Social Security: Disability Rolls Keep Growing, While Explanations Remain
Elusive (Letter Report, 02/08/94, GAO/HEHS-94-34).

More people are applying for and being awarded Social Security
disability benefits than ever before, and these beneficiaries are
remaining on the disability rolls for longer periods of time. As a
result, disability payments have burgeoned. Changes in beneficiary
characteristics have accompanied this growth: the average age of new
beneficiaries is now below 50, mental impairment awards to younger
workers have risen substantially, and more and more new beneficiaries
receive such low disability insurance benefits that they get additional
income from the Supplemental Security Income program. These low benefit
levels suggest that the new beneficiaries had limited work histories.
Higher unemployment probably contributes to increasing applications, and
policy changes have produced changes in the numbers and types of
beneficiaries. Quantitative data on the impact of these factors are
lacking, however, and important questions remain. The upshot is that the
Social Security Administration's (SSA) ability to predict future growth
and change in the rolls is limited. Better information would also help
SSA to determine whether improvements in program management are needed.

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

 REPORTNUM:  HEHS-94-34
     TITLE:  Social Security: Disability Rolls Keep Growing, While 
             Explanations Remain Elusive
      DATE:  02/08/94
   SUBJECT:  Handicapped persons
             Federal social security programs
             Supplemental security income program
             Information gathering operations
             Beneficiaries
             Disability insurance
             Demographic data
             Social security benefits
             Disadvantaged persons
             Disability benefits
IDENTIFIER:  Social Security Disability Insurance Trust Fund
             Social Security Disability Insurance 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.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


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


Report to the Chairman, Committee on Finance, U.S.  Senate, and the
Chairman, Committee on Ways and Means, House of Representatives

February 1994

SOCIAL SECURITY - DISABILITY ROLLS
KEEP GROWING, WHILE EXPLANATIONS
REMAIN ELUSIVE

GAO/HEHS-94-34

Disability Trends


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

  ALJ - administrative law judge
  ASPE - Assistant Secretary for Program Evaluation
  CDR - continuing disability review
  CRS - Congressional Research Service
  DDS - disability determination service
  DI - Disability Insurance
  HHS - Department of Health and Human Services
  OASI - Old Age and Survivors Insurance
  SSA - Social Security Administration
  SSI - Supplemental Security Income
  VR - vocational rehabilitation

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


B-254029

February 11, 1994

The Honorable Daniel P.  Moynihan
Chairman, Committee on Finance
United States Senate

The Honorable Dan Rostenkowski
Chairman, Committee on Ways and Means
House of Representatives

This report responds to your requests for information about the
recent growth in the number of persons receiving benefits under the
Social Security Disability Insurance (DI) program.  Since 1985,
beneficiary rolls in the DI program have grown by over 30 percent, or
more than twice the growth of the insured population. 

You asked us to review the DI program to (1) determine what is known
about the reasons for this growth and (2) identify what key questions
need to be answered to better understand the program's potential for
future growth.\1


--------------------
\1 In April 1993, GAO testified before the House Committee on Ways
and Means' Social Security Subcommittee on the program's growth and
changes in the characteristics of those who are coming on the rolls. 
See Social Security:  Rising Disability Rolls Raise Questions That
Must Be Answered (GAO/T-HRD-93-15, Apr.  22, 1993). 


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

In 1993, the Social Security Administration (SSA) actuary forecasted
that DI rolls would continue growing and would nearly double to over
6 million disabled workers in the next 10 years.  These rolls have
already grown substantially.  In the 3 years between 1989 and 1992,
applications rose by a third, and almost half the applicants in 1992
succeeded in obtaining benefits.  Once on the rolls, beneficiaries
have been staying longer.  Between 1985 and 1992, the number of
beneficiaries who had been on the rolls more than 15 years grew by 93
percent. 

Changes in the characteristics of new beneficiaries have accompanied
this growth.  The new beneficiaries' average age is generally
decreasing and is now below 50.  Mental impairment awards to younger
workers increased more than 500 percent between 1982 and 1992,
helping to pull down the average age.  Also, an increasing percentage
of new beneficiaries receive such low DI benefits that they get
additional income from the Supplemental Security Income (SSI)
program.  Their low DI benefit levels indicate that these new
beneficiaries have mostly had limited work histories. 

Several reasons for the growth and change in the DI rolls have been
identified.  For example, higher unemployment probably contributes to
increasing applications, and policy changes have contributed to
changes in the numbers and types of beneficiaries.  However,
quantitative data on the impact of these reasons are lacking, and
important questions remain open.  For example, SSA lacks adequate
data on how many people in the general population suffer disabilities
that might qualify them for benefits if they applied.  As a result,
the agency has limited ability to predict future growth and change in
the rolls. 

Without better information, neither SSA nor the Congress can be sure
whether the current growth will continue, or whether current trends
might reverse, as they have done in the past.  Also, better
information may assist SSA to determine, in view of these trends,
whether actions are needed to better manage the program.  The
Department of Health and Human Services (HHS) and others are
developing research programs.  Although the planned research will
address the key questions we identify in this report (see pp.  7-8),
it is too early to assess whether these efforts will provide adequate
answers.  Results from initial research efforts will not be available
until mid-1994. 


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

The DI program provides over $27 billion annually in cash benefits to
about 3.5 million workers who have become totally unable to work due
to disability.\2 When a worker who is insured for disability under
Social Security applies for benefits, state disability determination
services\3 (DDS) examine medical information to decide whether an
applicant is disabled.  If initially denied benefits, applicants may
pursue their claims through several levels of appeal.  Once awarded
benefits, a person may remain on the rolls until (1) death, (2)
conversion to regular retirement benefits at age 65, or (3) medical
recovery and/or return to work. 

DI is the nation's primary source of income replacement for disabled
workers insured under Social Security.  A parallel program, SSI,
provides benefits for aged and disabled indigent persons.  Some
persons, whose work histories are so limited that they qualify for
very small DI benefits, receive benefits from both programs.  DDSs
use the same standards and procedures for determining disability in
both programs. 


--------------------
\2 Dependents of disabled workers may also receive benefits. 

\3 Although these are state agencies, SSA funds and oversees them. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :3

Our analysis of program trends was generally limited to available
information, such as SSA's regularly generated program reports.  We
also reviewed studies of the program conducted by HHS at the request
of the DI Trust Fund's Board of Trustees, and by the Congressional
Research Service (CRS) at the request of the Senate Finance
Committee.\4 To obtain supplementary information about awardee
characteristics, we examined SSA's 831 file--a computerized
database--on the results of disability decisions.  We did most of our
work at SSA headquarters in Baltimore, Maryland, from April 1992
through August 1993 in accordance with generally accepted government
auditing standards. 


--------------------
\4 See The Social Security Disability Program:  An Analysis, HHS
(Dec.  16, 1992) and CRS Report for Congress:  Status of Disability
Programs of the Social Security Administration, CRS, 92-691 EPW
(Sept.  8, 1992). 


   DISABILITY ROLLS GROWING IN
   SIZE, CHANGING IN CHARACTER
------------------------------------------------------------ Letter :4

Today, more persons receive disability benefits than ever before. 
With the exception of a few years in the late 1970s and early 1980s,
the number of disabled worker beneficiaries has been increasing since
the program began in 1957.  In addition, as shown in table 1, the SSI
program is experiencing similar growth. 



                                     Table 1
                     
                       Adults Receiving Federal Disability
                      Benefits Compared to the U.S. Working-
                             Age Population, 1970-92

                              (Numbers in thousands)

                1970      1975      1980      1985      1990      1991      1992
----------  --------  --------  --------  --------  --------  --------  --------
U.S.         113,502   125,988   137,242   146,884   153,707   155,278   156,831
 working-
 age
 populatio
 n
Number of      1,493     2,489     2,859     2,657     3,011     3,195     3,468
 DI
 beneficia
 ries\a
Percent of     1.32%     1.98%     2.08%     1.81%     1.96%     2.06%     2.21%
 population
Number of         \b     1,678     1,743     1,841     2,418     2,600     2,843
 SSI
 recipient
 s
Percent of               1.33%     1.27%     1.25%     1.57%     1.67%     1.81%
 population
--------------------------------------------------------------------------------
Note:  Data include persons aged 18 through 64.  Concurrent
beneficiaries, who receive benefits from both programs, are reflected
in both DI and SSI data. 

\a Excludes disabled adult dependents. 

\b The SSI program did not pay benefits until 1974. 

While the number of beneficiaries is rising, the type of person
receiving DI benefits is changing as well.  Today, beneficiaries are
coming on the rolls at earlier ages, even though the average age of
the insured population has been rising.  As a result, average
beneficiary age has been dropping, and these younger beneficiaries
are staying on the rolls longer than in the past.  Between 1980 and
1992, the proportion of beneficiaries on the rolls for 15 years or
more almost tripled, rising from 5 to 14 percent. 

Mental impairment awards, which often go to younger persons, are also
increasing.  In 1992, more than 160,000 such awards were made, or
more than a quarter of all awards.  They are now the largest single
category of disability.  Mentally impaired applicants are also more
frequently successful in obtaining benefits.  They received awards 58
percent of the time in 1992, while the average success rate for
physically impaired applicants was 39 percent.  (See app.  II.)

Increasingly large numbers of applicants are eligible for both SSI
and DI.  Applicants for such concurrent awards formed nearly half of
all DI applicants in 1992, up from a little more than a third in
1980.  The fact that these new applicants need supplementary SSI
benefits suggests that they are less well off and may have less
extensive and less highly paid work histories than the DI-only
applicants who predominated in the past. 

Changes in the number of persons receiving disability benefits are
not new.  As table 1 indicates, beneficiary counts have risen and
fallen in the past.  The "incidence rate" (the annual number of new
benficiaries per thousand insured workers) in 1992 (approximately 5.2
awards per thousand insured workers) was the same as it was in 1978. 
This number is well below the peak of 7.1 in 1975 (see fig.  II.1). 

However, the disability insurance program has now grown beyond
previous experience.  Increasing applications, increasing awards, and
decreasing terminations have worked together to swell the rolls.  In
addition, should the trends continue toward more young, mentally
impaired beneficiaries who receive benefits from both programs, the
character of DI rolls will change significantly.  These changes raise
critical questions about the future direction of the DI program. 


   REASONS FOR CHANGE NOT FULLY
   UNDERSTOOD
------------------------------------------------------------ Letter :5

Changes in a variety of social, economic, legal, and other conditions
can affect the number and type of persons on the rolls.  For example,
when legislative requirements for determining disability are relaxed,
a higher percentage of applicants will receive awards.  Moreover, if,
as a result, more young persons apply and are awarded benefits, the
termination rate will likely decline, since younger beneficiaries can
be expected to stay on the rolls longer. 

Thus, changes in conditions can affect the application, award, and
termination rates--which in turn determine the size of the rolls. 
These rates and the major conditions that appear to affect them are
discussed in table 2.  Detailed explanations of the rates and the
changes in conditions appear in appendixes I through III. 



                           Table 2
           
           Disability Insurance Application, Award,
             and Termination Rates and Conditions
                        Affecting Them

Rates         Conditions affecting rate
------------  ----------------------------------------------
Application


Insured workers are applying for benefits a higher rate
(app. I
------------------------------------------------------------
              The economy: High unemployment may increase
              applications, but not under all circumstances.

              Outreach efforts: SSI outreach appears to have
              also increased DI applications.

Award


Increasing percentage applications result i awards (app.
II).
------------------------------------------------------------
              Changes in adjudicative standards: Legislative
              and regulatory changes have generally made it
              easier to obtain benefits.

              Court decisions: Court decisions have changed
              some eligibility policies and may influence
              the climate in which decisions are made.

              Erroneous awards not a factor: Known errors in
              decisions have not caused increases in awards.

              Appellate awards play a small role: Increasing
              awards by administrative law judges (ALJ) have
              contributed slightly to program growth.

Termination


Beneficiaries are lea the rolls at a lower (app. III).
------------------------------------------------------------
              Beneficiary demographics: Almost all
              terminations are due to death or attaining
              normal retirement at age 65.

              Disability reviews: Continuing disability
              reviews (CDR)\a have had a very small impact.

              Rehabilitation: Vocational rehabilitation has
              made a minimal contribution to terminations.
------------------------------------------------------------
\a In CDRs, SSA reassesses the disability status of beneficiaries. 
Those who have sufficiently improved are removed from the rolls. 

Much remains to be understood about the causes for program growth and
the future outlook.  For example, although court decisions appear to
have a significant impact on the award rate, the mechanism and the
extent of this impact remain unclear.  Even if research leads to a
better understanding of the conditions contributing to growth in the
rolls, it will be an additional task to forecast when each of these
underlying factors will stop adding to the rolls. 


   SSA HAS LITTLE INFORMATION ON
   WHICH TO BASE ITS PROJECTIONS
   OF THE PROGRAM'S FUTURE
------------------------------------------------------------ Letter :6

Some observers believe that the recent increases in the application
and award rates are unlikely to continue for long.  For example,
SSA's actuary expects that the rates of disability for most age
groups of workers will experience slight declines in the future. 
Even with these declines, the actuary projects that the rolls will
nearly double--to over 6 million people--in the next 10 years. 

The actuary has limited information, however, on which to base these
projections.  To project future new beneficiary counts, the actuary
starts with projections of the future insured population, based on
current demographic trends.  To predict how many persons from the
insured population will receive benefits, the actuary projects future
incidence rates.  These incidence rates are projected based largely
on overall historical experience more so than on results of research
into the reasons these rates have changed.  Since these rates change
often (see fig.  II.1), the projection of future incidence rates has
been problematic. 


   SSA'S PLANNED RESEARCH PROGRAM
------------------------------------------------------------ Letter :7

The current growth in the rolls and the changes in their composition
raise basic questions about the DI program.  Without better answers
to these questions, it will remain difficult to predict the program's
future course.  Recognizing the need for better information on the
causes of program trends, the trustees of the DI Trust Fund
recommended that HHS conduct the "best possible" research into the
reasons for DI program changes. 

HHS has started research to analyze these program trends.\5 By
mid-1994, SSA, working with HHS' Assistant Secretary for Planning and
Evaluation (ASPE), hopes to have completed a thorough analysis of its
administrative data and a review of the available literature.  In
preparation, SSA is reviewing and consolidating data on disability
applicants. 

As part of this research, SSA and ASPE plan to conduct a medical
examination study--a study of the population based on medical
examinations of selected individuals.  This study will identify
persons who would be considered disabled under Social Security
program criteria, were they to apply for benefits.  In addition to
identifying potential enrollees, this study would allow SSA to
determine how and why some disabled individuals continue to work
despite their impairments. 


--------------------
\5 In addition, the National Academy of Social Insurance is currently
studying DI at the request of the House Committee on Ways and Means. 


   KEY RESEARCH QUESTIONS HAVE
   BEEN IDENTIFIED, BUT IT IS TOO
   EARLY TO ASSESS PROGRESS
------------------------------------------------------------ Letter :8

Based on our review of the DI program and SSA's research plan, we
believe that the key questions to be addressed in a disability
research program have been identified.  They include the following: 

Do current increases in applications reflect real increases in the
level of disability?  If disability is increasing in the population
as a whole, then current increases in applications may be
appropriate.  On the other hand, if disability in the population is
stable or falling, the increase in applications may be due to other
factors.  SSA plans to start addressing this question with a review
of existing literature on disability prevalence, as well as through
the medical examination study. 

Who applies for disability?  Has this changed?  An application for
disability benefits generally represents a decision on the part of
the applicant to seek such benefits.  If proportionally more
persons--or different types of persons--are making this decision,
understanding their motivations may prove useful in forecasting
future growth. 

SSA plans to address these questions by (1) analyzing its newly
consolidated administrative data on applications, (2) supplementing
this information with a survey of new applicants, and (3) researching
trends in the labor market that may be influencing new applicants. 
To better project the future of disability applications, the agency
plans to synthesize this information in a statistical analysis. 

What affects the award rate?  Has this changed?  The award rate
summarizes the results of millions of disability decisions.  As such,
it can be influenced by factors ranging from changes in program
policy to changes in disability adjudicators' subjective attitudes. 
Better understanding of these factors and their relative impact on
the award rate should help to forecast future growth in the rolls. 

To identify factors affecting the award rate, SSA plans to (1)
identify trends in award and appeal rates at differing levels of
adjudication, (2) identify characteristics of applicants likely to
win awards, and (3) analyze the effects of policy changes by
comparing award rates before and after the changes. 

Do the standards for determining disability result in benefits being
awarded only to those unable to work?  When making a disability
award, decisionmakers apply program standards to determine whether a
person is unable to work and thus is entitled to benefits.  If the
standards are correct and are properly applied, relatively few
working persons would have conditions similar in severity to those
suffered by beneficiaries.  However, if more than a "few" workers
suffer from similarly severe conditions, yet continue to work, it may
be necessary to review the standards and their application. 

SSA's work on the medical examination study will provide information
for assessing the extent to which the working population has
impairments similar to those of DI beneficiaries.  This work is a
long-range effort, however. 

What causes changes in the DI termination rate?  Although we know
that the termination rate is heavily influenced by increasing awards
to younger mentally impaired persons, we cannot be sure about the
rate's future direction. 

To provide better information on terminations, SSA plans include (1)
conducting research on beneficiary characteristics, including their
earnings histories; (2) studying beneficiaries who have medically
recovered to determine if they later returned to the DI rolls; and
(3) assessing changes in education and workforce participation of the
handicapped, including the effects of legislative change on workplace
access and hiring. 

It is too early to tell whether SSA's efforts will result in adequate
answers to these questions; some uncertainty over future growth is
likely to remain, even with research.  Early results from SSA's
program will not become available until mid-1994.  However, we
believe that these questions provide a good framework for the
research effort.  To the extent that these questions are left
unanswered, predicting the future course of the disability program
will remain problematic. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :9

We requested written comments on a draft of this report from HHS, but
none were received within the stated period for incorporation in the
report.  However, we discussed the draft with SSA staff and CRS
staff.  We incorporated their comments where appropriate. 


---------------------------------------------------------- Letter :9.1

As agreed with your offices, unless you publicly announce the
contents of this report earlier, we plan no further distribution of
it until 30 days from its issuance date.  We will then send copies to
SSA, HHS, and other interested parties.  Copies of the report will be
made available to others upon request. 

Please contact me on (202) 512-7215 if you have any questions about
this report.  Other major contributors are listed in appendix V. 

Sincerely yours,

Jane L.  Ross
Associate Director,
 Income Security Issues


RISING APPLICATIONS FOR DISABILITY
INSURANCE BENEFITS
=========================================================== Appendix I

Today, more people are applying for Disability Insurance benefits
than ever before.  The rate at which the insured apply has varied
over the years.  Between 1989 and 1992, however, the rate of
application increased 28 percent, from 8.5 to 10.9 applicants per
thousand insured persons. 

Economic factors may account for some of this increase, although
research on this issue has not been conclusive.  In times of high
unemployment, when impaired persons lose their jobs, they may apply
for DI.  Other evidence suggests that this relationship does not
always hold.  For example, applications did not increase during the
high unemployment rates prevailing in the early 1980s.  During that
time, very stringent program administration (low award rates and high
termination rates) may have dissuaded applications. 

The population mix of applicants is changing also.  Most of the
recent increase in DI applications comes from those who are eligible
for SSI supplements to their DI benefits.  These concurrent
applicants are generally less well off than those whose benefits are
entirely paid from DI funds. 


   APPLICATION RATES HAVE VARIED
   BUT NOW ARE RISING
--------------------------------------------------------- Appendix I:1

The number of persons per thousand insured who decide to apply for
benefits is reflected in the application rate.  For example, in 1992,
10.9 persons per thousand insured applied for DI benefits. 

As shown in figure I.1, the application rate has varied over the past
22 years.  From 1982 through 1986, the application rate appeared
relatively flat.  After a drop to a low point in 1989, this rate
started rising sharply. 

   Figure I.1:  History of DI
   Application Rate, 1970-92

   (See figure in printed
   edition.)

Note:  SSA techniques for reporting the number of applications
changed in October 1981, resulting in a drop of about 11 percent in
the reported number of applications.  The adjusted rate (dashed line)
shows the effect of such an 11-percent drop in prior years.  However,
since the reporting change may have affected prior years' data to a
greater or lesser extent, caution should be used in relying on this
adjustment. 


   DOES UNEMPLOYMENT AFFECT THE
   APPLICATION RATE? 
--------------------------------------------------------- Appendix I:2

Some data suggest that rises in unemployment may explain the increase
in application rates.  Persons with impairments may find it difficult
to obtain and keep jobs during periods of high unemployment.  As a
result, they may find disability benefits relatively more attractive. 

However, research on this point has not been conclusive.  HHS points
out that past quantitative studies by SSA researchers had "disagreed
on the existence of such a relationship between unfavorable levels of
unemployment and the number of disability applications.  Where a
statistically significant relationship has been found, it generally
has been weak."

Other data suggest that the relative difficulty applicants experience
in obtaining and keeping a disability award may also play a role in
the application and unemployment rates since 1970.  As shown in
figure I.2, the unemployment and application rates were particularly
far apart in the 1981 to 1984 period. 

   Figure I.2:  DI Application
   Rate Compared to Unemployment
   Rate, 1970-92

   (See figure in printed
   edition.)

Some of the difference in the rates during 1981 to 84 may be due to
stringent program conditions that prevailed at that time, during
which SSA was awarding benefits to relatively few applicants.  (See
app.  II, fig.  II.2.) Also, SSA was removing large numbers of
persons from the rolls through continuing disability reviews.  (See
app.  III.) These initiatives were well


known and publicized.  As a result, SSA believes that the stringent
conditions of that time discouraged people from applying for DI. 

We identified a 1991 study that also reached the same conclusion.\1


--------------------
\1 This study, Self-Screening in Target Public Assistance Transfer
Programs, by Donald O.  Parsons of Ohio State University (Journal of
Political Economy, Vol.  99, No.  4, (1991)), found that a 10-percent
decrease in the initial allowance rate induces a 4-percent reduction
in applications. 


   CONNECTION BETWEEN APPLICATIONS
   FOR SSI AND DI
--------------------------------------------------------- Appendix I:3

As shown in figure I.3, much of the growth in the number of persons
applying for DI benefits comes from those who were applying for both
DI and SSI.  In 1980, concurrent DI/SSI applications made up about
one-third of the DI total.  By 1992, concurrent applications made up
almost half. 

   Figure I.3:  Concurrent SSI/DI
   Applications Rose Faster than
   DI-Only Applications, 1980-92

   (See figure in printed
   edition.)

This change suggests that factors that increase SSI applications may
also be contributing to the DI rolls.  For example, SSA has recently
conducted outreach efforts for its SSI program, publicizing the
availability of benefits and explaining how they can be obtained. 
SSA believes these efforts served to increase SSI participation. 
Because of the overlap between SSI and DI, the outreach efforts may
also have helped to increase DI applications. 

This SSI/DI connection was also evident in 1974, when the SSI program
first started paying benefits.  At that time, the DI application rate
reached its highest point in the period we examined.  (See app.  I,
fig.  I.1.)


APPLICANTS ARE MORE FREQUENTLY
SUCCESSFUL
========================================================== Appendix II

During the last decade, the percentage of applicants who were
successful in obtaining DI awards increased each year.  These
increases occurred at both the initial and appellate levels of
decisionmaking.  In 1992, about 48 percent of DI applicants were
found eligible to receive benefits.  This number represented a
substantial increase in the rate of awards from the low levels of the
early 1980s.  Award rates in 1992 appeared similar to those
prevailing in the mid-1970s. 

The growth in the award rate should be seen in perspective, however. 
For example, if application rates were to decline, higher award rates
would have a limited input on the rolls.  In 1992, although
applications were rising, the application rate was not as high as it
had been in the past. 

One measure, the "incidence rate," or number of new awards per
thousand insured persons, combines the effects of the award and
application rates into one index.  In 1992, this rate was not as high
as in the 1970s.  As shown in figure II.1, higher incidence rates
occurred from 1971 through 1977.  The 1992 incidence rate was
equivalent to that prevailing in 1978. 

   Figure II.1:  History of DI
   Incidence Rate, 1970-92

   (See figure in printed
   edition.)

Increases in DI awards are due in part to changes mandated by the
Congress and the courts.  Changes in the adjudicative climate--the
subjective attitudes of decisionmakers--may also play a role.  A
small portion of the increases in the award rate may be due in part
to more persons' winning their appeals before administrative law
judges, who now make awards in more than two-thirds of their cases. 

As shown in figure II.2, the award rate has changed in two directions
since 1970.  The rate declined from a high of 48.1 percent in 1972 to
its low of 29.3 percent in 1982.  Since 1982, the rate has been
rising, reaching 47.7 percent in 1992. 

   Figure II.2:  History of DI
   Award Rate, 1970-92

   (See figure in printed
   edition.)

Note:  SSA's change in technique for reporting applications also
affects these data.  (See app.  I, fig.  I.1.)


   HOW HAVE LEGISLATIVE AND
   REGULATORY CHANGES AFFECTED THE
   AWARD RATE? 
-------------------------------------------------------- Appendix II:1

"Disability" is defined in law and regulation; then further policy
guidance is provided in SSA operational instructions.  Changes in any
of these criteria can be expected to have an impact on the rate of DI
awards.  Generally, recent legislation and policy changes have tended
to liberalize the requirements for determining disability. 
Amendments to the Social Security Act adopted in 1984 required SSA to
(1) place greater emphasis on the opinions of the applicants'
treating physicians, (2) focus more attention on the combined effect
of multiple impairments, and (3) increase attention to the role of
pain in restricting a person's ability to work. 

These amendments also required SSA to develop new criteria for
evaluating mental impairment disabilities to better judge an
applicant's ability to work in a competitive environment.  Available
data suggest that policy changes regarding how mental impairments
should be evaluated have had a significant effect on the award rate. 


      CHANGES IN MENTAL IMPAIRMENT
      STANDARDS HAVE RESULTED IN
      INCREASED AWARDS
------------------------------------------------------ Appendix II:1.1

Mental impairment awards have increased substantially.  In 1992,
nearly 26 percent of all disability awards were made on this basis,
compared to nearly 11 percent in 1982.\1 Mental impairment awards are
now the largest single category of awards.  As shown in figure II.3,
this increase started in 1982, when SSA first began considering
revisions to its mental impairment regulations. 

   Figure II.3:  Mental Impairment
   Awards:  An Increasing
   Percentage of All DI Awards,
   1981-92

   (See figure in printed
   edition.)

When the new SSA standards were adopted in August 1985, the increase
accelerated.  SSA attributes the "spike" in 1986 mental impairment
awards shown in figure II.3 to processing of pending applications,
which had been stockpiled waiting for the new standards to take
effect.  After the spike, mental impairment awards continued an
upward trend, although at a slower rate. 

This increase in mental impairment awards may reflect (1) an increase
in the number of mentally impaired persons seeking awards and/or (2)
an increase in the rate at which applicants are awarded benefits. 
Available data did not permit us to determine how much each of these
possibilities contributes to the overall increase.  SSA data,
however, indicate that applicants whose disability is based on mental
impairment have a higher allowance rate than many other types of
applicants. 

Specifically, in 1992, when the overall award rate on all claims was
47.7 percent, mental impairment claims had a higher success rate.  DI
applicants with mental impairments were successful nearly 66 percent
of the time.  Concurrent DI/SSI applicants with such impairments
achieved a 54-percent success rate.  Overall, persons applying on the
basis of mental impairments were successful 58 percent of the time,
whereas persons applying on the basis of physical impairments were
successful only 39 percent of the time.\2


--------------------
\1 These data include only cases where mental impairment is the
primary cause of disability. 

\2 Data on success rates are based on SSA's 831 file.  Our analysis
included 93 percent of awards reported by SSA in 1992.  This file
provides a "snapshot" of the decisions made in that year,
irrespective of whether the application was filed in that year or a
previous year. 


      IMPACT OF OTHER LEGISLATIVE
      AND PROGRAM CHANGES
      DIFFICULT TO ASSESS
------------------------------------------------------ Appendix II:1.2

The impacts of legislative, regulatory, and other program changes are
more difficult to assess.  For example, no data or studies have
quantified the impact of the mandates for increased emphasis on the
opinions of treating physicians, multiple impairments, and pain.  The
effects of such changes on decisionmaking can be subtle.  However,
they can also be significant, especially in borderline cases where
much subjective judgment is needed. 

Subjective judgment is an important part of disability
decisionmaking, despite SSA efforts to make decisions more
objectively.  HHS explains that even with the extensive operating
guidance provided by SSA, "...deciding whether a person is able to
perform other work in the economy, given his or her age, education,
and work experience, frequently involves a certain degree of judgment
on the part of the disability adjudicator.  Similarly, disabilities
involving pain or mental impairments are inherently more difficult to
evaluate than conditions with more overt physical manifestations."

In these judgmental cases, the "adjudicative climate" can play an
important role.  This climate is defined by SSA as "...the
perceptions of individual disability adjudicators, based on the
prevailing national attitudes regarding disability, that may affect
how they apply existing formal policy in instances where some
judgment is required within the specified evaluation procedures."

In such a decisionmaking environment, subtle changes in regulation
may have a significant effect.  For example, it may not be possible
to directly assess the effect of a requirement to give "increased
emphasis" to the opinion of a treating physician.  Nonetheless, to
the extent it alters the adjudicative climate, such a requirement can
have a significant impact. 


   WHAT HAS BEEN THE IMPACT OF
   COURT DECISIONS? 
-------------------------------------------------------- Appendix II:2

Since more than half a million new DI awards are now being made each
year, the number of awards to successful litigants appears relatively
inconsequential.  For example, according to SSA, the two largest DI
class action decisions against the agency have resulted in 8,440 and
2,621 new awards, respectively, to class members.  However, SSA
indicates that court decisions may have had a significant impact in
increasing awards by causing SSA to liberalize policies and rulings
in favor of applicants. 

SSA materials indicate that the mental impairment regulations
discussed above were influenced by a 1982 decision of the U.S. 
District Court in Minnesota, Mental Health Association of Minnesota
v.  Schweiker\3 .  In other situations, SSA has responded to court
decisions by changing regional policies--by issuing acquiescence
rulings to comply with a decision only within the applicable judicial
circuit.  Today, 13 such rulings pertaining to DI are in effect. 

And SSA records show that adjudicators in the Seattle region
significantly increased their rate of allowances, from 39 to 52
percent, in the 2 years following the implementation of a decision of
the U.S.  District Court there.  Such a major change took place
despite the fact that the ruling directly benefitted only 300 DI
applicants--representing less than 2 percent of a total of 17,000 DI
applicants in the Seattle area in that year. 


--------------------
\3 554 F.  Supp.  157 (D.  Minnesota 1982). 


   KNOWN ERRORS IN DECISIONMAKING
   DO NOT CONTRIBUTE TO INCREASED
   AWARDS
-------------------------------------------------------- Appendix II:3

One of our requesters asked us to examine whether errors made by
decisionmakers were causing the growth in awards.  Our analysis of
available information points in the other direction.  That is, if
decisionmaking errors had never been made, more awards could have
resulted. 

In our analysis, we started with decisionmaking at the initial
disability determination services decision level, since about
two-thirds of benefit awards and denials are made at this level.  SSA
reviews these decisions to assure their quality.\4 SSA conducts these
reviews on a statistical sample of cases, then uses the results to
measure DDS performance.  Generally, these reviews find that DDSs
achieve more than 96-percent accuracy in award decisions.  In benefit
denials, DDSs usually reach about 93-percent accuracy. 

Because DDSs produce about 40 percent more denials than awards, the
higher error rates on DDS denials may have a significant effect in
reducing the number of awards.  Specifically, an analysis of quality
assurance data indicates that about 30,500 of the 713,300 DDS denials
in 1992 would have been awards if no errors had been made.  In
addition, about 7,000 of the 503,100 awards would have been
denials.\5 Thus, if DDSs had made no errors at all, a net increase in
awards would have resulted.  Analysis of data from other years
reaches similar conclusions. 


--------------------
\4 We are currently working on a request from the Senate Committee on
Finance to review SSA's quality assurance process. 

\5 Quality assurance data indicated that about 6.8 percent of denials
had to be returned to DDSs for correction.  When developed and
corrected, about 63 percent of these denials were reversed.  About 3
percent of awards were returned, of which 47 percent were reversed. 
The denial data do not include the effects of eventual appeals. 


   WHAT HAS BEEN THE IMPACT OF
   INCREASED ALLOWANCES OCCURRING
   AT THE ALJ LEVEL? 
-------------------------------------------------------- Appendix II:4

Although ALJs currently award benefits to more than 70 percent of
applicants who appeal, and although the number of appeals has
increased, available data do not suggest that ALJ decisions are the
major cause of recent award rate increases.  Instead, most of the
increase in total awards occurred at the DDS level. 

As shown in figure II.4, the award rate increased at both the initial
determination and reconsideration stages in the DDSs.  The rate also
increased at the ALJ level, but more rapidly. 

   Figure II.4:  DI Award Rate
   Rose at Both the DDS and ALJ
   Levels, 1985-92

   (See figure in printed
   edition.)

The slower change at the DDS level, however, had a greater impact on
the rolls because DDSs provide roughly three-and-a-half times as many
awards as ALJs.  Figure II.5 shows the sources of awards. 

   Figure II.5:  Number of DI
   Awards Made by DDSs and ALJs,
   1985-92

   (See figure in printed
   edition.)

Note:  Data include awards for disabled dependents.  Awards made by
SSA's Appeals Council and federal district courts, which represent
less than 2 percent of the total, are not included. 

Thus increases at all levels of the decisionmaking process are
playing a role in increasing total awards. 


MORE BENEFICIARIES REMAIN ON THE
ROLLS
========================================================= Appendix III

The rate at which beneficiaries leave the DI program has generally
been declining for more than 20 years.  Since 1985, over 90 percent
of terminations from the DI rolls have occurred when beneficiaries
died or converted to retirement payment under Social Security's Old
Age and Survivors Insurance fund at age 65.  These types of
terminations can be expected to further decrease in the future, since
new awardees are coming on the rolls at younger ages and staying
longer.  Increasing numbers of awards to younger persons with mental
impairments have contributed to this decrease in awardee age. 

In recent years, persons recovering\1 from disability have
represented less than 5 percent of terminations.  CDRs and vocational
rehabilitation have had little impact on the DI rolls.  For example,
VR reduced the DI rolls by less than 0.2 percent in 1992. 


--------------------
\1 Persons are considered "recovered" when they leave the rolls
because of a CDR, rehabilitation, or voluntary return to work. 


   HOW BENEFICIARIES LEAVE THE
   ROLLS
------------------------------------------------------- Appendix III:1

DI beneficiaries leave the rolls (benefits are terminated) under
several circumstances.  About 5 percent of beneficiaries die each
year.  This is a higher rate than the rate of death in the population
aged 18 to 64 as a whole, because DI beneficiaries are usually
severely physically impaired and thus have lower life expectancies
than the average person. 

Second, persons who reach age 65 on the DI rolls are automatically
converted to retirement payments under the OASI fund.  This is the
same fund that pays benefits to all Social Security retirement
beneficiaries.  Thus, a person can process into general retirement
status from the disability rolls. 

Third, persons leave the rolls when SSA determines they are no longer
disabled.  SSA is required to make such CDR determinations every 3
years on all cases where medical improvement is expected or possible. 

Also, some persons return to work as a result of receiving VR
services or finding work without such services. 


   TERMINATION RATES HAVE
   GENERALLY BEEN DECLINING
------------------------------------------------------- Appendix III:2

As shown in figure III.1, the rate at which beneficiaries leave the
program has generally been declining, with the exception of the
period from 1981 through 1983.  During this period, SSA conducted
over a million CDRs and attempted to remove about 442,900 persons
from the DI rolls.\2

   Figure III.1:  History of DI
   Termination Rate, 1970-92

   (See figure in printed
   edition.)

Note:  1991 data omit terminations for reasons other than death or
retirement. 

This initiative was not sustainable, however.  Because of opposition
to it, the Secretary of HHS halted CDR activity in October 1984. 
Legislation also prevented SSA from terminating benefits unless the
agency could find medical improvement in a beneficiary's condition. 
Since 1984, the DI termination rate has resumed its decline. 


--------------------
\2 As of 1987, about two-thirds of former beneficiaries who were
determined by SSA, between 1981 and 1984, to be ineligible for
benefits had been reinstated on the benefit rolls.  See Social
Security Disability:  Denied Applicants' Health and Financial Status
Compared with Beneficiaries' (GAO/HRD-90-2, Nov.  6, 1989). 


   AWARDEE AGE HAS SIGNIFICANT
   ROLE IN DECLINE OF TERMINATION
   RATES
------------------------------------------------------- Appendix III:3

The age of new awardees has been dropping.  Because more than 90
percent of terminations are due to death or conversion to retirement,
this continuing decline in awardee age is significant. 

Beneficiaries are coming on the rolls at a younger age and staying
longer.  The average awardee age dropped from 52.3 years in 1970 to
47.8 years in 1992.  (See fig.  III.2.) At the same time,
beneficiaries are staying on the rolls longer.  SSA data indicate
that 14 percent of beneficiaries in 1992 had been on the rolls for 15
years or more.  The percentage is a sharp increase from 1980, when
the comparable figure was 5 percent. 

Several factors contribute to the decline in awardee age.  Through
1981, the decline could be explained by a parallel decline in the
average age of the insured population.  More "baby-boom" workers were
achieving insured status.  As more younger workers came into the
insurance pool, the pool's average age declined.  As the pool grew
younger on average, awardees did also. 

Beginning in 1982, the situation changed.  The average age of the
insured started to increase.  This event occurred because the
baby-boom group was now starting to work its way through the insured
population.  Between 1981 and 1992, the insured pool's average age
grew by more than a year, from 36.4 to 37.7.  Despite this increase
in the age of the insured, the age of new awardees continued to drop. 
Specifically, the average awardee age fell by 3 years, from 50.8 in
1981 to 47.8 in 1992.  (See fig.  III.2.)

   Figure III.2:  Age Trends in
   Insured and New Awardee
   Populations, 1970-92

   (See figure in printed
   edition.)

Differing rates of awards to younger and older persons explain some
of the decline in average awardee age since 1982.  As shown in table
III.1, incidence rates (awards per thousand insured persons) for
younger workers have been rising faster and for a longer period than
incidence rates for older workers.  In the 12 years between 1980 and
1992, incidence rates for younger workers rose 51.5 percent, from 1.7
persons per thousand to 3.3 per thousand.  For older workers, the
rate has generally been declining, with increases coming only in 1991
and 1992.  Between 1990 and 1992, the older workers' rate rose 25.5
percent, from 10.6 to 13.3. 



                         Table III.1
           
           Incidence Rate Trends Differ for Younger
                  and Older Workers, 1970-92

                          1970  1980  1985  1990  1991  1992
------------------------  ----  ----  ----  ----  ----  ----
Awards per 1,000 workers
 under 50                  2.0   1.7   1.9   2.3   2.6   3.3
Awards per 1,000 workers
 age 50 or older          13.1  12.2  11.0  10.6  11.3  13.3
------------------------------------------------------------

   WHAT HAS CAUSED THE RECENT
   DECLINE IN AVERAGE AWARDEE AGE? 
------------------------------------------------------- Appendix III:4

The divergence between the disability experience of those younger and
older than age 50 has two causes.  For younger workers, increased
mental impairment awards appear to be the major factor.  For older
workers, decreases in awards to persons with cardiovascular problems
seem to play an important role. 

As shown in figure III.3, mental impairment awards account for most
of the increase in awards to persons younger than 50.  The changes in
mental impairment standards discussed in appendix II appear to be the
major cause of this increase, although other factors may also play a
role. 

   Figure III.3:  Many Awards to
   Younger Persons Are Due to
   Mental Impairment, 1981-92

   (See figure in printed
   edition.)

For older workers, a decline in DI awards based on cardiovascular
impairment appears to play a role in decreasing their incidence
rates.  These types of awards dropped from 177,300, or 30 percent of
all awards in 1975, to 89,800, or 14 percent of all awards in 1992. 
Since most cardiovascular awards go to persons older than 50,
declines in such awards have coincided to some extent with reductions
in awards to older persons. 

However, the overall causes of the decline in awards to older persons
are not clear.  New, more rigorous, standards for evaluating
cardiovascular impairments were established in 1979, which
contributed to the decline in this type of award.  The shift from a
manufacturing to a service economy may also contribute to declining
disability among the older population.  As more workers spend time in
less strenuous service jobs, they may experience lower physical
demands from employment, and thus suffer less work-induced
disability. 


   MINIMAL RECENT IMPACT OF CDRS
------------------------------------------------------- Appendix III:5

Although SSA is required to take steps to remove ineligible
beneficiaries from the DI rolls, SSA has experienced difficulty in
doing so.  Funding and legal constraints have hampered its efforts. 

The Social Security Act requires that every 3 years SSA review
disability cases where medical improvement is expected.  SSA has been
unable to meet this requirement for more than 6 years, due in part to
heavy workload demands caused by the recent increase in applications. 
SSA estimates that it now has over 1 million CDRs backlogged. 

We believe CDRs remain necessary.  For example, in our November 1989
report on denied applicants\3 , we found that 15 percent of persons
removed from the rolls during the 1981 to 1984 period of heavy
emphasis on CDRs had returned to work and were still working in 1987. 
SSA has estimated that the net cost of not performing CDRs in fiscal
years 1990 through 1993 will be $1.4 billion, projected through
1997.\4

However, it should be realized that CDRs, even if fully pursued, have
limited ability to affect the size of the rolls.  SSA has estimated
that performing overdue CDRs could remove around 30,000 persons from
the rolls.  This figure is less than 1 percent of the 3.5 million
beneficiaries on the rolls. 


--------------------
\3 See GAO/HRD-90-2, Nov.  6, 1989. 

\4 This figure includes amounts saved by removing from the rolls
those whose medical condition had improved to the extent that they
were no longer disabled.  The figure is net of the increased
administrative costs necessary to perform CDRs. 


   REHABILITATION CONTRIBUTES
   LITTLE TO TERMINATIONS
------------------------------------------------------- Appendix III:6

Rehabilitation has had very little impact on the disability rolls. 
The Social Security Act provides that DI applicants be referred to VR
agencies.  However, in 1992, less than 6,300 DI beneficiaries, or
less than 0.2 percent of the rolls, were returned to work via this
route.  In that year, SSA spent only $32 million on VR, or about 0.1
percent of benefit costs. 

In 1987, we reported that VR agencies were having little success with
DI beneficiaries.\5 In that study, rehabilitation counselors told us
VR efforts were unsuccessful mostly because (1) the disabilities of
DI recipients were generally too severe to warrant rehabilitation and
(2) DI recipients generally feared losing their DI and Medicare
benefits if rehabilitation resulted in a permanent return to work. 


--------------------
\5 Social Security:  Little Success Achieved in Rehabilitating
Disabled Beneficiaries (GAO/HRD-88-11, Dec.  1987). 


STATISTICAL DATA FOR FIGURES USED
IN THIS REPORT
========================================================== Appendix IV



                                    Table IV.1
                     
                     History of DI Application Rate, 1970-92
                                   (Figure I.1)

                              (Numbers in thousands)


Calend                                                    Based on      Based on
ar            Reported      Adjusted    DI-insured        reported      adjusted
year              data        data\a     workers\b            data        data\c
------  --------------  ------------  ------------  --------------  ------------
1970             868.2         772.7        72,400            12.0          10.7
1971             924.4         822.7        74,500            12.4          11.0
1972             947.8         843.5        76,100            12.5          11.1
1973           1,066.9         949.5        77,800            13.7          12.2
1974           1,330.2       1,183.9        80,400            16.5          14.7
1975           1,285.3       1,143.9        83,300            15.4          13.7
1976           1,232.2       1,096.7        85,300            14.4          12.9
1977           1,235.2       1,099.3        87,000            14.2          12.6
1978           1,184.7       1,054.4        89,300            13.3          11.8
1979           1,187.8       1,057.1        93,700            12.7          11.3
1980           1,262.3       1,123.4        98,000            12.9          11.5
1981           1,161.3       1,033.6       100,000            11.6          10.3
1982           1,020.0                     102,600             9.9
1983           1,017.7                     104,500             9.7
1984           1,035.7                     105,400             9.8
1985           1,066.2                     107,100            10.0
1986           1,118.4                     109,600            10.2
1987           1,108.9                     111,600             9.9
1988           1,017.9                     113,500             9.0
1989             984.9                     115,800             8.5
1990           1,067.7                     118,500             9.0
1991           1,208.7                     120,300            10.0
1992           1,335.1                     122,100            10.9
--------------------------------------------------------------------------------
\a The application rate is the number of applications per 1,000
workers insured for disability. 

\b The numbers of DI-insured workers are rounded to hundreds of
thousands. 

\c Computed by GAO.  GAO reduced the number of applications by 11
percent to compensate for 1981 changes in SSA's reporting techniques. 

Source:  Annual statistical supplements to the Social Security
Bulletin. 



                          Table IV.2
           
               DI Application Rate Compared to
           Unemployment Rate, 1970-92 (Figure I.2)


Calendar                  Percent of              Percent of
year                Rate  base year\        Rate   base year
------------  ----------  ----------  ----------  ----------
1970                12.0       100.0         4.8       100.0
1971                12.4       103.3         5.8       120.8
1972                12.5       104.2         5.5       114.6
1973                13.7       114.2         4.8       100.0
1974                16.5       137.5         5.5       114.6
1975                15.4       128.3         8.3       172.9
1976                14.4       120.0         7.6       158.3
1977                14.2       118.3         6.9       143.8
1978                13.3       110.8         6.0       125.0
1979                12.7       105.8         5.8       120.8
1980                12.9       107.5         7.0       145.8
1981                11.6        96.7         7.5       156.3
1982                 9.9        82.5         9.5       197.9
1983                 9.7        80.8         9.5       197.9
1984                 9.8        81.7         7.4       154.2
1985                10.0        83.3         7.1       147.9
1986                10.2        85.0         6.9       143.8
1987                 9.9        82.5         6.1       127.1
1988                 9.0        75.0         5.4       112.5
1989                 8.5        70.8         5.2       108.3
1990                 9.0        75.0         5.4       112.5
1991                10.0        83.3         6.6       137.5
1992                10.9        90.8         7.3       152.1
------------------------------------------------------------
Note:  Percentages calculated based on current year rate divided by
1970 rate. 

\a Unemployed as percent of labor force, including resident armed
forces. 

Source:  Application rate as shown in table IV.1.  Unemployment rate
from Economic Report of the President, Jan.  1993. 



                          Table IV.3
           
             Concurrent SSI/DI Applications Rose
           Faster Than DI-Only Applications, 1980-
                       92 (Figure 1.3)

                    (Numbers in thousands)


                          Percent of              Percent of
Fiscal year       Number   base year      Number   base year
------------  ----------  ----------  ----------  ----------
1980                 803       100.0         395       100.0
1981                 748        93.2         371        93.9
1982                 651        81.1         349        88.4
1983                 631        78.6         381        96.5
1984                 622        77.5         414       104.8
1985                 621        77.3         429       108.6
1986                 653        81.3         502       127.1
1987                 604        75.2         463       117.2
1988                 593        73.8         407       103.0
1989                 580        72.2         396       100.3
1990                 604        75.2         440       111.4
1991                 648        80.7         519       131.4
1992                 661        82.3         603       152.7
------------------------------------------------------------
Note:  Data include all disability decisions made by DDSs, including
cases involving disabled dependents.  Percentages calculated based on
current year numbers divided by 1970 numbers. 

Sources:  SSA's State Agency Operations Reports and CRS Report for
Congress:  Status of the Disability Programs of the Social Security
Administration (Sept.  8, 1992). 



                          Table IV.4
           
            History of DI Incidence Rate, 1970-92
                        (Figure II.1)

                    (Numbers in thousands)

                                                  Awards per
                                               1,000 insured
                                                     workers
Calendar           Number of      DI-insured      (Incidence
year                  awards         workers           rate)
------------  --------------  --------------  --------------
1970                   350.4          72,400             4.8
1971                   415.9          74,500             5.6
1972                   455.4          76,100             6.0
1973                   491.6          77,800             6.3
1974                   536.0          80,400             6.7
1975                   592.0          83,300             7.1
1976                   551.5          85,300             6.5
1977                   568.9          87,000             6.5
1978                   464.4          89,300             5.2
1979                   416.7          93,700             4.4
1980                   391.6          98,000             4.0
1981                   345.3         100,300             3.5
1982                   298.5         102,600             2.9
1983                   311.5         104,500             3.0
1984                   357.1         105,400             3.4
1985                   377.4         107,100             3.5
1986                   416.9         109,600             3.8
1987                   415.8         111,600             3.7
1988                   409.5         113,500             3.6
1989                   425.6         115,800             3.7
1990                   468.0         118,500             3.9
1991                   536.4         120,300             4.5
1992                   636.6         122,100             5.2
------------------------------------------------------------
Source:  Annual statistical supplements to the Social Security
Bulletin. 



                                    Table IV.5
                     
                        History of DI Award Rate, 1970-92
                                  (Figure II.2)

                              (Numbers in thousands)


Calend                                                    Based on      Based on
ar            Reported      Adjusted     Number of        reported      adjusted
year              data        data\a        awards            data          data
------  --------------  ------------  ------------  --------------  ------------
1970             868.2         772.7         350.4            40.4          45.3
1971             924.4         822.7         415.9            45.0          50.6
1972             947.8         843.5         455.4            48.0          54.0
1973           1,066.9         949.5         491.6            46.1          51.8
1974           1,330.2       1,183.9         536.0            40.3          45.3
1975           1,285.3       1,143.9         592.0            46.1          51.8
1976           1,232.2       1,096.7         551.5            44.8          50.3
1977           1,235.2       1,099.3         568.9            46.1          51.8
1978           1,184.7       1,054.4         464.4            39.2          44.0
1979           1,187.8       1,057.1         416.7            35.1          39.4
1980           1,262.3       1,123.4         391.6            31.0          34.9
1981           1,161.3       1,033.6         345.3            29.7          33.4
1982           1,020.0                       298.5            29.3
1983           1,017.7                       311.5            30.6
1984           1,035.7                       357.1            34.5
1985           1,066.2                       377.4            35.4
1986           1,118.4                       416.9            37.3
1987           1,108.9                       415.8            37.5
1988           1,017.9                       409.5            40.2
1989             984.9                       425.6            43.2
1990           1,067.7                       468.0            43.8
1991           1,208.7                       536.4            44.4
1992           1,335.1                       636.6            47.7
--------------------------------------------------------------------------------
\a Adjusted application data as shown in table IV.1. 

Source:  Annual statistical supplements to the Social Security
Bulletin. 



                          Table IV.6
           
           Mental Impairment Awards: An Increasing
              Percentage of All Awards, 1981-92
                        (Figure II.3)

                    (Numbers in thousands)


                    Total awards
                     to disabled                  Percent of
Calendar year            workers        Number         total
------------------  ------------  ------------  ------------
1981                       345.3          36.3          10.5
1982                       298.5          31.5          10.6
1983                       311.5          50.6          16.2
1984                       357.1          64.1          18.0
1985                       377.4          68.6          18.2
1986                       416.9         124.0          29.7
1987                       415.8          81.2          19.5
1988                       409.5          85.8          21.0
1989                       425.6          88.5          20.8
1990                       468.0         105.2          22.5
1991                       536.4         126.2          23.5
1992                       636.6         164.1          25.8
------------------------------------------------------------
Source:  Annual statistical supplements to the Social Security
Bulletin. 



                                      Table IV.7
                       
                       Award Rate Rose at Both the DDS and ALJ
                            Levels, 1985-92 (Figure II.4)

                                (Numbers in thousands)




Fi
sc
al   Actions                     Initial                     Actions
ye  taken on                     actions                    taken on
ar   appeals  Number    Rate     taken\a    Number    Rate   appeals  Number    Rate
--  --------  ------  ------  ----------  --------  ------  --------  ------  ------
19     168.1    92.1    54.8     1,016.1     318.2    31.3     286.4    37.9    13.2
 85
19     151.7    78.7    51.9     1,138.5     381.7    33.5     282.0    43.9    15.6
 86
19     171.9    98.2    57.1     1,096.1     357.2    32.6     325.7    44.2    13.6
 87
19     188.7   111.7    59.2     1,010.5     331.8    32.8     303.6    39.8    13.1
 88
19     196.0   122.1    62.3       962.0     331.6    34.5     297.0    42.0    14.1
 89
19     192.3   127.7    66.4     1,013.5     363.3    35.8     336.0    49.6    14.8
 90
19     209.4   144.9    69.2     1,101.6     421.7    38.3     344.8    54.4    15.8
 91
19     231.3   164.2    71.0     1,268.8     503.1    39.7     400.0    62.6    15.7
 92
------------------------------------------------------------------------------------
Note:  Data include all disability decisions made by DDSs, including
cases involving disabled dependents. 

\a Indicates clearances by DDSs at initial stage. 

Sources:  SSA's State Agency Operations Report and data from SSA's
Office of Hearings and Appeals. 



                          Table IV.8
           
             Number of DI Awards Made by DDSs and
                 ALJs, 1985-92 (Figure II.5)

                    (Numbers in thousands)


Fiscal year     Made by ALJs    Made by DDSs           Total
------------  --------------  --------------  ==============
1985                    92.1           356.1           448.2
1986                    78.7           425.6           504.3
1987                    98.2           401.4           499.6
1988                   111.7           371.6           483.3
1989                   122.1           373.6           495.7
1990                   127.7           412.9           540.6
1991                   144.3           476.1           621.0
1992                   164.2           565.7           729.9
------------------------------------------------------------
Note:  Data include awards to disabled dependents.  Awards made by
SSA's Appeals Council and federal courts, which represent less than 2
percent of the total, are not included. 

Sources:  SSA's State Agency Operations Reports and data from SSA's
Office of Hearings and Appeals. 



                                     Table IV.9
                      
                      History of DI Termination Rate, 1970-92
                                   (Figure III.1)

                               (Numbers in thousands)


Ca
le
nd
ar     On rolls at
ye    beginning of      Terminated         Percent                         Percent
ar            year     during year      terminated          Number      terminated
--  --------------  --------------  --------------  --------------  --------------
19         1,394.3           260.4            18.7           208.7            15.0
 70
19         1,492.9           266.5            17.9           216.9            14.5
 71
19         1,647.7           261.7            15.9           215.0            13.0
 72
19         1,832.9           304.8            16.6           261.0            14.2
 73
19         2,016.6           321.0            15.9           277.8            13.8
 74
19         2,236.9           329.5            14.7           297.4            13.3
 75
19         2,488.8           351.5            14.1           310.1            12.5
 76
19         2,670.2           401.3            15.0           334.6            12.5
 77
19         2,837.4           413.6            14.6           337.6            11.9
 78
19         2,879.8           422.5            14.7           347.2            12.1
 79
19         2,870.6           408.1            14.2           342.8            11.9
 80
19         2,858.7           434.2            15.2           333.4            11.7
 81
19         2,776.5           483.8            17.4           319.5            11.5
 82
19         2,603.6           453.6            17.4           328.0            12.6
 83
19         2,569.0           371.9            14.5           319.7            12.4
 84
19         2,596.5           340.0            13.1           322.9            12.4
 85
19         2,656.6           341.3            12.8           321.5            12.1
 86
19         2,728.5           331.5            12.1           313.7            11.5
 87
19         2,785.9           346.3            12.4           315.4            11.3
 88
19         2,830.3           336.3            11.9           318.6            11.3
 89
19         2,895.4           327.8            11.3           319.5            11.0
 90
19         3,011.3           320.3            10.6           319.2            10.6
 91
19         3,194.9           345.9            10.8           334.7            10.5
 92
----------------------------------------------------------------------------------
Note:  1991 data omit terminations for reasons other than death or
retirement.  Retirement equates with conversion to Old Age and
Survivors Insurance benefits. 

Sources:  Annual statistical supplements to the Social Security
Bulletin and The Social Security Disability Insurance Program:  An
Analysis, HHS, (Dec.  6, 1992). 



                         Table IV.10
           
            Age Trends in Insured and New Awardee
             Populations, 1970-92 (Figure III.2)


Calendar year               New awardees  Insured population
--------------------  ------------------  ------------------
1970                                52.3                38.5
1971                                52.2                38.2
1972                                52.7                38.2
1973                                52.9                38.2
1974                                52.2                37.9
1975                                51.7                37.6
1976                                51.8                37.4
1977                                51.6                37.4
1978                                51.4                37.3
1979                                51.4                37.3
1980                                51.2                36.6
1981                                50.8                36.4
1982                                50.8                36.6
1983                                50.1                36.9
1984                                49.9                37.1
1985                                50.0                37.2
1986                                48.7                37.2
1987                                49.2                37.2
1988                                49.2                37.3
1989                                49.0                37.4
1990                                48.2                37.4
1991                                48.1                37.5
1992                                47.8                37.7
------------------------------------------------------------
Source:  Analysis of data in annual statistical supplements to the
Social Security Bulletin. 



                         Table IV.11
           
            Many Awards to Younger Persons Are Due
            to Mental Impairment, 1981-92 (Figure
                            III.3)

                    (Numbers in thousands)



Calendar                  Percent of              Percent of
year              Number   base year      Number   base year
------------  ----------  ----------  ----------  ----------
1981                24.6       100.0        98.5       100.0
1982                21.7        88.2        87.6        89.0
1983                35.3       143.5        85.3        86.6
1984                43.7       177.6        97.1        98.6
1985                49.3       200.4       108.8       110.5
1986                89.8       365.0        98.5       100.1
1987                59.0       239.8       125.5       127.5
1988                63.3       257.3       119.7       121.6
1989                64.1       260.6       125.6       127.6
1990                77.9       316.7       137.5       139.6
1991                94.3       383.3       161.1       163.6
1992               125.6       510.6       187.6       190.5
------------------------------------------------------------
Note:  Percent of base year computation based on unrounded number of
awards, with 1981 as the base year. 

Source:  Annual statistical supplements to the Social Security
Bulletin. 


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V

HEALTH, EDUCATION, AND HUMAN
SERVICES DIVISION,
WASHINGTON, D.C. 

Barry D.  Tice, Assistant Director, (410) 965-8920
David F.  Fiske, Evaluator-in-Charge
Ellen S.  Habenicht, Evaluator
John W.  Wood, Actuary
Vanessa R.  Taylor, Evaluator (Computer Science)
Joanne R.  Newman, Information Processing Assistant