Despite the many limitations that can be noted about the federal commitment to data gathering on persons with mental retardation and related conditions in institutional settings, there have been a number of useful recent sources of national statistics, all conducted or substantially supported by Federal government agencies. The most significant and recent of these studies are reviewed below.
Since 1980 three censuses, or complete enumerations, have attempted to include all institutional and other residential settings for persons with mental retardation and related conditions. These were the Decennial Census of Population and Housing in 1980, the National Census of Residential Facilities (NCRF) for people with mental retardation in 1982 (Lakin, Hill & Bruininks, 1985), and the Inventory of Long-Term Care Places (ILTCP) in 1986 (Sirrocco, 1989). These are briefly discussed below.
Each ten years the Bureau of the Census conducts the Census on the entire population, regardless of residential setting, and publishes data on those living in places that it categorizes as institutions or noninstitutional group quarters. Specific questions vary somewhat from census to census, but always include demographic and basic housing items. Health questions, if included, appear in the "long" version of the census form, which only a sample of the population is asked to complete. A complete enumeration of persons in all types of institutions and special settings is conducted with each Decennial Census of Population and Housing. However, the purpose of the Decennial Census is reapportionment and statistics covering the entire population. Accordingly, the attempts on the part of the Bureau of the Census to systematically classify the types of facilities have been less thorough than the actual population count In all the 1980 Census identified almost 50,700 institutions with about 2.5 million residents. The largest single category of both facilities and residents was the "home for the aged" grouping, which includes nursing homes and personal care facilities for elderly persons. The category "mental hospitals and residential treatment centers" included about a quarter of a million people. The count in the 1980 Census of facilities for mental handicapped, individuals was 5,410 facilities and 149,421 residents. NCRF surveys of facilities for persons with mental retardation undertaken three years before and two years after the 1980 Census (i.e., in 1977 and 1982) found 11,025 and 15,633 facilities, respectively, and 247,796 and 243,669 residents with mental retardation, respectively (Lakin, Hill, & Bruininks, 1985). While the frequent small size and "typical household" nature of many group homes for persons with mental retardation may have contributed to this discrepancy, they cannot fully account for it. As noted the 1980 Census identified 5,410 "homes and schools for the mentally handicapped." In the 1982 mail census there were identified 5,164 facilities of 7 or more residents, and even this subset of mental retardation facilities in 1982 had populations that outnumbered the 1980 Census facility populations by 210,481 to 149,421. While the number of "facilities" for persons with mental retardation that are of a "typical household" size and/or nature (i.e., foster family care models) may have contributed to undercounting mental retardation facilities in the U.S. Census, other factors are obviously involved. One such factor could possibly be that many mental retardation facilities are misclassified as mental health facilities, nursing homes, or homes for persons with physical handicaps, although there is no evidence of the greater than expected number of these other facilities which would be expected to result from such misclassification.
The other two general census surveys of residential settings for persons with mental retardation and related conditions conducted during the 1980s (the 1982 NCRF and the 1986 ILTCP) are discussed in some detail in the next section of this paper which describes the methodology and limitations of the Institutional Population Component of the 1987 National Medical Expenditure Survey. Therefore, they are only briefly mentioned here. The University of Minnesota conducted a census type survey of all state licensed, contracted or operated residential facilities for people with mental retardation in the U.S. as of June 30, 1982 (15,633 facilities). The 1982 NCRF surveyed registries of facilities constructed within each state which were compiled from (a) state, regional, and county mental retardation program licensing agencies, state offices reimbursing contracted services, and other state or regional offices maintaining listings of licensed or contracted providers, (b) the 1982 Directory of Public Residential Facilities for the Mentally Retarded maintained by the National Association of Superintendents of Public Residential Facilities for the Mentally Retarded, and (c) facilities surveyed by the Center for Residential and Community Services (CRCS) in its earlier 1977 NCRF survey. As noted above the 1982 NCRF counted nearly 244,000 persons with mental retardation in facilities licensed or contracted to serve persons with mental retardation and related conditions. In addition to number of residents, facility level data were gathered on resident characteristics, facility administration and costs, resident movement and in other areas. The methods and findings of the 1982 NCRF, with some comparative findings from the 1977 NCRF, can be found in the survey's summary report (Lakin, Hill, & Bruininks, 1985).
In 1986, the National Center on Health Statistics (NCHS) conducted a first-time survey called the Inventory of Long-Term Care Places. The content and approach used for the ILTCP was largely based on the National Master Facility Inventory, a mail census of nursing and related care facilities, periodically updated by NCHS. However, the scope of facilities in the ILTCP was expanded to include facilities for persons with mental retardation. The ILTCP was designed specifically as the sampling frame for the Institutional Population Component of the 1987 National Medical Expenditure Survey, and will be discussed in considerable detail in that regard later in this report
In addition to the census surveys since 1980 there have been two sample based surveys including residents with mental retardation in different types of institutions have been conducted over the years. These include the National Nursing Home Survey and the National Medical Expenditure Survey.
The National Center for Health Statistics has conducted sample surveys of residents of long-term care facilities since 1963. The earlier surveys were known as the Resident Places Surveys (1963, 1964, and 1969). Later, NCHS initiated the National Nursing Home Survey as an ongoing data collection system. It is based on a facility sample, and a resident sample drawn from sampled facilities. The sample frame has been the National Master Facility Inventory. The National Nursing Home Survey has been conducted by NCHS three times, in 1973-74, 1977, and 1985. Each of the surveys has collected data that describe the facilities and data on a sample of the current residents, which includes typically 125-200 sample members indicated to have mental retardation or a related condition. In 1977 and again in 1985, the survey included an additional component that described people discharged from the nursing home during the previous calendar year, providing useful information on the outcomes of nursing home stays. Outcomes of discharge included whether sample members returned to a community residence (i.e., their own homes), or were transferred to another health care facility or hospital.
Although conducted prior to 1980, there are two sample based surveys deserving attention as precursors to the National Medical Expenditure Survey. In 1976, the Bureau of the Census conducted the Survey of Institutionalized Persons for the Department of Health and Human Services on persons in all kinds of institutions, including those for persons with mental retardation. The survey included detailed sample data on persons living in a wide range of long-term care facilities, including nursing homes, facilities for children, facilities for persons with physical handicaps, facilities for persons with mental illness, facilities for persons with mental retardation, and persons in chronic disease hospitals. Data were collected about the institution, sampled residents, and the resident's family. The survey findings have not been widely used, in part because of a significant flaw that was discovered in the sample frame, resulting in a substantial underrepresentation of persons in mental retardation and mental health facilities. However, this survey still represents the most recent data on persons in certain kinds of specialized long-term care facilities.
In 1978-1979 the University of Minnesota conducted a sample survey of 236 public and private residential facilities for persons with mental retardation and approximately 2,000 individual residents. The 1977 NORF served as the sample frame for that study. Detailed data were gathered on resident demographic, functional, medical and behavioral characteristics, programs and services received, daily experiences and relationships, and on facility characteristics and costs. The study also gathered extensive data on residents in movement among facilities and gathered useful data on persons providing care in the residential settings surveyed (Hauber, Bruininks, Wieck, Sigford, & Hill, 1981).
The Institutional Population Component of the National Medical Expenditure Survey, the subject of this report, was conducted in 1987 by the National Center for Health Services Research (now the U.S. Agency for Health Care Policy and Research). It represents the most recent effort to gather national data on populations of residential settings for persons with mental retardation and related conditions. The methodology and instruments used in this survey are described in some detail elsewhere (Edwards and Edwards, 1989). Generally, the National Medical Expenditure Survey was Intended to respond to the need for national information on access to medical care, health insurance, health and disability-related losses of productive activity, and utilization of and expenditures for a range of medical care including physician visits, other medical provider visits, hospitals stays, and drugs, equipment and supplies. it focused on gathering nationally representative statistics on health care utilization and expenditures in the United States. It was the third such effort since 1977. The two early studies, the National Medical Care Expenditure Survey (NMCES) and the National Medical Care Utilization and Expenditure Survey (NMCUES), were conducted in 1977 and 1980, respectively. The 1987 NMES survey was similar to these earlier studies in its gathering of a wide range of health care utilization and expenditures data on members of approximately 14,000 households in the United States. However, because of the rapidly growing expenditures for care in institutional and related settings under Medicaid and other public and private programs, an "Institutional Population Component," with large samples of nursing homes and mental retardation facilities and their residents were also drawn (3,347 and 3,618 current residents, respectively). In all during 1987 data were collected on samples of persons living in about 14,000 private households, 800 nursing and personal care homes, and 700 facilities for persons with mental retardation. Notably absent from the NMES Institutional Population Component were mental health facilities and people living in them.
In general, then, there have been a range of studies including persons with mental retardation and related conditions in institutional settings in recent years. A major strength of the National Medical Expenditure Survey was specifically its effort to provide comparable data on persons with different types of conditions in different types of settings. As more of these data become available, the benefits of the integrated inclusion of persons in a range of long-term care settings will undoubtedly prove useful. On the other hand, as discussed in the following pages, the effort to broaden the coverage of settings and individuals as represented in NMES also dramatically increases the challenges of doing so well.
This report contains only the data obtained in NMES on mental retardation facilities and their residents. Its purpose is to provide a summary of the findings obtained on mental retardation facilities and their residents in the first phase of the NMES (the only data available at this writing). As part of this analysis, the report also examines certain aspects of the NMES design and sample frame which effect the national estimates obtained.