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 15
Improving the Social Security Disability Decision Process: Interim Report INTRODUCTION Overview of Social Security Disability Programs The Social Security Administration (SSA) provides Social Security Disability Insurance (SSDI) benefits to disabled persons of less than full retirement age and to their dependents. SSA also provides Supplemental Security Income (SSI) payments to disabled persons who are under age 65. For both programs, adult disability is defined as a “medically determinable physical or mental impairment” that prevents an individual from engaging in any substantial gainful activity and is expected to last at least twelve months or result in death. Childhood disability under SSI is defined as a physical or mental condition or combination of conditions that causes “marked and severe functional limitations” and is expected to last at least twelve months or result in death. As of December 2004, SSA was making disability payments under the SSDI and SSI programs to 9.8 million people aged 18–64 and 1.0 million children based on their own disability, as well as payments to an additional 1.5 million dependents of disabled individuals in the SSDI program (SSA, 2005c:31–33). SSDI and SSI disability account for 22. percent of SSA’s benefits payments each year (SSA, 2005a:147). Yet, they account for 57 percent of SSA’s administrative costs, because determining disability status is much more complicated than verifying a worker’s work records to compute Social Security retirement benefits (SSA, 2005a:161). In fiscal year (FY) 2005 SSA expected to process approximately 2.5 million initial claims for disability benefits, 1.6 million continuing disability reviews, and 600,000 appeals hearings (SSA, 2004a:4–5). The annual number of claims, which was 2.1 million in FY 2000, is expected to continue to increase as baby boomers age into their 50s and early 60s and become more prone to disabling impairments. Claimants file applications through one of 1,300 SSA field offices, over the telephone, by mail, or on the Internet. Interviewers in the field offices apply the nonmedical eligibility criteria to determine insured status and ask claimants to provide supporting
OCR for page 16
Improving the Social Security Disability Decision Process: Interim Report medical documentation. The case files are then sent to a state Disability Determination Services (DDS) agency, which makes the initial decision as to whether the individual is disabled based on medical and, if needed, vocational criteria provided by SSA. The DDSs use a five-step decision process, called the sequential evaluation process by SSA, for each claim for disability benefits, whether under SSDI or SSL1 The first decision is whether the applicant is currently engaged in substantial gainful employment, which is defined as earning more than a certain amount per month ($830 a month in 2005 except for statutorily blind beneficiaries).2 If yes, the application is denied. Step two is to determine whether the applicant has a medically determinable impairment that is severe, i.e., significantly limits the applicant’s physical or mental ability to engage in basic work activities, and that will last 12 months or longer. If no, the claim is denied. If yes, step 3 is a determination of whether the applicant’s impairment meets or equals one of the more than 100 conditions listed in a Listing of Impairments (Listings), which are assumed to be so severe that SSA accepts that the applicant cannot work. If yes, the claim is allowed. If not, step 4 involves an assessment of the applicant’s residual functional capacity to do past work. If yes, the claim is denied. If no, step 5 looks to see if the applicant can engage in any employment in the U.S. economy. If yes, the claim is denied. If no, it is allowed. On average, the DDSs allow approximately 37 percent of the claims they adjudicate through the five-step process. The share of claims that are allowed at step 3 by meeting or equaling the Listings has declined over the years, to about half currently. This means that a growing percentage of cases must undergo intensive medical-vocational evaluation during steps 4 and 5, which require more time and DDS staff resources to perform. Currently, approximately 70 percent of the allowances that are eventually made are made at the initial level, a process that takes about 97 days (SSA, 2004a:Appendix A). About a third (36 percent) of the applicants whose claims are initially denied decide to file an appeal. Of those, approximately 67 percent result in allowances at some point in the multi-step appeals process, which averages more than a year to complete. Ultimately, slightly more than half of the claims (52.3 percent) are approved.3 On September 25, 2003, the Commissioner of Social Security told the Subcommittee on Social Security of the House Committee on Ways and Means of her intent to develop a “new approach” to disability determination (Barnhart, 2003). The goals of the new approach are to (1) make the correct decision as early in the claims process as possible and (2) foster return to work at all stages of the process. On July 27, 2005, after consulting with a variety of groups, SSA published a notice of proposed rulemaking (NPRM) describing specific regulatory changes it plans to make to implement the Commissioner’s new approach to the disability decision-making process (SSA, 2005e). Among the initiatives proposed in the NPRM was the establishment of a Federal Expert Unit that would set up and administer a national network of medical, psychological, and vocational experts to support the disability decision process at the initial decision point and subsequent levels of appeal. The NPRM also proposes establishment of a Quick Disability Determination Process that would use a predictive statistical model to identify cases likely to be 1 For overviews written for physicians, see Nibali (2003) and Robinson and Wolfe (2000). 2 The statutorily blind are subject to different earnings rules. 3 These calculations are based on FY 2003 data.
OCR for page 17
Improving the Social Security Disability Decision Process: Interim Report allowed and send them to a special unit to be adjudicated within 20 days. The NPRM includes other changes that are not addressed in this report. IOM Study Charge SSA asked the Institute of Medicine (IOM) to help in two areas related to its initiatives to make the disability determination process more efficient and accurate. First, SSA asked IOM to recommend improvements to the Listing of Impairments, which contains impairments and associated medical findings that SSA considers so severe that individuals with them (or with conditions equal to them in severity) are considered disabled without additional evidence of their inability to work. Among other things, the IOM will look at alternative conceptual models for organizing the Listings (currently based on 14 body systems), processes for determining when the Listings require revision, and the advisability of integrating functional assessment into the Listings. Second, SSA asked IOM to recommend ways to improve the use of medical expertise in the disability determination process, including the appeals process. Subsequently, in the July 2005 NPRM, SSA announced it is looking to the IOM for advice on the qualifications of the medical and psychological experts to be recruited for the national network. SSA’s requests to IOM are broken down into 10 specific tasks. The first seven pertain to the Listings, and the final three—tasks 8, 9, and 10—pertain to presumptive disability categories, organization of medical expertise, and training and certification of consultative examiners, respectively (see Appendix A). The last three tasks are addressed in this interim report. NOTE: Tasks 9 and 10 are presented before task 8 because, due to the NPRM, the recommendations on the qualification standards for medical experts are higher priority than those on presumptive disability, which is not the subject of rulemaking at this time. IOM Committee The IOM formed a committee to conduct the study. The 16 committee members are experts in clinical decision making, physical medicine and rehabilitation, orthopedic surgery, occupational medicine and nursing, psychiatry and psychology, pediatrics, public health, functional assessment, occupational rehabilitation, legal and economic aspects of disability, social security disability administration, claimant advocacy, and private disability insurance. The committee expects to issue its final report in 2006. The committee was asked by SSA to operate under certain assumptions: The existence of a national electronic disability records system that allows claims folder documents to be transmitted electronically and viewed by all employees and medical personnel, regardless of their location; The existence of a national video hearing capability; An unchanged statutory definition of disability; and
OCR for page 18
Improving the Social Security Disability Decision Process: Interim Report That beneficiary return-to-work, vocational rehabilitation, and work incentives for applicants are outside of the scope of the committee’s deliberations. Interim Committee Report SSA asked the committee to prepare preliminary recommendations, prior to the final report, addressing the three tasks that relate to medical expertise issues, with a special focus on the appropriate qualifications of medical and psychological experts involved in disability decision making. Therefore, the committee agreed to issue this interim report, which provides the committee’s initial impressions on those tasks. This report, based on preliminary information gathering and analyses, and deliberations at the committee meetings, is in some respects an interim report. After further information gathering and analyses of the effectiveness of the disability decision process in identifying those who qualify for benefits and those who do not, the committee may refine its recommendations concerning medical and psychological expertise in the final report. The final report will address a number of issues with potential implications for the qualifications of the medical experts involved in the disability decision process. These include the extent of knowledge about differences in decision outcomes depending on the qualifications of decision makers; research on inter-rater reliability of decision criteria; comparisons of evaluations of samples of cases by different groups of SSA adjudicators or by adjudicators compared with outside medical experts; results of long-term follow-ups of applicants who were allowed and denied; evaluations of alternative decision making models, such as the single decision maker model; studies of sources of variation in allowance rates among DDSs and Office of Hearings and Appeals hearing offices; and in-depth analyses of program statistics about the outcomes of applications for benefits at different levels of decision making. The final report will summarize what is known about how well the disability determination process serves as a screening tool to avoid false positives and false negatives and the factors that affect its sensitivity and specificity, in epidemiologic terms.4 In the absence of such information and analysis, this report is based mostly on the judgment of the committee of the qualifications, training, and certification that should be expected of medical participants in the disability decision process. SSA’s Proposed Disability Decision Process Changes SSA’s requests to IOM are closely related to the agency’s plans to institute a new approach to improve the SSDI and SSI disability process. For purposes of the medical expertise issues addressed in this interim report, the relevant aspect of the plan is the intent to establish a national network of medical and psychological consultants who would be used by all adjudicators, including disability examiners in the DDSs and administrative law judges (ALJs) in the Office of Hearings and Appeals (OHA). In the July 2005 NPRM, SSA proposed establishing “a Federal Expert Unit to augment and strengthen medical and vocational expertise for disability adjudicators at all levels of the disability 4 Sensitivity is the measure of a screen’s ability to detect true positives, i.e., those the test is seeking to screen in, and specificity is the measure of its ability to detect true negatives, i.e. those the test is seeking to exclude.
OCR for page 19
Improving the Social Security Disability Decision Process: Interim Report determination process,” and creating “a national network of medical, psychological, and vocational experts who will be available to assist adjudicators throughout the country” (SSA, 2005e:43593). SSA explained that: …[U]nder our current disability adjudi cation process, medical, psychological, and vocational experts are not consistently available to all adjudicators at every level or in all parts of the country. We are therefore proposing to establish and operate a Federal Expert Unit, which we believe will help to ensure the full development of the record, enable adjudicators to make accurate determinations or decisions as early in the process as possible, and facilitate subsequent review should a case be appealed to a higher level We propose to create a national network of medical, psychological, and vocational experts who will be available to assist adjudicators throughout the country. This national network may include experts employed by or under contract with the State agencies; however, all experts affiliated with the national network must meet qualifications prescribed by the Commissioner. The Federal Expert Unit will organize and maintain this network comprised of medical, psychological, and vocational experts who will provide medical, psychological, and vocational expertise to State agencies, reviewing officials, administrative law judges, and the Decision Review Board. We want to ensure that the right set of medical eyes reviews medical records and answers questions about the wide variety of impairments seen in disability claims. We believe that the expert network affiliated with the Federal Expert Unit will help ensure that a medical, psychological, and vocational expert who has the qualifications required by the Commissioner assists in adjudicating disability claims. With the assistance of the Institute of Medicine, we plan to develop standards that define the medical and psychological expertise necessary for experts to qualify for participation in the national network (SSA, 2005e:43594). In discussions with SSA staff, the committee was asked to assume that, under the new plan, there would be a national pool of medical experts and that there would be flexibility in payment rates for expert medical advice. The committee was not to assume that the experts would necessarily be centrally located. The plan for achieving process improvements is also predicated on successful implementation of SSA’s electronic disability system (eDib), which is currently being rolled out nationwide. SSA expects all DDSs and OHA hearing offices to be processing their workloads with electronic disability folders on a regular basis by the end of 2006 (Barnhart, 2005). Although SSA refers to eDib as an electronic disability system, it is not a true electronic medical record, in which all information would be stored as structured data. In such a system, the data could be manipulated for purposes such as policy analysis or management information. SSA’s eDib system includes a mixture of structured data and images, with the medical evidence portion of the file consisting of images of paper medical records. This will make it possible for medical experts in remote locations to review medical records, assuming that arrangements to keep claimant files secure are made.
OCR for page 20
Improving the Social Security Disability Decision Process: Interim Report TRENDS IN THE DISABILITY DECISION PROCESS The Social Security disability programs have grown rapidly in recent years, and several problems have come with this growth—problems that promise to become worse as the baby boomer generation reaches the age when disability becomes more likely (the oldest baby boomers will turn 60 in 2006). These problems, which have been documented in a number of reports,5 include: significant growth in applications; variability in decision making at each stage in the adjudication process; high rate of appeals and of reversals of initial decisions to deny benefits; and long length of time to reach a final decision. Program Growth Historically, both the SSDI and SSI disability programs have experienced steady, although not consistent, program growth. The causes for this growth and fluctuation are complex and not fully understood, and may include the broader socioeconomic and demographic environment as well as programmatic actions and court decisions (IOM, 2002:42). Even in the context of historical program growth, recent increases in application rates have had significant workload implications for SSA. According to SSA statistics, 1,041,362 workers applied for SSDI in 1998. The number of applications grew steadily to 1,485,482 in 2004, an increase of 42.6 percent in 6 years (Zayatz, 2005:Table 2). Applications have also increased in the SSI program. Applications by adults for SSI disability payments increased from 1,108,957 in 1998 to 1,438,992 in 2004 (29.8 percent increase), and applications for SSI childhood disability payments went from 335,732 in 1998 to 402,218 in 2004 (20.1 percent increase) (SSA, 2005b:Appendix C).6 Variability in Disability Decisions As noted above, the initial disability allowance rate is approximately 37 percent. However, this rate varies significantly from DDS to DDS, and from region to region. In calendar year (CY) 2004, the initial allowance rate for individual states ranged from 25.3 percent to 61.1 percent. An SSA study of initial allowance rates in the late 1990s analyzed a similar spread of about 30 percentage points across DDSs. It found that statistically controlling for economic conditions, demographic factors, prevalence of SSDI and SSI beneficiaries, health status, and other external factors explained only half of the variance across the states (Strand, 2002). This finding suggests that up to half of the variance in allowance rates among the states may be due to differences in state administrative practices (e.g., use of consultative examinations, involvement of doctors in making dis- 5 See reports of the Social Security Advisory Board (SSAB) and the Government Accountability Office (GAO), for example, GAO, 2003, 2004; SSAB, 1998, 2001a. 6 The three categories of applications total more than 2.5 million—the number of applicants—because some adults apply for SSDI and SSI benefits concurrently.
OCR for page 21
Improving the Social Security Disability Decision Process: Interim Report ability decisions, payment amounts for medical evidence of record and consultative examinations, salaries and qualifications of disability decision makers, and training practices), quality assurance practices of state agencies and SSA regional offices, and/or state program policies (e.g., requiring individuals to file for SSA disability as a condition of eligibility for state benefits).7 Allowance rates at the ALJ hearing level also vary widely from state to state. In FY 2002, the overall hearing allowance rate was about 66 percent but, at the state level, the rate varied from 35 percent to 86 percent (SSAB, 2001b:70). Appeals and Allowances on Appeal About a third of the claims denied initially by the DDSs are appealed to the ALJ hearing level, after being denied at reconsideration level. By way of comparison, 8 percent of the claims for disability compensation from the Department of Veterans Affairs (VA) are appealed to the hearing level, even though VA, like SSA, allows less than half the claims initially.8 At UnumProvident, a provider of private long-term disability insurance, 10 to 15 percent of cases initially denied are appealed (Lewin, 2001:42). In Canada, approximately 20 percent of the disability applicants denied by the Canada Pension Plan appealed for a hearing in FY 2003.9 The appeal rate of one-third contributes to the lengthy average time it takes to make a final decision. Many applicants must wait an average of more than one year from the time they are denied by a DDS until an ALJ decides their case (see next section, below, on timeliness). There also are financial implications of the high rate of appeals. According to the Government Accountability Office (GAO), the average cost of processing an appeal hearing ($2,157 in 2001) is much greater than the average cost of processing an initial claim ($583 in 2001) (GAO, 2004:11). The 1,200 ALJs who hold hearings on disability cases are employees of SSA, and although they operate separately from the DDS adjudicators, they are supposed to apply the same rules and regulations. Nevertheless, ALJs allow 61 percent of the claims that are appealed to them and deny or dismiss the rest. At the VA, 22 to 26 percent of the appeals are successful, 29 to 36 percent are remanded back for reconsideration, and 40 to 42 percent of the initial decisions are upheld (Lewin, 2001:42). At UnumProvident, 7 to 12 percent are reversed (Lewin, 2001:42). About half of the denials for disability benefits from the Canada Pension Plan are reversed (CPP/OAS, 2004:Figure 8). Possible reasons for the high allowance rates by ALJs in appeals cases include (SSAB, 2001b:5–6): The fact that most claimants are never seen in person by an adjudicator until they have an ALJ hearing; Rules that allow claimants to introduce new evidence and allegations at each stage of the appeals process; 7 These factors are listed in SSAB, 2001b:5. 8 VA pays for partial permanent disability, and the 8 percent includes those appealing for a higher disability rating as well as those denied any benefits. 9 In 2002–2003, 31,200 applicants (51.4 percent of all applicants) were denied at the initial or reconsideration levels. During the same period, there were 6,300 (20.2 percent of the denials) appeals for a hearing before a hearing tribunal (Canadian Parliament, 2003 :Ch. 4).
OCR for page 22
Improving the Social Security Disability Decision Process: Interim Report The time lag between the initial denial and the hearing, during which a claimant’s condition may worsen; Administrative differences between the DDSs and hearing office, including their access to and use of medical and vocational expertise; Greater involvement of attorneys and other claimant representatives at the ALJ hearing, who help claimants assemble a more complete case record; Differences in the quality assurance procedures applied to initial decisions and hearings-level decisions, which may bias DDS examiners to deny claims and ALJs to allow claims; and Differences in the training given to ALJs and state DDS examiners. Decision Timeliness In FY 2004, average processing times for disability claims at each level of the decision process were (Figure 1): 95 days for initial disability claims; 97 days for reconsiderations; 394 days for hearings; and 251 days for decisions on appeals of hearings at the Appeals Council. Although the DDSs processed 2.6 million initial disability cases in FY 2004, the backlog was 625,000. ALJs issued 497,000 hearing decisions, but the number of pending hearing requests was 636,000 (SSA, 2005a:45–50). Figure 1 Processing time for disability claims in days, CY 2004 SOURCE: SSA, 2005a: 17.
OCR for page 23
Improving the Social Security Disability Decision Process: Interim Report Nevertheless, most of the disability allowance decisions in FY 2004 were made at the initial level of the decision process (Figure 2): 71.0 percent of the claims eventually allowed were allowed at the initial level; 6.2 percent of the claims eventually allowed were allowed at reconsideration; 22.6 percent of the claims eventually allowed were allowed at an ALJ hearing; and 0.2 percent of the claims eventually allowed were allowed at the Appeals Council. Given that nearly a quarter of the claims were allowed after a hearing or an appeal of a hearing, the overall weighted average time until an allowance was made was 204 days. As noted above, appeals add substantially to the administrative costs of the program. Making the correct decision initially will require more resources at the front end of the disability decision-making process, but these costs could be offset by savings from lower appeals rates. Figure 2 Percentage of allowed claims by decision level, CY 2004 SOURCE: SSA, 2005d. Implications of Trends for the Interim Report The increasing caseload makes it desirable to make the disability decision process as efficient as possible. One way to increase efficiency would be to find ways to make the Listings a more effective screening tool at step 3 of the sequential decision process, which will be addressed in the second phase of the study and the final report. A more effective screening tool could reduce the number of false negatives, i.e., cases that, al-
OCR for page 24
Improving the Social Security Disability Decision Process: Interim Report though they meet the definition of social security disability, fail the screen and therefore have to undergo a more intensive and time-consuming medical-vocational analysis at steps 4 and 5 before they are allowed. The variability in allowance rates from DDS to DDS and between DDSs and OHA might be reduced if all had access to the same or equivalent expertise. If cases could be better evaluated and the medical record more fully developed at the beginning of the process, there might be fewer appeals and fewer allowances on appeal. Equalizing access to medical expertise across decision making units could help ameliorate the trends in case loads, processing time, appeal and reversal rates, and variability in decision making, but it is also justified on the grounds that every applicant for disability benefits should have the benefit of the expertise needed to evaluate their case regardless of where they live. Although it cannot be proven with available information, the committee finds it logical to expect that this will result in improved evaluations of complex cases, leading to more accurate decision, which in turn will have beneficial effects by reducing the need for lengthy appeals and improving the case record for appeals that are filed. In the final report, the committee will analyze existing data and past research and identify research and experiments that SSA could sponsor to see what difference that different types and amounts of medical expertise make in decision outcomes. The committee also notes in this interim report that several other factors are important in improving decision making. These include fuller development of cases before the initial decision and a quality assurance system with incentives that balance the need for making the right decision with the need for making decisions as quickly as possible. CONSTRAINTS ON THE DISABILITY DECISION PROCESS The committee supports the primary goals of SSA’s new disability plan—to make the right decision as early in the claim process as possible, and to improve the accuracy, consistency, and timeliness of disability decisions at all levels of the disability process. Because the agency has not adopted the final version of its new plan, it is too early to reach any conclusions about the plan itself. Nevertheless, the committee believes that several factors significantly limit SSA’s ability to make the correct decision early in the process, and these factors contribute to error, inconsistency, and delay in decision making. The development of a full record at the beginning of the disability decision process and evaluation of the record by appropriate experts, including medical experts, may not be essential for an accurate initial disability decision in every case. However, fuller case development at the front end of the process should reduce the impetus for appeals, reduce the number of allowances on appeal, and shorten the average length of time before reaching final adjudication. Full case development at the front end of the process has been recommended in a number of reports.10 This is difficult to accomplish, however, because disability claim processing personnel, especially in DDSs, are subject to strong pressures to make decisions quickly and to reduce administrative costs (including medical costs) per case. 10 See, for example, ACUS, 1989; Bloch et al, 2003.
OCR for page 25
Improving the Social Security Disability Decision Process: Interim Report These time and cost pressures are reflected in SSA’s Performance and Accountability Report for FY 2004, wherein SSA identifies four strategic goals. The first goal is “To deliver high-quality, citizen-centered service.” Within each strategic goal, the report identifies specific strategic objectives. The first strategic objective under this first strategic goal is “To make the right decision in the disability process as early as possible.” To measure achievement in reaching this goal and this objective, SSA identifies six key performance indicators. Yet, despite the emphasis on high-quality service and making the right decision early in the process, none of these key performance indicators measures either the accuracy of the decisions or whether the decision was made as early in the process as possible. All six performance indicators focus on making decisions quickly and reducing the caseload (SSA, 2005a:45–51): Number of initial claims processed by DDSs; Number of SSA hearings processed; Average processing time for initial disability claims; Average processing time for hearings; Number of initial claims pending; and Number of hearings pending. Within the context of these pressures to move cases quickly, DDSs and ALJs are also subject to a contrasting set of incentives that have the effect of pushing decision outcomes in different directions at different levels of adjudication. In addition to tight time and resource constraints, DDSs are subject by law to a 50 percent “pre-effectuation” review by SSA regional offices of all decisions to allow claims in SSDI cases before payment is made. These factors create a disincentive to performing complete claim development because the chances of having a denial decision subsequently reviewed are slight. SSA’s quality assurance (QA) system reviews both SSDI and SSI cases, including allowances and denials, but the sample size is approximately 1 percent of the cases.11 About a third of the denials are appealed for review by an ALJ, but the costs of decisions reversed by the ALJ are not internal to the DDS. In addition, there is no information feedback loop from the appeals process on the results of hearings, especially on the characteristics of cases allowed on appeal that were denied initially. In contrast, ALJs have incentives to allow claims. First, they are permitted to make use of several processing expedients that apply only to allowance cases (e.g., shortform decisions, “bench” decisions, on-the-record decisions). These make allowances easier and faster to process than denials, which require full development and a comprehensive written decision. Second, the chances of having an allowance decision subsequently reviewed are very small. On the other hand, more than half of all denial decisions are reviewed by the Appeals Council at the claimants’ requests. The new disability plan recently set forth by the Commissioner describes SSA’s intent to “create and operate a comprehensive and multidimensional approach to quality assurance” in order to improve quality and accountability throughout the disability proc- 11 In FY 2004, just over 40,000 cases (both allowances and denials) were reviewed for quality assurance (QA), of which 1,500 (3.6 percent) were returned to correct errors or for better documentation; 335,000 cases of allowances were subjected to preeffectuation review (PER), of which 13,000 (3.8 percent) were returned due to error or inadequate documentation. QA reviews cover initial claims, reconsiderations, and determinations of continuing eligibility; PER covers favorable initial and reconsideration decisions in title II and concurrent claims (SSA, 2005a:215–216).
OCR for page 54
Improving the Social Security Disability Decision Process: Interim Report dures, and other questions concerning the condition in question. These would serve as a checklist to remind DEs what is typically needed in each type of case and help them identify which questions and requests need to be made in each specific case, depending on what is already in the record. The unneeded items could be deleted and the now individualized letter could be printed. However, the DE should include a narrative paragraph in the request letter describing what the DE is concerned about and thinks is needed, to provide further guidance to the CE provider. The cooperative effort with AUCD to develop an electronic database of all appropriate assessment instruments for assessing development and functioning in children should help DEs and MCs determine what test or tests are needed in childhood cases. Similar databases could be developed of appropriate tests for other common medical conditions that are the subjects of disability claims. Many DDSs have developed standard request letters and forms. SSA should establish a process to identify best practices currently in use among the DDSs for facilitating useful CE examinations and reports. Experienced DEs and MCs should be involved, along with outside medical experts knowledgeable about current clinical practice and research. PRESUMPTIVE DISABILITY CATEGORIES Task 8: “Review SSA’s list of presumptive disability conditions and suggest revisions, additions, and/or deletions. The committee will recommend essential criteria for establishing candidate conditions for presumptive disability and recommend the level of professional expertise needed to certify that a presumptive diagnosis is adequately established.” An adult or child applying for SSI based on disability or blindness may receive up to 6 months of payments prior to the final determination of disability or blindness if he or she is determined to be “presumptively disabled” or “presumptively blind” (hereafter, presumptive disability) and meets all other (i.e., financial) eligibility requirements. SSA field offices are authorized only to make presumptive disability determinations for certain specified impairments. DDSs can make presumptive disability findings in any case in which the available evidence indicates a “strong likelihood” that the claim will be allowed after going through the regular formal determination process. Initially, the list of presumptive disability categories, because it was meant for use by field office personnel with no medical training, was confined to conditions that were either (1) easily verifiable by direct observation during the claims interview (e.g., amputation) or (2) easily confirmed by a telephone call to an appropriate authority, such as the treating physician or school official (e.g., total deafness). Over time, as presumptive disability categories were added to the original three, the evidence requirements became more complex. In some categories, a document is now needed. For example, in low birth weight cases, a birth certificate or hospital admission summary is required. In end stage renal disease cases, Medicare form CMS-2728 is needed. For allegations of HIV infection, SSA developed a special check-block form for the treating source or CE provider to
OCR for page 55
Improving the Social Security Disability Decision Process: Interim Report complete. Field office interviewers use the completed form to confirm the diagnosis and see if the checked items indicate that disease manifestations are of listing-level severity. The impairment categories for which field offices may make presumptive disability decisions currently are: Amputation of a leg at the hip. Allegation of total deafness. Allegation of total blindness. Allegation of bed confinement or immobility without a wheelchair, walker, or crutches, allegedly due to a long-standing condition, excluding a recent accident and recent surgery. Allegation of stroke (cerebral vascular accident) more than three months in the past and continued marked difficulty in walking or using a hand or arm. Allegation of cerebral palsy, muscular dystrophy or muscular atrophy and marked difficulty in walking (e.g., use of braces), speaking or coordination of the hands or arms. Allegation of Down syndrome. Allegation of severe mental deficiency made by another individual filing on behalf of a claimant who is at least seven years of age. Birth weight below 1200 grams (2 pounds, 10 ounces) for a child claimant who is less than one year old, according to the birth certificate or other evidence, such as the hospital admission summary. Human immunodeficiency virus (HIV) infection confirmed by a medical source. Available evidence such as the hospital admission summary which shows a gestational age at birth and a certain corresponding birth weight for a child claimant who is less than one year old (e.g., 35 weeks and 1700 grams of less). Confirmation from physician or hospice official that an individual is receiving hospice services for a terminal illness. Allegation of spinal cord injury producing the inability to ambulate without the use of a walker or bilateral hand-held assistive devices for more than 2 weeks which is confirmed by an appropriate medical professional. End stage renal disease (ESRD) with ongoing dialysis where file contains an ESRD Medical Evidence Report-Medicare Entitlement and/or Patient Registration. Allegation of amyotrophic lateral sclerosis (ALS, Lou Gehrig’s disease). According to SSA statistics, field offices and DDSs granted presumptive disability status in 135,603 cases in CY 2004, constituting approximately 7.7 percent of the 1.76 million SSI claims made that year. Field offices accounted for 17,191 (13 percent) of the 135,603 presumptive disability decisions made in CY 2004 (the rest were made by DDSs). The field office presumptive disability cases constituted 1.0 percent of the applications for SSI benefits in 2004. The rate of presumptive disability varied from field office to field office, from 0.1 percent to 3.5 percent (a difference of 35 times). Looking at field offices as a group, half of all presumptive disability decisions were in the two low-birth-weight categories (categories 9 and 11 in Table 1, Column 2). Three other categories accounted for another 25 percent: category 6 (allegation of cere-
OCR for page 56
Improving the Social Security Disability Decision Process: Interim Report TABLE 1 Field Office Presumptive Disability Decisions, by Presumptive Disability Category, CY 2004 (1) Number of FO PD Decisions (2) Percentage of All FO PD Decisions (3) Reversal Rate (4) Percentage of All FO PD Reversals 1. Amputation of a leg at the hip 342 2.0% 48.5% 9.4% 2. Allegation of total deafness 479 2.8% 12.1% 3.3% 3. Allegation of total blindness 320 1.9% 9.1% 2.6% 4. Allegation of bed confinement or immobility without aid 674 3.9% 9.1% 3.5% 5. Allegation of stroke more than 3 months in the past & continued marked difficulty… 230 1.3% 16.5% 2.2% 6. Allegation of cerebral palsy, muscular dystrophy or muscular atrophy & marked difficulty in… 1,342 7.8% 12.1% 9.2% 7. Allegation of Down syndrome 1,863 10.8% 1.3% 1.4% 8. Allegation of severe mental deficiency… 717 4.2% 7.1% 2.9% 9. Birth weight less than 1200 grams… 5,691 33.1% 1.9% 6.2% 10. HIV infection confirmed… 1,235 7.2% 42.1% 29.4% 11. Low birth weight for gestational age… 3,092 18.0% 14.8% 26.0% 12. Hospice services for a terminal illness confirmed… 433 2.5% 6.7% 1.6% 13. Allegation of spinal cord injury causing inability to ambulate without… 266 1.5% 9.4% 1.4% 14. ESRD with ongoing dialysis and Medicare Form… 441 2.6% 1.8% 0.5% 15. Allegation of ALS 19 0.1% 5.3% 0.1% Total 17,191 100.0% 10.3% 100.0 SOURCE: Unpublished tables provided by the Office of Disability and Income Support Programs, SSA. bral palsy, muscular dystrophy, or muscle atrophy), category 7 (allegation of Down syndrome), and category 19 (HIV infection). Ultimately, 10.3 percent of the field office presumptive disability cases were not allowed at the initial decision level after going through the regular disability determination process (Table 1, column 3). However, some categories had very low “reversal rates,” between 1 percent and 2 percent (e.g., categories 7, 9, and 14), while several others had very high reversal rates, between 40 percent and 50 percent (categories 1 and 10).
OCR for page 57
Improving the Social Security Disability Decision Process: Interim Report According to SSA statistics on all title SSI disability determination decisions in CY 2004 (including concurrent SSI-DI and both presumptive disability and non-presumptive disability cases), seven impairment codes had allowance rates of 95 percent or higher (Appendix Table 7). Of these, three correspond to current presumptive categories (ALS, birth weight under 1,200 grams, and Down syndrome). The other four were cancers with high fatality rates (e.g., pancreatic and liver cancer). Another five codes had allowance rates of 90 to 94.9 percent. These included one code that corresponds to a presumptive disability category (developmental disabilities including autism). The others were cancers (e.g., lung cancer). Setting the cutoff at 85 percent would yield six more codes, all malignancies except chronic renal failure and childhood origin psychosis (adult). Almost 90 percent of DDS cases granted presumptive disability status were eventually allowed after going through the regular disability determination process, but they did not achieve this rate in every type of case. The DDSs experienced allowance rates of 95 percent or more in 47 impairment codes, accounting for 38 percent of the presumptive disability cases initiated by them (Table 2). These cases included organic mental disorders, cerebral palsy, schizophrenia, leukemia, mental retardation, chronic renal failure, birth weight under 1,200 grams, lung, liver, and pancreatic cancers, cancer, and developmental disabilities including autism.40 Reversal rates were higher than 50 percent for some impairments, but they only involved a third of one percent of the cases. There were reversal rates of between 10 percent and 50 percent in 27 percent of DDS presumptive disability cases. TABLE 2 DDS Presumptive Disability Decisions, by Impairment Code, CY 2004 Allowance Rate Number of Codes Number of Decisions Percentage of Decisions Cumulative Allowance Rate Number of Codes Number of Decisions Percentage of Decisions 95–100% 47 44,951 38.0% 95–100% 47 44,951 38.0% 90–94.9% 42 39,373 33.3% 90–100% 89 84,324 71.2% 85–89.9% 35 13,729 11.6% 85–100% 124 98,053 82.8% 80–84.9% 29 10,294 8.7% 80–100% 153 108,347 91.5% 70–79.9% 33 7,023 5.9% 70–100% 186 115,370 97.4% 60–69.9% 10 856 0.7% 60–100% 196 116,226 98.2% 50–59.9% 6 206 0.2% 50–100% 202 116,432 98.3% 0–49.9% 8 340 0.3% 0–100% 210 116,772 98.6% Unknown 1,643 1.4% 118,415 100.0% SOURCE: Unpublished tables provided by the Office of Disability and Income Support Programs, SSA. 40 Appendix Table 6 lists all 47 impairment codes for presumptive disability decisions that had reversal rates less than 5 percent in CY 2004.
OCR for page 58
Improving the Social Security Disability Decision Process: Interim Report The annual number of presumptive disability decisions has increased greatly since 2001, when there were 60,543. The increase was entirely due to DDSs; field offices granted presumptive disability in slightly fewer cases in 2004 than in 2001, 17,191 compared with 18,862. Although the number of presumptive disability decisions more than doubled, the overall rate of reversals fell from 13 percent to 9 percent. DDSs lowered their reversal rate from 12 percent to 9 percent and the field offices from 17 percent to 10 percent. These differences seem to be due to management practices more than to medical policies. Revising the Presumptive Disability Categories with Explicit Criteria Presumptive disability is primarily a social policy. Congress included it in the 1972 act establishing the SSI program because it recognized that applicants for SSI have little or no income or assets and probably need immediate assistance with the costs of living. Claimants must meet the low income and assets criteria for SSI even to be considered for presumptive disability. Therefore, the committee is unable to recommend specific categories to include or delete because the selection criteria are not solely medical. We can, however, analyze the impacts of setting different levels of potential reversal rates (i.e., presumptive disability cases that turn out not to meet the criteria for allowances) and suggest that SSA adopt explicit criteria and use them to revise the presumptive disability categories. Recommendation 3–1. SSA should consider dropping some current presumptive disability categories, and perhaps adding some, after deciding on explicit criteria for including categories. The current presumptive disability categories were created for use by nonmedical personnel in SSA field offices, which means they must have two characteristics. They must be capable of being applied by a layperson, and they must have a high probability of meeting SSA’s definition of disability. To be applied by an interviewer in a field office, the categorical condition must be either directly observable during the claims interview or quickly confirmable by contacting the treating physician or other appropriate authority. To have a high probability of meeting SSA’s definition of disability, a categorical condition must be one in which the diagnosis alone correlates highly either with inability to work for a year or more or the presence of a terminal illness. The nonmedical criterion in selecting presumptive disability categories is the degree of risk, i.e., the reversal rate, that SSA is willing to pay for in order to increase the number of true positives. Early in the SSI program, an SSA ruling suggested that a reversal rate of more than 10 percent a month over several months made a category a candidate for deletion.41 Perhaps this is too stringent. Only nine of the 15 current categories would meet this criterion (Table 3). A reversal rate of 15 percent or less would include 12 of the current categories, and a rate of 20 percent or less would add one more, bringing the total to 13. A reversal rate of 20 percent or less would reach more needy claimants 41 In Social Security Ruling 80–36 (1980). The ruling was rescinded later when new presumptive disability categories made it obsolete.
OCR for page 59
Improving the Social Security Disability Decision Process: Interim Report than one of 10 percent or less, but the costs of cases ultimately found to be ineligible would be higher. Table 3 Approximate Costs of Presumptive Disability Cases Ultimately Disallowed, CY 2004 Data PD category allowance rate Number of PD categories Number of disallowances Cost of disallowances Cumulative cost of lower allowance rates 90% or more 9 338 $906,447.75 $906,447.75 85–89.9% 3 679 $1,820,940.90 $2,727,388.65 80–84.9% 1 38 $101,908.33 $2,829,296.98 50–79.9% 2 686 $1,839,713.49 $4,669,010.47 SOURCE: Unpublished tables provided by the Office of Disability and Income Support Programs, SSA. At the end of 2003, the average SSI benefit for the blind and disabled under age 65 was $446.97. Taking this as the average cost of a presumptive disability decision and assuming that reversed presumptive disability cases were paid for the full 6 months, or $2,681.80 each, moving the reversal bar from 10 percent to 15 percent (thus adding 679 presumptive disability cases) would have cost approximately $1.8 million in 2004 (Table 3). Moving it to 20 percent (adding another 38 cases) would have cost another $102,000. Note also that removing the two presumptive disability categories with reversal rates between 40 percent and 50 percent would have reduced costs by approximately $1.8 million. The committee also notes that the presumptive disability categories do not include conditions that consistently have high allowance rates, which therefore on equity grounds could be considered as candidates for inclusion as presumptive disability categories. As noted in the findings section, above, title SSI claims having one of 12 primary impairment codes had a 90 percent chance of approval in CY 2004. Of the 12, four correspond closely to existing presumptive disability categories. The other eight were types of malignancies. Currently, a claim presenting with one of the malignancy codes could only be granted presumptive disability status by a field office if the claimant were in hospice care and had less than six months to live. SSA could consider granting presumptive disability to claimants who know they have one of these cancers, and the diagnosis can be confirmed by contacting their treating physician. Dropping the bar to impairments with 85 percent approval rates would add four more cancers, chronic renal failure and childhood-origin psychosis in adults. An 80 percent allowance rate adds nine more categories—six cancers, mental retardation, other disorders of the spinal cord, and multiple body dysfunction. In 2004, the 12 impairment codes with 90 percent allowances rates accounted for 50,171 cases, 2.9 percent of all SSI claims made. Of these, less than 17,000 were actually granted presumptive disability. If the allowance rate cutoff were 85 percent, it would have been 18 codes and 70,486 cases, 4.0 percent of all SSI claims. Of these, 19,434 were
OCR for page 60
Improving the Social Security Disability Decision Process: Interim Report presumptive disability cases. If the cutoff were 80 percent, the number of codes would have increased to 31 and the number of cases to 139,072, or 7.9 percent of all SSI claims. Of these, 31,045 were presumptive disability cases. If SSA decides to adopt a uniform approval rate for presumptive disability cases, Table 3 needs to be expanded to include the costs of adding the cases with impairment codes that meet the established approval rate standard that are not now included. For example, if SSA decided to include conditions in which the approval rate averages 90 percent, the costs of presumptive disability cases in categories with lower approval rates would be eliminated, but the costs of new conditions meeting the 90 percent allowance rate criterion would be added (Table 4). Table 4 Approximate Costs of Adopting Different Allowance Rates to Determine Presumptive Disability Cases, CY 2004 Data PD category allowance rate Number of PD categories Number of disallowances Cost of paying disallowed cases Number of impairment codes Number of disallowances Cost of paying disallowed cases 90% or more 9 338 $906,447 12 1,182 $3,169,083 85–89.9% 3 679 $1,820,940 6 2,324 $6,232,503 80–84.9% 1 38 $101,908 *9 11,293 $30,285,567 50–79.9% 2 686 $1,839,713 48 96,929 $259,944,192 * Includes mental retardation, with an allowance rate of 80.1 percent in 2004. SOURCE: Unpublished tables provided by the Office of Disability and Income Support Programs, SSA. These calculations indicate that there are a number of conditions in which the probability is high that a claimant who alleges he or she has one will be allowed. The question the committee cannot answer is whether that probability should be 90 percent, 85 percent, 80 percent, or some other number. The answer depends on the tradeoff society wishes to make between helping groups of low-income persons with given probability of being determined disabled by SSA and the cost of paying for cases that end up being disallowed. The preceding discussion assumes that the goal of the presumptive disability policy is to identify early those who are likely to be determined disabled after the regular evaluation process and start paying them. This is consistent with congressional intent, namely, to enable claimants who have a high likelihood of being awarded benefits eventually to begin receiving benefits immediately. However, the result of the current system is that claimants with the same condition may be treated differently, depending on if they do or do not have sufficient medical evidence or an established diagnosis. And as we have discussed above, it will result in claimants with an equally severe but not explicitly categorized presumptive disability impairment being ineligible for presumptive disability status, if their impairment cannot be easily observed or verified. Other approaches are possible. SSA could decide to grant presumptive disability in cases where the claimant’s condition is worsening and he or she has little or no funds
OCR for page 61
Improving the Social Security Disability Decision Process: Interim Report or insurance to pay for medical treatment that might reduce the level of impairment or prevent it from worsening further. Another approach would be to stipulate that certain conditions are potentially so serious that presumptive disability is warranted, even if the award rates for those conditions are far less than 100 percent. Increasing Consistency in the Use of Presumptive Disability The percentage of SSI claims that are granted presumptive disability varies widely across SSA field offices and DDSs, a situation that SSA could improve. The percentage of field office presumptive disability cases ranges from 0.1 percent to 3.5 percent. The range for DDSs is from 0.6 percent to 34.6 percent. Field offices had presumptive disability cases in all 15 categories in 11 states. In two states, however, field offices used only one of the 15 categories, accounting for two cases each. Some categories with high allowance rates overall (i.e., highly likely to be allowed) were not used at all in some states, such as Down syndrome, birth weight less than 1,200 grams, ESRD, and ALS (Table 5). Table 5 Field Office Use of Presumptive Disability Categories for High-Allowance-Rate Impairments, CY 2004 Impairment Overall allowance rate* Number of states without PD cases in the category Down syndrome 97.6% 3 Birth weight less than 1,200 grams 98.0% 5 Chronic renal failure (ESRD) 88.8% 13 ALS 98.3% 40 *Overall means decisions on all title SSI disability applications, PD and non-PD. SOURCE: Unpublished tables provided by the Office of Disability and Income Support Programs, SSA. Recommendation 3–2. SSA should mandate use of the presumptive disability procedure by, and take other administrative steps to achieve more consistency among, the field offices and DDSs in the implementation of the presumptive disability policy. Currently, the use of the presumptive disability procedure is voluntary, which leads to large differences in practice from office to office that are not the result of inherent difficulties in assessing cases. If SSA has a presumptive disability process, and it is part of national social policy, it should be applied uniformly from place to place. This is primarily a matter for administrative policy to achieve.
OCR for page 62
Improving the Social Security Disability Decision Process: Interim Report Learning from Terminal Illness (TERI) Procedures PD is a policy to expedite payments. TERI (TERminal Illness) cases involve expedited decisions. According to the SSA program operations manual, “Cases where there is an indication of a terminal illness (TERI) are to be handled in an expeditious manner because of the sensitivity involved.” Teleservice and field office interviewers are supposed to be on the alert for potential TERI cases in certain situations, for example, when the claimant or other person—friend, family member, or medical provider—says the illness is terminal, the claimant is receiving hospice care, the claimant has AIDS, or any other situation in which the medical condition is untreatable and expected to end in death. According to SSA’s program operations manual, examples of potential TERI conditions include claimants: with a chronic dependence on a cardiopulmonary life-sustaining device; awaiting a heart, heart/lung, liver or bone marrow transplant (excluding kidney and corneal transplants); with chronic pulmonary or heart failure requiring continuous home oxygen and who is unable to care for personal needs; with a malignant disease (e.g., cancer) who is home confined or institutionalized, unable to care for personal needs and unresponsive to therapy; with diabetes and one or more of the following: multiple amputations due to diabetic gangrene; recurrent cardiovascular events (e.g., infarction or failure); or recurrent cerebrovascular events with neurological deficit; with chronic liver disease (e.g., cirrhosis or hepatitis) and a history of massive gastrointestinal hemorrhage; who is comatose for 30 days or more; or who is a newborn with a lethal or severe genetic or congenital defect. Recommendation 3–3. SSA should look at TERI procedures for lessons in making expedited decisions on cases that must meet specific medical criteria. These include uniform special procedures throughout the decision process that promote consistency as well as speed. TERI cases are subject to special procedures through which the case is expedited through every step of the disability determination process, including the appeals process and the payment process. They are also sent immediately from the field office to the DDS for development and quick decision. The issue is whether field office interviewers are able to identify TERI cases with some accuracy. This would involve determining the percentage of cases flagged as TERI cases that turn out to meet the criteria and, taking a sample of all cases, the percentage of cases that were not flagged that should have been. Differences between types of cases and from field office to field could be analyzed for better understanding of TERI decision making and ways to improve it, and also of ways to improve presumptive disability procedures.
OCR for page 63
Improving the Social Security Disability Decision Process: Interim Report AFTERWORD This interim report has addressed three of the 10 tasks that the committee is charged with addressing. It focuses on one important part of improving SSA’s disability decision process: the qualifications of medical personnel participating in the decision process. It also makes recommendations for improving the implementation of the presumptive disability policy. The committee is addressing other seven tasks in its next, and final, report, which is due in 2006. These are tasks 1 through 7 in Appendix A, which focus on ways to improve other important parts of the process: the medical criteria and procedures that are used to screen applicants for eligibility.
OCR for page 64
Improving the Social Security Disability Decision Process: Interim Report This page intentionally left blank.
Representative terms from entire chapter: