who make up about one-half of all SSI
recipients due or overdue for CDRs. For SSI adult recipients, SSA
selected from among those who were classified as MIE or MIP and under
the age of 59 in that year. SSA's 1996 fiscal year plan called for
conducting about 100,000 full medical CDRs and 107,900 mailer CDRs of
adult MIE and MIP SSI recipients. For children, SSA limited its
selection to low-birth-weight babies, who totaled about 7,200.
According to SSA, it did not select other children in anticipation of
the requirement to conduct disability eligibility redeterminations on
between 300,000 and 400,000 children receiving SSI. Among
18-year-old SSI recipients, SSA selected 18,000 MIE and MIP
recipients, which is about one-third of SSI recipients attaining age
18.
For adult and 18-year-old recipients, SSA used different approaches
to select recipients for CDRs. To select adult recipients, SSA used
formulas developed for DI beneficiaries to estimate the likelihood of
benefits being terminated as a result of a CDR. As is currently done
under the DI program, SSA then selected a portion of those with the
highest and lowest estimated likelihood of benefit termination for
full medical and mailer CDRs, respectively. SSA did not select
recipients in the middle range, which contains the majority of
recipients included in the estimation process, because in this range,
the formulas are less helpful in identifying recipients who are more
likely to have their benefits terminated as a result of a CDR and,
therefore, to warrant a full medical CDR. SSA has only developed
formulas for use in selecting adult recipients. Among 18-year-old
recipients, SSA selected a judgmental sample of those it believed
would be most likely to have their benefits terminated as a result of
a CDR based on some of the characteristics used to select adult cases
for CDRs, such as impairment type and length of time receiving SSI.
SSA tested the validity of using the DI formulas to estimate the
likelihood of benefit termination for adult SSI recipients in its
1995 study of SSI CDRs and its analysis of SSI and DI population
characteristics. The SSI and DI programs are subject to the same
eligibility requirements, and SSA's 1995 study of 5,000 adult SSI
CDRs found that the formulas differentiated between cases most and
least likely to result in benefit terminations about equally well for
both disability populations. Since relatively few SSI recipients
have ever undergone a CDR, SSA did not use length of time since the
last CDR and the number of previous CDRs, which are variables
normally included in the DI formulas, when estimating the likelihood
of benefit termination for SSI recipients.
In fiscal years 1997 and 1998, required disability eligibility
redeterminations on children will count toward the requirement to
conduct at least 100,000 SSI CDRs in each year. In addition, SSA
will be required to conduct CDRs annually on (1) low-birth-weight
babies and (2) other children under age 18 at least once every 3
years starting on the date of enactment of the 1996 amendments. In
fiscal years 1997 and 1998, therefore, CDRs on children could
dominate SSI CDRs. SSA plans to continue to develop and modify the
formulas and SSI selection process as it learns more about conducting
CDRs on this population. SSA's current plans for broad CDR process
improvements include expanding the use of the formulas to children
and certain recipients classified as MINEs in order to select
individuals for full medical and mailer CDRs from these recipient
categories on the basis of their estimated likelihood of benefit
termination. SSA also plans to (1) develop a new type of mailer CDR
for gathering information on recipients' medical conditions directly
from their physicians and other treating sources and (2) obtain
Medicaid data and integrate the data into the statistical formulas to
increase the validity of the estimated likelihood of benefit
termination. These latter improvements would allow SSA to better
predict which recipients in the middle range of estimated likelihood
of benefit termination are more likely to have their benefits
terminated as a result of a CDR and, therefore, to warrant full
medical CDRs. (We discuss these plans further in app. III.)
SSI CDRS ARE COST-EFFECTIVE FOR
CERTAIN CATEGORIES OF
RECIPIENTS
------------------------------------------------------------ Letter :6
Reviewing recipients' disability status, especially those most likely
to improve, is one component of a well-managed program. Few
recipients voluntarily report medical improvement and leave the
rolls. SSA currently estimates that CDRs will remove only about 5
percent of SSI recipients from the rolls in the long run.\11 However,
if the CDR process was not in place, recipients' continuing
disability eligibility would be uncertain and the number of
ineligible recipients would likely increase over time. On the basis
of SSA's estimate that 5 percent of SSI recipients would have their
benefits terminated as a result of CDRs, we estimate that about
95,000 of the approximately 1.9 million SSI recipients currently due
or overdue for CDRs are no longer medically eligible for benefits.
In fiscal year 1996 alone, these recipients would have received about
$481 million in federal SSI benefits and about $418 million in
federal and state Medicaid benefits.
SSI CDRs on some categories of recipients appear to be
cost-effective. Benefit terminations result in SSI and Medicaid
savings at both the federal and state levels (see app. IV for
information on savings). SSA calculates cost-effectiveness for
various recipient categories by comparing (1) the estimated present
value of benefit savings due to benefit terminations resulting from
CDRs on a category with (2) the estimated total costs of conducting
CDRs for that category. Because SSA has little experience conducting
SSI CDRs, SSA cautions that estimates of savings resulting from SSI
terminations are somewhat tentative.
SSA estimates that CDRs on adults SSA has classified as MIE or MIP
save about $3 in federal SSI and Medicaid benefits for every $1 spent
conducting CDRs on those categories. State savings increase this
ratio to $4 saved for every $1 spent. SSI CDRs on low-birth-weight
babies are more cost-effective than CDRs on adults--saving about $14
in program benefits for every $1 spent conducting CDRs; however,
these children constitute less than 1 percent of the SSI disabled
population. SSA estimates that, in general, CDRs on recipients
classified as MINE are not cost-effective, and, at best, break even.
--------------------
\11 Because SSA has little experience conducting CDRs on the SSI
population, SSA cautions that any estimates of the percentage of SSI
recipients who would have their benefits terminated after a CDR are
tentative.
OPTIONS EXIST FOR MAKING CDRS
MORE COST-EFFECTIVE AND USING
CDRS TO HELP RECIPIENTS MOVE
OFF DISABILITY
------------------------------------------------------------ Letter :7
Increased SSI CDR activity comes at a time when both the Congress and
SSA have sought a CDR strategy that is more cost-effective. In the
Contract With America Advancement Act, the Congress emphasized
maximizing the combined savings from CDRs under the SSI and DI
programs. SSA has been working to improve its ability to identify
recipients for whom conducting CDRs is cost-effective.
Options exist for making SSI CDRs more cost-effective and helping SSA
meet the challenge of conducting all required CDRs. In companion
reports, we identified two options for improving the CDR process--one
that could make CDRs more cost-effective and one that would
strengthen return-to-work efforts. In addition to these options, to
increase service to the public and more efficiently use resources,
SSA is exploring coordinating CDRs with redeterminations of
recipients' financial eligibility.
REVISING REQUIREMENTS CAN
IMPROVE CDRS'
COST-EFFECTIVENESS
---------------------------------------------------------- Letter :7.1
In our companion reports, one option we proposed for improving the
CDR process was for SSA to adopt less rigid requirements for
scheduling CDRs in order to shift the emphasis from periodic reviews
to a system that is more cost-effective. The current system, in
which periodic CDRs are scheduled for all SSI recipients, including
those with virtually no potential for medical improvement, is a
costly approach to identifying the approximately 5 percent of
recipients who are likely to have improved to the point of being
found ineligible for benefits. Furthermore, the frequency of CDRs is
currently based on medical improvement classifications that do little
to identify those most likely to have their benefits terminated as a
result of a CDR. We found that the estimated likelihood of benefit
termination was very similar for recipients classified as MIE and
MIP. In addition, although millions of dollars are spent annually to
conduct periodic CDRs, some individuals, especially DI beneficiaries
for whom SSA is not conducting CDRs, have received benefits for years
without having any contact with SSA regarding their disability or
their ability to return to work despite continuing disability.
We recommended in these reports a three-pronged approach to
increasing the cost-effectiveness of CDRs while maintaining program
integrity. Specifically, we recommended that SSA replace the routine
scheduling of CDRs with a new process that, if extended by the
Congress to all recipients, would (1) be cost-effective by selecting
for review individuals with the greatest potential for medical
improvement and subsequent benefit termination, (2) correct a
weakness in SSA's current CDR process by reviewing a random sample of
all other recipients, and (3) improve program integrity by
instituting contact with those not selected for CDRs or financial
eligibility redeterminations.\12 As part of this effort, we also
recommended that the Commissioner of Social Security develop a
legislative package to obtain the authority the agency needs to enact
this new process for those portions of the SSI and DI populations
that are subject to routinely scheduled CDRs.
Less rigid requirements regarding the frequency of CDRs are necessary
if CDRs are to be conducted primarily on those recipients whose cases
are most cost-effective to review--that is, those recipients with the
greatest potential for medical improvement. But to maintain program
integrity, SSA must keep abreast of the potential for medical
improvement of all recipients. Currently, SSA excludes MINE
recipients and those aged 59 and older from the selection process
altogether. We believe this weakness in the current process could be
addressed by conducting CDRs on a random sample of recipients in
these or other categories that SSA decides in the future are less
cost-effective to review.
Instituting periodic contact with recipients who are not chosen for
CDRs or financial eligibility redeterminations can help protect
program integrity by reminding recipients that their medical
conditions are being monitored and serving as a deterrent to abuse by
those no longer medically eligible for benefits. More specifically,
we believe that a new type of brief mailed contact would, at a
minimum, allow SSA to contact a majority of recipients with overdue
CDRs in the year it is implemented to remind them of their
responsibility to report medical improvements. SSA could also use
such contact to gather information to support ongoing or planned
initiatives, such as SSA's return-to-work initiatives or planned
improvements to the CDR process.
Some SSA officials expressed concern about the cost of this new type
of mailed contact with recipients. Although some administrative
funds would be used for the contact that might have been used for
CDRs or other activities, the contact should result in significant
program savings because of the considerable number of recipients who,
on the basis of SSA's experience, can be expected to refuse
repeatedly to provide the requested information and, as a result,
have their benefits terminated after a prescribed due-process
procedure is followed. On the basis of SSA's experience with CDRs
and financial eligibility redeterminations, we assume that about 1
percent of the SSI recipients who were contacted would have their
benefits terminated for noncooperation. This benefit termination
rate represents a onetime net federal savings of about $230 million
from contacting SSI recipients due or overdue for CDRs in fiscal year
1996. (See app. II for a further discussion of estimated savings.)
--------------------
\12 In order to minimize the burden placed on recipients to provide
SSA with information, those who would receive financial eligibility
redeterminations are excluded from the proposed contact.
SSA COULD BETTER UTILIZE THE
CDR PROCESS TO ENCOURAGE
RECIPIENTS TO WORK
---------------------------------------------------------- Letter :7.2
Another option we proposed for improving the use of CDR-related
resources was to support return-to-work efforts by better using the
CDR process to assess recipients' work potential, even if there is no
medical improvement, and encouraging recipients to obtain vocational
rehabilitation (VR) services. With medical advances and new
technologies creating more opportunities for disabled people to work,
some recipients who do not medically improve may nonetheless be able
to engage in substantial gainful activity. In an April 1996 report,
we recommended that the Commissioner of Social Security take
immediate action to place greater priority on return to work,
including designing a more effective means to identify and expand
recipients' work capacities and better implementing existing
return-to-work mechanisms.\13
In our companion reports, we recommended that SSA use CDR contacts to
identify recipients' productive capacities, inform them about VR
services, and encourage them to work. Currently, through contacts
during the CDR process, SSA generally provides little support and
assistance to help recipients become self-sufficient. When
conducting full medical CDRs, SSA obtains information on VR services
received since the initial application or last CDR. However, SSA and
DDS staff are neither required nor instructed to assess recipients'
work potential, make recipients aware of rehabilitation services, or
encourage them to seek VR services. SSA provides limited
encouragement through mailer CDRs by asking respondents to indicate
whether they are interested in rehabilitation or other services that
could help them obtain work. Those respondents who indicate an
interest and appear to be reasonable candidates\14 for rehabilitation
are to be referred to state VR agencies. However, on average, only
about 8 percent of all SSI recipients and DI beneficiaries are
referred for VR services.
--------------------
\13 SSA Disability: Program Redesign Necessary to Encourage Return
to Work (GAO/HEHS-96-62, Apr. 24, 1996).
\14 According to SSA's guidelines, applicants should not be referred
for VR services if they have terminal illnesses or severe or rapidly
progressive impairments not responding to treatment. VR referrals
are also limited by restrictive state policies that screen out
applicants who are not considered reasonable candidates for
rehabilitation.
SSA IS EXPLORING MORE
COORDINATION BETWEEN CDRS
AND FINANCIAL ELIGIBILITY
REDETERMINATIONS
---------------------------------------------------------- Letter :7.3
SSA is exploring the potential for better coordinating SSI CDRs with
redeterminations of recipients' financial eligibility. Each year,
SSA reviews the income, resources, and living arrangements of about 2
million SSI recipients to ensure that they still meet SSI's financial
eligibility requirements. Because staff involved in conducting CDRs
and financial eligibility redeterminations either are located in the
same place or are the same, SSA is hoping to expand coordination to
conserve its resources and provide better service to the public.
Currently, the only coordination that takes place is on the part of
SSA's field office staff, who are instructed when conducting a CDR to
gather financial eligibility redetermination information if the
recipient is also due for such a redetermination.
In exploring opportunities for coordination, SSA will have to resolve
procedural issues that, in the past, served as obstacles to pursuing
greater coordination. Over the past 10 years, interest in
coordinating the two activities has been thwarted by (1) different
schedules for conducting CDRs and financial eligibility
redeterminations throughout the year and (2) the lack of compatible
databases for SSA field office staff to determine who is scheduled
for both CDRs and financial eligibility redeterminations. SSA
believes that increased numbers of SSI CDRs and large demands on
staff resources will serve as added incentives to overcoming these
and other potential obstacles.
CONCLUSIONS
------------------------------------------------------------ Letter :8
Congressional action in 1994 prompted an increase in SSI CDR activity
that should help SSA identify and remove more ineligible recipients
from the program. In 1996, the Congress further increased the number
of required CDRs and disability eligibility redeterminations and also
increased funding that SSA can use to conduct SSI CDRs in the future.
However, SSA will likely face challenges from competing priorities
for staff resources, including required DI CDRs.
Because of increases in the required number of SSI CDRs; the large
backlog of required DI CDRs; and the Contract With America
Advancement Act, which emphasizes cost-effectiveness, we identified
in companion reports two options that could make the CDR process more
cost-effective. We recommended that a more cost-effective approach
for determining who receives CDRs may be to (1) review recipients
with the greatest potential for medical improvement and subsequent
benefit termination, (2) correct a weakness in SSA's CDR process by
reviewing a random sample of all other recipients, and (3) ensure
program integrity by instituting contact with recipients not selected
for CDRs or financial eligibility reviews. However, for this
approach to be cost-effective, SSA needs to be able to accurately
estimate the likelihood of benefit termination for all recipients,
which it can now only do for portions of those recipients classified
as MIE or MIP. Furthermore, using CDR contacts to assess recipients'
potential for and promote VR services and coordinating CDRs with
financial eligibility redeterminations could increase the efficient
use of CDR resources.
AGENCY COMMENTS AND OUR
EVALUATION
------------------------------------------------------------ Letter :9
In commenting on our draft report, SSA generally agreed with our
conclusions regarding its progress in conducting SSI CDRs and stated
that this report, along with the companion reports, provided valuable
information that would be helpful to the agency in achieving its CDR
goals in the future. The agency agreed that SSA should continually
seek ways to maintain stewardship of the disability program in the
most cost-effective manner and begin to consider which legislative
changes, if any, will produce such a result. The agency also stated
that it would (1) test using CDR contacts to assess recipients'
potential for and promote VR services and (2) continue to explore
options for coordinating CDRs with financial eligibility
redeterminations. We also received technical comments from SSA,
which we incorporated where appropriate. SSA's comments are
reprinted in appendix V.
---------------------------------------------------------- Letter :9.1
As agreed with your office, we will send copies of this report to the
Commissioner of Social Security. We will also make copies available
to others on request.
Please contact me at (202) 512-7215 if you or your staff have any
questions about this report. Other GAO contacts and staff
acknowledgments are listed in appendix VI.
Jane L. Ross
Director, Income Security Issues
SCOPE AND METHODOLOGY
=========================================================== Appendix I
This appendix provides additional details concerning our methodology.
This information includes the databases and sample used in analyzing
characteristics of SSI recipients due or overdue for CDRs in fiscal
year 1996. Also included is information on our estimates of (1)
savings resulting from benefit terminations after CDRs and (2)
onetime savings from our proposed new type of mailed contact. We
used Supplemental Security Income Record Description (SSIRD) data as
provided and did not evaluate the data's accuracy. We did our work
between September 1995 and August 1996 in accordance with generally
accepted government auditing standards.
DATABASE AND SAMPLE USED TO
ANALYZE CHARACTERISTICS OF SSI
RECIPIENTS DUE OR OVERDUE FOR
CDRS
--------------------------------------------------------- Appendix I:1
To determine the number of SSI recipients currently due or overdue
for CDRs, we used the SSA Office of Disability's (OD) CDR database.
This database contained records on all recipients SSA had determined
were due or overdue for a CDR in fiscal year 1996. For purposes of
our analyses, we took a random sample of 15 percent of those
recipients stratified by whether the (1) recipient was an adult or a
child and (2) DDSs had classified as medical improvement expected
(MIE), possible (MIP), or not expected (MINE). We eliminated from
our sample recipients whose CDR due dates were after fiscal year 1996
or who were over age 65.\15 On the basis of our sample data, we
estimated the size of the population with these exclusions. Table
I.1 contains initial population and sample sizes and final sizes
after adjustments.
Table I.1
Initial and Final Population and Sample
Sizes for SSI Recipients Due or Overdue
for CDRs in FY 1996
MIEs and
MIPs MINEs MIEs MIPs MINEs Total
----------------- ---------- ---------- ---------- ---------- ---------- ----------
Population 998,671 641,697 114,464 348,516 92,167 2,185,515
provided by OD
15% random sample 148,300 96,253 17,170 52,275 13,825 327,823
CDR due date 251 32,213 35 54 5,822 38,375
after FY 96
Age over 65 2,233 804 0 0 0 3,037
Final sample 145,816 63,236 17,135 52,221 8,003 286,411
Adjusted 972,111 421,580 114,231 348,156 53,354 1,909,432
population
-----------------------------------------------------------------------------------------
For the final sample, we obtained information on characteristics from
SSA's SSIRD and OD's CDR database. From the SSIRD, we obtained
information on age, gender, race, impairment, length of time
receiving benefits, and length of time overdue for a CDR. Because
information obtained from OD did not always differentiate between
adult MIE and MIP recipients, we used SSIRD data to classify adults
into the two categories.\16 From OD's CDR database, we obtained
information on (1) medical improvement classifications for all
children and for adults classified as MINE and (2) estimates of the
likelihood of benefit termination for adult MIE and MIP recipients,
the only recipient categories for whom likelihood of benefit
termination estimates were available.
Because we used a sample to estimate characteristics of the universe
of recipients due or overdue for CDRs in fiscal year 1996, the
reported estimates in tables II.2 through II.4 have sampling errors
associated with them. Sampling error is variation that occurs by
chance because a sample was used rather than the entire population.
The size of the sampling error reflects the precision of the
estimate--the smaller the sampling error, the more precise the
estimate. In appendix II, the tables in which we report recipients'
characteristics contain sampling errors for reported estimates
calculated at the 95-percent confidence level. This means that the
chances are about 95 out of 100 that the range defined by the
estimate, plus or minus the sampling error, contains the true
percentage. With few exceptions, the sampling errors were less than
1 percentage point. This means that for most percentages, there is a
95-percent chance that the actual percentage falls within plus or
minus 1 of the estimated percentage.
--------------------
\15 We excluded from our analysis recipients receiving disability
benefits who are over age 65. If CDRs were conducted and these
recipients were found to be no longer disabled, they would continue
to qualify for SSI benefits on the basis of their age. At age 65,
individuals receiving SSI disability benefits also become eligible
for SSI aged benefits. Such individuals can choose to continue
receiving disability benefits or switch to aged benefits.
\16 We classified as MIE any beneficiary clearly identified as such;
we classified all other recipients in OD's MIE/MIP category as
MIPs--this includes 236 of the total records in our sample.
ESTIMATES OF SAVINGS RESULTING
FROM BENEFITS BEING TERMINATED
AFTER CDRS
--------------------------------------------------------- Appendix I:2
We obtained information from a variety of sources to estimate the
present value of savings to federal and state governments resulting
from benefits being terminated after CDRs. The present value of
savings is the current value, in constant 1996 dollars, of benefits
that would have been paid over a recipient's lifetime had benefits
not been terminated. Appendix IV contains our estimates of the
present value of savings resulting from SSI CDRs.
To calculate savings, we (1) obtained estimates of federal and state
SSI and Medicaid savings and (2) calculated increased benefits that
would be paid by other programs after SSI benefits had been
terminated. From SSA, we obtained an estimate of the present value
of federal SSI savings and a formula\17 for estimating the present
value of state SSI supplement savings. From the Health Care
Financing Administration we obtained estimates of the present value
of federal and state Medicaid savings.
To calculate offsetting costs from benefits paid by other programs,
we used assumptions provided by the Congressional Budget Office
regarding increased benefits under the Food Stamp and Aid to Families
With Dependent Children (AFDC) programs that former SSI recipients
would receive once they no longer qualified for SSI benefits. In
calculating increases in Food Stamp program benefits, we assumed that
(1) about 50 percent of SSI recipients terminated as a result of a
CDR would be receiving food stamps and (2) without SSI benefits,
which count as income when determining Food Stamp benefit levels,
Food Stamp benefits would increase by about one-third of recipients'
former SSI benefit levels. In calculating offsetting AFDC costs, we
assumed that about 50 percent of children who were terminated from
the SSI program would be eligible for AFDC and that, on average,
families' AFDC benefits would increase by $70 per month, the marginal
per-child AFDC cost.
--------------------
\17 According to SSA, the present value of state SSI supplement
savings equals the average monthly state supplement times 75. This
means that, for every $1 of average monthly state SSI supplement,
states save $75.
SAVINGS ESTIMATE FOR PROPOSED
CONTACT WITH RECIPIENTS DUE OR
OVERDUE FOR CDRS
--------------------------------------------------------- Appendix I:3
The new type of brief mailed contact proposed in the companion
reports would result in program savings because we expect a
considerable number of recipients to repeatedly refuse to provide
requested information and, as a result, have their benefits
terminated.\18 As a condition of receiving benefits, recipients are
required to respond to reasonable requests for information. When
recipients do not respond, SSA first attempts to contact recipients
to determine their reasons for nonresponse. If a recipient refuses
to cooperate, SSA then follows procedures to ensure due process in
terminating benefits.
In calculating savings, we estimated the (1) number of recipients who
would be contacted and the percentage who would fail to cooperate,
(2) savings per termination, and (3) cost per contact. Table I.2
presents these estimates and summarizes assumptions used in making
the estimations. As the table shows, of the approximately 1.9
million recipients who are currently due or overdue for a CDR, we
propose that SSA contact the approximately 1,121,000 recipients who
we estimate would not be scheduled for either a CDR or a financial
eligibility redetermination in that year. We estimated that the
number of recipients scheduled for a CDR would be about 236,000
recipients, the number planned for fiscal year 1996. According to
SSA, about one-third of SSI recipients receive financial eligibility
redeterminations annually, and we estimated that about 552,200
disabled SSI recipients would be scheduled for such redeterminations
in fiscal year 1996.
Table I.2
Estimated Costs and Savings of Proposed
Mailed Contact
Number
or
dollar
amount
-------------------------------------------------------------- ------
Estimated number of recipients terminated as a result of mailed
contact
----------------------------------------------------------------------
Recipients due or overdue for CDR in FY 1996 1,909,
432
Recipients not receiving a CDR or financial eligibility 1,121,
redetermination and who would receive mailed contact 199
Recipients receiving mailed contact who would fail to 11,212
cooperate (at a noncooperation rate of .01)
Estimated savings per termination
----------------------------------------------------------------------
SSI and Medicaid savings $26,49
5
Food Stamp benefits increase that would offset savings $3,100
Net savings per beneficiary terminated $23,39
5
Estimated total savings to the federal government
----------------------------------------------------------------------
Savings $262,3
04,740
Total cost for initial mailed contact (at $25 per contact) $28,02
9,975
Program savings less costs $234,2
74,765
----------------------------------------------------------------------
On the basis of SSA's experience with mailer CDRs and financial
eligibility redeterminations, about 1 percent of the recipients who
received the mailed contact would have their benefits terminated for
continual noncooperation. We estimated that the mailed contact would
be responsible for only the first 5 years of savings resulting from
terminating SSI recipients' benefits because of failure to cooperate.
We used 5 years for our period of savings because, given SSA's system
for scheduling financial eligibility redeterminations, all SSI
recipients would have been contacted at least once within 5 years of
the mailed contact. To estimate savings, we used SSA and Health Care
Financing Administration estimates of the present value of SSI and
Medicaid savings, respectively, that would be realized each year
after benefits were terminated as a result of a CDR (see app. IV).
We assigned a cost for the initial mailed contact of $25, the lower
range of SSA's estimate for the cost of the current mailer CDR.
Because this figure overestimates the costs of the scannable mailed
contact, it provides a conservative estimate, including some
administrative and developmental costs.
--------------------
\18 See Social Security Disability: Improvements Needed to
Continuing Disability Review Process (GAO/HEHS-97-1, Oct. 16, 1996)
and Social Security Disability: Alternatives Would Boost
Cost-Effectiveness of Continuing Disability Reviews (GAO/HEHS-97-2,
Oct. 16, 1996).
ADDITIONAL DATA ON SSI CDRS
========================================================== Appendix II
Table II.1
Number of CDRs Conducted on Individuals
Receiving Both DI and SSI, and SSI Only,
FY 1988-95
DI and SSI
Fiscal year SSI only Total
---------------------------------------------- ------ ------ ------
1988 48,345 32,573 80,918
1989 42,675 86,364 129,03
9
1990 21,286 39,500 60,786
1991 8,297 18,830 27,127
1992 10,202 14,715 24,917
1993 5,335 8,517 13,852
1994 15,445 10,743 26,188
1995 25,408 34,664 60,072
======================================================================
Total 176,99 245,90 422,89
3 6 9
----------------------------------------------------------------------
Source: SSA state agency operating reports.
Table II.2
Characteristics of Adult, Child, and 18-
Year-Old SSI Recipients Due or Overdue
for CDRs in FY 1996, in Percentages
18-
Childr year-
Adults en olds Total
------------------------------------ ------ ------ -------- ------
Total estimated CDR population 1,393, 478,59 37,141 1,909,
693 8 432
Largest sampling error in column at 1.1 0.4 1.3 0.2
the 95-percent confidence level
Age in years
----------------------------------------------------------------------
Under 5 10 2
5-9 32 8
10-14 44 11
15-17 9 2
18-21 5 5 100 7
22-29 17 13
30-39 24 17
40-49 23 17
50-59 21 15
60 and over 11 8
Average age (mean) 42 11 18 34
Average age (median) 42 11 18 34
Diagnostic group
----------------------------------------------------------------------
Infectious and parasitic diseases 1 0 0 1
Neoplasms 1 2 1 1
Endocrine, nutritional, and 5 8 5 6
metabolic diseases
Disorders of blood and blood- 0 2 1 1
forming organs
Mental disorders, excluding mental 34 18 24 30
retardation
Mental retardation 19 34 48 23
Neurological and sensory disorders 5 13 8 8
Circulatory disorders 4 1 0 3
Respiratory disorders 1 3 1 2
Digestive disorders 1 0 0 1
Genitourinary disorders 1 0 0 1
Skin and subcutaneous tissue 0 0 0 0
disorders
Musculoskeletal disorders 5 1 1 4
Congenital anomalies 0 5 1 2
Injuries 2 1 1 2
Other 0 9 1 2
Not identified 20 2 6 15
Medical improvement classification
----------------------------------------------------------------------
MIE 13 23 15 16
MIP\a 56 67 69 59
MINE 30 10 16 25
Estimated likelihood of benefit termination
----------------------------------------------------------------------
Subpopulation with likelihood 862,76 \c \c \c
estimated 4\b
Under 5% 75 \c \c \c
5-24% 22 \c \c \c
25-49% 3 \c \c \c
50-74% 0 \c \c \c
75% and over 0 \c \c \c
Average likelihood (mean) 5 \c \c \c
Average likelihood (median) 2 \c \c \c
Number of years receiving benefits
----------------------------------------------------------------------
Under 4 9 28 16 14
4-5 23 37 33 27
6-7 18 16 15 18
8-9 15 8 10 13
10 and over 34 10 25 28
Average years (mean) 9 6 8 8
Average years (median) 8 5 6 6
CDR maturity
----------------------------------------------------------------------
Due in FY 1996 17 31 25 21
Due 1 year ago 17 26 25 19
Due 2 years ago 15 15 15 15
Due 3 years ago 12 8 8 11
Due 4 years ago 11 6 7 10
Due 5-10 years ago 22 10 16 19
Due over 10 years ago 3 1 3 2
Not identified 2 2 1 2
Average years (mean) 3 2 3 3
Average years (median) 3 2 2 2
Gender
----------------------------------------------------------------------
Female 56 37 37 51
Male 44 63 63 49
Race
----------------------------------------------------------------------
Black 28 34 39 29
White 47 35 44 44
Other 11 12 10 11
Not identified 14 18 7 15
----------------------------------------------------------------------
Note: Estimates are based on a 15-percent sample. For the
percentages in each column, the second row contains the largest
sampling error at the 95-percent confidence level. Because of
rounding during the estimation process, row entries may not sum to
row totals.
\a We classified 236 of the sample records for adult recipients as
MIPs because a MIE or MIP classification was not specified.
\b The total number with an estimated likelihood of benefit
termination is less than the total number for the column because SSA
provided estimates of the likelihood of benefit termination only for
MIEs and MIPs aged 59 and under. This includes recipients 1 year
older than the current cutoff SSA uses when selecting recipients for
CDRs. SSA currently limits its CDR selection to recipients under age
59. Furthermore, SSA does not estimate the likelihood of benefit
termination for children or MINEs.
\c SSA does not estimate the likelihood of benefit termination for
children.
Source: GAO analysis of SSIRD records and files supplied by OD.
Table II.3
Characteristics of SSI Adult and Child
Recipients Due or Overdue for CDRs in FY
1996, by Medical Improvement
Classification, in Percentages
MIE MIP\b MINE Total MIE MIP MINE Total MIE MIP MINE Total
-------------- -------- -------- -------- ------------ -------- --------- -------- ---------- -------- ------------ -------- ------------
Total 186,727 785,383 421,580 1,393,693 114,231 348,156 53,354 515,739 300,958 1,133,539 474,934 1,909,432
estimated CDR
population
Largest 0.6 0.3 0.4 0.2 0.7 0.4 1.1 0.3 0.5 0.2 0.4 0.2
sampling
error in
column at the
95-percent
confidence
level
Age in years
-----------------------------------------------------------------------------------------------------------------------------------------------------
Under 5 28 3 3 9 11 1 0 2
5-9 29 31 20 30 11 10 2 8
10-14 29 44 47 41 11 14 5 11
15-17 5 9 11 8 2 3 1 2
18-21 5 5 3 5 8 13 18 12 7 7 5 7
22-29 15 13 26 17 9 9 23 13
30-39 25 21 29 24 15 14 25 17
40-49 28 25 17 23 18 17 15 17
50-59 20 23 16 21 12 16 15 15
60 and over 7 12 9 11 4 9 8 8
Average age 42 44 40 42 9 12 13 11 29 34 37 34
(mean)
Average age 42 44 37 42 9 12 13 11 30 35 34 34
(median)
Diagnostic group
-----------------------------------------------------------------------------------------------------------------------------------------------------
Infectious and 0 1 1 1 0 0 0 0 0 1 1 1
parasitic
diseases
Neoplasms 1 1 0 1 5 1 1 2 3 1 0 1
Endocrine, 6 7 1 5 10 7 4 7 7 7 2 6
nutritional,
and metabolic
diseases
Disorders of 0 0 0 0 1 2 1 2 1 1 0 1
blood and
blood-
forming
organs
Mental 54 41 11 34 19 20 8 19 41 35 11 30
disorders,
excluding
mental
retardation
Mental 12 17 25 19 20 41 27 35 15 24 25 23
retardation
Neurological 5 4 9 5 9 13 23 13 6 7 10 8
and sensory
disorders
Circulatory 2 4 3 4 1 1 1 1 2 3 3 3
disorders
Respiratory 1 2 1 1 5 3 2 3 3 2 1 2
disorders
Digestive 1 1 0 1 1 0 0 0 1 1 0 1
disorders
Genitourinary 0 0 1 1 0 0 0 0 0 0 1 1
disorders
Skin and 0 0 0 0 0 0 0 0 0 0 0 0
subcutaneous
tissue
disorders
Musculoskeleta 5 7 2 5 2 1 1 1 4 5 2 4
l disorders
Congenital 0 0 1 0 5 4 8 5 2 2 1 2
anomalies
Injuries 4 2 2 2 1 1 1 1 3 2 2 2
Other 0 0 0 0 19 5 6 8 8 2 1 2
Not identified 6 11 43 20 1 1 16 2 4 8 40 15
Estimated likelihood of benefit termination
-----------------------------------------------------------------------------------------------------------------------------------------------------
Subpopulation 174,194 688,570 862,764\c \d \d \d \d
with
likelihood
estimated
Under 5% 70 76 75 \d \d \d \d
5-24% 26 21 22 \d \d \d \d
25-49% 4 3 3 \d \d \d \d
50-74% 0 0 0 \d \d \d \d
75% and over 0 0 0 \d \d \d \d
Average 5 4 5 \d \d \d \d
likelihood
(mean)
Average 2 2 2 \d \d \d \d
likelihood
(median)
Number of years receiving benefits
-----------------------------------------------------------------------------------------------------------------------------------------------------
Under 4 23 10 2 9 52 23 5 27 34 14 2 14
4-5 20 33 7 23 22 45 18 37 21 37 8 27
6-7 16 23 11 18 9 17 18 16 13 22 12 18
8-9 15 14 16 15 8 7 18 8 13 12 16 13
10 and over 25 20 65 34 8 8 41 11 19 16 62 28
Average years 7 7 13 9 5 6 9 6 6 7 12 8
(mean)
Average years 7 6 12 8 4 5 9 5 5 6 11 6
(median)
CDR maturity
-----------------------------------------------------------------------------------------------------------------------------------------------------
Due in FY 1996 10 20 15 17 21 37 12 31 14 25 15 21
Due 1 year ago 14 20 13 17 26 27 13 26 19 22 13 19
Due 2 years 20 16 13 15 18 15 13 15 19 15 13 15
ago
Due 3 years 9 12 12 12 9 7 11 8 9 11 12 11
ago
Due 4 years 8 11 12 11 6 5 10 6 7 9 12 10
ago
Due 5-10 years 29 17 29 22 15 6 31 11 24 14 29 19
ago
Due over 10 5 2 3 3 2 1 6 1 4 1 3 2
years ago
Not identified 4 2 2 2 4 1 4 2 4 2 3 2
Average years 4 3 4 3 3 2 4 2 3 3 4 3
(mean)
Average years 3 2 4 3 2 1 4 2 4 2 4 2
(median)
Gender
-----------------------------------------------------------------------------------------------------------------------------------------------------
Female 55 57 54 56 38 36 41 37 49 51 52 51
Male 45 43 46 44 62 64 59 63 51 49 48 49
Race
-----------------------------------------------------------------------------------------------------------------------------------------------------
Black 27 30 25 28 28 37 27 34 27 32 25 29
White 47 45 52 47 33 35 47 36 42 42 51 44
Other 13 11 9 11 11 12 14 12 12 12 10 11
Not identified 13 14 14 14 27 15 12 18 19 14 14 15
-----------------------------------------------------------------------------------------------------------------------------------------------------
Note: Estimates are based on a 15-percent sample. For the
percentages in each column, the second row contains the largest
sampling error at the 95-percent confidence level. Because of
rounding during the estimation process, row entries may not sum to
row totals.
\a Includes 18-year-olds.
\b We classified 236 of the sample records for adult recipients as
MIPs because a MIE or MIP classification was not specified.
\c The total number with an estimated likelihood of benefit
termination is less than the total number for the column because SSA
provided estimates of the likelihood of benefit termination only for
MIEs and MIPs aged 59 and under. This includes recipients 1 year
older than the current cutoff SSA uses when selecting recipients for
CDRs. SSA currently limits its CDR selection to recipients under age
59. Furthermore, SSA does not estimate the likelihood of benefit
termination for children or MINEs.
\d SSA does not estimate the likelihood of benefit termination for
children.
Source: GAO analysis of SSIRD records and files supplied by OD.
Table II.4
Characteristics of Selected SSI Adults
Due or Overdue for CDRs in FY 1996, by
Estimated Likelihood of Benefit
Termination, in Percentages
Largest
sampling
error in
Total row at the
estimated 95-percent
75% and CDR confidence
Under 5% 5-24% 25-49% 50-74% over population level
-------------- ---------- ---------- ---------- ---------- ---------- ---------- -----------
Total 646,623 188,067 25,320 2,427 327 862,764
Age in years
---------------------------------------------------------------------------------------------------
18-21 97 1 2 0 0 50,860 0.4
22-29 73 17 10 1 0 132,907 0.6
30-39 60 35 4 0 0 210,314 0.5
40-49 68 31 1 0 0 251,008 0.5
50-59 94 6 0 0 0 217,568 0.3
Not identified 94 6 0 0 0 107 11.9
Average age 42 39 30 29 36 41
(mean)
Average age 43 40 29 28 33 41
(median)
Diagnostic group
---------------------------------------------------------------------------------------------------
Infectious and 62 31 6 1 0 9,500 3
parasitic
diseases
Neoplasms 34 52 13 1 0 9,560 3
Endocrine, 78 19 3 0 0 54,974 1
nutritional,
and metabolic
diseases
Disorders of 43 29 21 7 0 4,287 4
blood and
blood-
forming
organs
Mental 80 18 2 0 0 396,135 0
disorders,
excluding
mental
retardation
Mental 97 2 0 0 0 150,381 0
retardation
Neurological 66 34 1 0 0 35,440 1
and sensory
disorders
Circulatory 85 14 0 0 0 23,067 1
disorders
Respiratory 67 31 2 0 0 10,420 2
disorders
Digestive 29 68 4 0 0 7,173 3
disorders
Genitourinary 43 46 9 2 0 3,407 4
disorders
Skin and 30 46 19 5 0 1,647 6
subcutaneous
tissue
disorders
Musculoskeleta 60 36 3 0 0 46,247 1
l disorders
Congenital 56 27 12 4 0 2,980 5
anomalies
Injuries 38 51 10 0 0 20,347 2
Other 52 35 11 1 1 2,627 5
Not identified 43 47 8 1 0 84,574 1
Medical improvement classification
---------------------------------------------------------------------------------------------------
MIE 70 26 4 0 0 174,194 1
MIP 76 21 3 0 0 688,570 0
Number of years receiving benefits
---------------------------------------------------------------------------------------------------
Under 4 66 29 4 0 0 117,747 0.7
4-5 73 24 3 0 0 270,948 0.4
6-7 76 21 3 0 0 181,527 0.5
8-9 76 20 4 1 0 114,561 0.6
10 and over 81 16 2 0 0 177,881 0.5
Average years 8 7 7 7 13 7
(mean)
Average years 7 6 6 7 12 6
(median)
CDR maturity
---------------------------------------------------------------------------------------------------
Due in FY 1996 73 23 3 0 0 165,647 0.6
Due 1 year ago 73 24 3 0 0 165,954 0.6
Due 2 years 75 22 3 0 0 140,614 0.6
ago
Due 3 years 76 21 2 0 0 98,540 0.7
ago
Due 4 years 80 18 2 0 0 88,540 0.7
ago
Due 5-10 years 77 19 3 0 0 160,987 0.5
ago
Due over 10 78 19 2 0 0 20,313 1.5
years ago
Not identified 60 35 5 1 0 22,167 1.7
Average years 3 3 3 3 5 3
(mean)
Average years 3 2 2 2 6 2
(median)
Gender
---------------------------------------------------------------------------------------------------
Female 76 21 3 0 0 476,482 0.3
Male 74 23 3 0 0 386,135 0.4
Not identified 91 5 5 0 0 147 12.0
Race
---------------------------------------------------------------------------------------------------
Black 75 22 3 0 0 254,748 0.4
White 74 22 3 0 0 394,255 0.4
Other 74 23 3 0 0 97,900 0.7
Not identified 78 20 2 0 0 115,861 0.6
---------------------------------------------------------------------------------------------------
Notes: Estimates are based on a 15-percent sample. For the
percentages in each row, the last column contains the largest
sampling error at the 95-percent confidence level. Because of
rounding during the estimation process, row entries may not sum to
row totals.
Data in this table include recipients 1 year older than the current
cutoff SSA uses when selecting recipients for CDRs because SSA
provided estimates of the likelihood of benefit termination for MIEs
and MIPs aged 59 and under. SSA currently limits its CDR selection
to recipients under 59. Furthermore, SSA does not estimate the
likelihood of benefit termination for children or MINEs.
Source: GAO analysis of SSIRD records and files supplied by OD.
SSA PLANS FOR CDR PROCESS
IMPROVEMENTS
========================================================= Appendix III
SSA plans to expand and enhance its procedures for selecting SSI
recipients and DI beneficiaries for CDRs and conducting the reviews.
More specifically, SSA plans to (1) expand the use of formulas for
estimating the likelihood of benefit termination to children and
certain recipients classified as MINE and (2) obtain medical
treatment information about recipients and integrate the data into
the process for selecting recipients for CDRs.
SSA plans to expand the use of statistical formulas for estimating
the likelihood of benefit termination to children and a portion of
both the SSI recipients and DI beneficiaries classified as MINE. To
develop the formulas to estimate the likelihood of benefit
terminations for child SSI recipients, SSA plans to conduct reviews
of children by selecting cases from across the range of impairments.
According to SSA, this process expansion is not expected to begin
until about fiscal year 1998 because of new legislation eliminating
the individualized functional assessment (IFA) component of
disability eligibility criteria for child recipients. An SSA
official explained that the agency is close to validating the use of
the formulas for MINEs and plans to begin conducting CDRs on this
group in fiscal year 1997. Included in this process expansion will
be MINEs who are classified as such because they are older rather
than because of their impairment. SSA believes that these
age-classified MINEs may be cost-effective to review because some of
them may have improved medically to the extent that they are no
longer disabled. At this time, SSA does not have any plans to
include the MINEs who are classified as such because they are
believed to have permanent disabilities.
SSA also plans to pursue two approaches for the collection of medical
treatment information about recipients. First, SSA has plans to
develop a new type of low-cost mailer CDR to be sent to recipients'
physicians and other treating sources. At this time, SSA only
selects individuals for CDRs from among the groups with the highest
and lowest estimated likelihood of benefit termination for full
medical and mailer CDRs, respectively. SSA officials explained that
they do not conduct CDRs on SSI recipients or DI beneficiaries with
likelihood of benefit termination estimates in the middle range
because they believe the formulas do not adequately distinguish
between these individuals for purposes of determining who in this
group should receive full medical CDRs. According to SSA, if it
conducted mailer CDRs on the middle group, this would likely result
in more beneficiaries being subsequently referred for full medical
CDRs than would be cost-effective. Similarly, if it conducted full
medical CDRs on the middle group, it would be using a higher-cost
process than SSA believes is necessary for some in this group. SSA
believes the new mailer CDR to physicians and other treating sources
would provide information about medical conditions and treatments
received that would help SSA to determine who in the middle group has
a likelihood of benefit termination warranting a full medical CDR.
Second, SSA plans to obtain Medicaid data and integrate the data into
the statistical formulas to increase the validity of the estimated
likelihood of benefit termination. SSA expects that the additional
information will also allow it to better identify the appropriateness
of a mailer or full medical CDR for recipients with estimates of the
likelihood of benefit termination in the middle range. Given that
the majority of SSI recipients and DI beneficiaries for whom
likelihood of benefit termination is estimated fall into the middle
range of estimates, these CDR process enhancements are particularly
critical to SSA's ability to meet its CDR goals over the next 7
years.
FEDERAL AND STATE SAVINGS
RESULTING FROM SSI CDRS
========================================================== Appendix IV
Our calculations of present value savings are based on estimates
provided by SSA and the Health Care Financing Administration and
assumptions provided by the Congressional Budget Office on offsetting
Food Stamp and AFDC costs. For SSI CDRs, savings result from SSI and
Medicaid benefits being terminated for recipients who no longer meet
the program's definition of disability. Table IV.1 contains the
present value of federal savings per CDR termination. As the table
indicates, the present value of savings to the federal government per
CDR termination is, for adults, $42,000 and for children, $33,000.
This means, for example, that for every adult for whom a CDR results
in a termination, the federal government could expect to save, on
average, $42,000 (in constant 1996 dollars) that it would have paid
over the recipient's lifetime had benefits not been terminated.
The savings for children are less than those for adults, primarily
because, even after being terminated from the SSI program, a majority
of children would continue to qualify for Medicaid benefits on the
basis of their families' economic status or their participation in
AFDC. As the table shows, these estimates also take into account
offsetting costs resulting from increases in AFDC and Food Stamp
benefits that some former SSI recipients would receive once they no
longer qualified for SSI.
Table IV.1
Present Value of Federal Savings From
SSI CDRs
Childr
Program Adults en
------------------------------------------------------ ------ ------
Savings
SSI $31,50 $33,00
0 0
Medicaid 16,000 7,000
Total 47,500 40,000
Offsets
Food Stamp 5,500 6,000
AFDC \a 1,000
Total 5,500 7,000
Net savings $42,00 $33,00
0 0
----------------------------------------------------------------------
\a Not applicable.
Sources: SSA's Office of the Actuary, Health Care Financing
Administration's Office of the Actuary, and Congressional Budget
Office's Human Resources Cost Estimate Unit.
Because many states pay SSI state supplements and all states share in
Medicaid, SSI CDRs also result in states realizing SSI and Medicaid
savings.\19 Because benefit levels vary across states, the present
value of savings also varies. Table IV.2 shows, for the five states
with the largest total state supplement payments in fiscal year 1994,
a range of over $10,000 in the present value of SSI supplement
savings. States' savings also vary because the size of their
recipient populations differ. Table IV.2 shows the wide variation in
potential state savings based on (1) the number of disabled
individuals currently receiving state SSI supplements and due and
overdue for CDRs and (2) SSA's current estimates of a 5-percent
benefit termination rate. The present value of state Medicaid
savings would average about $11,100 for adults and $4,800 for
children.
Table IV.2
Present Value of SSI State Supplement
Savings and Potential Total Savings in
Five States
Fiscal
year
1994 Present Number of Present
total value of recipients value of
suppleme SSI state due or potential
nts paid savings overdue SSI state
(in per for CDRs savings\a
millions terminatio in March (in
State ) n 1996 millions)
------------ -------- ---------- ---------- ----------
California $963.0 $13,100 235,500 $154.5
New York 260.5 5,100 164,000 41.8
Wisconsin 100.3 6,300 51,100 16.1
Massachusett 74.9 5,800 41,100 12.0
s
Pennsylvania 71.6 2,800 80,700 11.5
----------------------------------------------------------
\a Assumes that CDRs would result in benefit terminations for 5
percent of recipients due or overdue for CDRs.
Sources: SSA's Office of the Actuary and Office of Research,
Evaluation, and Statistics.
(See figure in printed edition.)APPENDIX V
--------------------
\19 In 1995, over 2 million disabled individuals received state SSI
supplements.
COMMENTS FROM THE SOCIAL SECURITY
ADMINISTRATION
========================================================== Appendix IV
(See figure in printed edition.)
(See figure in printed edition.)
(See figure in printed edition.)
GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================== Appendix VI
GAO CONTACTS
Robert L. MacLafferty, Assistant Director
Susan E. Arnold, Evaluator-in-Charge, (415) 904-2000
STAFF ACKNOWLEDGMENTS
In addition to those named above, the following persons made
important contributions to this report: Chris C. Crissman,
Assistant Director; Kerry Gail Dunn, Senior Evaluator; Julian M.
Fogle, Senior Evaluator; Ann Lee, Senior Evaluator; Elizabeth A.
Olivarez, Evaluator; Susan K. Riggio, Evaluator; and Ann T. Walker,
Evaluator (Database Manager).
RELATED GAO PRODUCTS
=========================================================== Appendix 0
Social Security Disability: Alternatives Would Boost
Cost-Effectiveness of Continuing Disability Reviews (GAO/HEHS-97-2,
Oct. 16, 1996).
Social Security Disability: Improvements Needed to Continuing
Disability Review Process (GAO/HEHS-97-1, Oct. 16, 1996).
Supplemental Security Income: Some Recipients Transfer Valuable
Resources to Qualify for Benefits (GAO/HEHS-96-79, Apr. 30, 1996).
SSA Disability: Program Redesign Necessary to Encourage Return to
Work (GAO/HEHS 96-62, Apr. 24, 1996).
PASS Program: SSA Work Incentives for Disabled Beneficiaries Poorly
Managed (GAO/HEHS-96-51, Feb. 28, 1996).
SSA Rehabilitation Programs (GAO/HEHS-95-253R, Sept. 7, 1995).
Supplemental Security Income: Disability Program Vulnerable to Fraud
When Middlemen Are Used (GAO/HEHS-95-116, Aug. 31, 1995).
Social Security Disability: Management Action and Program Redesign
Needed to Address Long-Standing Problems (GAO/HEHS-95-233, Aug. 3,
1995).
Supplemental Security Income: Growth and Changes in Recipient
Population Call for Reexamining Program (GAO/HEHS-95-137, July 7,
1995).
Disability Insurance: Broader Management Focus Needed to Better
Control Caseload (GAO/T-HEHS-95-164, May 23, 1995).
Supplemental Security Income: Recipient Population Has Changed as
Caseloads Have Burgeoned (GAO/T-HEHS-95-120, Mar. 27, 1995).
Social Security: Federal Disability Programs Face Major Issues
(GAO/T-HEHS-95-97, Mar. 2, 1995).
Social Security: Rapid Rise in Children on SSI Disability Rolls
Follows New Regulations (GAO/HEHS-94-225, Sept. 9, 1994).
Social Security: New Continuing Disability Review Process Could Be
Enhanced (GAO/HEHS-94-118, June 27, 1994).
*** End of document. ***