House Ways and Means Subcommittee on Social Security Hearing

Chairman Johnson, Ranking Member Becerra, and Members of the Subcommittee:

I am pleased to be here today to discuss our preliminary observations on efforts by the Social Security Administration (SSA) to modernize its disability criteria. In the last 20 years, SSA disability benefit payments to beneficiaries and their dependents have increased almost fivefold, from $35 billion in 1990 to over $164 billion in 2010. The number of working-age beneficiaries also grew from about 9 million in 2000 to almost 12 million in 2010, and, as more baby boomers enter their disability-prone years, SSA expects the number of claims to grow. n1 Given the extensive size and cost of its disability programs, SSA must have current and appropriate criteria by which to assess whether an applicant's medical conditions affect his or her ability to perform work in the national economy. GAO has designated SSA's federal disability program as a high-risk area; in 2008, we reported that the agency's medical criteria did not fully reflect medical and technological advances, and the occupational criteria it used from the Department of Labor (Labor) had not been updated since 1991, despite the trend away from manual labor jobs and towards service- and knowledge-based jobs.

My statement today focuses on initial observations from our ongoing review for this subcommittee and examines the status and management of SSA's efforts to (1) update its medical listings to reflect current medical knowledge and (2) develop its own occupational information system (OIS) to reflect labor market changes. To examine these issues we reviewed prior GAO and SSA Office of Inspector General reports; relevant federal laws and regulations; program documentation including policies, procedures, strategic goals, and supporting project plans and cost estimates; and position papers and testimonies from disability groups and commissions. We interviewed SSA officials, key project contractors and stakeholders, disability experts, and representatives of other agencies that administer disability programs. In our ongoing work, we will further evaluate SSA's plan to develop its own OIS against best practices for cost estimating and scheduling. We will also review more broadly other steps SSA is taking to incorporate modern concepts of disability into its programs. We plan to issue our final report later in 2012 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Background

SSA administers two of the largest disability programs: the Disability Insurance (DI) program, enacted in 1956, and the Supplemental Security Income (SSI) program, enacted in 1972. n2 In order to be eligible for DI or SSI benefits based on a disability, an individual must meet the definition of disability for these programs--that is, they must have a medically determinable physical or mental impairment that (1) prevents the individual from engaging in any substantial gainful activity, and (2) has lasted or is expected to last at least one year or result in death. n3

To determine eligibility, SSA uses a five-step sequential process that is intended, in part, to expedite disability decisions when possible and limit administrative costs by conducting less intensive assessments at earlier steps (see fig. 1). n4 At steps 1 and 2 of the process, SSA determines whether an applicant is working and meets income thresholds as well as the medical severity of impairments. If so, the applicant moves to step 3 of the process. At this step, SSA examiners assess the applicant's medical impairment(s) against the Listings of Impairments, also known as the medical listings, which are organized into 14 major body systems for adults and reflect medical conditions that have been determined by the agency to be severe enough to qualify an applicant for benefits. n5 If the individual's impairment meets or is equal in severity to one or more of those in the listings, the individual is determined to have a disability. If not, SSA performs an assessment of the individual's physical and mental residual functional capacity.those in the listings, the individual is determined to have a disability. If not, SSA performs an assessment of the individual's physical and mental residual functional capacity.those in the listings, the individual is determined to have a disability. If not, SSA performs an assessment of the individual's physical and mental residual functional capacity. n6 Based on this assessment, SSA determines whether the individual is able to perform past relevant work (step 4) or any work that is performed in the national economy (step 5). To inform determinations at steps 4 and 5, SSA uses a Department of Labor database--known as the Dictionary of Occupational Titles (DOT)--for an inventory of occupations performed in the national economy.

Since 2003, SSA's and other federal disability programs have remained on our high-risk list, in part, because their programs emphasize medical conditions in assessing work capacity without adequate consideration of work opportunities afforded by advances in medicine, technology, and job demands. Since the 1990s, we, along with SSA's Office of Inspector General and the Social Security Advisory Board, have expressed concerns that the medical listings being used no longer provide sufficient criteria to evaluate disability applicants' inability to work and that SSA was simply extending the listings instead of periodically updating them. In 2008, we reported that SSA had established a new process for revising the listings--referred to by SSA as the "business process"--to better incorporate feedback into its continuous updates. n7 This process, which has been in effect since 2003, includes incorporating feedback from multiple parties, including medical experts and claims examiners, to update their medical criteria. SSA should also gather external feedback through comments associated with regulatory actions, such as the publication of advanced notices of proposed rulemaking (advanced notices) and notices of proposed rulemaking (notices) in the Federal Register. n8 In addition, one year after a revision is made, SSA should conduct a study reviewing the changes. According to SSA documentation, this internal case study, now referred to as the postimplementation study, should involve surveying the field regarding the results of the regulation and areas to improve, as well as reviewing the data to determine whether expectations from the revision have been proven.

Figure 1: SSA's Five-Step Sequential Evaluation Process for Determining Disability

With respect to information on jobs in the national economy that supports SSA's occupational criteria, we and others have reported that the DOT, which SSA still relies on to assess eligibility at steps 4 and 5 of the process, is outdated. The DOT has not been updated since 1991, and Labor has since replaced the DOT with a new database called the Occupational Information Network (O*NET). n9 However, SSA has determined that O*NET is not sufficiently detailed for evaluating DI and SSI disability claims and therefore has begun developing its own OIS in order to better reflect the physical and mental demands of work in the national economy.

SSA Has Improved Its Process for Updating Medical Listings, but Faces Challenges Completing Updates

SSA Has Improved Its Strategy to Update Medical Listings

Since our last review in 2008, SSA has made several changes that hold promise for improving medical listings updates. First, the agency is using a two-tiered system for ongoing revisions to the listings. Under this system, SSA first completes a comprehensive listings update for a body system that reviews all the diseases and disorders listed within that system and makes revisions it determines are needed. For subsequent updates of listings for a body system that underwent a comprehensive revision, SSA will pursue a more targeted approach--that is, SSA will conduct ongoing reviews with the expectation of making targeted revisions for a small number of medical diseases or disorders that need to be updated. Agency officials told us that targeted updates should be completed more quickly than comprehensive updates, allowing them to focus on the most critical changes needed. However, officials also noted that these ongoing reviews could result in major or even no changes, as appropriate. As of early March 2012, SSA had begun the ongoing review process to consider opportunities for targeted revisions for 8 out of 14 adult body systems that were recently comprehensively revised. Also as of early March 2012, the agency had not yet completed comprehensive revisions for the six remaining systems, which the agency expects to do before they conduct subsequent reviews under the targeted approach.

Another change, according to agency officials, is that in 2010 the SSA Commissioner set a 5-year cycle time for updating listings for each body system. n10 Previously, SSA set expiration dates for periodically updating listings according to each body system, ranging from 3 to 8 years, but frequently extended them. SSA officials believe that conducting targeted reviews will generally allow the agency to conclude any necessary revisions prior to the 5-year expiration period. Additionally, they expect that using the "business process," which requires early public notification of changes and obtaining necessary data and feedback from internal and external parties, should help keep continuous reviews on track. See figure 2 for the status and expiration dates of listings for the 14 adult body systems, undergoing review for either comprehensive or possible targeted revisions, as of early March 2012.

Figure 2: Status of Revisions for SSA's Adult Body System Listings

a For the two-tiered revision process, the beginning point is the date of the last comprehensive or targeted revision.

b SSA officials told us they have grouped the disorders in the special senses and speech listings into two broad sections--vision and hearing--and two standalone medical conditions. SSA has not yet comprehensively revised the two standalone medical conditions that cover speech and disturbances of labyrinthine-vestibular function. The two sections have undergone comprehensive revisions and are undergoing reviews that began in 2007 for the vision section and in 2010 for the hearing section. Special senses and speech listings are set to expire in 2015.

c SSA published limited revisions for the hematological disorders listings in 1988, the mental disorders listings in 2000, and the musculoskeletal system listings in 2002.

SSA has made another change by more extensively engaging the medical community to identify ways to improve the medical listings. For example, SSA contracted with the Institute of Medicine to study its medical criteria for determining disability and to make recommendations for improving the timeliness and accuracy of its disability decisions, resulting in a 2007 report with recommendationsmedical n11 11 and a symposium of experts in 2010. SSA has addressed some of the institute's recommendations, such as making better use of its administrative data to update criteria and creating a standing committee through the institute to provide recommendations for listings revisions.

SSA Has Experienced Delays with Its Revision Process

SSA continues to face delays in completing both comprehensive and other ongoing updates. For example, as of early March 2012, SSA officials told us they still needed to complete comprehensive revisions for listings of six body systems that have been ongoing for the last 19 to 33 years, after numerous extensions beyond the original expiration periods (see table 1). Two of the remaining six body system listings--mental and neurological disorders, which are among those SSA uses most frequently in its eligibility determination process--have not been comprehensively revised for 27 years. n12 Four of the remaining body system listings are set to expire in 2012. Of these four, SSA is developing a notice of proposed rulemaking for three of them and has issued a notice on the fourth. However, it is unclear whether SSA will complete the revisions before they are set to expire.

Table 1: Key Dates for Comprehensive Revisions to Body System Listings

Body system and section Year comprehensive revision last published Years since comprehensive revision last publisheda Current expiration dateb

Special senses and speechc * Speech 1979 33 2015

* Disturbances of labyrinthine-vestibular function

Neurological disorders 1985 27 2012

Mental disorders 1985 27 2012

Hematological disorders 1985 27 2012

Musculoskeletal system 1985 27 2013

Respiratory system 1993 19 2012

Source: GAO analysis based on information provided by SSA officials.

While ongoing reviews involving SSA's targeted approach seem to be moving at a faster pace than the comprehensive revisions, SSA has already extended the expiration date for its review of the cardiovascular system. SSA may also need to extend expiration dates for listings under two other body systems set to expire in 2012, according to agency officials. Extensions may be needed because SSA has not yet published the notice of proposed rulemaking for those two body system listings and may not have time to publish these notifications, respond to comments, and complete the final updates of the listings by their current expiration date.

At the same time that SSA has been experiencing delays completing timely revisions, agency officials reported challenges with other steps in the business process. Whereas they always intended to review data at the one year mark to determine whether expectations from a revision were met, SSA officials told us they only began conducting these reviews in 2010. Specifically, they completed one postimplementation study in 2011 that involved a targeted sample of 175 cases, and they are conducting and planning to conduct two more studies at the one year mark. While disability experts we interviewed spoke highly of SSA's business and targeted review processes to obtain feedback early on and update the listings more promptly, recent delays raise questions regarding the agency's ability to follow its current business process while completing continuous and timely revisions for all 14 body systems.

SSA officials offered two key reasons for the delays in updating the listings: limitations in the number and expertise of staff and the complexity and unpredictability of the regulatory process. According to SSA officials, revising the medical listings requires research, deliberation, testing, regulatory review, and consensus with many stakeholders, and consequently is difficult and time-consuming to achieve. In addition, according to an SSA official in the Office of Medical Listings Improvement, the office is short-staffed and some staff lack the expertise necessary to perform this work. To address these constraints, SSA has contracted with the Institute of Medicine to review and develop recommendations for revising two of the body system listings. n13 However, SSA officials told us that the agency does not currently have a plan to renew this contract after it expires in 2012, nor use any other contractor to address staffing shortfalls due to budget constraints. Also contributing to delays is the time required for internal review and public comment under the regulatory process that depends largely on the number and the substance of comments received, according to an SSA official. For example, SSA officials told us that the advanced notice of proposed rulemaking to the listings of mental disorders in 2003 resulted in 500 comments, which took SSA 5 years to incorporate into a draft notice of proposed changes for regulatory review.

SSA Has Begun an Ambitious Project to Develop Its Own Source of Occupational Information

SSA is in the Preliminary Stage of Developing a New Up-to-Date Source of Occupational Information

In 2008, SSA began a multiyear project to develop a new source of occupational information that will replace the outdated information currently being used to determine if claimants are able to do their past work or any other work in the national economy. Since the 1960s, SSA has been using the DOT, which contains a list of job titles found in the national economy and was last updated in 1991. n14 The DOT provides SSA with descriptions of the physical demands of work--such as climbing, balancing, and environmental requirements--for each of the more than 12,000 occupations listed. According to SSA, these descriptions have been essential to its evaluations of how much a claimant can do despite his or her impairment and whether this level of functioning enables the claimant to do his or her past work or any other work. After its last limited update, Labor decided to replace the DOT with O*NET, which has far fewer job titles compared with the DOT, but has served Labor's purposes more efficiently. According to an SSA report, after investigating potential alternatives, SSA decided that O*NET and other existing databases with occupational information were not sufficiently detailed and able to withstand legal challenges for use in its decision-making process. SSA further decided to develop its own occupational information system, which would contain detailed information as in the DOT, but would also include additional information, such as the mental demands of work. In addition, the OIS should (1) meet SSA's legal, program, and data requirements; (2) be flexible enough to incorporate changes in SSA's policies and processes; and (3) be able to be updated to reflect the evolving workplace environment.

In 2008, SSA began taking several steps to guide the development of its OIS. SSA created an internal office and working group, as well as an Occupational Information Development Advisory Panel, comprised of external experts in areas related to the development of occupational information systems. n15 The advisory panel holds quarterly public meetings and has several subcommittees that review material and make recommendations to SSA on developing various components of the OIS. For example, in a 2009 report, the advisory panel supported the need for SSA to develop a new source of occupational information, rather than adapt O*NET, and recommended the type of data SSA should collect, as well as suggested ways to classify occupations.

To further inform its efforts, SSA has sought input from agencies or organizations that either collect occupational information or also use the DOT. For example, SSA officials held initial meetings with Labor and U.S. Census Bureau officials to gain information on sampling methods used for the O*NET, the Occupational Employment Statistics program, and Census Bureau's household surveys. n16 Additionally, SSA and Labor are in the process of completing a Memorandum of Understanding that will formalize their collaboration efforts on the new OIS. According to an SSA official, as the OIS project progresses, SSA plans to convene ad hoc roundtables with experts and other agency officials to explore specific subject areas, such as sampling issues. Besides working with Labor and Census Bureau officials, SSA officials and panel members have sought input from other experts and current users of the DOT, such as SSA disability adjudicators and external rehabilitation professionals, by conducting a user needs analysis in 2009 and presenting the OIS project at events and conferences.

In July 2011, SSA published a detailed research and development plan outlining all activities related to researching, developing, and testing the

key components of the OIS in order to implement the OIS by 2016 at an estimated cost of $108 million. For example, the plan includes several baseline activities to identify and study other occupational information systems and various approaches for analyzing occupations that may inform or could be leveraged in SSA's OIS data collection. The plan also includes activities to identify the primary occupational, functional, and vocational characteristics of current beneficiaries. Other key components of the plan include developing descriptions of work requirements, such as the physical and mental demands for jobs, and data collection and analysis strategies. SSA also plans to develop a strategy for piloting data collection nationwide within this time frame.

SSA Has Made Progress on Its OIS, but It Is Too Soon to Know if Project Timelines Will be Met

As of February 2012, SSA had made progress on many of the baseline activities outlined in its research and development plan for the OIS. n17 For example, according to an SSA official, its investigation of existing occupational information systems, now complete, has resulted in useful information about design issues other organizations have confronted and mitigated when creating their own system. Additionally, SSA's preliminary analysis of its own administrative data identified the most frequently cited occupations and functional and vocational characteristics of disability applicants. SSA officials told us the agency will target the occupations identified in this analysis for its pilot studies of the OIS. Also in 2011, SSA completed a comprehensive framework for assessing an individual's capacity to work--key to informing the OIS content, according to SSA officials--which was based on recommendations of outside experts as well as SSA's policy and program requirements.

While SSA has made progress on several key activities, agency officials delayed 2011 completion dates for certain activities and anticipate making additional changes to its timeline as a result of not meeting its staffing goals for fiscal year 2011. For example, the activities that were delayed by several months included finalizing reports for the baseline studies and conducting a literature review that would inform how occupations might be analyzed for the OIS. SSA officials told us that they would have needed to have the full complement of projected 2012 staff by September 2011 to complete all of the 2012 planned activities within the estimated schedule. However, SSA officials said they did not have the budget to hire new staff in September 2011. To address this challenge, SSA officials hired consultants to meet some of their needs. SSA officials also met with the Office of Personnel Management to explore the possibility of an interagency agreement that would allow SSA to use one or two of the Office of Personnel Management's industrial organizational psychologists to help on a part-time basis.

As part of our ongoing work, we are assessing SSA's current OIS project schedule and cost estimates against best practices, and have preliminarily identified some gaps in SSA's approach. For example, best practices require cost estimates to be comprehensive and include information about life cycle costs--that is, how much the project is expected to cost over time. However, while SSA has estimated the cost to research and develop the OIS, the estimate does not project the future costs to implement or maintain the system. The cost of sustaining an OIS could be significant, based on other agencies' experiences maintaining their systems for collecting national occupational information. We preliminarily identified other gaps, such as lack of documentation describing step by step how the cost estimate was developed so that those unfamiliar with the program could understand how it was created. For our final report due later in 2012, we plan to deliver more comprehensive findings on how well SSA is managing the development of its OIS against best practices, such as estimating costs of the OIS and ensuring that the project schedule reliably estimates related activities, the length of time they will take, and how they are interrelated. We will also identify any mitigation strategies the agency may have to address project risks, such as the risk of the agency not receiving full funding.

Chairman Johnson, Ranking Member Becerra, and Members of the Subcommittee, this concludes my prepared statement. I will be happy to respond to any questions.

n1 In fiscal year 2010, SSA received 6.3 million benefit claims from either workers with disabilities or their dependents, up from 3.1 million in 2000.

n2 The DI program provides monthly cash benefits to eligible applicants who have a sufficient work record. The SSI program provides monthly benefits to eligible applicants with limited income and resources who are disabled, blind, or age 65 or older.

n3 42 U.S.C. [Subsec.] 423(d)(1)(A) and 1382c(a)(3)(A). Substantial gainful activity is generally work activity involving significant physical or mental activities that are done for pay or profit, whether or not a profit is realized. 20 C.F.R. [Subsec.] 404.1572 and 416.972. In 2012, the substantial gainful activity threshold is $1,690 per month for blind recipients and $1,010 per month for individuals with other disabilities.

n4 See 20 C.F.R. [Subsec.] 404.1520 and 416.920.

n5 According to agency officials, there are almost 300 medical impairments organized in 119 listings under the 14 major body systems for adults.

n6 SSA's physical and mental residual functional capacity assessments establish the extent to which an individual's medically determinable impairment(s), including any related symptoms, such as pain, may cause physical or mental limitations or restrictions that may affect his or her capacity to do work-related physical and mental activities.

n7 GAO, Federal Disability Programs: More Strategic Coordination Could Help Overcome Challenges to Needed Transformation, GAO-08-635 (Washington, D.C.: May 20, 2008).

n8 Any changes to the medical listings have to proceed according to an established process for rulemaking outlined in federal law. As such, when changes are made to the listings, a notice of proposed rulemaking must generally be published in the Federal Register with a public comment period before the final rule is issued. 5 U.S.C. [Sec.] 553.

n9 Labor launched O*NET, a general purpose occupational information database used for workforce development, economic development, career exploration, academic and policy research, in 1998.

n10 The 5-year expiration period will be applied to listings under a body system upon completion of their current revision. SSA began applying the 5-year expiration period in 2011 when it comprehensively updated the endocrine body system listings. For other body systems listings updated prior to 2011, SSA generally assigned expiration periods extending beyond 5 years.

n11 Institute of Medicine, Improving the Social Security Disability Decision Process (Washington, D.C.: 2007).

n12 According to SSA officials, the agency made limited, but not comprehensive, revisions to the mental disorders listings in 2000.

n13 The institute provided recommendations for changes to the cardiovascular and the Human Immunodeficiency Virus listings in two reports. For the reports see, Institute of Medicine, Cardiovascular Disability: Updating the Social Security Listings (Washington, D.C.: 2010) and HIV and Disability: Updating the Social Security Listings (Washington, D.C.: 2010).

n14 The DOT provides a wide range of occupational information that could be used for job placement, occupational research, career guidance, labor-market information, curriculum development, and long-range job planning.

n15 The Occupational Information Development Advisory Panel was formed in December 2008 under the Federal Advisory Committee Act.

n16 The Occupational Employment Statistics program produces employment and wage estimates for approximately 800 occupations. The Census Bureau's household surveys include (1) the American Community Survey, which is an ongoing survey that provides annual data on demographics such as age, education, and disabilities, and (2) the Current Population Survey , which is primarily a labor force survey, conducted every month by the Census Bureau for the Bureau of Labor Statistics and provides data such as the national unemployment rate.

n17 Some of the activities that were part of the research and development plan were completed in years prior to the plan's issuance.

Read this original document at: http://waysandmeans.house.gov/UploadedFiles/Embargoed_Bertoni_Testimony.pdf