The feigning of disabling illness for the purpose of disability compensation, or “malingering,” is common in Social Security Disability examinations, occurring in 45.8%–59.7% of adult cases. In this study, we estimated the costs of malingering based on mental disorder data published by the Social Security Administration. At the most widely accepted base rate of malingering in medicolegal cases involving external incentive, costs were high, totaling $20.02 billion in 2011 for adult mental disorder claimants. Moreover, these figures clearly underestimate the costs of the larger problem with feigned disability in both adults and children. We urge a change in Social Security policies to allow the use of validity testing in the examination for disability claims.
We neither know how to receive benefits, nor how to bestow them.
Seneca, Lucius Annaeus (1887). Seneca on Benefits.
The Social Security Administration (SSA) provides disability payments for individuals who can prove that they are unable to work in “any substantial gainful activity” (SSA, 2012b). Those alleging mental disabilities may present with a variety of psychological and neuropsychological disorders that include mental retardation, mood disorders, organic mental disorders, and psychotic disorders. Although SSA has been recently limiting psychological testing, the Psychological Consultative Examination (PCE) for these disorders has traditionally included a thorough interview, mental status examination, and psychological testing such as the Wechsler Adult Intelligence Scale-IV or the Minnesota Multiphasic Personality Inventory-2 (MMPI-2).
Authentication of the accuracy of the claimant's report of symptoms by tests of Symptom Validity (SVTs) and determination of the accuracy of the levels of measured cognitive ability by tests of Performance Validity (PVTs) are necessary to determine if the findings truly represent the claimants' self-reported problems (Larrabee, 2012). In Social Security Disability (SSD) claims, where there are external incentives in the form of disability benefits, the use of validity methods is critical (Chafetz, 2010), as the claimant may feign symptoms or intentionally under-perform on tests of physical or cognitive abilities in order to obtain benefits.
Feigning or exaggeration of problems for an external incentive is more generally known as malingering. Malingering is a common finding in civil litigation where the incentive involves damages in personal injury lawsuits, in disability cases where the incentive involves reliable compensation, and in criminal settings where the defendant might be motivated to avoid punishment. Malingering is not a new concept, as Stevens (1986) cited reports of rabbis in the second century B.C. noticing people taking unwarranted advantage of relief facilities.
Having a strong external incentive figures prominently as the first criterion in both the original diagnostic guidelines of Slick, Sherman, and Iverson (1999), and the updated revisions published by Slick and Sherman (2013). In these guidelines, evidence for invalid test performance and/or symptom report, or compelling inconsistencies, in the absence of alternative explanations, provides sufficient evidence for a determination of malingering. This determination is evaluated using probabilistic criteria, whereby definite malingering is defined by significantly worse-than-chance performance on forced-choice test procedures, and probable malingering is defined by failure of two or more PVTs, or one PVT and one SVT.
High rates of invalid performance considered as malingering have been found by numerous investigators. Larrabee (2003) summed over 11 studies of base rates of malingering, the majority claiming mild Traumatic Brain Injury (mTBI), showing an overall frequency of 548/1,363 subjects (40%), identified with motivated performance deficits suggestive of malingering. Ardolf, Denney, and Houston (2007) used the formal criteria proposed by Slick and colleagues (1999) to classify 32.4% of criminal evaluees as showing probable malingering, and 21.9% as showing definite malingering, for a combined probable/definite rate of 54.3%. Larrabee, Millis, and Meyers (2009), appreciating the consistency in base-rate estimates of malingering over numerous studies, proposed a “magical number” of 40 ± 10 to represent the average base rate (or prevalence) of malingering in subjects with an external incentive.
Converging lines of evidence have suggested that malingering is also a significant factor in the PCE for SSD. In a base rate study by Chafetz (2008), cumulatively 45.8% of disability claimants failed the Test of Memory Malingering (TOMM; Tombaugh, 1996) at below-chance or chance levels or failed both the TOMM and the Symptom Validity Scale (SVS) for Low Functioning Individuals (Chafetz, Abrahams, & Kohlmaier, 2007). In a separate sample, cumulatively 59.7% failed the Medical SVT (MSVT; Green, 2004) at below-chance or chance levels or failed both the MSVT and the SVS. Moreover, other investigators studying SSD cases have found high rates (50+%) of single PVT failure (Miller, Boyd, Cohn, Wilson, and McFarland, 2006; J. King, personal communication) on the PCE.
The goal of the present study is to estimate the cost of malingering in SSD evaluations for adults claiming psychological and mental disability, using the published statistical records from SSA for 2011 (SSA, 2008). In this paper, estimates of cost will be made for adults only, as the widespread studies of malingering have been done mostly with adults. Also, conditions involving chronic pain will not be considered, as there are no listings specifically for pain in the determinations of disability. In estimating costs, we will use base rates in 10% increments, from 10% to 90%, to cover the range of measured and hypothetical base rates in the Mittenberg, Patton, Canyock, and Condit (2002) investigation. First, we identify the relevant SSA programs.
Characteristics of the Programs
Old-Age, Survivors, and Disability Insurance
The Old-Age, Survivors, and Disability Insurance (OASDI) programs ensure a basic level of monthly income upon retirement eligibility, death, or disability by insured workers (www.ssa.gov). Two separate programs of OASDI pay benefits to workers and their families: (a) Old-age and Survivors Insurance for retired or deceased workers (and their families); and (b) Disability Insurance (SSDI) for disabled workers and their families. The SSDI Trust fund pays benefit to disabled workers who satisfy the legal requirements, are unable to engage in substantial gainful activity due to a medically determinable physical or mental impairment severe enough to meet the Listing requirements of the program, and have not yet attained retirement age (Board of Trustees, Federal OASDI Trust Funds, 2012).
According to the 2012 “Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds” (Board of Trustees, Federal OASDI Trust Funds, 2012), the OASDI programs were providing benefits to 55 million people in 2011 (54 million in 2010). Of these 55 million, 38 million were retired workers and dependents of retired workers, 6 million were survivors of deceased workers, and “11 million were disabled workers and their dependents.” Also, in 2011, 158 million people had earnings covered by Social Security and paid payroll taxes.
Total OASDI expenditures in 2011 amounted to $736 billion. Total Trust Fund income in 2011 amounted to $805 billion, out of which $691 billion were in non-interest income, with the difference of $114 billion in interest earnings. In 2011, total OASDI assets held in U.S. Treasury securities amounted to $2.7 trillion. It was noted that the assets of the SSDI Trust Fund (under “intermediate assumptions”) were projected to decline steadily and would fall below 100% of the annual cost by the beginning of 2013, continuing to decline until the Trust Fund was exhausted in 2016.
Focusing on the SSDI Trust Fund in 2011, total receipts were $106.28 billion, and total disbursements were $132.33 billion. Of these disbursements, benefit payments amounted to $128.95 billion, and total administrative expenses were $2.92 billion, which amounted to 2.2% of the total disbursements spent on administration of the program.
Concerning the distribution to disabled workers and their families in 2011, $119.563 billion went to disabled workers (92.7%), $0.608 billion went to spouses (0.5%), and $8.765 billion went to children of disabled workers (6.8%).
Supplemental Security Income
The Supplemental Security Income (SSI) program is a national federal assistance program administered by SSA that guarantees a minimum level of income for needy aged, blind, or disabled individuals (SSA, 2012a, b). This program is primarily administered at the federal level and supplemented at the state level. In December 2011, 8.1 million individuals received federally administered SSI benefits (including federally administered state supplementary payments). These payments averaged $502 monthly. The SSI program provides a uniform federal income floor, with optional state programs that supplement the income.
In 2011, the SSI made a total payments of $49.520 billion. Of this total, Federal SSI accounted for $45.999 billion and State supplementation (federally administered) accounted for $3.521 billion. Payments to Disabled individuals by the SSI amounted to $43.667 billion (88.2% of total), of which Federal SSI accounted for $41.103 billion (89.4% of total) and State supplementation paying $2.564 billion (72.8% of total). As SSI is funded by general tax revenues and not Social Security taxes, there is no balance sheet dealing with income and outflow per se.
In 2011, SSI payments went to 8,112,773 recipients, and of these 6,861,634 (84.6%) were considered Disabled, the other categories being Aged and Blind.
SSA publishes quarterly statistics on their activities and the individuals served. The information SSA collects includes the number of individuals served in the federal program, the diagnostic category of those served, the average monthly pay per diagnostic category, gender, geographical area, and classification as a worker widow or adult child. Disabled workers are classified according to SSA's unique taxonomy, which is outlined in the “Disability Evaluation Under Social Security” (i.e., The Blue Book; US Dept. of Health and Human Services, 1994, revised 2008). This taxonomy provides the listings for the various disabilities, with guidelines for how each disability is determined (http://www.ssa.gov/disability/professionals/bluebook/). We used the SSA statistics compiled as of December 2011 to identify all cases of mental disability that were being provided for the SSDI program (SSA, 2012a) and the SSI program (SSA, 2012b). It is important to keep in mind that these tables provide only a static representation of a dynamic system as of December 2011, with beneficiaries entering and leaving the system on a daily basis.
Listing Inclusions and Exclusions
How is impairment determined in this taxonomy? The SSA publishes a guideline named Listing of Impairments (http://www.socialsecurity.gov/disability/professionals/bluebook/listing-impairments.htm). This listing describes impairments of each major bodily system that are considered severe enough to prevent an individual from doing ANY gainful activity. The Mental Disorders classification includes Intellectual Disability, Mood Disorders, Organic Mood Disorders, Psychotic disorders, and Other (mental) disorders. Pain disorders are categorized within the musculoskeletal, inflammatory, and immune system listings and are not a separate category in the Mental Listings, precluding the use of these data.
There were 8,575,544 disabled workers (removing dependent beneficiaries) in the SSDI system as of December 2011. Of these beneficiaries, there were 369,093 Intellectual Disability cases, 1,304,851 with Mood Disorders, 302,036 Organic Mental Disorders, 435,929 Psychotic Disorders, and 333,063 “Other” cases, which according to the Blue Book includes mTBI and eating disorders. There were also 11,269 disabled workers with Autistic Disorders, 6,750 with Developmental Disorders, and 5,937 Childhood-Adolescent Disorders Not Elsewhere Classified (including ADHD). The sum is 2,768,928 claimants for mental disorders, which is 32.3% of the total cases. Considering only these mental disorders, and multiplying the number of beneficiaries in each category by the monthly benefit for each disorder (SSA, 2012a, pp. 25–27), then totaling these monthly costs and dividing by the total number of claimants, gives an average monthly benefit for all mental disorders of $965.11. (By contrast, the largest Listing category is: Musculoskeletal System & Connective Tissue Disorders, with 2,488,374 beneficiaries receiving an average monthly benefit of $1,186.96, totaling $2.95 billion/month and $35.44 billion/year.) We note that this average monthly amount for Mental Listings is lower than the average for all Listings ($1,110.50). The total monthly cost for these mental Listing beneficiaries is $2.67 billion, and the total yearly cost is $32.07 billion. Using these figures, the total number of malingering individuals and the costs are shown in Table 1.
|Level (%)||# Disabled Workers 2,768,928||2011 Total Cost $32,067,993,684|
|Level (%)||# Disabled Workers 2,768,928||2011 Total Cost $32,067,993,684|
Notes: The 40% rate is bolded as the probable rate of malingering given in Larrabee, Millis, and Meyers (2009). For the SSDI Total, the number of disabled workers is used, removing spouse and child beneficiaries. Costs were estimated by multiplying the average disability figure for each mental condition by the December 2011 number of individuals with that condition, summing over all conditions, and then multiplying by 12 for the yearly estimated amount.
SSI Adult Data
As of December 2011, there were 6,054,132 disabled beneficiaries in the SSI system, 4,777,010 of whom were adults (ages 18–64). Of these, 67,018 were cases of Autistic Disorder, 31,816 cases of Developmental Disorders, 46,921 Childhood-Adolescent Disorders Not Elsewhere Classified, 944,778 cases of Intellectual Disability, 803,079 with Mood Disorders, 193,943 Organic Mental Disorders, 436,997 Psychotic Disorders, and 273,191 “Other” cases. The total number of SSI adults with mental disorder listings was 2,797,743, which is 58.6% of the SSI adult cases. The mean monthly benefit for the mental disorders in adults was $535.77. The total monthly cost for these beneficiaries was $1.5 billion, and the total yearly cost was $17.99 billion. Table 2 shows the estimated number of malingering individuals and the costs.
|Level (%)||# Adults < age 65 2,797,743||2011 Total Cost $17,987,337,450|
|Level (%)||# Adults < age 65 2,797,743||2011 Total Cost $17,987,337,450|
Notes: The 40% rate is bolded as the probable rate of malingering given in Larrabee, Millis, and Meyers (2009). The SSI figures include the number of adults (less than age 65) minus the children as of December 2011. Costs were estimated by multiplying the average disability figure for each mental condition by the December 2011 number of individuals with that condition, summing over all conditions, and then multiplying by 12 for the yearly estimated amount.
These data show that the 2011 estimated costs of malingered mental disorders in the two main SSD programs are noteworthy. Using the widely-reported malingering base rate of 40% (Larrabee et al., 2009), and combining the adult data from the SSDI and SSI programs, the 2011 combined estimated cost of malingered mental disorders was $20.02 billion. It is also interesting to note that in the SSDI program the mental disorder cases comprise 32.3% of the cases, but in the SSI system they comprise 58.6% of the adult cases (child cases not analyzed here).
Prior reports of the costs of malingered disability have shown higher annual figures (∼$180 billion) over all programs (Chafetz, 2011). Noting the smaller figures from analyzing only malingered mental disorders in the present study, it becomes instructive to look at reasons for the differences. The inability to separately classify payments by pain disability is important, as musculoskeletal injuries are a major source of SSD claims, with 2,488,374 beneficiaries receiving a total annual benefit in 2011 of $35.44 billion. These claims are frequently based upon what claimants say they can and cannot do, as well as their performance on functional capacity examinations, symptom report (including pain rating scales), and physical examination, all of which can be feigned. It is not surprising that PVT and SVT failure rates are high in individuals claiming impairment due to chronic pain (Gervais, Rohling, Green, & Ford, 2004; Greve, Etherton, Ord, Bianchini, & Curtis, 2009; Meyers, Millis, and Volkert, 2002) or that 42% failure of a PVT is found in physicians' offices for private disability claims involving soft tissue or orthopedic injuries. If we calculate malingering at a 40% rate for the 2011 musculoskeletal category of SSDI, we see that another $14.18 billion would be added to the estimate.
Malingering-by-proxy occurs when a vulnerable individual acts under the guidance, direction, or control of others and typically occurs when a parent or guardian causes a child to produce false symptoms or underperform on a test (Slick & Sherman, 2013). Although this problem has been studied in the SSD framework (Chafetz, 2008; Chafetz & Prentkowski, 2011) in which approximately 50% of children fail one or two PVTs or score significantly below chance on a PVT, the problem of secondary gain in children in other medicolegal settings has not been widely studied. Considering that the reporting of symptoms and problems in children comes predominantly from the guardians, we are likely dealing with rates of symptom/problem inflation in the range of adult malingering. If we take the total mental disorder costs for children in SSI in 2011 as $6.03 billion, then at the 40% rate we estimate that approximately $2.41 billion were spent on feigned mental disability in children.
When we analyze the costs of malingering for the entire program (either SSI or SSDI), we are in no way suggesting that an individual might be malingering cancer, liver disease, or a kidney disorder. We are instead considering that a claim for disability is ultimately about the behavioral inability to work (adults) or about functional limitations (children). The SSI/SSDI definition of disability (SSA, 2012b, p. 2) is instructive: “To be considered disabled, an individual must have a medically determinable physical or mental impairment that is expected to last (or has lasted) at least 12 continuous months or to result in death and (1) if 18 or older, prevents him or her from doing any substantial gainful activity, or (2) if under 18, results in marked and severe functional limitations.”
Thus, an adult with a medically determined illness or condition must still prove that he/she cannot do ANY substantial gainful activity. This is not an “own occupation” disability program. If the adult, who has a severe yet treatable kidney disease, cannot climb utility poles to perform electrical work, he/she must still prove that he/she cannot perform any other occupation for which he/she is qualified including work that is not physically demanding (e.g., clerical work). If this individual were unhappy with the supervisor, or disliked co-workers, there could well be another motivation for secondary gain. Given that a high percentage of claimants in physicians' offices fail PVTs (Richman et al., 2006) and that secondary gain validity issues result in approximately 40% PVT and SVT failure in many different venues of medicolegal cases (Larrabee et al., 2009), it is likely that a similarly high proportion of these “purely medical” SSD cases are feigning disability from the medical-based impairment.
It must be considered that the base rates of malingering used in this study come from comprehensive reviews of medicolegal work (Larrabee et al., 2009) and from studies of SSD claimants (Chafetz, 2008). There is no way of knowing the percentage of claimants who are successful in fraudulently obtaining benefits through malingered disability claims. As analyzed in Chafetz (2011), the suppression of the use of validity methods by SSA entails a bias that leads to a higher false-positive rate for assessments of disabling impairment. Claimants who can successfully feign disabling problems are not likely to get caught with this bias in place.
It is also instructive to think of secondary costs. Many beneficiaries also receive Medicare (SSDI) or Medicaid (SSI/SSDI) benefits. One might also raise the question as to the expenditures on spouses and children of beneficiaries who had malingered the examination (which were left out of the current calculations). Moreover, the calculations for a single year provide only a cross-section of the problem, whereas beneficiaries may receive benefits for decades. This long-view shows how the problem of malingering may compound costs over a long period of time, and one might also consider costs of the loss of labor for the same period. Other costs not calculated here would include lump sum payments for back pay that are provided when a beneficiary enters the system. Total disbursement figures include these monies, and thus, one must consider the larger costs of malingering include these payments to individuals who had feigned disabilities.
Rewarding a claim for a disability that does not truly exist may create a debilitating system for an individual who otherwise has the ability to work or go to school. For example, Cassar, Hales, Longhurst, and Weiss (1996) presented the case study of a 13-year-old eighth grader whose mother disagreed with the psychiatric diagnosis of uncomplicated ADHD, insisting that the child had a bipolar disorder. She discontinued treatment with the first provider and had her child admitted. During hospitalization and under the therapeutic milieu, the child was weaned off of methylphenidate without emergence of any symptoms of any psychiatric diagnosis. The mother was very unhappy about this development, as she had applied for disability benefits on the grounds of “chronic mental illness.”
Behavioral evidence also indicates that rewarding diminished behavioral activation may result in depression (e.g., Kimbrel, Nelson-Gray, & Mitchell, 2007), at least in part by removing the opportunity for positive life events (Hundt, Nelson-Gray, Kimbrel, Mitchell, & Kwapil, 2007). Thus, failure of the DDS to perform due diligence and screen individuals for malingering may eventually result in true disability in individuals who were not previously disabled.
According to the Board of Trustees, Federal OASDI Trust Funds (2012), in 2011 Social Security's cost continued to exceed both the program's tax income and its non-interest income; the deficit of tax income to cost was $148 billion. News accounts suggest that economic factors are an influence in terms of how many individuals even apply for disability (http://online.wsj.com/article/SB10001424052970204296804577121392750460030.html). Although many factors are involved (e.g., a temporary reduction in the Social Security payroll tax for 2011 and 2012), our findings are suggestive that a significant financial payout is going to individuals who are not truly disabled. In a letter to SSA Commissioner Astrue by Dr. Katherine Nordal (2010), Executive Director of the APA Practice Directorate, citing both the National Academy of Neuropsychology and the American Academy of Clinical Neuropsychology recommendations, the use of validity measures was strongly recommended to improve the accuracy of disability determinations. More recently, a letter from U.S. Senator Tom Coburn (2013) to Commissioner Astrue, supported by all of neuropsychology's national organizations, strongly urged the funding of validity testing in SSD examinations. We further suggest that the mandated use of validity measures will help improve accuracy and efficiency of disability determinations and be less wasteful of the program's resources.
Conflict of Interest
The authors wish to thank Dr. Glenn Larrabee for his many insights and his careful reading of this manuscript.