Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 30
Improving the Social Security Disability Decision Process 3 Individuals with Disabilities The purpose of the Social Security Disability Insurance (SSDI) and Supplementary Security Income (SSI) programs is to provide cash assistance to individuals who, because of physical or mental impairment, are unable to engage in “substantial gainful activity.” This is currently defined by the Social Security Administration (SSA) as earning no more than $900 a month ($1,500 for blind individuals). The purpose of the Listing of Impairments (the Listings) is to provide a standardized test or tool to quickly “screen in” claimants whose medical condition is of such a severity that few, if any, individuals having that condition could be expected to engage in substantial gainful activity. Assessing the effectiveness of the Listings as a quick screening tool requires information about who is screened in, especially those who should not be, and who is not screened in, especially those who should be. This chapter summarizes what is known about individuals in the population with impairments and functional limitations that make it very difficult or impossible for them to work and what is known about the characteristics of the SSDI and SSI applicant pools and of those who are awarded benefits and those who are denied. INFORMATION FROM SURVEYS ON INDIVIDUALS WITH DISABILITIES From time to time, SSA has conducted surveys to obtain information on individuals with disabilities, those who do and do not apply to Social Security for disability benefits, and on what happens to those who are denied. In addition, a number of national population surveys include questions about disability.
OCR for page 31
Improving the Social Security Disability Decision Process Social Security Surveys SSA has sponsored three major surveys to provide information useful for understanding and projecting enrollment trends in the disability programs (Hu et al., 1997). They were the: 1966 Social Security Survey of the Disabled 1972/1974 Survey of Disabled and Nondisabled Adults 1978 Survey of Disability and Work The surveys were used in a number of analyses of disability enrollment patterns, but they are dated. Program enrollment increased significantly and unexpectedly in the late 1980s to early 1990s. Studies of this growth identified a number of factors that might have caused it, primarily socioeconomic and demographic trends, court decisions, and changes in program eligibility (IOM and NRC, 2002:Chapter 2). Lack of information about the pool of potential applicants and the factors influencing their decision whether or not to apply for benefits made it impossible to understand or predict program dynamics. Accordingly, SSA began to develop another national sample survey in the 1990s, the Disability Evaluation Study (DES). The DES, later renamed the National Study of Health and Activity, was to collect more detailed information from respondents about health conditions and limitations affecting ability to work, and this self-reported information was to be supplemented by medical examinations of respondents. SSA fielded a pilot in 2000 and a revised pilot in fiscal year (FY) 2002. SSA found that the need to conduct in-person screening to achieve an adequate response rate and to increase the sample size to make the results representative would have doubled the original estimate of costs. SSA decided to discontinue the effort in FY 2003, because “although extensive testing of the National Study of Health and Activity developed a workable methodology, … the benefits of the survey would not justify the substantial projected costs” (SSA, 2002:141). National Surveys Several ongoing national surveys include questions about impairments, functional limitations, employment limitations, and other disability-related topics. None provide information on the number and characteristics of individuals who are prevented from earning more than a minimal amount by engaging in productive work because of identifiable chronic physical or mental impairments that would meet the Social Security definition of disability. Little is known about individuals who meet SSA’s disability criteria in every respect except that they are engaged in substantial gainful activity despite their medical condition. The surveys rely on self-report and so at
OCR for page 32
Improving the Social Security Disability Decision Process best can only estimate the number of people who say they are unable to work because of a disability. Most of the surveys simply ask the respondent to report any limitation on their ability to work and, in some cases, the nature of the limitation and its cause. Mostly, however, surveys separate questions about work limitations from those about functional limitations, such as activities of daily living (ADLs) and instrumental activities of daily living (IADLs), and about actual employment at the time of the survey and during the previous year. The main surveys with information about disability are the: Survey of Income and Program Participation (SIPP) National Health Interview Survey American Community Survey (ACS) Panel Study of Income Dynamics (PSID) Decennial Census Each of these surveys and their findings are described below. Table 3-1 summarizes recent results from the surveys. The surveys estimate that indi- TABLE 3-1 Number and Prevalence of Individuals Ages 25-61 with and without Disabilities and Number and Percentage of Those Employed, by Disability Category and by Survey Disability Category 2003 American Community Survey Total Population Employed Number Percent Number Percent All 143,796,355 100.0 107,425,071 74.7 With any disability 17,146,845 11.9 6,738,710 39.3 With a sensory disability 3,944,398 2.7 1,968,255 49.9 With a physical disability 10,819,521 7.5 3,656,998 33.8 With a mental disability 5,745,569 4.0 1,620,250 28.2 With a self-care disability 2,925,715 2.0 535,406 18.3 With a go-outside-home disability 4,227,427 2.9 756,709 17.9 With an employment disability 9,854,223 6.9 1,862,448 18.9 With no disability 126,649,510 88.1 100,686,360 79.5 Disability Category 1994-1995 National Health Interview Survey-Disability Supplement Total Population Employed Number Percent Number Percent All 128,104,461 100.0 100,914,424 78.8 With any disability 21,457,049 16.7 11,543,892 53.8 With a sensory disability 5,400,346 4.2 3,477,823 64.4 With a physical disability 4,660,568 3.6 1,868,888 40.1
OCR for page 33
Improving the Social Security Disability Decision Process With a mental disability 5,402,293 4.2 2,387,814 44.2 With a self-care disability 1,534,615 1.2 329,942 21.5 With a go-outside-home disability 1,613,015 1.3 277,439 17.2 With an employment disability 14,903,532 11.6 6,334,001 42.5 With no disability 106,647,412 83.3 89,370,531 83.8 Disability Category 2002 Survey of Income and Program Participation Total Population Employed Number Percent Number Percent All 142,520,000 100.0 108,518,780 76.1 With any disability 26,620,000 18.7 13,017,180 48.9 With a sensory disability 6,490,202 4.6 3,472,258 53.5 With a physical disability 18,790,000 13.2 8,718,560 46.4 With a mental disability 4,394,330 3.1 1,625,902 37.0 With a self-care disability 3,362,523 2.4 766,655 22.8 With a go-outside-home disability 4,931,252 3.5 1,001,044 20.3 With an employment disability 14,420,000 10.1 3,994,340 27.7 With no disability 115,900,000 81.3 95,501,600 82.4 Disability Category 2003 Panel Study of Income Dynamics Total Population Employed Number Percent Number Percent All 143,132,000 100.0 115,235,114 80.5 With any disability 30,566,000 21.4 19,103,750 62.5 With a sensory disability na na na na With a physical disability na na na na With a mental disability 13,897,000 9.7 8,574,449 61.7 With a self-care disability 9,395,000 6.6 4,237,145 45.1 With a go-outside-home disability 12,375,000 8.6 5,927,625 47.9 With an employment disability 19,304,000 13.5 10,018,776 51.9 With no disability 112,566,000 78.6 96,131,364 85.4 Disability Category Decennial Census 2000 Total Population Employed Number Percent Number Percent All 138,500,000 100.0 103,955,780 75.1 With any disability 14,006,000 10.1 5,854,508 41.8 With a sensory disability 3,346,000 2.4 1,743,266 52.1 With a physical disability 9,448,000 6.8 3,363,488 35.6 With a mental disability 5,218,000 3.8 1,575,836 30.2 With a self-care disability 2,628,000 1.9 570,276 21.7 With a go-outside-home disability na na na na With an employment disability na na na na With no disability 124,494,000 89.9 98,101,272 78.8 NOTE: The percentage of those with specific disabilities adds up to more than the percentage of those with any disability, because some people report having more than one disability. SOURCE: Burkhauser et al., 2006:Tables 11, 13.
OCR for page 34
Improving the Social Security Disability Decision Process viduals ages 25-61 with disabilities constitute between 12 and 21 percent of the total population in that age group. These estimates are due to different survey procedures, definitions of disability, and time periods. Generally, the more questions that a survey has about disability, the higher the estimates of population and prevalence (PSID is an exception) (Weathers, 2005:28). The percentage of individuals reporting limitations on employment due to a disability ranges from 7 percent in the 2003 ACS to 14 percent in the PSID. According to ACS, which has the strictest definition of work disability, 19 percent of those who reported a work limitation were employed at the time of the survey (9 percent reported full-time employment during the preceding year). Those reporting functional limitations, such as the inability to perform one or more ADLs or IADLs, worked less than those reporting impairments, whether mental, physical, or sensory, according to all the surveys. Between 18 and 20 percent of those with IADLs were employed at the time of the survey. For those with ADLs, the employment rate was between 18 and 23 percent. Among those reporting an impairment, those with a mental impairment were least likely to be employed (between 28 and 37 percent), the physically impaired somewhat more likely (between 34 and 46 percent), and the sensory impaired most likely (between 50 and 59 percent). Survey of Income and Program Participation According to the most recent report on disability based on SIPP, “In 2002, 18.1 percent of the 282.8 million people in the civilian noninstitutionalized population reported having a disability. Among the 51.2 million people with a disability, 32.5 million (11.5 percent of all people) had a severe disability.”1 Approximately 10.7 million people 6 years old and older (4.1 percent of this age group) needed assistance with one or more ADLs or IADLs (U.S. Census Bureau, 2006). The rates of nonsevere disability, severe disability, and needing personal assistance were higher for women than men overall, although the relation- 1 The Census Bureau report is based on responses to questions about the ability to perform various activities in the fifth wave of the SIPP (U.S. Census Bureau, 2006). The respondent was classified as having a nonsevere disability if he or she had difficulty performing one or more functional activities (seeing, hearing, speaking, lifting/carrying, using stairs, walking, or grasping small objects), one or more ADLs, or one or more IADLs, or if he or she had certain conditions, such as a learning disability or some other type of mental or emotional condition. They were classified as having a severe disability if they were unable to perform one of more of the functional activities, ADLs, or IADLs; used a wheelchair, cane, crutches, or walker; or had one or more specified conditions, such as mental retardation or Alzheimer’s disease; or had any other mental or emotional conditions that seriously interfered with everyday activities; or had a condition that limited the ability to work around the house or made it difficult to remain employed.
OCR for page 35
Improving the Social Security Disability Decision Process ship was reversed in age groups under 25 years. The disability rate also went up with age. The rate of severe disability was 3.4 percent for those under 15 years old, 12.6 percent for those ages 45-54, and 19.3 percent for those ages 55-64. The rate of disability also varied by race and ethnic group. The rate of severe disability was 7.2 percent among Asians and Pacific Islanders, 8.8 percent among Hispanics, 11.8 percent among non-Hispanic whites, and 14.0 percent among non-Hispanic blacks. The rates of need for personal assistance in these groups were 2.5, 2.9, 3.9, and 4.7 percent, respectively. Persons with disabilities were more likely to be in poverty. Among 25-64-year-olds, 11 percent of those with nonsevere disability and 26 percent of those with severe disability were poor, compared with 8 percent of those who reported not having any disability. About 77 percent of those ages 25-64 with severe disability had an annual income under $20,000, and 38 percent were in households with an annual income less than $20,000, compared with 39 and 12 percent, respectively, of nondisabled persons in that age group. Individuals 25 to 64 years old with severe disability were much more likely than nondisabled individuals to receive some form of public assistance (53 percent compared with 7 percent). This included Social Security and SSI. Approximately 29 percent of those in the severe disability group were receiving Social Security benefits and 21 percent were receiving SSI, compared with less than 3 percent and 0.3 percent of the nondisabled.2 Among individuals 21 to 64 years old, 88 percent of those without a disability were employed at some point during the preceding year, compared with 82 percent of those with nonsevere disability and 43 percent of those with severe disability. The median annual earnings of those with a severe disability were barely half those of people without a disability ($12,800 compared with $25,000). Among people in the labor force during the preceding year, 53 percent of the nondisabled worked full-time all year, compared with 13 percent of the severely disabled. About 12 percent of the nondisabled were unemployed all year, compared with 58 percent of the severely disabled. Employment and annual earnings varied by type of disability in terms of functional domains (communication, physical, and mental), limited ability to perform ADLs or IADLs (or both), and specific diagnoses. Of individuals 21-64 years old with a disability in one functional domain, 61 percent were employed for at least part of the preceding year, with median annual earnings of $18,300, compared with median annual earnings of $25,046 among those without disability. About 52 percent of those with disability in two 2 This age bracket includes those 62-64 years of age who have taken their Old-Age and Survivors Insurance benefits early as well as those receiving SSDI benefits and/or SSI payments.
OCR for page 36
Improving the Social Security Disability Decision Process domains were employed, with median annual earnings of $15,364. About 32 percent of those with disability in all three domains were employed, with median earnings of $8,520. About 29 percent with an ADL limitation requiring personal assistance were employed at some point during the year, with median annual earnings of $9,902. The percent employed varied according to which ADL was limited from 17 percent (difficulty eating— required personal assistance) to 41 percent (difficulty dressing—did not need personal assistance). About 31 percent with an IADL limitation requiring personal assistance were employed, with median earnings of $8,953. Like ADLs, the percentage employed varied somewhat according to which IADL was limited. About 47 percent of those with a mental disability were employed during the year, with median annual earnings of $13,771. Survey respondents who reported an activity limitation or being in fair or poor health were asked about the cause or causes. The largest group (consisting of 8.1 million individuals) were those with back or spine problems, of which 61 percent were employed during the preceding year, with $17,825 in median annual earnings. The next largest group (5.6 million) reported arthritis or rheumatism as a cause of activity limitation or of fair or poor health, of which 58 percent were employed, with median annual earnings of $18,000. The group with the lowest employment rate were those with mental or emotional problems. About 29 percent of the 1.7 million in this group were employed, with median annual earnings of $8,383.3 Researchers have used a data set of individuals who reported applying for SSDI and/or SSI in the 1990 SIPP and were matched with SSA administrative records to obtain detailed information on their interactions with the disability decision process and decisions at each step. The matched sample consists of adults who applied for SSDI or SSI disability benefits and whose claims were decided by DDSs during calendar years 1986 through 1993 at the initial or reconsideration level. Lahiri et al. (1995) and Hu et al. (1997) compiled the data set and used it to model the sequential decision process that the DDSs use to determine disability. They identified several health and activity limitation variables that increased the likelihood of meeting or equaling the Listings, which is decided at step 3 of the five-step decision process. These were having three or more severe limitations in their ability to perform ADLs, two or more IADL limitations, a mental disability, and no history of work. Individuals limited in working due to musculoskeletal impairments, however, were more likely to be moved on to the later steps of the decision process, where vocational factors are considered in conjunction with medical factors. The model predicted an overall allowance rate for the 3 Other groups reporting employment rates under 30 percent were those with paralysis and stroke, but their numbers were too small for a reliable estimate. The 90 percent confidence intervals were 14.4 and 10.4, respectively.
OCR for page 37
Improving the Social Security Disability Decision Process sample of applicants of 38.9 percent, which was 1.5 percentage points less than the actual allowance rate of 39.5 percent. Dwyer et al. (2001) used the model parameters from Hu et al. (1997), after making some adjustments in procedures involving the sample selection cutoff method and sample restrictions, to predict eligibility among the population of individuals not currently receiving benefits. They estimated that “4.4 million people, or 2.9 percent of the nonbeneficiary population aged 18-64, would meet SSA’s medical criteria for disability” (Dwyer et al., 2001:27). Of these, 3 million had average earnings less than the substantial gainful activity amount ($500 a month at that time). They predicted that, when a recession occurs, some of the 1.4 million who were medically eligible but had earnings larger than $500 a month would be likely to lose their jobs and apply for disability benefits. American Community Survey In the 2003-2005 ACS, respondents 5 years old or older were asked six disability questions. People ages 16-64 were classified as having a disability if they reported at least one of the following six conditions: Sensory Impairment. Question 15a asked respondents to mark yes or no as to whether they had any of the following long-lasting conditions: “Blindness, deafness, severe vision or hearing impairment.” Physical Disability. Question 15b asked if they had a long-lasting “condition that substantially limits one or more basic physical activities such as walking, climbing stairs, reaching, lifting, or carrying.” Mental Disability. Question 16a asked if they had a physical, mental, or emotional condition lasting six months or more that made “learning, remembering, or concentrating”difficult. Self-care Disability. Question 16b asked if they had a physical, mental, or emotional condition lasting six months or more that made “dressing, bathing, or getting around inside the home” difficult. Going-Outside-Home Disability. Question 17a asked those 16 years old and older if they had a physical or mental condition that made it difficult to “go outside the home alone to shop or visit a doctor’s office.” Employment Disability. Question 17b asked if they had a physical, mental, or emotional condition lasting six months or more that made “working at a job or business” difficult. According to the 2005 ACS, 12 percent of the population 16-64 years old reported a disability (Table 3-1). About 3 percent reported having a sensory disability, 7 percent a physical disability, 5 percent a mental disability, 2 percent a self-care disability, 3 percent a go-outside-home disability, and
OCR for page 38
Improving the Social Security Disability Decision Process 6.8 percent an employment disability, using the ACS definitions. Overall, women were slightly more likely to report disabilities than men (12.2 percent compared with 12.0 percent). About 38 percent of 16-64 year-olds who reported any disability were employed at the time of the survey, compared with 74 percent of the same age group without disabilities. The employment percentages ranged from about 17 percent among those with employment or go-outside-home disabilities to about 30 percent among those with a mental or physical disability to 47 percent with a sensory disability. Approximately 49 percent worked at some time during the year, compared with 85 percent of the nondisabled population. SOCIAL SECURITY DISABILITY APPLICANTS In calendar year (CY) 2005, slightly more than 2.1 million workers applied for SSDI benefits, compared with 1.3 million in 1995. The increase reflects in part an increase in the number of workers insured for disability and in part a higher incidence of claims (from 10.6 per 1,000 covered workers in 1995 to 14.6 per 1,000 covered workers in 2005).4 In CY 2005, almost 1.5 million adults ages 18-64 applied for SSI disability benefits, as did approximately 0.4 million children. In 1995, the numbers of SSI applications for adults and children were 1.3 million and 0.5 million, respectively.5 The total of 4 million applications for SSDI and SSI benefits in 2005 represented approximately 3 million individuals, because more than a quarter of the adults who applied filed for both SSDI and SSI benefits. The four populations of applicants—SSDI-only adults, concurrent SSDI/ SSI adults, SSI-only adults, and SSI children—differ in terms of disabilities claimed (Figure 3-1). In 2004, musculoskeletal disorders were the dominant diagnoses among adult applicants, especially those with a substantial work history (SSDI-only adults). Mental disorders were prominent among SSI children and second in frequency among adults without enough work history to qualify for Social Security benefits (i.e., SSI-only adults). SSA expects the aging of the baby-boom generation to increase the rate of applications for musculoskeletal disability disorders in the coming years (Zayatz, 2005). 4 Internet table: “Disabled Workers: Applications for Disability Benefits & Benefit Awards.” Available: www.ssa.gov/OACT/STATS/table6c7.html (accessed October 4, 2006). 5 Table V.C1, “Disabled Adult Claims,” and V.C2, “Disabled Child Claims,” in 2006 Annual Report of the SSI Program. Available: www.ssa.gov/OACT/SSIR/SSI06/AllowanceDate.html (accessed October 4, 2006).
OCR for page 39
Improving the Social Security Disability Decision Process FIGURE 3-1 Distribution of applications for Social Security disability benefits among program and selected diagnostic groups, 2004 (percentage of all applications). SOURCE: Appendix Table 3-1. Recent Social Security Disability Awardees In 2004, the state Disability Determination Services (DDSs) made more than 2.3 million initial decisions on claims, allowing benefits in 944,000 cases (41 percent) and denying 1,385,000 (59 percent). The allowance rate varied by program segment, however. It was highest for SSDI-only applicants, at 48 percent, and lowest for concurrent SSDI/SSI applicants, at 30 percent (Table 3-2). The mix of impairments among beneficiaries allowed benefits at the initial decision differed somewhat from the mix among applicants (Figure 3-2). Although 36 percent of SSDI-only applicants claimed a musculoskeletal problem, only 24 percent of those allowed initially had a musculoskeletal problem. Similarly, 30 percent of the concurrent SSDI/SSI claims were musculoskeletal, but only 16 percent of the allowances were. The percentage of SSI child claims based on mental problems was also lower among awards than applications—52 percent compared with 55 percent—and the percentage of awards based on respiratory disorders among children (primarily asthma) was also lower among awards than applications—2 percent compared with 9 percent. The largest percentage of SSDI/SSI claims
OCR for page 40
Improving the Social Security Disability Decision Process was musculoskeletal, but the largest percentage of awards was for mental disorders. Mental disorders other than retardation are the largest basis for both awards and claims in SSI child and SSI-only adult cases. TABLE 3-2 Initial Allowance Rate for Claims by Program Segment, 2004 Program Segment Number of Decisions Number of Allowances Percentage of Claims Allowed SSDI-only claims 741,471 356,696 48.1 Concurrent SSDI/SSI claims 725,011 219,198 30.2 SSI-only adult claims 466,182 187,603 40.2 SSI child claims 395,714 180,045 45.5 All SSDI and SSI claims 2,328,378 943,542 40.5 NOTE: SSDI applicants and beneficiaries are mostly workers but also include disabled adult children of workers and disabled widows and widowers of deceased workers. SOURCE: Table provided by SSA/ODISP/ODP/ODPIS, November 2005. FIGURE 3-2 Distribution of initial awards of Social Security disability benefits among program and selected diagnostic groups, 2004 (percentage of all awards). SOURCE: Appendix Table 3-2.
OCR for page 41
Improving the Social Security Disability Decision Process Impairment Trends Among Awardees SSA reports annual statistics on disabled worker awards by diagnostic group since 1961 (Figure 3-3).6 The series is not exactly comparable to the data reported above, because it combines SSDI-only and concurrent SSDI/ SSI awards and does not break out retardation from other mental disorders until 1995. The figure shows a substantial shift in the mix of impairments among SSA disabled worker beneficiaries over time, as the number of annual awards increased by 170 percent. The changing mix of diagnostic groups reflects in part changes in program eligibility and take-up rates, in part the aging of the insured worker population, and in part advances in medical knowledge and treatments, especially in reducing heart attacks and stroke. Cardiovascular claims have dropped from more than 30 percent of awards in the early 1970s to barely 10 percent currently. This reflected a decrease in incidence from approximately 2.0 allowances per 1,000 insured workers from 1973 to 1977 to less than 0.6 since 1997. The percentage of awards for mental disabilities, meanwhile, increased from about 10 percent in the 1960s and 1970s to about 25 percent in the 1990s (the incidence per 1,000 insured workers increased from 0.5 to between 1.0 and 1.3). The share of musculoskeletal claims increased from between 15 and 20 percent during the 1970s and 1980s to about 25 percent in the 1990s (the incidence per 1,000 workers was about 0.6 in 1981 and 1.4 in 2003). Among SSI adults, awards for mental disability have increased from 22 to 32 percent of all awards since 1985 (Figure 3-4). Musculoskeletal cases have experienced a similar relative increase but at a lower level: from 7 to 16 percent of all awards. The percentage of infectious disease cases increased from near 0 to more than 4 between 1985 and 1990, then declined after 1995, probably due to the advent and increased prevalence of HIV infection and AIDS and the subsequent development of effective therapies. The annual number of awards in SSI child cases increased by 285 percent from 1985 to 2003. Although the number of annual awards increased in all the diagnostic categories in Figure 3-5, the awards for mental disabilities other than retardation increased so much relative to the other categories that they accounted for more than half the awards in 2003, compared with less than 10 percent in 1985. SSA’s Office of the Actuary anticipates that the aging of the baby boomers will result in a growing proportion of awards based on arthritic, back, and bone disorders in the adult groups. The actuaries also expect to see a temporary increase in annual SSDI awards for mental disability continue for several years because of an effort beginning in 2001 to identify SSI ben- 6 Disabled widows and widowers and disabled adult children of workers are not included in this data series.
OCR for page 42
Improving the Social Security Disability Decision Process FIGURE 3-3 Distribution of annual disabled worker (DI) awards among selected diagnostic groups, 1961-2003 (percentage of all awards). NOTE: The table includes SSDI-only and SSDI/SSI concurrent awards. SSA reported the distribution of awards among diagnostic groups on the basis of initial awards until 1995, when it began to report the distribution among awards made at all stages of the decision process, including the various appeals levels. Because the rate of successful appeals is relatively high for musculoskeletal cases, the percentage of awards based on musculoskeletal disability jumped and depressed the percentage of awards for mental disabilities. SOURCE: Appendix Table 3-3. eficiaries whose SSDI eligibility has not been recognized (Zayatz, 2005). In the latest annual report on the SSI program, SSA predicts that the number of SSI beneficiaries will continue to grow, largely due to growth in the U.S. population, but this projection is based on historical disability incidence rates by age and sex, not on any analysis of how factors that affected those rates might be changing (SSA, 2006). Denied Applicants Little is known about applicants who are not allowed, although it would be useful in assessing the Listings to know about the rate of false negatives (i.e., applicants who actually meet the requirements but are turned down). Information about subsequent employment and earnings would also be helpful in assessing the validity of the Listings. It might become evident
OCR for page 43
Improving the Social Security Disability Decision Process that individuals with a certain type of condition and examination results are rarely, if ever, employed after being denied, which could indicate that the criteria in the Listings are not measuring what they are supposed to measure or the wrong tests are being used. Denied applicants have been studied in the past, but recent studies are few. In 1976, Treitel used SSA administrative records to determine the subsequent fate of men denied benefits on medical-vocational grounds. By 1973, 14 percent had died (compared with 7 percent of the same-age general population); another 16 percent had reached retirement age, and 40 percent had not worked in the four years after denial. Less than half (47 percent) of those alive and of working age in 1972 had worked at some point during the year (based on Bound, 1989:490). Treitel then studied applicants denied at the hearing level. He found that: Persons persisting in claims rarely returned to substantial employment. While many of those denied at hearings in 1970 had some form of earnings in the next 5 years, only about 10 percent had substantial earnings of FIGURE 3-4 Distribution of annual SSI adult awards among selected diagnostic groups, 1985-2003 (percentage of all awards). SOURCE: Appendix Table 3-4.
OCR for page 44
Improving the Social Security Disability Decision Process FIGURE 3-5 Distribution of annual SSI child awards among selected diagnostic groups, 1985–2003 (percentage of all awards). SOURCE: Appendix Table 3-5. $3,600 or more 5 years later. Most claimants for benefits appear to remain out of the labor market whatever the decision on their disability insurance claim (Trietel, 1979). Bound (1989) used data on denied applicants from the 1972/1974 Survey of Disabled and Nondisabled Adults and the 1978 Survey of Disability and Work matched with Social Security earnings records to compare applicants awarded benefits with those denied benefits. Bound found that 33 percent of the denied applicants were employed at the time of the 1972 survey and 29 percent at the time of the 1978 survey, compared with 92 and 87 percent, respectively, of the same-age general population. More did some work during the past year, 45 percent in 1972 and 40 percent in 1978, but their earnings were less than half of the same-age population overall— $4,000 compared with $9,000 in 1972 and $5,300 compared with $14,000 in 1978. About 51 percent of the denied applicants in 1972 reported having severe work limitations, compared with 93 percent of the beneficiaries and
OCR for page 45
Improving the Social Security Disability Decision Process 12 percent of the general population. The comparable percentages in 1978 were 64, 97, and 14 percent. The percentages with specific health conditions were much higher than those reported by the general population and nearly as high as those who were awarded benefits. Bound also cited a study of Nagi (1969) in which independent multidisciplinary clinical teams consisting of a physician, psychologist, occupational therapist, social worker, and vocational counselor evaluated the medical status and work capacity of a sample of SSDI applicants in person. The teams found 68 percent of those later allowed benefits and 36 percent of those denied benefits to be “not fit for work.” The teams determined that another 25 percent of those allowed were only fit for work under “special conditions,” “sheltered conditions,” “at home only,” or “part-time under normal conditions.” Nearly a third of the denied applicants (31 percent) were similarly categorized, and another third (34 percent) were deemed fit for work under normal conditions (about half of them at their former job, the rest at other jobs). More recently, Bound et al. (2003) used the same matched SIPP data set as Hu et al. (1997) and Dwyer et al. (2001) to track the change in earnings and income before and after the disability decision of both those allowed and those denied. The average monthly earnings of SSDI applicants were about 15 percent of what they were three years earlier. Three years later, applicants who were denied benefits earned about 35 percent of what they did six years before (the authors did not report how many were employed full-time, part-time, or not at all). The pattern was similar for those denied SSI benefits. REFERENCES Bound, J. 1989. The health and earnings of rejected disability insurance applicants. American Economic Review, 79(3):482-503. Bound, J., R.V. Burkhauser, and A. Nichols. 2003. Tracking the household income of SSDI and SSI applicants. Research in Labor Economics, 21:113-158. Burkhauser, R.V., R.R. Weathers II, and M. Schroeder. 2006. A guide to disability statistics from the Panel Study of Income Dynamics. Rehabilitation Research and Training Center on Disability Demographics and Statistics, Cornell University. Available: digitalcommons. ilr.cornell.edu/edicollect/1207/ (accessed October 4, 2006). Dwyer, D., J. Hu, D.R. Vaughn, and B. Wixon. 2001. Counting the disabled: Using survey self-reports to estimate medical eligibility for Social Security’s disability programs. ORES Working Paper #90. Office of Research, Evaluation, and Statistics, SSA. Available: www.ssa.gov/policy/docs/workingpapers/wp90.pdf (accessed October 4, 2006). Hu, J., K. Lahiri, D.R. Vaughn, and B. Wixon. 1997. A structural model of Social Security’s disability determination process. Working Paper No. 72. Office of Research, Evaluation, and Statistics, SSA. IOM and NRC (Institute of Medicine and National Research Council). 2002. The dynamics of disability: Measuring and monitoring disability for Social Security programs, edited by G.S. Wunderlich, D.P. Rice, and N.L. Amado. Washington, DC: National Academy Press.
OCR for page 46
Improving the Social Security Disability Decision Process Lahiri, K., D.R. Vaughan, and B. Wixon. 1995. Modeling SSA’s sequential disability determination process using matched SIPP data. Social Security Bulletin, 58(4):3-42. Nagi, S.Z. 1969. Disability and rehabilitation: Legal, clinical, and self-concepts and measurements. Columbus, OH: Ohio State University Press. SSA (Social Security Administration). 2002. FY 2002 performance and accountability report. Available: www.ssa.gov/finance/2002/fy02PAR.pdf (accessed October 18, 2006). SSA. 2006. 2006 Annual report of the SSI program. Available: www.ssa.gov/OACT/SSIR/SSI06/index.html (accessed October 5, 2006). Treitel, R. 1979. Disability claimants who contest denials and win reversals through hearings. ORS Working Paper No. 2. Baltimore, MD: Office of Research and Statistics, SSA. U.S. Census Bureau. 2006. Americans with disabilities: 2002. Current Population Report P70-107. Available: www.census.gov/prod/2006pubs/p70-107.pdf (accessed October 5, 2006). Weathers II, R.R. 2005. A guide to disability statistics from the American Community Survey. Rehabilitation Research and Training Center on Disability Demographics and Statistics, Cornell University. Available: digitalcomons.ilr.cornell.edu/cgi/viewcontent.cgi?article=1123&context=edicollect (accessed October 4, 2006). Zayatz, T.A. 2005. Social Security Disability Insurance Program worker experience. Actuarial Study No. 114, Office of the Actuary, SSA. Available: www.ssa.gov/OACT/NOTES/s2000s. html (accessed October 5, 2006).
Representative terms from entire chapter: