An experimental investigation of an innovative community treatment model for persons with a dual diagnosis (DD/MI).

Habilitative services for persons with mental retardation who suffer from significant mental health problems are deficient in availability, accessibility, and adequacy. An active treatment case management model was developed and compared with the model of service delivery used by the involved agency for persons with dual diagnoses at pretreatment and at one, two, and three years of treatment. This study used a randomized experimental design, with participants assigned to two community treatment model to determine whether the active treatment model was more effective in increasing independent functioning skills and decreasing the maladaptive behaviors of participants. Results revealed increment improvement in treatment outcome with the active treatment model as compared to the traditional model of case management services. The active treatment model was more effective, with increased functional behaviors and decreased maladaptive behaviors. The active model showed promise in helping participants to move to less staff intensive day and residential settings. Implications for rehabilitation counseling practice and future applied research in rehabilitation were discussed.

As persons with mental retardation continue to move from institutional settings to the community, rehabilitation service providers are increasingly aware that flexible and innovative service alternatives are essential to address their varied needs. This growing awareness has revealed that persons with mental retardation are a heterogeneous group, requiring an array of habilitative programs delivered in a broad range of supportive environments (Coelho, 1990). One notable result of the shift in service delivery environments is that many persons with mental retardation are in need of community mental health services for psychiatric problems (Fletcher, 1988).

Persons with mental retardation develop mental illness or emotional disturbances about twice as frequently as persons not similarly labeled (Eaton & Menolascino, 1982; Menolascino & Stark, 1984; Menolascino & Potter, 1989). Estimates suggest that 20-39% of persons with mental retardation suffer from some form of diagnosable emotional disturbance compared to a general population's prevalence rate of 16-20% (Menolascino, 1989; Menolascino, Newman, & Stark, 1983; Reiss, 1990).

The complexity of a dual diagnosis of mental retardation and mental illness presents a persistent challenge to rehabilitation service delivery systems, requiring specialized services to meet appropriately the needs of these persons. Generally, persons who are among the dually diagnosed population represent a significant group of persons who are not only underserved, they are poorly served by the mental health service system (Campbell & Malone, 1991; Galligan, 1990). There are well-documented incidents of abuse, over-medication, and in many sectors of the mental health system, a general lack of interest in working for this population (Rinck, Guidry, & Calkins, 1989). Thus, the full inclusion of persons with mental retardation into the community is increasingly challenging the field of rehabilitation (Coelho & Kelley, 1990).

There is a need to develop innovative models for the provision of habilitative services to persons within the dually diagnosed population if their needs are to be appropriately met (Coelho & Kelley, 1989; Galligan, 1990). Services and supports in the community must consider psychiatric, medical, educational, vocational, residential and recreational needs. Traditional methods of treatment proven effective in treating either group separately (e.g., psychotherapy, behavior modification, psychotropic drugs, specialized residential and day programs and combined approaches) will not automatically work in treating the unique needs of persons who are dually diagnosed (Menolascino & Stark, 1984). There is no prevalent consensus on the merit of one treatment strategy over another (Goff, 1986). Perhaps the problem in treatment efficacy may rest, not in the strategies and techniques used, but in the extent and amount of active treatment provided.

The present study was designed to evaluate experimentally the effectiveness of an innovative intensive support services model for persons with a dual diagnosis of mental retardation and mental illness. This active treatment model was compared with the standard model of service delivery used by the involved agency for persons with dual diagnoses at pretreatment and at one, two, and three years of treatment. A purpose was to determine whether the active treatment model was more effective in increasing adaptive functioning skills and decreasing the maladaptive behaviors of participants.

Method

The study was conducted at a comprehensive, community mental health (CMH) agency serving about 1,160 adults with developmental disabilities and 3,441 adults with mental illness in a three-county catchment area in Michigan. The CMH agency directly operates or contracts services for community living arrangements for about 220 adults with developmental disabilities and 141 adults with mental illness in either group homes or foster homes. The agency also provides a variety of day programming services (i.e., adult daily living, prevocational work activity, psychiatric day treatment), along with supported employment and outpatient mental health services for about 569 persons with developmental disabilities and 486 persons with mental illness.

Announcements about the study were made throughout the community mental health system. All case managers serving adults labeled as mentally ill or developmentally disabled were notified. Material was also shared with administrative staff. Meetings were scheduled with the project staff, referral source, and the potential participant for all referrals who met the project's entry criteria. Also, die primary caregivers or guardians, depending on the status of the individual, were included in these meetings. At each meeting, a complete explanation of the project was carefully reviewed with all parties. Written consent was obtained to participate in the random assignment procedures and in the project. Pre-test data collection instruments were then administered.

Research Design

The experimental design for the study was a pretest-posttest control group design. Participants were stratified by high or low level of maladaptive behavior and then assigned randomly to either the active treatment model or the standard case management treatment model. The independent variable was the type of treatment. The dependent variables were a measure of adaptive functioning and two measures of maladaptive functioning.

Entry Criteria

All participants received services from the comprehensive community mental health agency. Participants were required to meet the following entrance criteria: (a) have an open case in the CMH system; (b) have a developmental disability with moderate to mild characteristics; (c) resided independently, with family or significant other, or in a community group home not operated directly by the CMH or contracted by the CMH; (d) be eighteen years of age or older; (e) have a DSM-III-R diagnosis of mental illness or behavioral complications concerning mental illness (e.g., changed sleep patterns, eating habits, emotional affect, mood/motivation, or increased confused thinking).

The entrance criteria were established to secure persons that the published literature and the involved agency have deemed to be at a high risk for institutionalization. These persons have typically received the least amount of planned support from the service delivery system, and yet in times of crisis, have required costly resources to maintain them in the community.

Participants

Fifty-seven persons were referred for participation in the study. Of this number, 47 persons (28 men, 19 women) met the entrance criteria and consented to participate in the study. Twenty-four persons were randomly assigned to the active treatment model and 23 persons to the traditional treatment model. One person in the active treatment died, leaving a final sample of 46 participants for analysis.

Measures

Three instruments were completed yearly on each participant. The data were collected by two trained mental health staff in the CMH, who interviewed participants or staff members able to provide the necessary information. Thus, the participant, the participant's own case manager, with staff familiar with the participant (e.g., day program and residential staff) assisted in the completion of assessments and returned them to the agency's Evaluation Office for analyses.

Intake Questionnaire. The questionnaire was comprised of 98 items within the following sections: demographic, family and advocate information, medical status, and service usage. This questionnaire was designed specially for the study and was completed by agency staff familiar with the participant.

AAMD Adaptive Behavior Scale. (Nihira, Foster, Shellhaas & Leland, 1975.) This third-party behavior-rating scale, designed for use with persons who are developmentally disabled, describes a person's degree of personal independence in daily living and his/her ability to meet the social demands of the environment.

The scale, assessing adaptive behavior and maladaptive behavior, is completed by a staff person familiar with the participant. Part One has 66 items across ten domain related to achieving independent living (e.g., independent functioning, economic activity, language development, domestic activity, self-direction, and socialization). Part Two focuses on maladaptive social and personal characteristics and contains 44 items in 14 domains (e.g., antisocial behavior, rebellious behavior, violent and destructive behavior). Scores for the 10 domains of the adaptive behavior section were summed to derive a total scale score. Scores from the 14 domains of the maladaptive section were also summed to derive a maladaptive behavior total score.

In the present study, total scores for Parts I and R of the AAMD instrument were computed. Cronbach's alphas for the summed Part I scores at pretreatment, and at one, two, and three years of treatment were .90, .88, .89 and .88, respectively; for the summed Part 11 scores, the alphas were .81, .80,.78 and .79. These results suggest that the summed scores have adequate internal consistency.

Michigan Maladaptive Behavior Scale. (Michigan Department of Mental Health, 1983.) This instrument has 20 behavior areas (e.g., Physical Assault, Self-Injurious Behavior, Pica). For each participant, the rater decides which of the behaviors are problems. For each problem behavior identified, the rater determined, based on one or more statements, the frequency of occurrence of the maladaptive behavior. The rater identified one of the five levels of intervention required to avert the occurrence of a maladaptive episode, which provides frequency and intensity to a maladaptive behavior. A total scale score was developed by adding the frequency rating to the level of intervention rating for each behavior described. All 20 scores were totaled. This sum was divided by the maximum rating to obtain a total score for each participant. Cronbach's alphas for the summed scores at pretreatment, and at one, two, and three years of treatment were .77, .80,.78 and .8 1, respectively. These results suggest that the summed scores have adequate internal consistency.

Treatment Conditions

Traditional Treatment

The community mental health agency has the authority to operate and administer a range of mental health services. Since services are centralized, it provides direct case management services to persons with developmental disabilities and persons with mental illness. The case manager provides a variety of direct treatment services (e.g., counseling, crisis intervention, and consultation) to individuals with developmental disabilities residing in a range of community placement options. Responsibilities include counseling, advocacy, individual program plan development and coordination, behavior management, and placement services. The case manager coordinates the development of treatment plans, programs, and provides consultation and assistance to service providers and staff. Case managers monitor individual treatment plans, services and maintain extensive case records and related reports. All services are coordinated and planned using an interdisciplinary team approach.

Active Treatment Model

Case management in this treatment model was similar to the traditional treatment model noted above, with planned differences in the frequency and intensity of participant involvement. The model was designed to provide for greater direct contact services with participants in their natural environment. Contact with each participant was expected to occur twice a week. This was considerably different from the traditional model's average of two contacts per month. It was anticipated that this intense contact would enable both the reviews of the physical surroundings and knowledge of immediate resources and deficits (e.g., food, clothing, financial resources) and facilitate greater knowledge and understanding of stressors. Actual settings have included day programs, group homes, family homes, stores, restaurants, and walking within the community.

Individualized treatment included: (a) intensive participant care coordination, management and advocacy with comprehensive service planning in all areas and across disciplines; b) individualized planning with each participant involved in the selection and planning of their own services; utilization of existing community resources with facilitation and support by staff, as needed; (d) training, education and support in acquisition of prosocial adaptive skills and functioning; (e) behavioral programming to reduce or eliminate maladaptive behaviors, consistent with agency policies; (f) development of measurable objectives for each participant and data-based decision making regarding progress; (g) on-site consultation, training and support for care providers in day and residential settings; (h) medication monitoring provided by the agency's psychiatrist; (i) access to 24-hour crisis intervention and support; and ) brief individual psychotherapy.

Staff

The active treatment team consisted of three mental health staff (2.5 FTE), two professionals (masters level training in rehabilitation counseling), and one paraprofessional. One professional provided services 50% of the time and the remainder in administrative functions. On the average, the caseload within the active treatment model was 7-10 participants. A typical caseload within the traditional model was one case manager for 35 participants.

Clinical experience for staff in the active treatment model ranged from five years to twelve years before the study. This range of clinical experience was also similar for staff in the traditional case management services.

On an average, the active treatment model provided direct contact services twice a week. Persons in the traditional case management services were provided direct contact once a month if the person was attending a day program. However, if the participant was attending supported employment services, not attending a day program, or attending a special education program, contacts averaged once a quarter.

Results

Demographics

Although the participants were randomly assigned, both groups were examined for differences on selected demographic variables. None of the demographic variables showed statistical significance between groups before treatment. The sample characteristics are presented in Table 1.

                Selected Demographic Characteristics

                                 Active             Traditional
Characteristic                   Treatment          Treatment

AGE (years)                         34                   33

SEX
  Male                              16                   12
  Female                             7                   11

RACE
  White                             22                   20
  Black                              1                    2
  Hispanic                           0                    1

MARITAL STATUS
  Never married                     22                   22
  Ever Married                       1                    1

LEVEL OF RETARDATION
  Mild                              14                   15
  Moderate                           9                    8

TIME SPENT IN INSTITUTION
  None                              17                   14
  1 to 5 years                       1                    2
  6 to 10 years                      1                    1
  11+ years                          4                    6

TYPE OF DAY PROGRAMMING
  No Day Program                     3                    5
  Adult Daily Living                 6                    3
  Work Activity Center              12                   11
  Adult Education                    0                    1
  Senior Day Center                  0                    1
  Public School                      2                    2

PERSONS WHO HAVE BEEN ARRESTED
  None                              17                   19
  1 to 2 times                       6                    2
  3 to 4 times                       0                    2

Sixty-three percent of the participants were labeled with mild mental retardation and 37% were moderate retardation. The participants' average age was 34 years (range=20 to 67, standard deviation [SD]=10.30). Most of the participants (67%) had never resided in an institution. However, 22% had resided within an institution for 11 years or more.

Adaptive Functioning

The AAMD Adaptive Functioning Scale was analyzed using a 2 X 4 (group by time) analysis of variance with repeated measures. Table 2 summarizes the analysis of adaptive functioning over time.

Table 2 shows a significant effect for treatment by time interaction, F(1,3)=5.76, p<.001). Table 3 shows that adaptive functioning increased most for the active treatment (199.35 to 211.86) while the traditional treatment group dropped (205.96 to 201.30). These results show the effectiveness of the active treatment model in increasing functional adaptive behavior, as contrasted with the traditional treatment model.

Maladaptive Functioning

The AAMD Maladaptive Behavior Scale was analyzed using a 2 X 4 (group by time) analysis of variance with repeated measures. Table 2 summarizes the analysis of maladaptive behaviors over time.

Repeated-Measures ANOVA

     Measure

AAMD-Adaptive
     Functioning      Treatment                   .12    .727
                      Time                       1.59    .194
                      Treatment X Time           5.76    .001

AAMD-Maladaptive
     Behavior         Treatment                  1.42    .239
                      Time                       6.80    .000
                      Treatment X Time           6.05     001
Michigan Maladaptive
Behavior Scale        Treatment                  2.68    .109
                      Time                       6.30    .001
                      Treatment X Time           5.57    .001

Note: Number of participant: Active Treatment-23;
Traditional Treatment-23.

Table 2 shows significant effects for both time [F(3) = 6.80, p<.001] and treatment by time interaction, F(1,3) = 6.05, p<.001 Table 3 shows that maladaptive behaviors decreased most for the active treatment (53.74 to 22.50), while the means for the traditional treatment group remained constant (53.43 to 53.00).

[TABULAR DATA OMITTED]

Similar significant effects were found for the Michigan Maladaptive Behavior scale. Table 2 shows significant effects for both time[F(3) = 6.30, p<.001] and treatment by time interaction, F(1,3) = 5.57, p. The traditional treatment remained constant (11.96 to 11.48); the active treatment showed decrease (12.65 to 6.45). Tables 2 and 3 summarize findings from that measure. These results show the effectiveness of the active treatment model in decreasing maladaptive behaviors, as contransted with the traditional model.

Day/Residential

Five participants in the active model moved to a residential setting that was less staff intensive in programming (e.g., group home to an apartment), and one participant moved to a more staff intensive living setting. Nine participants in the active model moved to less staff intensive day programming settings (e.g., adult activity to work activity program). Of this number, four participants entered supported employment settings.

Participants in the traditional model moved two persons into a less staff intensive residential setting, and four participants had to be placed in settings requiring more staffing. Day programming showed that four persons moved to less staff intensive day staff intensive day programming.

Psychiatric Inpatient Usage

During the study, participants used 4,848 hours of acute community inpatient psychiatric services and 635 days in state inpatient hospitalization. Usage by model showed that persons in the active model used 61% of the acute community inpatient psychiatric services and no participant required state psychiatric inpatient services. Persons in the active model were more likely to use planned respite/crisis services compared to persons in the traditional model of case management services. Persons in the traditional model used less acute community inpatient days and all of the state inpatient hospital days.

Discussion

Results of the present investigation revealed incremental improvement in treatment outcomes with the active treatment model as compared to the traditional model of case management services. The active treatment model was more effective, with increased functional behaviors and decreased maladaptive behaviors. Findings from the study also suggested that the active model tended to facilitate more participant movement toward less staff intensive day and residential programming, compared to the traditional case management model. Although participants in the active model used more acute short-term inpatient services, no participant had to be hospitalized in long-term state inpatient facilities.

The active treatment model was designed specifically to allow for the provision of services in the natural environments of home and work environments. The goal was to establish positive clinical relationships and to provide more intensive individualized service programming on an as needed basis. Participant empowerment, through active participation in their own objectives and life goals were emphasized.

This proactive approach was expected to allow in-depth knowledge and understanding of each participant (e.g., their behaviors, thought patterns, actions, stated perceptions). Knowledge gained would then aid case managers more accurately identify mental illness issues, such as detecting decompensations. This active approach provided weekly direct contacts between staff and participants. Thus, regular occasions for observation allowed, in many instances, detection of initial or mild symptoms and immediate initiation of treatment to alleviate potential crises. This type of active programming seemed to render the time, resources, and coordination necessary to elicit effective intervention results for participants.

Staff in the active model were expected to provide individualized functional teaching, where participants were provided repertoires of skills to help them make informed decisions. Participants were taught to recognize problem situations, accept responsibility for their behavior(s), and move toward autonomy, eventually requiring less or no professional service intervention. Therefore, participants were in a better position to handle new problem situations with skills learned from previous opportunities to problem solve with the active treatment staff. Functional training in adult daily living skills (e.g., budgeting, cooking, cleaning, hygiene) in the participant's home and work environments, also seemed to add to the increased levels of independence and adaptive functioning found over time. The traditional treatment model, in contrast, may not have helped individuals to recognize potential problems, but provide intervention after crises had occurred. This approach may also have perpetuated dependence on professional support for problem identification and problem solving.

Unification of agency services (i.e., bridging the gap between the developmental disabilities and the mental illness components) was another positive effect of the active treatment model. Through joint networking, consultation, and problem solving, greater understanding and cooperation emerged between components and effective treatment resulted for participants.

Serving persons who are dually diagnosed with an active treatment team model requires a substantial financial investment. The cost for operating such a model was about 40% higher than traditional case management services. However, the use of community psychiatric options are much less expensive compared to serving the same people in an institution. More important, it allowed participants to enjoy a richer life by living in their own community.

This was a pilot-demonstration study; the resulting sample sizes for both treatment models were kept to a minimum. Also, participants were persons determined to be at a higher risk for institutionalization by the mental health system. There is no guarantee that the same results would be achieved with persons not deemed to be at risk for hospitalization or with similar persons in other geographic areas. Further research with "at-risk" and other persons who are dually diagnosed are needed. Pending replication of these results, the findings should not be viewed as definitive.

In summary, the active treatment model served four vital functions: (1) Needed services were provided for persons who would have otherwise fallen between the cracks in the traditional service delivery system; (2) The mental health system was educated about the varied needs of the dually diagnosed and available service options; (3) Participants were empowered to become active participants in their services; (4) Deflection from institutionalized occurred.

Habilitative services for persons with mental retardation who suffer from significant mental health problems remain deficient in availability, accessibility, and adequacy. More studies, using innovative programming, with this population are sorely needed. A long term commitment to comparative field research on the mental health aspects of mental retardation will help in developing more effective methods of prevention and treatment. They need, and warrant, our noticeable support.

A major challenge to the field of rehabilitation is to develop innovative solutions to disability related problems faced by consumers in various service delivery systems. If the field is to continue to progress, innovative problem solutions that have proved their efficacy, compared to traditional solutions, must take place. Thus, professionals in rehabilitation must continually challenge the traditional (re)habilitation approaches and cultivate new models. The new models will then require vigorous dissemination efforts so that all persons experiencing the problem situation may benefit.

References

Campbell, M., & Malone, R.P. (1991). Mental retardation and psychiatric disorders. Hospital and Community Psychiatry, 42, 374-379. Coelho, R.J. (1990). Job satisfaction of staff in unionized and non-unionized community residences for persons with developmental disabilities. Journal of Rehabilitation, 56, 57-62. Coelho, R.J., & Kelly, P.S. (1989, November). An Innovative Community-Based Model for the Treatment of Persons with a Dual Diagnosis (DD/MI). Paper presented at the Michigan Rehabilitation Conference, Flint, MI. Coelho, R.J., & Kelley, P.S. (1990, March). Implementing Community Treatment for Persons with a Dual Diagnosis. Paper presented at the Annual Conference of the Ontario Association on Developmental Disabilities, London, Ontario, Canada. Day, K.A. (1985). Psychiatric disorders in the middle aged and elderly mentally handicapped. British Journal of Psychiatry, 147, 660-667. Eaton, L., & Menolascino, F.J. (1982). Psychiatric disorders in the mentally retarded: Types, problems and challenges. American Journal of Psychiatry, 139, 1297-1303. Fletcher, R.J. (1988). A county systems model: Comprehensive-services for the dually diagnosed. In J. Stark, F. Menolascino, M. Albarelli, & V. Gray (Eds.), Mental retardation and mental health: Classification, diagnosis, treatment, services. New York: Springer-Verlag. Galligan, B. (1990). Serving people who are dually diagnosed: A program evaluation. Mental Retardation, 28, 353-358. Goff, G.A. (1986). An overview of issues affecting services to persons who are mentally-retarded and mentally ill. Behavior Management Quarterly, 2, 3-7. Menolascino, F.J. (1989). Model services for treatment/ management of the mentally retarded-mentally ill. Community Mental Health Journal, 25, 145-155. Menolascino, F.J., & Potter, J.F. (1989). Mental illness in the elderly mentally retarded. Journal of Applied Gerontology, 8, 192-202. Menolascino, F.J., Newman, R., & Stark, J.A. (1983). Curative aspects of mental retardation: Biomedical and behavioral advances. Baltimore: Paul H. Brookes. Menolascino, F.J. & Stark, J.A. (1984). The handbook of mental illness in the mentally retarded. New York: Plenum. Michigan Department of Mental Health (1983). Public Mental Health Manual, Maladaptive Behavior Scale. Author. Nihira, K., Foster, R., Shellhaas, M. & Leland, H. (1974). American Association on Mental Deficiency Adaptive Behavior Scale. Washington, D.C.: American Association on Mental Deficiency. Reiss, S. (1990) Prevalence of dual diagnosis in community-based day programs in the chicago metropolitan area. AmericaniJournal of Mental Retardation, 94, 578-585. Rinck, C., Guidry, J. & Calkins, C.F. (1989). Review of states' practices on the use of psychotropic medication. American Journal of Mental Retardation, 93, 657-668.