Disability Insurance (OASDI) program. The data in the CWHS are drawn from other master files—the Numident, MBR, and MEF—and from the IRS self-employed file. The CWHS also contains derived and constructed data elements that do not appear on any master file. Given the stringent restrictions on direct MEF access, the CWHS has become a substitute for the MEF in many instances. As of June 2000, access to the CWHS is very limited, but efforts are under way to make the file more widely available.

In its plans for a monitoring system SSA should include ad hoc studies on specific emerging policy issues as well as explore other questions that do not need continuing data. One example is a follow-up study of applicants for disability benefits to see whether some years later they are working (disabled or not), or not working (even if they were denied benefits). The ratio of denied who remain not working to the accepted who could be, or are, working, is some indication of the accuracy of the decisions. The total number of errors both ways could be some indication of the efficiency of the system. For example, Bound (1989) using data drawn from the 1972 and 1978 surveys of the disabled done by SSA, found that fewer than 50 percent of rejected male applicants work.

Another example is a study of employers as well as people with disabilities to develop information on employer tolerance in hiring workers with disabilities and on the willingness of employers to display the flexibility often required to deal with workers with disabilities. SSA should explore the experience of other agencies in conducting such surveys. The National Center for Health Statistics had conducted a survey of employers and the Medical Expenditures Panel Survey also surveys employers.

The workshops held by the committee and input from experts in the field led to a clear conclusion that key concepts in disability were subject to debate among scientists, policymakers, and disability interest groups. Comparisons of U.S. social legislation on disability and that of other nations, arguments about the role of the social and physical environment of a person in defining disability, and the impact of macroeconomic forces on self-identification as work disabled, all led the committee to concerns that the concepts and measurements of disability over time could (and perhaps should) undergo change.