Social Security Disability Insurance: Multiple Factors Affect
Beneficiaries' Ability to Return to Work (Letter Report, 01/12/98,
GAO/HEHS-98-39).

Pursuant to a congressional request, GAO reviewed the Social Security
Disability Insurance (DI) program, focusing on: (1) the self-reported
health and functional status of DI beneficiaries who work while still on
the rolls; (2) the occupations, earnings, and benefits of working
beneficiaries; (3) factors that working beneficiaries believe were
helpful in becoming employed; and (4) working beneficiaries' long-term
employment plans, including factors perceived as positively and
negatively affecting work plans.

GAO noted that: (1) in general, beneficiaries that GAO interviewed
achieved a range of work outcomes--some had substantial attachment to
the labor force, and others reported more modest gains; (2) respondents
achieved these outcomes despite indicating significant limitations or
difficulties associated with their impairments; (3) many respondents
rated their disability as severe or somewhat severe, reported
experiencing difficulty getting through the work day, and reported
having difficulty performing daily tasks and activities; (4)
nevertheless, beneficiaries were gainfully employed and, on average, had
moderate pay and benefits; most were satisfied in their positions; (5)
most beneficiaries GAO interviewed reported that financial need and the
desire to enhance self-esteem were the main reasons for attempting work;
(6) they indicated that a range of factors enabled them to return to
work; (7) those most prominently cited were improved functioning through
health care intervention and encouragement from family, friends, health
care providers, and coworkers; (8) to a somewhat lesser extent,
respondents told GAO that: (a) a flexible work schedule that allowed
them to receive health care services; (b) job-related training and
vocational rehabilitation services; and (c) high self-motivation also
helped facilitate employment; (9) DI work incentives and assistance from
Social Security Administration staff appeared to play a limited role in
helping beneficiaries become employed, although a number of respondents
said the program provision allowing them to work for a period of time
without losing cash and medical benefits, as well as the provision to
retain health care coverage for a limited time period after cash
assistance ends, was helpful; (10) about four of every ten respondents
told GAO that they planned to leave the rolls in the future; (11)
availability of worksite-based health insurance appears to differentiate
respondents who plan to leave the rolls in the future from respondents
who plan to stay; (12) many respondents--those planning to leave the
rolls as well as those planning to stay--regard their future health
status as an important factor affecting their plans; (13) many
respondents told GAO they had experienced impediments to employment such
as limited skills and training, or employers' not recognizing their
ability; and (14) such factors could affect respondents' future
attachment to the labor force.

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

 REPORTNUM:  HEHS-98-39
     TITLE:  Social Security Disability Insurance: Multiple Factors 
             Affect Beneficiaries' Ability to Return to Work
      DATE:  01/12/98
   SUBJECT:  Disability insurance
             Disability benefits
             Federal social security programs
             Social security benefits
             Persons with disabilities
             Employees with disabilities
             Beneficiaries
             Disadvantaged persons
             Eligibility determinations
             Vocational rehabilitation
IDENTIFIER:  Social Security Disability Insurance Program
             Medicare Program
             Supplemental Security Income Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to the Chairman, Subcommittee on Social Security, Committee on
Ways and Means, House of Representatives

January 1998

SOCIAL SECURITY DISABILITY
INSURANCE - MULTIPLE FACTORS
AFFECT BENEFICIARIES' ABILITY TO
RETURN TO WORK

GAO/HEHS-98-39

Working DI Beneficiaries

(106518)


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

  ADA - Americans With Disabilities Act
  CDR - continuing disability review
  DCI - data collection instrument
  DDS - Disability Determination Service
  DI - Disability Insurance
  HIV - human immunodeficiency virus
  PASS - Plan for Achieving Self-Support
  SSA - Social Security Administration
  SSI - Supplemental Security Income
  VR - vocational rehabilitation

Letter
=============================================================== LETTER


B-277977

January 12, 1998

The Honorable Jim Bunning
Chairman, Subcommittee on Social Security
Committee on Ways and Means
House of Representatives

Dear Mr.  Chairman: 

Social Security Disability Insurance (DI) is one of the largest
federal programs providing cash assistance to people with
disabilities.  Established in 1956, DI is an insurance program funded
by payroll taxes paid by workers and their employers into a Social
Security trust fund.  In 1996, about 4.4 million working-age people
(aged 18 to 64) received DI cash benefits.  The average monthly cash
benefit in 1996 was $704, and the overall amount of cash benefits
paid was about $40 billion. 

Not more than 1 in 500 DI beneficiaries leaves the rolls by returning
to work.  However, the Social Security Administration (SSA) estimates
that annually about 8,500 beneficiaries successfully complete a
9-month test of their ability to work in paid employment and enter an
extended period of eligibility intended to help ease their transition
to work.  Yet, relatively little is known about the confluence of
factors that helps beneficiaries overcome employment challenges and
disincentives, and the factors that inhibit them from achieving an
earnings level that leads to self-sufficiency. 

Recently, Members of the Congress and advocates for people with
disabilities have proposed various reforms, including tax incentives,
to help improve return-to-work outcomes.  These reforms include
changes that would allow beneficiaries who work while on the rolls to
keep more of their earnings, safeguard medical coverage, and enhance
vocational rehabilitation. 

To provide more information about the experiences of working
beneficiaries, we agreed to

  -- identify the self-reported health and functional status of DI
     beneficiaries who work while still on the rolls;

  -- identify the occupations, earnings, and benefits of working
     beneficiaries;

  -- report factors that working beneficiaries believe were helpful
     in becoming employed; and

  -- explore working beneficiaries' longer-term employment plans,
     including factors perceived as positively and negatively
     affecting work plans. 

To accomplish these objectives, we developed a structured interview
guide on the basis of a literature review and discussions with
experts and advocates.  We conducted survey interviews with 69 people
who were receiving DI benefits and working in the Washington, D.C.;
Atlanta; or San Francisco metropolitan areas.  We identified
potential respondents from a randomized list of DI beneficiaries who
reported earnings to SSA in 1995.  However, because neither the
metropolitan areas selected nor the people interviewed constituted a
random sample, our results are not generalizable to the entire
population of working DI beneficiaries.  The names of respondents
quoted in this report have been changed to protect their identities. 
Although we did not independently verify the data we used from SSA in
this report, the data are generally used by SSA researchers and
managers for program purposes.  Our work was performed in accordance
with generally accepted government auditing standards between January
and October 1997.  (For more detailed information on our scope and
methodology, see app.  I.)


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

In general, beneficiaries we interviewed achieved a range of work
outcomes--some had substantial attachment to the labor force, and
others reported more modest gains.  Respondents achieved these
outcomes despite indicating significant limitations or difficulties
associated with their impairments.  Respondents identified many
factors that they believe affected their ability to return to work: 
Services that improved health and functioning were paramount, while
assistance from SSA had limited impact.  In addition, beneficiaries
told us that their health status could affect their longer-term work
plans. 

More specifically, many respondents rated their disability as severe
or somewhat severe, reported experiencing difficulty getting through
the work day, and reported having difficulty performing daily tasks
and activities.  Nevertheless, beneficiaries were gainfully employed
and, on average, had moderate pay and benefits; most were satisfied
in their positions.  The typical beneficiary reported working 28
hours each week and receiving about $10.60 an hour; about one-third
had employer-based health insurance.  About 7 of every 10 said they
were engaged in professional, managerial, administrative support,
technical, or sales positions. 

Most beneficiaries we interviewed reported that financial need and
the desire to enhance self-esteem were the main reasons for
attempting work.  They indicated that a range of factors enabled them
to return to work.  Those most prominently cited were improved
functioning through health care intervention and encouragement from
family, friends, health care providers, and coworkers.  To a somewhat
lesser extent, respondents told us that (1) a flexible work schedule
that allowed them to receive health care services, (2) job-related
training and vocational rehabilitation services (especially
on-the-job training and help finding a job), and (3) high
self-motivation also helped facilitate employment.  DI work
incentives and assistance from SSA staff appeared to play a limited
role in helping beneficiaries become employed, although a number of
respondents said the program provision allowing them to work for a
period of time without losing cash and medical benefits, as well as
the provision to retain health care coverage for a limited time
period after cash assistance ends, was helpful. 

About four of every ten respondents told us they planned to leave the
rolls in the future.  Availability of worksite-based health insurance
appears to differentiate respondents who plan to leave the rolls in
the future from respondents who plan to stay.  Many
respondents--those planning to leave the rolls as well as those
planning to stay--regard their future health status as an important
factor affecting their plans.  Also, many respondents told us they
had experienced impediments to employment such as their limited
skills and training, or employers' not recognizing their ability. 
Such factors could affect respondents' future attachment to the labor
force. 


   BACKGROUND
------------------------------------------------------------ Letter :2

DI is designed to insure workers against loss of income due to a
disabling condition.  Workers who have worked long enough and
recently enough become insured for DI coverage.  To meet the DI
statutory definition of disability, an adult must be determined to be
unable to engage in any substantial gainful activity (in 1997, the
substantial gainful activity level was $1,000 a month for people who
are blind and $500 a month for people with other disabilities) by
reason of any medically determinable physical or mental impairment
that can be expected to result in death or that has lasted or can be
expected to last at least 1 year.  Moreover, the statutory definition
further specifies that, for a person to be determined to be disabled,
the impairment must be of such severity that the person not only is
unable to do his or her previous work but, considering his or her
age, education, and work experience, is unable to do any other kind
of substantial work that exists in the national economy. 

Of the 4.4 million DI beneficiaries in 1996, about 691,000 received
Supplemental Security Income (SSI) benefits.  SSI is a means-tested
income assistance program for disabled, blind, or aged individuals
regardless of their prior participation in the labor force.\1
Established in 1972 for individuals with low income and limited
resources, SSI is financed from general revenues.\2

People with disabilities concurrently receiving both DI and SSI
benefits have enough work credit to qualify for DI benefits and low
enough income and resources to qualify for SSI benefits as well. 

A wide range of impairments can qualify people for disability
benefits.  In 1996, 26 percent of adults receiving DI benefits had a
mental disorder (such as schizophrenia and anxiety disorders, but
excluding mental retardation).  Other common types of impairments
included musculoskeletal conditions (22 percent) and diseases of the
circulatory system (12 percent). 

Once a person is on the disability rolls, benefits continue until one
of three events occurs:  (1) the beneficiary dies; (2) the
beneficiary converts to Social Security retirement benefits at age
65; or (3) SSA determines that the beneficiary is no longer eligible
for benefits because either earned income exceeds the allowable limit
or SSA has decided that the beneficiary's medical condition has
improved to the point that he or she is no longer considered
disabled.  To determine whether medical conditions have improved, SSA
performs periodic continuing disability reviews (CDR).\3


--------------------
\1 The DI and SSI programs use the same statutory definition of
disability. 

\2 General revenues include taxes, customs duties, and miscellaneous
receipts collected by the federal government but not earmarked by law
for a specific purpose. 

\3 SSA is to conduct a CDR at least once every 3 years on DI
beneficiaries whose medical improvement is possible or expected. 
When medical improvement is not expected, SSA is to schedule CDRs at
least once every 7 years.  See Social Security Disability: 
Alternatives Would Boost Cost-Effectiveness of Continuing Disability
Reviews (GAO/HEHS-97-2, Oct.  16, 1996). 


      STATUTE PROVIDES FOR
      RETURNING BENEFICIARIES TO
      WORK
---------------------------------------------------------- Letter :2.1

The Social Security Act states that people applying for disability
benefits should be promptly referred to state vocational
rehabilitation (VR) agencies for services so that as many applicants
as possible can return to productive activity.  State Disability
Determination Service (DDS) offices, which act for SSA in making
disability eligibility determinations, decide whether to refer an
individual applicant to a state VR agency.  Despite this, as we
previously reported, DDSs referred for VR services on average only
about 8 percent of adults awarded disability benefits, and less than
10 percent of these beneficiaries were accepted as clients.\4

To reduce the risk a beneficiary faces in trading the guaranteed
monthly income and subsidized health coverage provided by the DI
program for the uncertainties of entering competitive employment, the
law provides various work incentives intended to safeguard cash and
health benefits while a beneficiary tries to return to work.  Several
of the major work incentives available to DI beneficiaries follow. 

The trial work period allows DI beneficiaries to work for a limited
time without their earnings affecting their disability benefits. 
Each month in which earnings are more than $200 is counted as a month
of the trial work period.  When the beneficiary has accumulated 9
such months (not necessarily consecutively) within a 60-month rolling
period, the trial work period is completed. 

The extended period of eligibility begins the month following the end
of the trial work period.  The extended period of eligibility is a
consecutive 36-month period during which cash benefits are reinstated
for any month the beneficiary's earnings are less than the
substantial gainful activity level.  Cash benefits may be paid for an
even longer period if a person is unable to perform any substantial
gainful activity. 

Another work incentive allows for continued Medicare coverage for at
least 39 months following a trial work period as long as medical
disability continues.  When this premium-free period ends, medically
disabled individuals may elect to purchase Medicare coverage at the
same monthly premium--more than $300 for full coverage in 1996--paid
by individuals aged 65 or older who are not insured for Social
Security retirement benefits.  Individuals entering the program
generally must complete a 24-month waiting period before they are
eligible for Medicare benefits at no cost; beneficiaries do not have
the option of purchasing Medicare during the waiting period. 


--------------------
\4 Social Security:  Little Success Achieved in Rehabilitating
Disabled Beneficiaries (GAO/HRD-88-11, Dec.  7, 1987). 


      ADVANCES HAVE FACILITATED
      EMPLOYMENT
---------------------------------------------------------- Letter :2.2

Many technological and medical advances, along with social changes,
have created more opportunities for some individuals with
disabilities to work.  These factors have sparked an increased
interest in public policy on the employment of people with
disabilities. 

Electronic communications and assistive technologies--such as
scanners, synthetic voice systems, standing wheelchairs, and modified
automobiles and vans--have given greater independence to people with
some disabilities.  Advances in the management of disability--such as
the development of medication to control mental illness or
computer-aided prosthetic devices--have helped reduce the functional
limitations associated with some disabilities.  Also, the shift in
the U.S.  economy toward the service industry may have opened new
opportunities for some people with physical impairments. 

Social change has also promoted the goals of greater inclusion of and
participation by people with disabilities in the mainstream of
society.  For instance, over the past 2 decades, people with
disabilities have sought to remove environmental barriers that impede
them from fully participating in their communities.  Moreover, the
Americans With Disabilities Act (ADA) supports the full participation
of people with disabilities in society and fosters the expectation
that people with disabilities can work and have the right to work. 
The ADA prohibits employers from discriminating against qualified
individuals with disabilities and requires employers to make
reasonable workplace accommodations unless doing so would impose an
undue hardship. 


      DESPITE BENEFICIARY
      POTENTIAL, UNDERSTANDING OF
      RETURN-TO-WORK EXPERIENCE IS
      LIMITED
---------------------------------------------------------- Letter :2.3

In light of advances and social changes, determinations by SSA that
disability applicants are unable to engage in gainful employment do
not mean that DI beneficiaries can never regain their capacity to
work.  In fact, SSA's determination--a decision at one point in
time--is probably a weak predictor of future work capacity.  Indeed,
some beneficiaries become employed while on the rolls, and other
people with disabilities--who could meet the medical criteria for
benefits if they were to apply--are able to remain in the workforce
and not apply for program benefits.  On the other hand, our past work
suggests that program design and operations make it difficult or
risky for DI beneficiaries to attempt work.\5 Some evidence even
suggests that receipt of DI benefits is associated with lower success
in return-to-work interventions.\6

While beneficiaries may face many challenges in attempting to return
to work, research suggests that successful transitions to work may be
more likely for younger people with disabilities and for those who
have greater motivation to work and stronger educational
backgrounds.\7 Studies have shown that a significant number of DI
beneficiaries possess these characteristics.  The DI rolls
increasingly are composed of a significant number of younger
individuals.  About one out of five is under the age of 40.\8 In
addition, in 1993, 35 percent of the 84,000 DI beneficiaries who
responded to an SSA questionnaire expressed an interest in receiving
rehabilitation or other services that could help them return to work,
which is an indication of motivation.  Moreover, a substantial
portion--almost one in two--of a cohort of DI beneficiaries who had
been on the rolls for a decade by the early 1990s had 12 or more
years of education.\9

Yet, there is limited understanding of the return-to-work experience
of DI beneficiaries, especially that of more recent DI beneficiaries. 
We reviewed return-to-work research, but little of it focused
explicitly on individuals receiving Social Security DI benefits.  Few
studies included DI beneficiaries as the primary research group or
collected DI status as a factor germane to study outcomes, although
some of the findings appear readily generalizable to the DI
population. 

We focused on how DI beneficiaries who are currently working were
able to return to the workforce.  To better understand the factors
identified in this report, additional studies involving beneficiaries
who are not working would be needed. 


--------------------
\5 See SSA Disability:  Program Redesign Necessary to Encourage
Return to Work (GAO/HEHS-96-62, Apr.  24, 1996) and SSA Disability: 
Return-to-Work Strategies From Other Systems May Improve Federal
Programs (GAO/HEHS-96-133, July 11, 1996.) In these reports, we
identified several weaknesses in the design and implementation of the
DI program that result in comparatively few DI beneficiaries
returning to work.  We have also reported on factors that led us to
believe that SSA could do more to identify and expand the productive
capacities of beneficiaries. 

\6 Frederick E.  Menz, "Lessons and Recommendations," in Frederick E. 
Menz, Julie Eggers, Paul Wehman, and Valerie Brooke, eds., Lessons
for Improving Employment of People With Disabilities from Vocational
Rehabilitation Research (Menomonie, Wisc.:  Rehabilitation Research
and Training Center, University of Wisconsin, 1997) and K.V. 
Straaton, R.  Maisiak, J.M.  Wrigley, M.B.  White, P.  Johnson, and
P.R.  Fine, "Barriers to Return to Work Among Persons Unemployed Due
to Arthritis and Musculoskeletal Disorders," Arthritis & Rheumatism,
Vol.  39, No.  1 (Jan.  1996), pp.  101-109. 

\7 J.C.  Hennessey and L.S.  Muller, "The Effect of Vocational
Rehabilitation and Work Incentives on Helping the Disabled-Worker
Beneficiary Back to Work," Social Security Bulletin, Vol.  58, No.  1
(spring 1995), pp.  15-28; R.J.  Butler, W.G.  Johnson, and M.L. 
Baldwin, "Managing Work Disability:  Why First Return to Work Is Not
a Measure of Success," Industrial and Labor Relations Review, Vol. 
48, No.  3 (Apr.  1995), pp.  452-67; and R.V.  Burkhauser and M.C. 
Daly, "Employment and Economic Well-Being Following the Onset of a
Disability:  The Role for Public Policy," in Jerry L.  Mashaw,
Virginia Reno, Richard V.  Burkhauser, and Monroe Berkowitz, eds.,
Disability, Work, and Cash Benefits (Kalamazoo, Mich.:  W.E.  Upjohn
Institute for Employment Research, 1996), pp.  59-101. 

\8 Annual Statistical Supplement to the Social Security Bulletin
(Washington, D.C.:  SSA, Sept.  1996). 

\9 J.C.  Hennessey and L.S.  Muller, "Work Efforts of Disabled-Worker
Beneficiaries:  Preliminary Findings From the New Beneficiary
Followup Survey," Social Security Bulletin, Vol.  57, No.  3 (fall
1994), pp.  42-51.  The cohort was DI beneficiaries who became
entitled to benefits for the first time between June 1980 and June
1981, were awarded benefits before May 1982, survived up to June
1992, personally participated in their interviews, and acknowledged
receipt of benefits near date of entitlement. 


   BENEFICIARIES REPORTED FAIR
   HEALTH AND ABILITY TO FUNCTION
------------------------------------------------------------ Letter :3

Many respondents reported limitations in their health and ability to
function.  Almost one-half rated their disability as severe or
somewhat severe.  Moreover, the majority rated their health as poor
or fair.  Generally, we did not find differences on these measures
between respondents with physical impairments and those with
psychiatric impairments.  (See table 1.) The beneficiaries we
interviewed were predominantly middle-aged; most were white; most
were educated beyond the high-school level; many lived alone; and
most were not married.  (App.  II contains information on the
individual characteristics of the respondent group.)



                                Table 1
                
                   Self-Rated Health Status of 69 DI
                             Beneficiaries

                                   Number of
                                  beneficiar  Excell        Fa
                                         ies     ent  Good  ir    Poor
--------------------------------  ----------  ------  ----  --  ------
Overall health rating of                  34       2     9  14       8
 beneficiaries with physical
 impairment\a
Mental health rating of                   36       2    12  14       8
 beneficiaries with psychiatric
 impairment
Physical health rating of                 36       1    12  16       6
 beneficiaries with psychiatric
 impairment\b
----------------------------------------------------------------------
\a Numbers do not add up to 34 because answers were not obtained from
all respondents. 

\b Numbers do not add up to 36 because answers were not obtained from
all respondents. 

Respondents also indicated that they experienced certain symptoms
that affected their capacity to work.  For instance, as table 2
shows, over one-half frequently or occasionally could not concentrate
or experienced fatigue, anxiety, or pain that made it difficult to
get through the workday. 



                                Table 2
                
                     Frequency of Symptoms of 69 DI
                             Beneficiaries

                                            Very
                                      frequently
                                               /        On
Symptom                               frequently  occasion       Never
------------------------------------  ----------  --------  ----------
Lack of concentration\a                       18        21          29
Fatigue                                       19        29          21
Anxiety                                       15        24          30
Pain                                          23        16          30
Depression\a                                  16        16          35
----------------------------------------------------------------------
\a Numbers do not add up to 69 because answers were not obtained from
all respondents. 

In addition to reporting difficulties at work, a slight majority of
respondents told us they had difficulty performing at least one daily
living activity, such as preparing meals or doing light housework. 
Table 3 shows the number of people who reported difficulties with
each of nine activities.  Overall, 14 respondents had difficulty
performing three or more activities. 



                                Table 3
                
                   Number of Beneficiaries Reporting
                 Difficulty Performing Daily Activities
                             by Themselves

Activity                                                        Number
--------------------------------------------------------------  ------
Keeping track of money and bills                                    18
Getting around outside one's home                                   15
Doing light housework                                               14
Preparing meals                                                     13
Getting in or out of a bed or chair                                 10
Bathing or washing                                                   7
Getting around inside one's home                                     6
Using or getting to the toilet                                       4
Dressing                                                             3
----------------------------------------------------------------------

   BENEFICIARIES REPORTED VARYING
   WORK OUTCOMES
------------------------------------------------------------ Letter :4

Respondents indicated a range of work outcomes:  Some appeared to
have substantial attachment to the labor force, and others had more
modest outcomes.  On average, respondents worked 28 hours over 4 days
a week.  Twenty-eight respondents worked 35 or more hours a week,
although 19 worked less than 20 hours a week.  The median length of
time--the value at which half of the values are above and half are
below--respondents were employed at their current jobs was 18 months. 
(These and other work status outcomes are presented--the value at
which half of the values are above and half are below--in greater
detail in app.  III.)

Although some earned much more than others, the respondent group
averaged moderate pay and benefits.  The average wage was about
$10.60 an hour, compared with a national hourly average of about
$12.20.\10 However, respondents' median wage was $8 an hour. 
Eighteen earned $12.50 an hour or more, while 26 earned $7 an hour or
less.\11 Some, but not all, received health insurance coverage (24),
paid sick time (30), and paid vacation time (42).  A large majority
(57) reported being very or somewhat satisfied with their jobs.  The
most satisfying aspects of their jobs were a general sense of
satisfaction, social benefit, and being productive.  The unhealthy or
stressful nature of their jobs, low pay, limited challenge, difficult
schedule, and lack of accommodations were the least satisfying
aspects. 

"Technical, sales, and administrative support" was the occupational
category in which the largest number of beneficiaries was employed
(28).  Other categories included "managerial and professional
specialty" (21), "service occupation" (14), and "operator,
fabricator, laborer, and other" (6).  About 1 of every 3 respondents
worked in a lower-skill position, such as cashier or food service
worker.  Nine of every 10 respondents held one job at the time of the
interview. 


--------------------
\10 We excluded the hourly wage reported by one respondent ($175)
when calculating the average hourly wage.  The national hourly wage
is based on earnings of production or nonsupervisory workers on
private nonfarm payrolls. 

\11 Given the average number of hours worked and the hourly wages
among the 69 beneficiaries we interviewed, it appears that many
respondents are probably in a trial work period or extended period of
eligibility phase.  If these people continue to work above
substantial gainful activity levels and do not have any
impairment-related work expenses, they will eventually lose all cash
and medical benefits. 


      REASONS FOR RETURNING TO
      WORK VARIED
---------------------------------------------------------- Letter :4.1

Financial needs and valuing work as a means to feel productive and
engaged in society were two prominent reasons given by beneficiaries
for wanting to return to work.  To a lesser extent, they were also
motivated by their expectation that they would work as well as by the
therapeutic and social benefits of work.  Over one-half envisioned
paid work as a future possibility when they first began receiving DI
benefits.  This underscores the importance of intervening early to
encourage return to work and promote work attempts when beneficiaries
may be most receptive to assistance.\12

Several respondents conveyed their motivations in the following ways: 

  -- Barbara--40s, fibromyalgia, consultant:  "Work gave me a purpose
     in life.  It was not for the money.  It was not [to] be around
     people.  It was for me and me only.  I've always worked.  I like
     to work!  After a couple years in bed I felt I had to do
     something other than to focus on my pain."

  -- Dave--40s, cerebrovascular disease, newspaper deliverer:  "Work
     has always been expected of me.  It was the way I was brought
     up.  Also, I had to have more money than I was drawing to
     [achieve] a .  .  .  lifestyle that would be decent. 
     Fluctuating benefits from SSA put pressure on me to do
     something."

  -- Ann--40s, thyroid condition, office support worker:  "A person
     has to have something to do--a purpose in life.  [The] amount
     provided on assistance is very minimal.  Although I have mixed
     feelings about work since it is a struggle, I still want to be
     productive and participate in the community."

  -- Bill--30s, psychotic disorder, chef's assistant:  "Work ethic--I
     am not lazy.  I love my job.  I don't want no one to take care
     of me.  I try to do that myself.  I do not want to be put aside. 
     SSA tried to keep me at home.  They thought I was unsuitable to
     work.  But I work to help my physical and mental condition."


--------------------
\12 GAO/HEHS-96-133, July 11, 1996. 


   MANY FACTORS FACILITATED WORK,
   ALTHOUGH VR SERVICES AND SSA
   PLAYED A LIMITED ROLE
------------------------------------------------------------ Letter :5

Beneficiaries reported a number of factors as helpful to becoming
employed.  Table 4 categorizes these factors into three
groups--primary, secondary, and tertiary--on the basis of how often
all respondents reported them.  In some instances, we combined
related areas of support and services in developing the factors and
assigning relative importance.  (We summarize the findings below and
present them in greater detail in app.  IV.)



             Table 4. Factors that Facilitated
          Working DI Beneficiaries' Employment, by
                   Frequency of Reporting

Factor              Description         Significance
------------------  ------------------  ------------------
Primary
----------------------------------------------------------
Health              Health              Early return to
intervention        interventions       work without
                    provided medical    health
                    stabilization and   intervention may
                    improved            be difficult for
                    functioning.        some.

Encouragement       Family, friends,    Desire to work can
                    coworkers, and      be influenced
                    health              positively, and
                    professionals       possibly
                    provided            negatively, by
                    encouragement and   social forces.
                    emotional support.


Secondary
----------------------------------------------------------
Flexible work       Number of hours     Typical 5-day, 40-
schedule            and work schedule   hour work week may
                    were responsive to  be unrealistic for
                    respondents' needs  some
                    and capabilities.   beneficiaries.

Job-related         Training and        Has implications
training and        services were       for retaining
services            directly related    workers in the
                    to finding and      labor force who
                    performing a job.   otherwise might
                                        apply for Social
                                        Security
                                        disability
                                        benefits.

Trial work period/  SSA provisions      Trial work period
extended period of  allowed             reported as
eligibility         beneficiaries to    useful, although
                    test their work     some felt that 9
                    capacity without    months is too
                    jeopardizing        short and $200
                    benefits and ease   earnings level is
                    transition to       too low.
                    workforce.

High self-          Respondents         Motivation to work
motivation          strongly wanted or  may develop over
                    needed to work,     time, as about 3
                    especially          in 10 did not
                    compared with       expect to work
                    disabled peers      upon program
                    without jobs.       entry.


Tertiary
----------------------------------------------------------
Religious faith     Religious faith     Interview did not
                    reported as         specifically
                    providing source    address religious
                    of strength and     faith; it may be
                    guidance.           more important
                                        than reported.

Job coaches         On-site job coach   Has implications
                    or similar          for retaining
                    specialist taught   workers in the
                    work skills.        labor force who
                                        otherwise might
                                        apply for Social
                                        Security
                                        disability
                                        benefits.

Assistive devices   Among most          Usefulness of
and equipment       frequently          assistive devices
                    mentioned items     and equipment is
                    were back/leg       largely limited to
                    braces, canes/      people with
                    crutches, adapted   physical
                    computers/          impairments.
                    keyboards, and
                    wheelchairs.

Enablements from    Respondents         About one-third
Americans With      reported that ADA   were aware of ADA,
Disabilities Act    provided rights,    and over one-half
                    accommodations,     of those who were
                    and hiring          aware said ADA was
                    opportunities.      not helpful.
----------------------------------------------------------
The two most frequently reported factors appear to have been the most
critical in helping beneficiaries become employed.  First, health
interventions--such as medical procedures, medications, physical
therapy, and psychotherapy--reportedly helped beneficiaries by
stabilizing their conditions and, consequently, improving
functioning.  Not only were health interventions perceived as
important precursors to work, but they were also seen as important to
maintaining ongoing work attempts.  Encouragement to work was also
critical.  Respondents told us they received encouragement from
family, friends, health professionals, and coworkers. 

Following are several beneficiaries' descriptions of the factors that
helped them return to work. 

  -- Carol--30s, bipolar disorder, administrative support worker: 
     "My family members were hoping and encouraging me to go to work
     and not rely on disability income.  They were helpful to me in
     assessing the merits and benefits of potential job offers.  .  . 
     .  I am using a combination of Prozac and Lithium medications to
     control my condition and [allow] me to work regularly where I
     don't use my sick days.  Therapy with my counselor for over 4
     years has really allowed me to work and function in a work
     environment."

  -- Mark--30s, epilepsy, maintenance worker:  "[The] medication[s]
     for [my] epilepsy help keep [my] condition under control, which
     minimizes seizures and risk of getting fired.  .  .  .  [My
     supervisor] check[s] from time to time to make sure everything
     is OK [and I'm] not burning out .  .  .  even suggests taking
     days off."

  -- Louis--20s, cancer, financial counselor:  "All my
     treatments--chemo, radiation, and my eye surgery--helped me to
     get well and become physically able to work.  If I did not have
     treatments, I would be
     dead.  .  .  .  [The ADA] keeps employers aware that employees
     cannot be dismissed because of .  .  .  disabilities."

  -- Stephen--30s, human immunodeficiency virus (HIV) infection,
     bartender:  "[My] infectious disease doctor [is] encouraging and
     is very supportive.  He wrote a letter to [my] employer
     explaining [my] condition and my capabilities.  .  .  .  [My]
     parents are very supportive.  .  .  .  [My] medications have
     made me physically able to work.  .  .  .  [Coworkers are]
     providing emotional support."

  -- Yvonne--40s, anxiety disorder, cashier:  "Psychotherapy and
     group therapy [have] been helpful.  Also, medication has been
     helpful.  .  .  .  My psychotherapist has gone out of his way to
     help me.  I can call him at any time.  The pastor of my church
     has also counseled me.  At the college I attended, a director of
     the disabled talks to my professors and tells them about my
     condition so that they can take this into account when assigning
     work and evaluating my performance.  .  .  .  ADA has helped
     because I believe that they would not have hired me because of
     my problems."

Other, less frequently reported factors also enabled beneficiaries to
work.  Although these factors were less prominent overall, they may
be very important to any one individual.  Indeed, any single factor
may be the key determinant in an individual's becoming employed. 
These factors include a flexible schedule (particularly to have time
off to visit a health professional), job-related training and
vocational rehabilitation services (especially job search and
on-the-job training), the trial work period/extended period of
eligibility, and high self-motivation.  To a somewhat lesser extent,
religious faith, job coaches, assistive devices and equipment, and
enablements provided under the ADA were useful.  In general, similar
proportions of respondents with physical impairments and those with
psychiatric impairments cited these factors as helpful to being
employed.  However, people with physical impairments found coworkers
and the trial work period more helpful than did those with
psychiatric impairments. 

Our study results are generally consistent with the literature
regarding factors associated with employment for people with
disabilities.  For instance, many of the respondents we talked to
reported a high motivation to work, were educated beyond high school,
or were in their 30s or 40s.  For many, work seemed to be
economically advantageous because they were earning at least
moderate-level wages and receiving very few program benefits--such as
housing assistance and food stamps--that are contingent upon low
earnings.  Consistent with other research, medical interventions,
technology, accommodations, and social support were found to
facilitate return to work.  Unlike other studies, transportation
appears to be neither a strong facilitator for nor an impediment to
employment (perhaps because our respondents were selected from major
metropolitan areas). 


      INFORMAL NETWORKS WERE
      IMPORTANT IN GETTING JOBS
---------------------------------------------------------- Letter :5.1

Respondents told us they found their jobs in one of several ways. 
The most common means--indicated by 28--was through an informal
network.  For instance, beneficiaries told us they heard about jobs
from family, friends, church members, and coworkers at prior jobs. 
Seventeen found their jobs through advertisements, while others
either heard about their jobs through formal service providers or
initiated the search on their own (for example, asking employers
about openings). 


      VR SERVICES DEEMED USEFUL
      WERE UNEVENLY OBTAINED
---------------------------------------------------------- Letter :5.2

Formal VR and training services played a role in beneficiaries'
return to work, although some services considered useful by those
respondents who received them were unevenly obtained by respondents
in general (see app.  IV).  As disability managers in the workplace
have told us in the past, providing the necessary return-to-work
assistance is one strategy to help people with disabilities return to
productive activity in the workplace.  On the one hand, the
beneficiaries we interviewed rated a number of services as very
useful.  For instance, about three-fourths of people who received
such assistance as training in activities of daily living or social
skills, training in a trade or business school, job placement, and
on-the-job training rated the service as very useful.  On the other
hand, some of these useful services were received by only a small
number of respondents.  Of course, some respondents may not have
needed these services.  But some of these services (for example, job
placement and on-the-job training) might possibly have been
beneficial to many who did not receive them--especially since 41
beneficiaries reported that their limited skills and abilities were
barriers to work. 


      SSA WORK INCENTIVES AND
      STAFF PLAYED LIMITED ROLE
---------------------------------------------------------- Letter :5.3

Overall, work incentives offered by the DI program to reduce the risk
of attempting work appear to have played a limited role in
beneficiaries' efforts to become employed.  Although the trial work
period was considered helpful by 31 respondents, several indicated it
had its shortcomings.  For instance, they indicated the amount
signifying a "successful" month of earnings ($200) was too low, an
all-or-nothing cutoff of benefits after 9 months was too abrupt,\13

and having only one trial period did not recognize the cyclical
nature of some disabilities.  Respondents' mixed views of the design
of the trial work period suggest that they value a transitional
period between receiving full cash benefits and losing some benefits
because of work yet might be more satisfied with a different design. 
Finally, over one-fifth were unaware of the trial work period and
therefore may have unknowingly been at risk of losing cash benefits. 

Many respondents were unaware of other work incentives as well. 
Consequently, fewer respondents reported these incentives as helpful
than might have had they been better informed.  For example, 41
respondents were unaware of the provision that allows beneficiaries
to deduct impairment-related work expenses from the amount SSA
considers the threshold for determining continued eligibility.\14
Using the deduction could make it easier for a beneficiary to
continue working while on the rolls without loss of benefits. 
Moreover, 42 respondents were unaware of the option to purchase
Medicare upon leaving the rolls.  As a result, some of these
beneficiaries may decide to limit their employment for fear of losing
health care coverage, while others, planning to leave the rolls, may
think they are putting themselves at risk of foregoing health care
coverage entirely upon program termination.  Various beneficiaries
provided the following comments on DI's work incentives. 


--------------------
\13 We testified on the need for more research to determine the costs
and benefits of changing work incentives, such as providing a gradual
benefit offset for DI beneficiaries.  See Social Security Disability: 
Improving Return-to-Work Outcomes Important, but Trade-offs and
Challenges Exist (GAO/T-HEHS-97-186, July 23, 1997). 

\14 Examples of expenses likely to be deductible include attendant
care services performed in the work setting, structural modifications
to a vehicle used to drive to work, wheelchairs, and regularly
prescribed medical treatment or therapy that is necessary to control
a disabling condition. 


         IMPAIRMENT-RELATED WORK
         EXPENSES
-------------------------------------------------------- Letter :5.3.1

  -- "[It] helped to maintain my health so I could stay physically
     ready
     to work ."

  -- "[It paid] for [my] job coach and allowed me to continue to
     receive benefits."

  -- "I deducted the cost of getting to and from work."


         TRIAL WORK PERIOD
-------------------------------------------------------- Letter :5.3.2

  -- "[It] allowed me to get back into the work force .  .  .  to see
     how far I could go."

  -- "It helped me save money at the beginning so I could [have]
     clothing and certain things needed for [the] office."

  -- "[It is] a wonderful provision.  I wish they didn't just offer
     it once."

  -- "When I first went to work, I wasn't scared of losing my
     benefits.  It relieved the pressure of losing benefits."


         EXTENDED PERIOD OF
         ELIGIBILITY
-------------------------------------------------------- Letter :5.3.3

  -- "[It] provides added security and encouragement for getting back
     into [the] work force."

  -- "I know I can work now and it won't affect [my] benefits unless
     I can make enough money to replace benefits."

  -- [It is helpful] when I have to miss work because of my health
     problems."

Generally, respondents told us SSA staff with whom they interacted
provided neither much help in nor hindrance to return-to-work
efforts.  Fifty-nine respondents answered "no" when asked if people
from SSA assisted them in becoming employed.  However, 52 respondents
told us that they did not have experiences with SSA that made it
difficult to become employed.  For the 17 people reporting
difficulties, the most common examples cited were the limited
assistance offered and poor information provided by SSA.  Also, some
beneficiaries noted that the $500 monthly earnings threshold, which
is used in the formula that determines if a person with a disability
other than blindness is working at gainful activity level (and
therefore no longer eligible for benefits), is set too low.  When
examining respondents' comments indirectly related to our questions,
we found that about one-third indicated frustration or
dissatisfaction with some aspect of SSA or the DI program.  For
example, some respondents told us they felt that the program was
humiliating and lost sight of people's needs.  Moreover, some
respondents indicated that SSA had suddenly informed them that they
needed to repay cash benefits mistakenly paid to them in the past. 
Following are comments from various beneficiaries, some of whom were
satisfied with SSA and the DI program and others who were not. 


         SATISFIED BENEFICIARIES'
         COMMENTS
-------------------------------------------------------- Letter :5.3.4

  -- "They were very encouraging to get employed.  They were helpful
     [in taking] away [the] threat of getting employed."

  -- "[SSA] explained what I could and could not do.  They took away
     the fear by explaining things to me."

  -- "They [helped me] with understanding questions."


         DISSATISFIED
         BENEFICIARIES' COMMENTS
-------------------------------------------------------- Letter :5.3.5

  -- "[SSA] need[s] people who [can] better assess [peoples'] needs."

  -- "[The] conditions [I'm] under all the time .  .  .  knowing
     [I'm] restricted to that amount when [I'm] trying to be part of
     society .  .  .  tough.  [I've] had to sell stuff to make ends
     meet."

  -- "They cut off funds [that were] used to purchase medication."

  -- "Frustration.  Bureaucracy.  Waiting.  .  .  .  Can't get SSA to
     call me back.  Can't find anything out."

  -- "The information they provide is not clear.  As a result, there
     is a huge fear factor associated with it."

  -- "They offer no information about training.  I was only told if I
     work and make money I will be dropped."

We previously reported that DI beneficiaries were confused by program
provisions and recommended that SSA better implement existing
return-to-work mechanisms.\15 SSA has told us that its strategy to
better promote return to work is evolving and that it envisions a
partnership between field office staff and the private sector.  SSA
noted it continues to train field office staff about work incentives
and to disseminate multimedia publications about work incentives.  In
addition, SSA said it has been using the private sector to help
inform beneficiaries and encourage them to work and expects to do so
more in the future.  Also, SSA has funded (in conjunction with the
Department of Education's Rehabilitation Service Agency) a research
project that developed models for training private sector disability
case managers about Social Security DI provisions and work
incentives.  Moreover, SSA expects that private VR providers,
participating under its experimental Alternate Provider Program and
other proposed initiatives, will provide beneficiaries information
and encourage them to work. 


--------------------
\15 See GAO/HEHS-96-62, Apr.  24, 1996. 


   LONGER-TERM PLANS WERE AFFECTED
   BY HEALTH CONCERNS
------------------------------------------------------------ Letter :6

Not surprisingly, personal health appears to be an overriding issue
as beneficiaries consider their future status in the DI program and
at the worksite.  Among the 44 respondents without employer-based
health insurance coverage, 29 plan to stay on the DI rolls into the
foreseeable future or are unsure of their future plans.  In contrast,
15 of 24 respondents with such coverage plan to exit the rolls. 
Moreover, when asked if anything would make it harder to work, about
one-half of the 46 respondents who responded affirmatively said that
poorer health would inhibit employment.  Similarly, some said that
improved health would facilitate work.  Again, we found little
difference in future work and program plans between people with
physical and psychiatric impairments. 

We asked beneficiaries whether they experienced certain impediments
to employment.  The most common impediments--which 18 to 21
respondents told us they experienced very frequently or
frequently--were "limited skills or training," "employers not
recognizing one's ability," and "lack of good jobs." (App.  V
provides further details.)

There is some indication that about one-third of respondents may have
been earning less than they were capable of earning.  When asked
about their decisions on how much to work, 18 people indicated they
were capable of working more but did not do so because they feared
additional earnings would jeopardize their DI benefits; another 7
respondents said they were capable of working more, but were unable
to find jobs offering more hours. 


   OBSERVATIONS
------------------------------------------------------------ Letter :7

DI provides protection to workers who are unable to work because of
injury, illness, or disease.  As technological and medical advances
and social change in this country alter the relationship between
impairment and work capacity, new opportunities are opening for
individuals with disabilities to become more fully integrated into
mainstream society.  Yet DI beneficiaries face challenges and,
possibly, disincentives to realizing their earnings potential. 

Consistent with the view that an individual's work life is central to
personal identity, beneficiaries indicated that having a job and
engaging in productive activities were valuable ends in and of
themselves.  While DI cash and medical benefits could be restructured
to give beneficiaries greater incentive to return to work, many of
the factors that beneficiaries report as helping them be employed are
beyond SSA's immediate responsibility.  Moreover, the eclectic array
of factors facilitating return to work--including medicine and
therapy, encouragement from various people, and worksite
accommodations--suggests that the return-to-work experience is unique
for each person.  Capitalizing on this uniqueness may increase the
success of programs aimed at helping beneficiaries engage and advance
in the labor force. 

Issues affecting health status and functioning were especially
prominent themes relating to return to work.  For example, the vast
majority of respondents told us medicine and therapy helped them
return to work by affording medical stabilization and improved
functioning.  Similarly, having a work schedule that allowed health
visits was seen as important.  Moreover, having employer-based health
insurance was common among those planning to leave the DI rolls in
the future, whereas not having coverage may have contributed to plans
to stay on the rolls.  Finally, beneficiaries reported that future
health status would affect whether it would be easier or more
difficult to remain working. 

The importance of DI program provisions and work incentives was more
ambiguous.  It appears that some beneficiaries are uninformed and
confused about program provisions and work incentives and do not
receive encouragement to attempt a return to work.  In fact, some
respondents felt that SSA was unresponsive to their desire to return
to work.  Through its growing partnerships with private service
providers, SSA now has an opportunity to implement more effective
strategies to inform beneficiaries about program provisions and work
incentives and to encourage them to return to work. 

Although the average respondent worked 28 hours a week, a sizable
portion may have been working below their capacity.  Factors that
might be limiting their work efforts could be fear of losing
accessible health care insurance if they leave the rolls; fear of
losing disability cash benefits if earnings are too high; inability
to find a job that offers more hours; and limited skills and
training, with few opportunities for direct job-related services to
overcome deficiencies.  Changes in the DI benefit structure and
enhanced employment and skills training opportunities might yield
higher efforts from those beneficiaries who are working below
capacity.  However, if such changes are not made, many beneficiaries
may continue to face disincentives to maximize their work effort. 
Such disincentives, when combined with beneficiary confusion and
discouragement, could mean that significant opportunities to
capitalize upon beneficiaries' stated interest to work would remain
unrealized. 

Our findings, though based on a small sample, not only illustrate the
commonly held belief that health coverage supports ongoing employment
efforts; they also suggest that access to health care during the
24-month waiting period for Medicare coverage can have the positive
effect of improving health and functioning so that work can be
attempted.  This underscores the potential importance of health care
coverage in obtaining and sustaining employment. 


   AGENCY COMMENTS AND OUR
   RESPONSE
------------------------------------------------------------ Letter :8

In commenting on a draft of this report, SSA agreed with our findings
that many DI beneficiaries are interested in returning to work and
that many factors affect their ability to return to work.  (See app. 
VI for SSA's comments.) SSA stated that it has placed a high priority
on helping beneficiaries return to work, citing the administration's
proposed Ticket to Independence Program (an initiative to help
beneficiaries obtain rehabilitation services, employment services, or
both from public and private VR providers) and SSA's forthcoming
research efforts in collaboration with other federal agencies.  Our
earlier reports noted the importance of a multifaceted approach to
program improvement.  A new VR service delivery system, for example,
would be likely to have the greatest effect if it was integrated into
a comprehensive return-to-work strategy that incorporated earlier
intervention, a focus on developing productive capacity, and changes
to the structure of benefits. 

SSA emphasized that our small sample size limited our ability to
generalize findings to the population of working beneficiaries.  In
particular, SSA expressed concern about the generalizability of
statements of dissatisfaction with the agency or the DI program.  We
recognize in the report that our findings are not generalizable. 
However, our findings are generally consistent with our past reports
and those of the 1988 Disability Advisory Council to the Department
of Health and Human Services and other researchers over the years. 

SSA also stated that our data might imply that receiving VR several
years after going on the disability rolls is more effective in
successful return to work than early referral at time of program
entry.  We did not ask respondents when they received VR
services--some may have received them soon after entering DI, and
others may have received them after receiving services to stabilize
their condition and regain functioning.  As we have reported
previously, return-to-work services should be provided at the
earliest appropriate time and when beneficiaries might be more
receptive to return-to-work assistance, such as during the continuing
disability review process, to encourage disabled workers to return to
the workplace.  Given that over one-half of respondents told us they
envisioned paid work as a possibility when they entered the program,
early intervention could play an important role in assisting some
beneficiaries. 

SSA also made several technical comments, which we incorporated where
appropriate. 


---------------------------------------------------------- Letter :8.1

We are sending copies of this report to the Commissioner of Social
Security and other interested parties.  Copies also will be available
to others on request.  If you or your staff have any questions
concerning this report, please call me on (202) 512-7215.  Other GAO
contacts and staff acknowledgments are listed in appendix VII. 

Sincerely yours,

Jane L.  Ross
Director, Income Security Issues


SCOPE AND METHODOLOGY
=========================================================== Appendix I


   IDENTIFYING THE TARGET
   POPULATION
--------------------------------------------------------- Appendix I:1

To address the research questions, we conducted in-person interviews
with 69 Social Security Disability Insurance (DI) beneficiaries who
were employed and who lived in the Washington, D.C.; Atlanta; or San
Francisco metropolitan areas.  To identify the target population, we
obtained from the Social Security Administration (SSA) a list of DI
beneficiaries who were on the rolls as of March 1997, who were aged
18 to 59, and who earned $50 or more in 1995.  From this list, we
identified 6,896 beneficiaries who lived within about a 40-mile
radius of the three metropolitan areas and whose 1995 earnings were
$4,320 or higher (an amount approximating the yearly salary of a
minimum-wage, half-time worker).\16 These metropolitan areas were
chosen to maximize the resources available to conduct interviews. 

Because the return-to-work process could vary meaningfully among
beneficiaries according to differences in certain characteristics
(for example, disability type and age), the study was designed to
involve a diverse group of working beneficiaries.  Yet, the small
sample size precluded us from making observations on subgroups, such
as younger workers with psychiatric impairments.  Nevertheless, if we
had selected randomly 69 respondents from the list of 6,896
beneficiaries, individuals with certain combined characteristics
might have been excluded.  Thus, we placed each of the 6,896
beneficiaries into 1 of 24 categories derived from the following
characteristics: 

  -- for geographic area, we used three levels:  the three
     metropolitan areas mentioned above;

  -- for disability type, we used two levels:  physical and
     psychiatric;

  -- for age, we used two levels:  18 to 39 and 40 to 59; and

  -- for 1995 earnings, we used two levels:  $4,320 to $5,999 and
     $6,000+.\17

Our target was to interview several beneficiaries from each category,
although the study's purpose did not require us to conduct the same
number of interviews from each category.  In most categories, we
interviewed two to five people.  However, we did not interview anyone
from one category, and we interviewed only one person from another 4
categories. 


--------------------
\16 This figure excludes beneficiaries dropped because their
diagnosis code was not valid--that is, the field was blank, the code
was 0000, or the code did not establish a medical diagnosis.  This
figure also excludes beneficiaries whose impairment was listed as
mental retardation, drug addiction disorder, or alcohol addiction
disorder.  We did not include these impairment groups in our target
population. 

\17 We selected $6,000 as the cut-off between earnings levels because
this amount is equal to the annualized Substantial Gainful Activity
level ($500) for non-blind beneficiaries. 


   CONTACTING POTENTIAL
   RESPONDENTS
--------------------------------------------------------- Appendix I:2

We mailed notification letters to beneficiaries (randomly listed) in
each category.  The letters conveyed the purpose of the study and
notified individuals that they would receive a call from a GAO
interviewer within a week to talk about participating in the study. 
Before mailing the letters, however, we confirmed beneficiaries'
telephone numbers (obtained from SSA) with directory assistance or a
telephone book.  We did not send notification letters to people whose
telephone numbers we could not confirm; neither did we send letters
when no number was provided by SSA and we could not obtain a number
through directory assistance or a telephone book. 

Upon initial telephone contact, we confirmed that beneficiaries were
currently enrolled in DI and working.\18 We asked about work status
because it was possible that someone's 1995 earnings were from
employment before coming onto the rolls or were actually unearned
(SSA's database does not differentiate between earned and unearned
income).  People were dropped from the study if they did not meet
these two screening criteria. 

We sent 418 notification letters, with the following results: 

  -- 15 letters were returned as undeliverable;

  -- 179 people were unreachable because, for instance, they did not
     return interviewer telephone messages, they had disconnected or
     otherwise inoperable phone numbers, they did not reside at the
     address we were given and we could not obtain a forwarding
     telephone number, or we were told by people answering the
     telephone that the person had died;

  -- 6 people were excluded because they told us they were no longer
     in DI;

  -- 107 were excluded because they told us they were not working;

  -- 42 refused to be interviewed; and

  -- 69 agreed to be interviewed. 


--------------------
\18 We considered individuals in the extended period of
eligibility--the 36-month period following a trial work period during
which people earning above $500 do not receive cash benefits but
retain Medicare coverage--to be enrolled in DI.  Also, to determine
work status, we asked, "Are you currently working at a job for pay?"


   DEVELOPING THE DATA COLLECTION
   INSTRUMENT
--------------------------------------------------------- Appendix I:3

We developed a data collection instrument (DCI) containing open-ended
and closed-ended questions to serve as an interview guide for this
assignment.  We developed the DCI after reviewing selected literature
(including instruments used in related research activities) and
discussing our research interests with experts and disability
advocates.  To help validate the instrument, we obtained and
incorporated comments from about 15 external reviewers representing a
federal agency, research organizations, and disability advocacy
groups.  In addition, we conducted 13 pilot interviews to test and
refine the instrument.  Pilot interviews demonstrated that
respondents could answer the questions and were generally comfortable
with the length of the interview (about 90 minutes).  The DCI is
available upon request. 


   DATA ANALYSIS
--------------------------------------------------------- Appendix I:4

To address the research questions, we conducted a descriptive
analysis of the data.  We converted responses to open-ended questions
into numerical data using a content analysis technique.  This
technique required us to first develop response categories for each
open-ended question (for instance, ways that families were helpful to
respondents in becoming employed).  Next, two to three interviewers
individually placed responses to open-ended questions into categories
and compared results.  About two-thirds were direct matches.  For
responses coded differently, interviewers discussed the rationale for
their decisions.  Consensus was achieved when a majority agreed,
which occurred on most occasions.  In a few instances, interviewers
were unable to achieve consensus and, consequently, their responses
were not used. 

A draft of our report was reviewed by disability researchers, whose
comments were incorporated as appropriate. 


DEMOGRAPHIC AND FINANCIAL
CHARACTERISTICS OF 69 RESPONDENTS
========================================================== Appendix II



                               Table II.1
                
                   Demographic Characteristics of 69
                              Respondents

                                                             Number of
Characteristic                                             respondents
--------------------------------------------------------  ------------
Gender
----------------------------------------------------------------------
Female                                                              35
Male                                                                34

Age
----------------------------------------------------------------------
20-29                                                               11
30-39                                                               21
40-49                                                               25
50-59                                                               12

Race
----------------------------------------------------------------------
White/caucasian                                                     50
African-American                                                    13
Hispanic                                                             2
Other                                                                4

Marital status\a
----------------------------------------------------------------------
Single and never married                                            34
Married                                                             21
Divorced                                                            13

Highest education\a
----------------------------------------------------------------------
High school or less                                                 23
College experience                                                  31
Postgraduate experience                                             14

SSA impairment category\b
----------------------------------------------------------------------
Psychiatric impairment (excludes mental retardation and             39
 drug and alcohol abuse)
Physical impairment
Neurological                                                         7
Musculoskeletal                                                      3
Neoplastic                                                           3
Immune system                                                        3
Special senses and speech                                            2
Genitourinary                                                        2
Hemic and lymphatic                                                  2
Endocrine and obesity                                                2
Digestive                                                            1
Cardiovascular                                                       1
Code not found in SSA codebook                                       4

Living arrangements
----------------------------------------------------------------------
Live alone                                                          27
Live with spouse                                                    21
Reside with someone other than spouse                               21

Years in DI program\c
----------------------------------------------------------------------
1-3                                                                  7
4-6                                                                 26
7-9                                                                 17
10+                                                                 19
----------------------------------------------------------------------
\a Numbers do not add up to 69 because answers were not obtained from
all respondents. 

\b SSA administrative records were the data source for impairment
types listed here.  We used self-reported information to determine
impairment type for table 1. 

\c SSA administrative records were the data source for years in DI
program. 



                               Table II.2
                
                    Financial Characteristics of 69
                              Respondents

                                                             Number of
Characteristic                                             respondents
--------------------------------------------------------  ------------
Household income\a
----------------------------------------------------------------------
$5,000-$9,999                                                        9
$10,000-$14,999                                                     15
$15,000-$19,999                                                      6
$20,000-$24,999                                                      7
$25,000 and over                                                    25

How well the money and support received took care of needs<
----------------------------------------------------------------------
Very adequately                                                     24
Somewhat adequately                                                 29
Somewhat inadequately                                                9
Very inadequately                                                    7

Source of benefits other than DI
----------------------------------------------------------------------
Government housing assistance                                        8
Workers' compensation                                                2
Private disability insurance                                         2
Veteran's Administration disability cash
 benefits                                                            1
Unemployment insurance                                               1
Food stamps                                                          1
Aid to Families With Dependent Children                              0
Other government program                                             2
----------------------------------------------------------------------
\a Numbers do not add up to 69 because answers were not obtained from
all respondents. 


WORK STATUS OUTCOMES OF 69
RESPONDENTS
========================================================= Appendix III

                                                             Number of
Question/response                                          respondents
--------------------------------------------------------  ------------
In a typical week, how many hours do you work at this job?\a
(mean=27.6;
median=28)
----------------------------------------------------------------------
5-19                                                                19
20-34                                                               18
35-53                                                               28

In a typical week, how many days do you work?\a (mean=4.1; median=5)
----------------------------------------------------------------------
1.5 days or less                                                     4
1.6-2.5 days                                                         9
2.6-3.5 days                                                         9
3.6-4.5 days                                                        10
4.6 or more days                                                    36

How long have you had this job?\ (mean=35; median=18)
----------------------------------------------------------------------
1-6 months                                                          18
7-12 months                                                         10
13-216 months                                                       41

How did you learn about your job?\a
----------------------------------------------------------------------
Informal network                                                    28
Advertisement                                                       17
Formal provider                                                     14
Self-initiated                                                       9

What aspects of your job are you most satisfied with? (multiple
responses
allowed)