Enhancing SSA’s Research Capacity
During the nearly six years that this committee met, it reviewed a large number of research reports, journal articles and government reports, and relevant internal documents and other unpublished reports provided by the Social Security Administration (SSA). It also heard presentations from SSA staff on various aspects of their work and progress made on the projects reviewed. Experts in the field addressed the committee during its meetings and also participated in two large workshops organized by the committee.
The committee’s recommendations in the preceding chapters and in its interim reports to SSA are intended to better inform public policy by developing a national data system for monitoring on a continuous basis the size and characteristics of the population eligible for Social Security disability benefits and enhancing research leading to improved assessment of work disability for the purpose of awarding benefits. The committee has recommended major research efforts, including research on the measurement of work disability in a survey context, evaluation of the role of the environment and vocational factors in determining work disability, evaluation of the functional capacity of applicants for disability benefits, and testing decision process models that emphasize rehabilitation and return to work. Such research cannot be accomplished without appropriate infrastructure and resources, in terms of both dollars and recruitment of qualified researchers. In the course of its study the committee noted several problems related to infrastructure and research capacity in SSA and going beyond a specific individual unit of SSA or the specific
subject matter under consideration. The successful reform of a disability decision process and the implementation of the national disability monitoring system depend on the resolution of these problems. The committee recognizes that the recommended enhancements would require substantial additional funds and qualified staff. This concluding chapter briefly addresses those issues.
As shown in the previous chapters, the number of disabled workers receiving Social Security Disability Insurance (SSDI) benefits and Supplemental Security Income (SSI) based on disability, as well as the costs of these programs, has grown substantially since the beginning of the programs. Continued growth is projected as the baby boom generation reaches the age of increased likelihood of disability. At the same time, disability policy has become more complex. Extensive research is needed to understand, estimate, and forecast growth to inform and guide public policy.
Over the years staff of the Office of Research, Evaluation and Statistics (ORES) in SSA has conducted a variety of excellent surveys and studies. Establishing and maintaining high-quality and relevant data systems for appropriate analysis and dissemination requires a sufficient and capable intramural research staff. The committee finds that there has been a loss of survey design and analytical capacity at the very time such work needs to be expanded. In the past two decades, downsizing has adversely affected both the ORES and the disability program (Institute for Health and Aging, 1997). A lesson learned from the experience with the National Study of Health and Activity (NSHA) is the importance of staffing to handle the issues that are critical in launching a large complex survey. The current impoverished research capability in SSA not only affects the timely analysis of data collected, but also leads to inability to anticipate important issues and respond to them. If not corrected, this situation will impair the ability of SSA to meet its policy needs in the twenty-first century.
The committee notes the limited resources allocated to all Social Security research activities. Two recent reports of the Social Security Advisory Board (SSAB, 1997, 1998) also noted the very small number of staff positions and budget amounts devoted to research and recommended that SSA increase its intramural and extramural research activities. A third report (Institute for Health and Aging, 1997) reviewed the mission, resources, and capabilities of SSA’s Office of Research, Evaluation, and Statistics and recommended that at least 50 new full-time positions be added to the ORES staff to strengthen the internal research and evaluation
capacity, to develop and support external resources for research, and to ensure adequate funding to support these programs. While these recommendations encompass all of SSA’s research activities and go beyond research in the disability program, the committee recognizes the need to revitalize and strengthen the research program of ORES and to encourage collaboration with other federal agencies in activities relevant to SSA. The committee fully endorses the recommendations made in these reports for increased resources and in-house capacity for research and commends SSA for its recent efforts to increase staff resources and research activities.
In response to these recommendations, SSA took some steps to increase staff levels in ORES from 132 positions in 1997 to an estimated 141 positions by 2000. Of this number, 99 in 1997 and 111 in 2000 were allocated to research evaluation and statistics. The remainder are distributed among publication activities, technological infrastructure, and management, administrative, and clerical functions. Clearly much more is needed to meet the demands for research and statistics in the coming years. SSA should ensure that an optimum mix of disciplines is represented on its staff. Some examples include survey methods, sampling statistics, economics, operational research, demography, epidemiologists, sociologists, cognitive psychologists, medicine, and the like.
Recommendation 7-1: The committee recommends that the intramural staff for disability research and statistics should be substantially expanded and diversified to implement the recommendations in this report.
In addition to the need for an expanded intramural research program, the committee believes that there is a major role for extramural research. A balanced program of intramural and extramural research is needed. “No amount of external research will replace the need for the agency to invest in the internal research capability, for it is essential in itself and inextricably linked with the capacity to implement and use an effective extramural program” (Institute for Health and Aging, 1997, p. 29). Moreover, an extramural research program places its own demands on the agency’s research staff. Oversight responsibility rests with the agency for careful evaluation of the work of external researchers to ensure the quality, adequacy, and appropriateness of the products, and for designing the approaches to testing and experimentation.
In addition, SSA has research grant authority under Section 1110 of the Social Security Act. Over the years, this authority has been the basis for a relatively small research grant program that has been managed by ORES. Grants were solicited for research in targeted areas, and in addi-
tion, investigator-initiated grants were peer reviewed and awarded. The funding of this research grant program has been erratic, with no funds allocated to the program during the past three decades.
Peer-reviewed extramural research programs have proved highly successful in the field of health services and clinical research. The Centers for Medicare and Medicaid Service (formerly the Health Care and Financing Administration), the Agency for Health Care Research and Quality, the National Institutes of Health, and the National Science Foundation have developed highly successful and sophisticated systems for review of investigator-initiated research in a wide variety of health areas. A similar, strong program is needed in the social insurance area and should be operated and managed by high-level qualified professional staff in ORES.
Recommendation 7-2: The committee recommends that the Social Security Administration (SSA) expand and diversify its extramural research program to include a balance of contracts, cooperative agreements, and investigator-initiated grants. This broadened research program would prepare SSA for the anticipated growth in the demands on the disability programs and help bring about the needed fundamental changes in its disability programs.
The committee notes, however, that although the grant authority has been unfunded in recent years, SSA has taken some steps in that direction by awarding cooperative agreements. Lacking adequate infrastructure at this time to operate an effective grant program, cooperative agreements with less demanding infrastructure could begin to serve some of the purposes similar to investigator-initiated research. Two such large agreements are the Disability Research Institute described earlier and the Retirement Research Consortium (RRC). These consortia draw researchers from several universities together. Their main goals are to foster research and evaluations, dissemination of information on retirement, and other SSA-related social policy including disability policy, training and education, and facilitating the use of SSA’s administrative data by outside researchers. To meet these goals, the centers perform many activities including research projects, policy briefs and working papers, annual conferences, and training. The RRC currently is composed of two, university-based, multidisciplinary centers, administratively based at Boston College and the University of Michigan.
SSA should view the ability to fund intramural research, external research—contracts, cooperative agreements, and grants—as separate tools to improve the functioning of the agency. Each can offer SSA leadership unique ways to learn of causes of external social and economic phe-
nomena that affect the applicant pool to SSDI, influences on how individuals make decisions about application to SSDI, the effectiveness of the processing of applications, the dynamic nature of eligibility, and what influences return to work among those eligible for SSDI. By judicious coordination of the three programs of research, SSA can greatly enhance management intelligence needed for assessing the desirability of change in policies.
Intramural research can most effectively be focused on internal information analysis, studying the effectiveness of administrative procedures in the SSDI program. In addition, intramural researchers can be statistical analysts of external data used to estimate key demand statistics for SSDI services. Finally, intramural researchers should supply key analysis of direct utility to SSA’s policymakers.
Research contracts are effectively used to collect well-specified data using standard techniques. For example, contracts might be used to provide ongoing estimates of key statistics of interest to SSA, collect data on an ongoing basis, or providing ongoing statistical support services. The value of a research contract is the assurance of quality and cost efficiency for ongoing work.
Cooperative agreements are best used when SSA has identified well-defined research products but there may be uncertainties about how best to obtain those products. With a cooperative agreement, as implied by the name, SSA staff can interact with external researchers to help shape methods and products throughout the agreement by working as a partner with these researchers. Thus, cooperative agreements seek new ideas from outside the agency for research information that has, at least, been sketched out prior to the agreement.
Research grants offer the greatest opportunity for innovative ideas but provide for little control by SSA management. They are reviewed by a set of peer scientists outside the agency. They are evaluated by the soundness of scientific thinking motivating them and the likelihood of advancement of understanding of problems facing SSA. SSA would define sets of key questions that it wanted to be addressed through the grant mechanism. Proposals would be initiated by external researchers. In comparing grants and cooperative agreements, grants are probably best used for high-risk, but high-payoff, domains of knowledge. If SSA exercised the grant mechanism, it is likely that real breakthroughs in the understanding of key population phenomena may be possible over time. These are the types of findings that could lead to new ways of administering the programs or new programs.
NEED FOR FUNDAMENTAL CHANGES
The previous chapters of this report make abundantly clear that SSA has been given a difficult task and dwindling resources to deal with it. The situation will get worse and not better in light of the anticipated growth in demands on the program as the baby boom generation reaches the age of increased likelihood of disabilities. In its recent reports the SSAB (2001, 2002) has reached similar conclusions and has recommended major rethinking of the disability program.
Little doubt exists that the current system is in need of improvement. It needs better understanding of the prevalence of disability in the population and the characteristics of this population, and better information about the job market, and about qualifications for jobs. The Department of Labor’s Dictionary of Occupational Titles (DOT) is no longer being updated, and as of now SSA has no replacement for the DOT, leaving a critical vacuum. The problems of Social Security’s disability decision process are deep and fundamental. This is not adequately reflected in the agency’s research agenda. Making small changes with the current system may not resolve the basic problems. Changes to O*NET (the Occupational Information Network), even if they are feasible, updated Listings of Impairments, and the like may help but will not necessarily solve the basic problems facing SSA. While the Listings can and should be updated in light of the changes in medical knowledge, methods to validate them are not yet in place. They need updating, however, even if we have no perfect instrument for their validation. Moreover, attempts to validate them will be confronted with the stark fact that so many persons who meet the Listings work at normal jobs in the national economy.
SSA must recognize that the present system for determining program eligibility may not be sustainable in the future and that it must think about different orientations and different ways in which the task of making these decisions is accomplished. It needs to have some mechanism to systematically give thought to these issues and initiate appropriate research. For example, SSA should initiate research on the costs and benefits of the current decision process and alternate innovative approaches. Without sufficient resources, however, SSA cannot accomplish this forward-looking agenda.
SSA recognized these problems in the early 1990s when it decided to rethink and fundamentally redesign the disability decision process. It stated that “the fragmented nature of the disability process is driven by and exacerbated by the fragmentation in SSA’s policymaking and policy issuance mechanisms. Policymaking authority rests in several organizations with few effective tools for ensuring consistent guidance to all disability decisionmakers. Different vehicles exist for conveying policy and
procedural guidance to decisionmakers at different levels in the process. … The organizational fragmentation of the disability process creates the perception that no one is in charge of it….” (SSA, 1994a, p. 11)
The SSAB (2001) also concluded that the disability policy and administrative infrastructure are weak and that constructive change and additional resources are required. It stated that “the problems with the administrative infrastructure begin at the top, where SSA’s current organizational structure diffuses responsibility over nearly every component of the agency. They continue throughout the disability system, where a fragmented and uncoordinated administrative arrangement makes consistency and fairness in decisionmaking difficult to achieve.”
“Problems in the area of policy are equally critical. For many years, disability policy has tended to be guided by court decisions and other pressures rather than by a well-thought-out concept of how the programs should be operating. Policy is articulated by too many voices with no single source of policy to which decisionmakers can turn for guidance and direction. Moreover it is inconsistent with the objectives of many disabled individuals to participate in the economic mainstream through employment” (SSAB, 2001, p. 29). In that report SSAB concluded that “… the issues facing the disability programs cannot be resolved without making fundamental changes. In our view these changes must be evaluated within the context of clear goals and objectives…” (p. 11).
The committee endorses the conclusions reached by SSA and the SSAB that underscored the need for fundamental change in the Social Security disability programs. SSA desperately needs a long-term, systematic research program to inform and guide (a) the anticipated growth in demands on SSA’s disability programs, and (b) improvements in the disability determination process.
In conclusion, the committee commends the SSA for initiating the daunting tasks of developing a national survey to improve the information base needed for monitoring and projecting the size and characteristics of the eligible population for guiding disability policy, and of attempting to overhaul the disability decision process to focus directly on developing new ways to assess the applicant’s functional ability or inability to work as a consequence of the medical impairment and to rely on these standardized functional measures to reach decisions. The ultimate goals of such a redesigned system were to make it simple to administer, to facilitate consistent application of rules at each decision level, to provide accurate and timely decisions, and to be perceived by the public as straightforward, understandable, and fair.
Although during the course of its study the committee identified much that needed changing, and it continues to be concerned about some of the decisions made by SSA, it recognizes, and is pleased, that SSA
made many modifications in response to its recommendations for improving the National Survey of Health and Activity. The committee believes that the blueprint for action it recommends for developing and implementing a disability monitoring system for Social Security programs, and for needed research relating to the redesign of the disability decision process will contribute toward a significantly improved and efficient system of measuring and monitoring work disability that will better inform public policy and serve the public. This blueprint is worthy of full funding and adequate staffing support by both the Congress and the executive branch of government.