U.S. Department of Health and Human Services
Design for a National Survey of Persons with Developmental Disabilities
Craig Thornton, Susan A. Stephens, Brian O. Burwell, James W. Conroy, Bradley K. Hill, and K. Charlie Lakin
This report was prepared under contract #100-88-0035 between the U.S. Department of Health and Human Services (HHS) and SysteMetrics/McGraw-Hill. For additional information, you may visit the DALTCP home page at http://aspe.hhs.gov/daltcp/home.htm or contact the Office of Disability, Aging and Long-Term Care Policy at HHS/ASPE/DALTCP, Room 424E, H.H Humphrey Building, 200 Independence Avenue, SW, Washington, DC 20201. The e-mail address is: webmaster.DALTCP@hhs.gov. The DALTCP Project Officer was Robert Clark.
A. RECOMMENDED SURVEY FOCUS
Government policy-makers and program managers need information about the characteristics of persons with mental retardation and developmental disabilities who are or could be eligible for participation in public programs. Such programs are increasingly treating persons with mental retardation and developmental disabilities as a single group where eligibility is being defined in terms of limitations in functioning rather than the presence of specific categorical conditions. This trend, combined with the move away from institution-based programs toward programs based in communities, has created an increasingly decentralized and complex service system. This dispersion of services has in turn led to an information gap that cannot be addressed by current government statistical programs. Specifically, information is needed about the number of persons who might meet various definitions of developmental disabilities; the needs, activities, and service use of these individuals; and the changing activities and needs of persons over the life cycle.
To meet this information need, we recommend that a national survey of persons with developmental disabilities be undertaken. This survey would support the development of a comprehensive population profile, including detailed information on functional capacity. Such a profile would provide accurate information about the characteristics of persons with developmental disabilities of all ages and all types of living arrangements, including persons living with their families, living independently, or living in community-based or other facilities.
The survey will be designed to yield national estimates. The survey will be based on a representative sample of the national population. Data from the survey can be used to make national estimates to support planning Federal policies and programs. The national data will also provide a reference point with which States can compare their data and will also provide a basis for making projections for subnational populations.
The survey will be designed as a representative cross-section of the population. We recommend that the survey be a cross-section of the United States population and that it provide a benchmark for assessing overall shifts in the number, characteristics and activities of persons with developmental disabilities.
B. RECOMMENDED SURVEY POPULATION AND SAMPLING APPROACH
The survey will represent all persons meeting the definition of developmental disabilities set out in the Developmental Disabilities Assistance and Bill of Rights Act Amendments of 1987. This definition defines developmental disabilities in terms of substantial chronic limitations in the ability to function in major life activities such as self-care, language, learning, mobility, self-direction, independent living, and economic self-sufficiency. The definition furthermore requires that these limitations manifest themselves prior to adulthood (specifically, age 22) and that they result in a need for long-term interdisciplinary services. This functioning-based definition departs from earlier definitions based on the presence of specific conditions such as mental retardation, cerebral palsy, epilepsy and autism.
The specific number of persons with developmental disabilities is unknown due to the general weakness of available prevalence evidence and the inherent difficulty in measuring the concept of substantial limitation set out in the definition. The Developmental Disabilities Act indicates that there are 2 million such persons. A review of the prevalence literature suggests a prevalence rate of 1.5 percent which would imply that there are about 3.6 million persons with developmental disabilities (including all ages). Finally, a recent U.S. Department of Health and Human Services working group reported that there were 1.2 million persons with severe limitations that were manifest prior to adulthood and required long-term interdisciplinary services. Overall, it is safe to assume that there are less than 4 million persons who would meet a categorical or functional definition of developmental disabilities and that somewhere around half of those persons would have severe limitations consistent with the more restrictive functional definitions.
The survey population will be identified on the basis of limitations in functioning, prior service use, and specific conditions. The sample selection procedures will screen a cross-section of the population to identify persons likely to be developmentally disabled on the basis of limitations in developmental disabilities, and specific conditions associated with developmental disabilities.
This three-part identification process is designed to include in the survey a broad range of persons who meet at least one of the major analytic or programmatic definitions of developmental disabilities. Additional information collected in the survey can then be used by researchers to identify persons meeting specific definitions of developmental disabilities.
The survey will cover persons of all ages. The survey will attempt to identify persons of all ages with developmental disabilities. However, given the difficulties in measuring the incidence of developmental disabilities among persons at the ends of the age distribution, the survey is likely to be most accurate for persons between the ages of 5 and 65.
Persons with mental illness will be included if they have limitations in functioning. While some persons with mental illness may meet the functional definition of developmental disabilities, they are often explicitly excluded from programs serving persons with developmental disabilities. We recommend that these persons be included in the survey, although no special steps are recommended to ensure that the sample will yield sufficient cases in this group to support separate analysis of this subpopulation.
Persons in institutions as well as community-based residential settings will be included. The survey will include a sample of the approximately 300,000 persons living in supervised facilities (which range from supervised apartments to nursing homes and large institutions). This supplemental sample will ensure that the proposed survey will provide a comprehensive perspective on persons with developmental disabilities. Given the billions of dollars spent annually on these persons and the policy interest in developing effective community-based programs, it is essential that the survey collect comparable data on all persons with developmental disabilities regardless of current living arrangement.
The survey will provide sufficient sample to analyze important subgroups of developmentally disabled persons. The survey will generate accurate estimates of the prevalence of most important subgroups of persons with developmental disabilities. It will also provide a basis for developing subgroup profiles and comparing policy-relevant subgroups such as persons with developmental disabilities who do not have mental retardation, and children, working-age adults, and elderly adults with developmental disabilities.
The National Health Interview Survey (NHIS) will be used as the basis for developing the survey sample. Because there is no registry of persons with developmental disabilities that could be used to select a nationally representative sample, it is necessary to screen households in an effort to identify persons with developmental disabilities. This is a potentially expensive process since it is estimated that less than 3 percent of the total population is developmentally disabled.
The most efficient avenue for conducting such a screening process is to add supplemental screening questions to the NHIS, a nationally representative health and disability survey that annually interviews approximately 132,000 persons. In order to insure an adequate sample of persons with developmental disabilities, we recommend that the screening questions be asked for two years of the annual NHIS. We estimate that almost 5,000 persons with developmental disabilities could be identified in this way.
Detailed data would be collected in a follow-up survey. A follow-up survey would be conducted with the persons identified through the NHIS screening in order to obtain detailed information about activities, service use, needs, and functional capacity. A separate follow-up is recommended rather than a longer more comprehensive NHIS supplement in order to collect the full range of information needed about persons with developmental disabilities and to improve the ability to identify and interview persons with developmental disabilities.
The sample identified from the NHIS will be supplemented to include persons in facilities. In order to include persons living in supervised facilities, it will be necessary to develop a supplemental sample frame, since such persons are generally excluded from the NHIS sample. Such a supplemental sample frame is feasible and is warranted by the need to collect information about the characteristics and experiences of persons with developmental disabilities who live in small group quarters and other facilities. The development of this supplemental sample poses several challenges, however, given the decentralized and often informal systems that provide residential services to persons with developmental disabilities.
C. CORE TOPICS TO BE ADDRESSED IN THE SURVEY
The following areas will be the primary focus of the survey:
D. POTENTIAL ANALYSIS TOPICS USING THE SURVEY DATA
The survey data will support analyses of many important program, policy, and cost-estimation issues. These include such issues as:
Alternative prevalence estimates. How many persons in the country would be defined as eligible for services under: (1) categorical definitions of developmental disabilities; (2) the definition contained in the Federal Developmental Disabilities Act; (3) the definition used by the Health Care Financing Administration programs that includes persons with mental retardation and related conditions, but excludes persons with mental illness; (4) definitions that focus on persons with severe limitations in functioning, such as persons with severe or profound retardation or similar level of functioning; and (5) definitions using alternative age-of-onset criteria.
Comparison of developmentally disabled persons with and without mental retardation. What are the different patterns of service use and needs between developmentally disabled persons with and without mental retardation? What implications do these differences have for the trend to combine services for all persons with developmental disabilities in a unified system?
Detailed analyses of functional limitations. What are the specific limitations persons face in the seven life areas listed in the functional definition of developmental disabilities? What kinds of services are persons with developmental disabilities using to overcome or cope with their limitations and are there unmet needs that could be addressed by government actions.
Comparison of persons in alternative living arrangements. How many of the persons with developmental disabilities who are living in the community have functioning limitations similar to persons in ICF/MR or large institutional facilities? What differences are there between these groups of persons living in different types of settings that might affect the effectiveness of those environments? (For example, activities, social relationships, use of services, and participation in community life.)
Comparison across groups defined by age and severity. How do the patterns of service needs and use differ for persons with developmental disabilities who are in different age groups; particularly children, working-age adults, and elderly adults? How do these patterns differ from persons with different levels of limitation?
Secondary conditions among persons with developmental disabilities. What are the specific health conditions that are correlated with developmental disabilities? How are these conditions related to living arrangements? Is there evidence of increased services and supports needed in certain settings to meet specific needs? To what extent and with what level of adequacy are the specialized needs of persons with medical and severe behavioral/emotional difficulties being met?
Analysis of access to services. How many persons with developmental disabilities are on waiting lists for residential, vocational or other services? What is the extent of unmet needs that can be inferred from the observed pattern of activities, limitations, and service use?
E. CURRENTLY AVAILABLE DATA ARE INADEQUATE FOR ADDRESSING THE CRITICAL ISSUES
In developing this design, data from many sources were examined in order to determine their adequacy for addressing the major information needs pertaining to developmental disabilities. These sources include:
While these data sources provide much available information they fail to provide a comprehensive picture of the population of persons with developmental disabilities either because of a lack of critical information about functional capacity, age of onset, or service use/need, or because peculiarities of the sampling process render the data set non-representative of the entire population.
This design was developed under a contract to Mathematica Policy Research, Inc. from the U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, Division of Long-Term Care and Aging Policy. The design team included researchers from Mathematica Policy Research, SysteMetrics/McGraw-Hill, Temple University, and the University of Minnesota. For additional information contact:
Robert Clark, Project Officer, U.S. Department of Health and Human Services, Room 410E, HHH Building, 200 Independence Avenue, S.W., Washington, D.C. 20201, 202-245-6443
Craig Thornton, Mathematica Policy Research, Inc., P.O. Box 2393, Princeton, New Jersey 08540, 609-275-6024