Deinstitutionalization:  Research Shows Multiple 

Major Benefits of Community Placement.

Revised: March 05, 2003 .

            In the past 20 years, a body of literature has developed on deinstitutionalization of people with developmental disabilities.  It shows what happens to the quality of life of people with developmental disabilities when they move from large congregate care settings to community living.  (Craig & McCarver, 1984; Haney, 1988; Larson & Lakin, 1989 and 1991.)  This body of literature is remarkably consistent.  Without contradiction, it demonstrates that people are “better off” in most ways when they leave large congregate care settings for community living in small, family-scale homes.  Correspondingly, the satisfaction and perceptions of quality among parents and other family members rises.

            Deinstitutionalization is far from new, and very far from untested.  The graph below shows what has been happening to institutional populations in the United States since 1850.

150 Years of Institutional Care in America

1000s of People in State Operated Public Institutions

As the graph shows, the past 30 years have seen the population of America’s public institutions for people with mental retardation reduced from about 190,000 to under 45,000.  The process is continuing.  Nearly every state mental retardation authority has concluded that no human being, regardless of degree of disability, “needs” to be isolated, segregated, or grouped with hundreds of “similar” people. 

            Since nearly 150,000 people have already experienced the move from institution to community, there has been ample opportunity to study the phenomenon.  Many research groups have been involved in this work.  Some of my own work will be summarized below, in order to provide the Court with the detailed reasons for my opinion in the current case.

            The measurable benefits from moving to the community can be summarized. The central question of studies of the outcomes of community placement has been: “Are people better off, worse off, or about the same?”  The phrase “better off” inherently implies the notion of “quality of life.”  However, nearly all people have their own complex of factors that they believe contribute to “quality of life.”  Usually their beliefs are not explicit, but rather, they form an internal set of values and judgments that are not always clearly defined.  In this situation, the best available scientific approach is to address as many aspects of “quality of life” as are reliably measurable.  Some of the dimensions of “quality of life,” or outcomes, that social scientists know how to measure reasonably well include:

·        independence

·        productivity

·        integration

·        access to the places and rhythms of mainstream American life

·        access to services when needed

·        health

·        health care utilization

·        health care satisfaction

·        mental health

·        mental health care utilization

·   mental health care satisfaction

·        friendships

·        physical comfort

·        privacy

·        individualized treatment

·        freedom from excessive restraints (physical, chemical, and authoritarian)

·        respect for dignity and human rights by staff and others

·        support for choicemaking and learning to make choices

·        personal satisfaction with multiple aspects of life

·        satisfaction of the family members and “circles of friends” who care about the person

·        the overall “locus of control” of the pattern of life (by paid professionals and/or by the person and non-professional relatives, friends, and advocates); power, control, choice, self determination.

            When multiple aspects of quality of life, or outcomes, are measured in a social program, the results are likely to be “mixed.”  A given social intervention may improve peoples’ lives in some areas, while diminishing them in others, and leaving still other areas unchanged.  This is a typical result, for example, in the field of substance abuse treatment programs.

            However, the research literature on community versus institutional living has not been “mixed.”  Through the assessment of all of these quality of life dimensions, my research in 18 states, and the research of other scientists in America, has consistently shown strong benefits associated with community placement.  Furthermore, the results have been extremely powerful, in that improvements have been documented in nearly every measurable outcome dimension.  Research in other nations (Australia, Canada, Denmark, England, France, Ireland, the Netherlands, New Zealand, Norway, Sweden) has revealed remarkably consistent findings associated with institutional closure (Mansell, J., & Ericsson, K. (Eds.), 1996.  Deinstitutionalization and Community Living: Intellectual Disability Services in Britain, Scandinavia, and the USA. London: Chapman and Hall.)

            In the following pages, I will briefly summarize the results of some of the largest and longest lasting studies of deinstitutionalization outcomes yet conducted:  the Pennhurst Longitudinal Study (Pennsylvania), and the Mansfield Longitudinal Study (Connecticut).  These two studies are of special interest because both culminated in total closure of the institution, with nearly all residents moving to community settings.  At the end of this section, I will summarize other large scale studies of community placement processes in New Hampshire, New Jersey, North Carolina, Oklahoma, Kansas, and Indiana. 

            Finally, I will provide some detail about the work that has been in progress in California since 1994, and is continuing right now.  In this California work, my team will be visiting every one of the more than 2,300 people who moved out of Developmental Centers under the terms of the Coffelt settlement.  We are measuring all of the quality of life dimensions above, and are also visiting day programs, and surveying families.  At the time of this writing, we have completed visits with 900 people.  We expect the visits to be completed by March.  At that time, we will be able and willing to report the findings to the Court, including any subgroups such as people who left Fairview Developmental Center.

 Pennhurst Longitudinal Study Results:  Pennsylvania

            In the Pennhurst Longitudinal Study, I was asked by the United States Department of Health and Human Services to study the effects of the district court’s orders in Halderman v. Pennhurst, 446 F.Supp. 1295 (E.D. Pa. 1978).  This Order resulted in the transfer of all of the people living in a large state institution in Pennsylvania to small, supervised community living arrangements (CLAs) in the communities from which they originally came.  Since 1978, I and my colleagues have individually monitored the well-being of each of the plaintiff class members -- more than 1,700 people -- every year.  Following is a summary of the results of the study through 1992 (the last year in which I directly supervised the project).

A.  The People:

            1154 people lived at Pennhurst on the date of Judge Raymond J. Broderick’s historic Order of March 17, 1978.  My team immediately visited every one of those people, and collected information about characteristics, abilities, behavior, health, and service needs.  These people (and other Pennhurst class members) are still being tracked.  Every person is visited every year, and every family is sent a survey.  We know more about their quality of life over the past decade and a half than any other group of people with disabilities in history.  Their characteristics at the beginning, in 1978, were as follows:

B.  The Community Model:

            Three person Community Living Arrangements (CLAs) were the predominant program model in Pennsylvania at that time.  These were either detached houses or apartments, and almost all were existing housing stock.

·        Some shift staffing and some live-in;

·        (later became almost entirely shift);

·        90% non-profit provider entities;

·        A day program was arranged before placement for every person.

C.  Additional Court Protections:

·        Special low-caseload Case Managers (1 to 30);

·        Consistent format IHPs, with multiple reviews;

·        Requirements for independent monitoring (including the Pennhurst Study itself).

D.  Development Toward Increased Independence (See Adaptive Behavior Development Graph)

·        In addition to the information in the graph, later study produced these findings:

·        Average gain in self-care skills upon placement 9%;

·        3 years after placement 12%;

·        Most recent measurement, 1992, 14%.

 Pennhurst Longitudinal Study:  

Adaptive Behavior Development 1978-1986
 

            QUOTE from the Pennhurst Study report:  “The adaptive behavior growth displayed by people who have moved to CLAs under this court order is literally 10 times greater than the growth displayed by matched people who are still at Pennhurst.”

E.  Challenging Behavior:  The improvements are shown on the graph headed “Improvements in Challenging Behavior.”  Moreover, the following findings accrued in later years:

·        Average improvement in challenging behavior area upon placement 1%;

·        3 years after placement 3%;

·        Most recent measurement, 1992, 6%.

 Pennhurst Longitudinal Study:  

Improvements in Challenging Behavior

F.  Qualities of Environments

            Scales measured included Normalization, Individualization, the 1979 version of the standards of the Accreditation Council for Mental Retardation and Developmental Disabilities, and Physical Quality.  All scales improved sharply and significantly upon community placement.

G.  Consumer Satisfaction

            The Pennhurst Study included repeated interviews with 56 people who were able to communicate.  About a third of those people reliably said they were happy at Pennhurst, and wanted to stay there.  After the moves, about two thirds reliably said they were happy in their new community homes, and wanted to stay in them.  The number of people reporting satisfaction with aspects of life in the community was approximately double what was found in the institution.  There were no areas of decreased satisfaction over the entire course of the study.

H.  Amount of Service

            People who moved to the community began to receive more hours of developmentally oriented service per month than similar people who stayed at Pennhurst (225 hours per month versus 189 hours per month).  Hence we concluded that, on an overall index of amount of service, the movers were better off.

I.  Health and Health Care

            Indicators of health remained stable across the entire length of the study. Use of medications decreased slightly after community placement.

J.  Day Activities

            The proportion of people taking part in an active day program increased from about a third at the beginning of the study at Pennhurst to practically 100% at the end in the community.

K.  Family Satisfaction:  Initially, in 1979, the families of the people at Pennhurst were very highly satisfied with the institution.

 Pennhurst Longitudinal Study:  

Initial Family Satisfaction with the Pennhurst Institution  

 These same families were also quite strongly opposed to community placement.

 Pennhurst Longitudinal Study:  

Initial Family Attitudes About Community Placement (1979)  

             At the beginning, 83% of families reported satisfaction with Pennhurst, and 72% opposed movement to the community. 

            When surveyed later, however, these families had radically changed their views.  When asked whether they thought their relatives were happier or less happy since moving, the 1991 survey results were as shown below.

Pennhurst Longitudinal Study:

“Has Your Relative’s General Happiness Changed Since Moving to the Community?”

Not a single family believed their relative was less happy in the community.  Other related analyses showed that the strong prevailing attitude had shifted to support for community living.

Pennhurst Longitudinal Study:

Agreement With the Idea of Community Living, “Before and After”

The graph shows a dramatic change in attitudes about relatives living in the community.  This analysis was performed with 134 families who completed our surveys both in 1979 and again in 1984.

            Some of the most compelling findings in the study were the verbatim comments of the families after deinstitutionalization.  These comments frequently included expressions of surprise that they (the parents) had ever opposed community placement in the first place, coupled with surprise at the magnitude of improvements in the qualities of their loved ones’ lives.

            In a 1991 community survey of these same families, the results from 420 responding families concerning their overall satisfaction with community living were:

L.  Neighbor Acceptance

            About three fourths of neighbors never find out there is a group home in their neighborhood.  Of those who do know, only about half had any negative reactions, and those tended to vanish by a year to a year and a half.

M.  Costs

            The total public cost of serving the people who moved to Community Living Arrangements (CLAs) was significantly less than for the matched people still at Pennhurst (about $110 per day versus $129 per day at Pennhurst).  However, the fiscal burden shifted sharply from Federal to state sources for the people who went to CLAs.  Because Federal funds were being used for Pennhurst but not for CLAs, the state contributed about $57 per day for people at Pennhurst, and about $98 per day for people in CLAs.

            Today, community programs are just as able to obtain Federal Medicaid funds as are institutions, primarily through the Waiver program.  California is a major participant in these Waiver programs.  Hence, the old disincentive for states to support community services is gone.

N.  Synopsis of the Pennhurst Longitudinal Study

            The 5 years of the Pennhurst Study led to the conclusion that, on the average, the people deinstitutionalized under the Pennhurst court order were better off in practically every way measured.  For the people who moved from Pennhurst to small community residences, results were conclusive.

            Since the end of the Federal study, my group continued to monitor the well-being of the Pennhurst class members with state, local, and University support.  The positive outcomes have not only been maintained, they have continued to increase.  For example, the class members have become continually more and more independent since moving to community homes.  Year after year, their challenging behaviors have decreased.

            The Pennhurst research led me to try to replicate the study in other states, with and without Court involvement.  Replication is at the heart of science.  I have been fortunate in this regard.  The Pennhurst research has been replicated, and extended, and refined, in many other states.  Some of these will be described below, particularly the Mansfield Longitudinal Study in Connecticut, the Applied Research Project in New Hampshire, the Quality Assurance Project in Oklahoma, the Winfield Closure Study in Kansas, the Quality Tracking Project on Institutional Closures in Indiana, and the Quality Tracking Project related to the Coffelt settlement in California.

 Mansfield Longitudinal Study Results:  Connecticut

            In Connecticut, I and my associates followed 1,350 class members in CARC v. Thorne, No. H-78-653(TEC) (D. Conn.) to measure their well-being.  A tracking project of this kind was required by the consent decree entered in that case in 1985.  At the beginning of the study, most class members were in congregate care settings:  state institutions, state regional centers, and private nursing homes.  Between 1985 and April, 1990, approximately 600 persons received community placements under the consent decree.

            The CARC class was typical of the population of public institutions in America.  The average age was 46; 53% were male; their average adaptive behavior score before placement was 45 on a scale of 1 to 100; and their average score in challenging behavior was 79 on a scale of 1 to 100.  About 7% of the class had a hearing loss, 15% had no vision, 21% had seizures, and 20% had serious medical needs.  Approximately 69% of the persons who received community placements under the Court order were labeled severely or profoundly retarded, compared to 75% of the CARC class as a whole.  This showed that community placement included people with the most intense needs, rather than being restricted to people gifted with high ability levels.

            We completed three major research designs in Connecticut. The three studies were (1) changes in well-being from pre-move to post-move; (2) comparison of changes in well being among people who moved versus extremely similar (matched) people who did not move; and (3) surveys of parents, other next of kin, next friends, and guardians.

            The overall results of five years of study of the people deinstitutionalized in Connecticut are summarized in the table on the following page.

Individual Outcomes Associatiate with Deinstitutionalization

Connecticut’s Mansfield Class Members, 1985-1991

             This table shows, from three separate studies over a 5-year period, that the people who moved from institution to community were significantly better off in most of the dimensions that we knew how to measure.

            On the average, class members in CARC v. Thorne who received community living arrangements under the Court’s order made significant gains in adaptive behavior after placement in the community.  Moreover, people labeled profoundly retarded made the greatest proportional gains:  more than 28 percent (Stull, Conroy, & Lemanowicz, 1991).  The longitudinal research design also showed that people who moved to community settings decreased their challenging behaviors during the years of the study.

            In the area of social integration, movers began with 3.1 events per week of being in the presence of nonhandicapped peers when living at the institutions. They increased to 23.0 events per week in the community.  We also found an inverse relationship between the size of a community living setting and the degree of social integration experienced by its residents.  Residents of smaller homes experienced more opportunities to be in the presence of nonhandicapped persons (Conroy, 1992).

            In 1988-89, only 29 members of the CARC class had experienced even a short-term placement in a psychiatric facility, regional center or institution. This suggested that there was little or no need to maintain the institutional setting; it was often suggested that the institution should be maintained as the “backup” for people whom the community had difficulty handling.

            Members of the CARC class who received community placement received medical care of an appropriate frequency.  The average class member had seen a physician within the preceding 5 months, and a dentist within 4 months.  These frequencies compared favorably with figures for the general public.

            During the course of our studies in Connecticut, we saw the cost of care at the Mansfield institution rise to $290 per person per day, more than double the cost of services in the community.  Even at that funding level, the quality of life in the institution could not come close to matching what was available in the community for very similar people.  We did find that quality had improved measurably in the institution, such as in the areas of social integration and increased earnings, but in no area were the gains as large as they were for people who moved from institution to community.

            We also found that people who had resided in community settings during the entire course of the study had made significant gains in many areas of quality of life dimensions, including adaptive behavior, challenging behavior, social integration, productivity, earnings, satisfaction, and family satisfaction.

            Just as in the Pennhurst Study, the families of the CARC class members opposed community placement at first, but later shifted to strong support.  The graph below shows the extent of attitude change among all of the Mansfield parents whose adult children moved from institution to community, and who answered both of our mail surveys in 1986 and 1990. 

            From all the results reported above, I concluded that many improvements occurred in the lives of people receiving services in Connecticut.  By far the greatest improvements were seen among the people who moved from institution to community.

Brief Reviews of Other Relevant Community Placement Studies

            From 1981 onward, I have been involved in studying the process of deinstitutionalization in New Hampshire (Bradley, Conroy, Covert, & Feinstein, 1986; Conroy, Dickson, Wilczynski, Bohanan, & Burley, 1992).  In January of 1991, the Laconia State School and Training Center closed.  New Hampshire thus became the first state in which no citizen with a developmental disability lived in a state institution.

            All of the people who remained at Laconia, a facility with a long and honorable history, are now living in community settings.  Most of the last remaining group of people had serious behavioral or medical/health challenges. Up until the final year, many state officials appeared to believe that the institution would always be necessary for some people.  In the end, New Hampshire elected to demonstrate the opposite.  Even the most “medically fragile” people are now living and thriving in small, homelike settings.  This achievement has an important place in the history of developmental disabilities.  New Hampshire was the first state to show that communities can support all people, regardless of the severity of their disabilities.

            I am continuing to perform studies and evaluations in New Hampshire to the present day.  I see compelling evidence that even the most “difficult” people have been afforded the necessary supports in community settings.  The overall evaluation of my 16 years of research in New Hampshire can only be that all people can, and do, live in the community, and that their lives are indisputably far better on the average.  Case studies, “stories,” and living examples are readily available.  I would recommend contacting Mr. Donald Shumway, current director of the human services agency, for further information.

            In New Jersey, the Johnstone Training and Research Center closed in 1992.  I headed a 3 year project to track the former residents and the qualities of their lives.  Two thirds of the Johnstone people went to other state developmental centers (institutions).  One third went to community settings.  The conclusions of the research were that both groups had experienced improvements in many dimensions of quality, but the movers to community settings were by far the most improved.  Moreover, the care for the people who moved to other institutions wound up costing more than Johnstone, while the care for people who moved to community homes cost less than Johnstone.  I wrote that “Future closure planning should, according to this and past research, employ deinstitutionalization rather than reinstitutionalization as its primary strategy” (Conroy & Seiders, 1994).

            Since 1990, I have been working on a statewide quality assurance system in Oklahoma that covers 3,700 people -- everyone receiving intensive services in the state.  Among these 3,700 people are approximately 1,000 Class Members in the Homeward Bound v. Hissom Memorial Center litigation and consent agreement.  In 1995, I found and reported that the outcomes for the 520 “Focus Class Members” (those who lived at Hissom on or after May 2, 1985) were in many ways the strongest and most positive I had ever studied (Conroy, 1996).  These extraordinarily positive outcomes were associated with a “new” kind of community living arrangement.  Nearly all of the Focus Class Members went from Hissom, not into “group homes,” but rather into individually designed “supported living” situations.  Practically no one had more than two roommates, and most had only one or none.  This method of deinstitutionalization was unique in America.  It also turned out to be the most successful.  A summary table of the outcome results is shown below.

Hissom Outcomes Study Summary & Interpretation:

“Are Focus Class Members Better Off Now Than They Were Before?”

The table clearly shows that the people who left Hissom are better off in nearly every way measured, and worse of in no dimension.

            More recently, we surveyed the families of these same people by mail (Conroy, 1999).  Just as in the other studies, families told us that their feelings about community living had changed.

Oklahoma Outcomes Study of Hissom Class Members:

Family Feelings About Community Living, “At First” and “Now”

 The graph shows that only a handful of these 301 families are now opposed to community living for their relatives (a total of 3, to be exact).  We also asked these families to rate their relatives’ qualities of life “Then” and “Now,” with “Then” referring to the time when they lived at the Hissom institution.  The results were the most dramatic we had yet seen in any of our studies, and they are depicted in the following graph.

Each one of the differences between perceived quality “Then” and “Now” was statistically significant, and also very large in magnitude.  This was powerful evidence of the eventual satisfaction experienced by families of people who moved from institutions to community homes.

            I have also been tracking the well-being of more than 1,200 people with dual diagnosis in North Carolina.  These members of the Thomas S. class are people who had mental retardation, but also had either a psychiatric diagnosis or a brush with the law that resulted in placement into a psychiatric facility.  Hundreds of the Thomas S. class members have moved to new community homes.  Despite widespread misgivings about their potential behavior problems, they are doing extremely well in their new community homes, with no evidence of criminal activity or “recidivism.”  In fact, they have made such progress that I am now working with the state to suggest that the Court’s supervision might be relaxed.  The Thomas S. class members are more integrated, more satisfied, better served, more independent, receiving less medications, and much more likely to be working and earning money.  This project has strongly suggested that serious behavior “problems,” even criminal histories, need not prevent people from flourishing in well supervised community homes.  The placement process was so successful that the judge recently dismissed his own Order, concluding that all the original goals of the Thomas S. action had been achieved.

            My company was selected to track the process of closing the Winfield State Hospital in Kansas.  For the 200 people who moved from institution to community in that effort, we measured qualities of life before and after the move.  The following table summarizes the results of our 2 years of study.

Verbal Summary of Kansas Outcomes at Year One