Data Trends and Policy Program: The Demographic Status of Americans with MentalRetardation and Related Disabilities

Glenn T. Fujiura, Ph.D.; Kiyoshi Yamaki, M.S.; *David Braddock, Ph.D.

Statement of Problem and Background
Research Questions or Hypotheses
Data Collection Strategies
Progress to Date
Key Findings and Potential Implications

Statement of Problem and Background

In a preliminary analysis of the demography of family care for adults with mental retardation (MR) (Fujiura, 1998), the juxtaposition of three major national trends was summarized: 1) the dramatic increase in public sector spending for long-term care for persons with MR, 2) low to moderate growth in residential system capacity, and 3) a large, home-based population. The analysis suggested critical questions about the character of demographics among Americans with MR. These are not new observations; the burgeoning attention to residential transitions from the homes of aging caretakers reflects a broad recognition of the aging of the American population and the concomitant aging of family caretakers (Freedman et al., 1997; Heller & Factor, 1991). The study of national data identified over one-quarter of those living in family settings (excluding living with a spouse and in one's own residence) were in households headed by a family member 60 years of age or older. Another 35% are adults in the households of middle-aged caretakers for whom transition issues are near term considerations. The size of the cohort suggest significant and as yet unrealized demands on the states' service systems. Identification of the scope and character of the potential demand from impending transitions-from-home is a research task of considerable importance in light of contemporary fiscal constraints on long-term care expansion and extant waiting lists (Hayden, 1994).

We are in a period of great uncertainty and significant experimentation. One central implication of these undercurrents is the need to monitor changes in state and federal policy and evaluate their impact on the lives of older Americans with MR. This is the context of the proposed program and the implicit issue addressed in the NIDRR priority on analyses of national data sets - to inform policy by measuring, tracking, and reporting on the extent to which society is progressing in addressing the health care, education, and employment needs of persons with MR. During the years of great reform in our nation's MR policy, researchers and policy makers decried the absence of quality data on the size or characteristics of the population (e.g., Bruininks, 1979; Conley, 1973; President's Panel on Mental Retardation, 1962; Rowitz, 1984). Decades after the early policy-making efforts of the President's Panel on Mental Retardation, access to "hard" data - that is, information on demography and service outcomes - continues to be an issue. The most basic of questions on the size and characteristics of the population are the most frequently asked of researchers by legislators, policy-makers, and consumer advocates (National Institute on Disability and Rehabilitation Research, 1993). In fact, impetus for the landmark 1994 National Health Interview Survey (NHIS) Disability Supplement was in large part due to the interest of federal agencies in gaining access to greater policy relevant information on MR.
back to top

Research Questions or Hypotheses

The Data Trends and Policy Program is conducting secondary analyses to:
1. synthesize and report data from relevant national data sets;
2. evaluate and recommend changes in the major survey;
3. establish and distribute a policy brief series through the RRTC clearinghouse;
4. publish original research on key socio-economic and demographic trends in MR; and,
5. participate in policy and research forums and other information exchanges.
back to top


The central thrust of the Data Trends and Policy Program is an extensive and systematic program of secondary analysis focused on the 1994-1995 National Health Interview Survey - Disability Supplement (NHIS-DS), 13-years of the NHIS core and family surveys, and the 1983-93 Surveys of Income and Program Participation (SIPP). This is a significant program of research. The combined 1994 and 1995 NHIS-DS is the most comprehensive survey of non-institutionalized persons with MR and other disabilities ever conducted in the US, and is the first national survey to include people with disabilities of all ages. The NHIS-DS Phase II data includes information about approximately 40,000 people with disabilities. The SIPP is the second major nationally representative probability-based survey explicitly incorporating Americans with MR and other disabilities. The emphasis of the SIPP is on the demographics and economic status of households and individuals. Although surveys have been conducted by the Bureau of the Census annually since 1983, the SIPP has been little used in MR research (in large part due to the complexity of the file system). The SIPP is notable for its extensive employment and income data that until recently has not been evaluated in the field of MR (Yamaki, 1998).

Four other health care data systems are of potential value in explicating access to health care: the National Ambulatory Medical Care Survey (NAMCS, Centers for Disease Control), the National Hospital Ambulatory Medical Care Survey (NHAMCS, Centers for Disease Control), the National Hospital Discharge Survey (NHDS, Centers for Disease Control), and the Medical Expenditure Panel Survey (MEPS, Agency for Health Care Policy). While none of these data sets were designed for disability-specific analyses, they are rich in data on accessing various components of the American health care system. Diagnostic information is available in varying quality across these data sets. We anticipate heavy reliance on ICD-9 codes for subject selection.

The National Ambulatory Care data set provides basic event data on the provision and use of ambulatory medical care, including office visits to specialists. The survey began in 1974, and has been fielded annually since 1989. Trained interviewers gather data on a one-week period. In parallel fashion, the National Hospital Ambulatory Medical Care Survey produces statistics on U.S. hospital emergency departments and outpatient clinics; the survey has been conducted since 1992. The National Hospital Discharge Survey, conducted annually since 1965, rounds out the portrait of high risk events through the characterization of inpatient utilization. Finally, the Medical Care Expenditure Survey (formerly referred to as the National Medical Expenditure Survey) is a representative survey of health care use, expenditures, sources of payment and insurance for the U.S. population.

This is a cost-effective project due to the extensive resources and experience of key researchers. The program has direct access to a national disability demography data CDROM archive established at the DHD (the Center on Emergent Disability). Holdings include the NHIS (1983 through 1994), SIPP (1984, 1988, 1990, 1991, 1992), Ntl Ambulatory Medical Care (1989-94), Ntl Hospital Ambulatory Care (1992-94), Ntl Survey of Family Growth (1973-95), Ntl Hospital Discharge (1986-93), Ntl Pediatric Trauma Registry (1988-94), Ntl Maternal & Infant Health (1988 & 1991), and the Ntl Longitudinal Transition Study. The archive has a special Internet account with the Inter-University Consortium for Political and Social Research at the University of Michigan and annually downloads updated data files.

Progress to Date

The Data Trends and Policy Program is a series of discrete studies designed to measure, track, and report on major national population trends among persons with MR and their families. Each analysis seeks to evaluate demographic trends among Americans with MR.

In the program's first four years, four distinct data series were evaluated and analyses completed on three: (a) the 1994-1995 National Health Interview Survey Disability Supplement, (2) the 18 core and supplemental surveys of the National Health Interview Survey for the years 1983 through 2000; (3) the Survey of Income and Program Participation (SIPP); and (4) health care service data sets of the CDC (National Ambulatory Medical Care Survey, National Hospital Ambulatory Medical Care Survey; and 1986-94 National Hospital Discharge Survey).

The core questions addressed to date are described in the following sections. A substantial proportion of early project effort was devoted to evaluation of data set utility which in turn affected the priorities of the analyses. The primary limitation encountered has been the lack of adequate samples of events - health care access data in the National Health Interview Survey -DS - or, lack of adequate sample sizes of persons with MR/developmental disabilities (in the CDC health care access data sets). Thus, the focus of our work in Years 1-4 has been on the NHIS-DS, the NHIS core series, and the SIPP.

The areas addressed are:
1. employment status of Americans with MR and related developmental disabilities.
2. construct validity of MR and developmental disabilities.
3. demography of Americans with MR and their families
4. access to health care for Americans with MR
5. ethnic & racial variations in MR status

All work has been based on secondary analyses of the 1994-1995 National Health Interview Survey - Disability Supplement (NHIS-DS), longitudinal series of the NHIS core surveys, and the SIPP.
back to top

Key Findings and Potential Implications

Project 1 Summary

Year 1 (October 1998 to September 1999) activities focused on the first comprehensive portrait of employment and economic status of persons with MR and related developmental disabilities (MRDD) based on national data (US Bureau of the Census, Survey of Income and Program Participation). The analysis was completed, a manuscript developed, and published:

Yamaki, K. & Fujiura, G.T. (2002) Employment and income status of adults with developmental disabilities living in the community. Mental Retardation, 40, 132-141.

The Yamaki & Fujiura (2002) study found approximately 40% of the identified MRDD population was 45 years or older (19.6% aged 45-54; 18.8% aged 55-65). Unemployment rates, low earnings, and dependence on income benefit programs tended to be more pronounced among the oldest cohort. Major findings included: (1) the general impoverishment of the MRDD population generally, and (2) major variations in economic well-being between sub-groups within the DD population - in particular, between those with and without intellectual disabilities.

Project 2 Summary

The original intent of Year 2 (October 1999 to September 2000) activities was to focus on family demographics, with emphasis on aging among persons with MRDD and among aging caretakers. Methodological emphasis was on operationalizing elements of the National Health Interview Survey, Disability Supplement for use in family and MRDD analysis. Preliminary work focused on the impact of different screening protocols for developmental disability. For example, the impact of including mild learning problems without multiple functional limitations, separation of Alzheimer's from generalized lifelong learning deficits, etc.

However, our findings of major substantive differences in economic profile across the two sub-groups of DD defined by the presence or absence of an intellectual disability in the Year 1 study by Yamaki and Fujiura (for example, an individual with severe motoric impairments who does not have any learning problems) have focused our attention on the "validity" of the Federal DD definition. Thus, prior to addressing family demography issues, the year 2 analysis focused on using the NHIS-DS data, in particular service use and need data, to look at the internal consistency of the DD population. This represents a classical measurement approach to the "construct" of developmental disabilities. The analysis was completed, a manuscript developed, and submitted for publication (currently under review):

Fujiura, G.T. (under review). The Continuum of Intellectual Disability: Demographic Evidence for the "Forgotten Generation". Mental Retardation.

Study results were consistent with the general proposition of the "forgotten generation," first introduced by the Presidents Committee on Mental Retardation in 1999. Data indicated a sizable population of Americans with the characteristics of "mild" intellectual disability (ID) that is clearly distinguishable from the general US population as well as those with specific learning disabilities. The profile of skills and service needs among persons identified as having an ID was generally consistent with a continuum of functioning ranging from "mild" to "severe" MR. Study implications are discussed in terms of substantial numbers of Americans at the margins of current diagnostic conceptions of MR who have substantial commonality with those formally labeled "MR" in terms of their social and economic vulnerability.

Project 3 Summary

Year 3 (October 2000 to September 2001) activities were originally to be directed to an analysis of access to health care. Preliminary runs on the NHIS-D indicated sub-par response rates among the MR sub-sample (and all groups generally) to the supplemental questions (which focused on services). As a consequence, the MR sample size was adequate only for the most rudimentary summaries. As an alternative method for profiling health related issues, we began a series of longitudinal summaries of chronic health conditions within the MR population. The focus was on detection of changes in health conditions over the past 16 years. Only one condition - obesity - showed a marked increase among Americans with MR during this period. Thus, the Year 3 analysis was refocused to better understand the obesity profile in the MR population. The analysis was completed and a draft manuscript has been developed. We plan on submitting it for publication during Year 4:

Yamaki, K. (2002). National Trends in the overweight status of adults with mental retardation in the community. RRTC on Aging and Developmental Disabilities.

In the Yamaki analysis, national-level trends on overweight status among the population with MR was compared to that of the general population. Longitudinal data from the National Health Interview Survey, encompassing the years 1985 through 2000, were used. Results indicated that both mean Body Mass Index and the proportion of the total population that was overweight was higher among the MR population than in the general population across study years. Differences were more prominent among females and young adults and generally showed an upward trend over the years. The implications of the results were discussed in terms of health promotion and the prevention of secondary conditions as a priority in health services.

Project 4 Plans and Summary to Date

Year 4 (October 2001 to September 2002) activities focus on the demographic analyses of the size of the home-based population of Americans with an intellectual disability. The intent of the project is to estimate the potential impact of the aging of the caretakers of the baby-boomer generation. To date we have drawn preliminary estimates from the 1994-95 NHIS DS and will extend the analysis into the Year 2000 NHIS (though the sampling methodology has changed significantly since the 1996 survey.

In our preliminary analysis using MR definitions based on the Year 2 Project (which included mild ID), a substantial proportion of persons with MR (61%) were identified in the 1994-95 sample that lived in the familial home. We will further refine the breakouts to include the age of the head of household which we use as the proxy for the primary parent or guardian for those living at home.

Project 5 Plans

Year 5 (October 2002 to September 2003) activities focus on a longitudinal analysis of MRDD across racial and ethnic lines. The intent of the project is to estimate changes on basic indicators of social and economic status (primarily employment and poverty indices) across the 1983-2000 period.

* Previously on the grant, no longer on this project

back to research project list