The Social Security Administration (SSA) is engaged in redesigning its disability determination process for providing cash benefits and medical assistance to blind and disabled persons under the Social Security Disability Insurance (SSDI) program and the Supplemental Security Income (SSI) program (Title II and Title XVI of the Social Security Act). The agency has undertaken a multiyear research effort to develop and test the feasibility, validity, reliability, and practicality of the redesigned disability determination process before making any decision about its national implementation.
SSA asked the National Academy of Sciences (NAS) to review and provide advice on this research. One of the major areas for review is the ongoing independent, scientific review of the scope of work, design, and content of the Disability Evaluation Study (DES) and the conduct of the study by the chosen survey contractor. NAS has been asked to identify statistical design, methodological, and content concerns and to address other issues as they arise. (See Appendix A for the study mandate.) Consistent with this charge is the need for the committee to ascertain that the methods proposed for measuring disability in the DES, particularly the approach to one of the DES objectives described below, are indeed adequate and correct.
SSA's stated plans regarding ongoing monitoring of the size of the pool of people who might be potentially eligible for benefits under the SSDI and SSI programs are to use existing survey vehicles sponsored by other agencies to supplement a periodic DES. The candidate surveys would be viewed as monitoring tools between periodic comprehensive surveys conducted by SSA. Thus, a partnership between SSA and other federal data collection agencies would be forged to provide ongoing monitoring of this population. Such a plan needs an
initial assessment of the current practices and future challenges for measurement of disability by use of household surveys.
Across the various surveys now in place, diversity exists in conceptual definitions of disability, in the nature of reference periods used for recall, in reporting rules for answering disability questions, in modes of data collection, and in levels of thresholds used for construction of disability statistics. Furthermore, many of the measures have not undergone extensive pretesting, such as state-of-the-art cognitive interviewing methods. Too few methodological inquiries have compared alternative measurement approaches.
The interim reports of the Committee to Review the Social Security Administration's Disability Decision Process Research have provided recommendations to SSA on the survey design issues faced by the DES. Although this short-run guidance will be beneficial to the work of SSA, it does not attempt to address medium- and long-term issues in survey measurement of disability.
As a step toward exploring these issues, the Committee to Review the Social Security Administration's Disability Decision Process Research (hereafter referred to as “the committee”) convened on May 27–28, 1999, a workshop titled “Survey Measurement of Work Disability: Challenges for Survey Design and Method.” The committee believed that a focused discussion among a wide range of disability researchers and survey methodologists could identify unanswered questions about measurement and provide a framework for a long-term research agenda in this area for SSA and others in this field. Such a workshop would be helpful for the committee's review of the design and content of the DES and enhance its ability to make informed recommendations to SSA on the subject. It also could be of great value to SSA in anticipating problems of using data from other agencies' surveys to monitor the size of the pool of eligible people.
Workshop participants included members of the committee, a wide range of researchers in survey design and methods as well as in disability research, and other invited experts. The workshop, though focused on SSA's needs, drew the attention of a large number of disability researchers from other agencies, academia, and members of the public. 1 The workshop agenda and a list of participants are shown in Appendix B and Appendix C, respectively.
The committee organized the workshop through a planning subcommittee composed of Robert Groves, Chair, Monroe Berkowitz, Ronald Brookmeyer, William Kalsbeek, Alan Jette, and Dorothy Rice. Nancy Mathiowetz served as consultant to the Institute of Medicine on the development and organization of the workshop. The full committee reviewed the plans for the workshop and this report, and modifications were made in response to the comments received. Thus, the workshop reflects the collective thinking of the committee regarding the issues discussed.
OBJECTIVES AND FOCUS OF THE WORKSHOP
The objectives of this workshop were to:
better understand the conceptual issues related to currently existing measures of work disability, the measurement error properties of these questions, and the essential survey conditions that affect the measurement of disability;
aid the committee in advising SSA on methods for measuring work disability in the DES and on ways to facilitate crosswalks between the DES and data collected in other federal household surveys; and
identify a research agenda for SSA in survey measurement of work disability.
The workshop opened with a session in which the authors of the two background papers presented a “point-counterpoint” dialog.2 The remaining sessions for the day focused on measurement and methodological issues within the framework of specific conceptual and programmatic themes. Participants then identified and discussed issues pertaining to:
translation of various conceptual models of disability and the disablement process to valid and reliable questions in a survey and their ability to address SSA's disability programmatic requirements;
measurement error properties of existing measures of work disability and the survey conditions that affect these measures;
the ability to crosswalk among measures of disability collected in a variety of settings, such as the DES, and other ongoing federal data collection efforts; and
gaps in the current set of disability measures and the inherent problems in attempting to fill those gaps.
In the final session of the workshop, participants identified the key issues that surfaced during the discussions and that could provide a framework for long-term research to address the gaps in survey methods and measurement of work disability.
This report includes the commissioned papers written for the workshop and a summary of the group discussions flowing from the presentations during the subsequent sessions outlined in the agenda (Appendix B). The report concludes with a research agenda identified by the participants during the last day of the workshop. The summary is limited to the views and opinions of those participating in the workshop and reflects the concerns and areas of expertise of the workshop participants. As such, it does not provide a comprehensive review of the research and current status of survey measurement of work disability. The issues and themes of the workshop provided the unifying focus for the various presentations and discussions that flowed over the course of the day-and-a-half workshop.