A Comparison Between the Effectiveness of Scattered versus Centralized Apartment Housing Units For Substance Abusing Mentally Ill (SAMI) Rural Clients: Implications for Outpatient Care

Paulette Marie Gillig M.D., Ph.D., Amy Clark Jomantas, M.S.,

Hilton Rodriguez, M.D., Jan Rhoades, M.S.W., LSW, CCDCIIIE,

& Amy Kissling, S.W.A


This study examined two different rural housing apartment options for the SAMI population: scattered and centralized, with respect to relapse rates and treatment issues.

A centralized housing unit for SAMI clients was developed because of the high relapse rates that seriously mentally disabled SAMI clients were having in a rural community. Prior to the centralized housing program, some SAMI clients were residing in buildings with other persons abusing substances who were not in treatment. This had resulted in peer pressure to begin using substances again. It was hoped that a centralized housing unit, where all clients were in treatment at the same outpatient setting, might overcome some of the previous problems, and that it could do so at a cost lower than providing all services on site within a staff-structured milieu. For this community, and for most rural communities, the staffing costs and lack of staff availability have made it impossible to create more traditional programs with on site staff, structured milieu and on-site services.

The only requirement for entry into the centralized program was that the person appear eligible for certificationion as seriously mentally disabled and have a concurrent diagnosis of an alcohol or other drug disorder. The housing was funded by two state agencies with the requirement that all services be delivered at other sites, not at the housing unit, and that the housing program not focus on service delivery. Treatment services were provided through the local mental health and drug treatment agency.


Housing rehabilitation programs typically have incorporated a number of “wrap-around” services. For example, housing rehabilitation models for chemically dependent persons, that in the past have been funded by the United States Department of Housing and Urban Development, have recommended that treatment be available both for chemical dependency and for mental illness. HUD has recommended that a vocational/jobs training program be in place, as well as a literacy program. HUD has recommended that child care be available, and that agreements are in place with local schools for an adult education program. Additionally, HUD has recommended that medical services such as home health care be available.

The Drug Abuse Treatment Outcome Studies, a collaborative national research program for evaluating the effectiveness of community-based drug abuse treatment in the United States, has compared several treatment modalities for effectiveness. The DATOS study examined (1) outpatient methadone treatment (two years or longer), (2) long term residential (ideally four months to two years with an average length of stay of three months, therapeutic communities), (3) outpatient drug-free (twelve-step or similar), and (4) short-term inpatient (thirty days).

The DATOS Study found that the long term residential treatment programs with structured milieus worked best, especially where the clients stayed more than 3 months. Clients who did better had developed a better relationship with the counselor, and missed fewer treatment sessions. Clients with a background of hostility/violence and clients with a history of cocaine use had the most difficulty developing a therapeutic relationship with a counselor. Successful CD treatment programs had a wider range of services available. Clients were more satisfied with the program and more successful in programs where there were more health and social services available, and where clients with similar needs were served. In terms of client demographics, the average age at first treatment admission was 30 years nationally, and the average interval between initiation of regular drug use and first treatment was seven years. One conclusion of the DATOS study was that treatment approaches should focus on strategic interventions that recognize and address the diversity of client treatment histories in order to maximize effectiveness .

A study by Siegal et al (1997) investigated the role of case management in retaining clients in substance abuse treatment. The authors found, as expected, that there was a positive relationship between length of time spent in substance-abuse treatment and improved outcomes from the treatment. The authors found that case management services were sometimes a useful alternative to conventional aftercare. This approach worked best when clients were allowed to select the intervention that best suited their specific needs. They found that the opportunity to “self-match” held people in treatment who would otherwise have dropped out.

SAMI (Substance Abuse/Mental Illness) population: Up to 60% of clients with schizophrenia may use illicit drugs (Addington & Duchak, 1997). Drugs are reportedly used by this population to increase pleasure, to “get high” and to reduce depression. However, subjective effects of increased depression and increased positive symptoms also are reported by people with schizophrenia who use illicit drugs.

There are gender differences in histories and social supports between men and women with schizophrenia and chemical dependency (Westreich, Guidj, Galanter & Baird, 1997). The men were more likely to be hospitalized for schizophrenia and to have used substances of abuse other than alcohol, and the women were more likely to be admitted with affective disorders. Also, the women were less likely to be homeless, but were more likely to report having been crime victims.

A study of dually-diagnosed patients in specialized treatment for people with severe mental illness and substance use disorders (Jerrell & Ridgely, 1995) identified outcome measures that correlated well with client progress when measured at six month intervals. These included client self-reports of social adjustment, life satisfaction, psychiatric and substance abuse symptoms, and current substance use, as well as interviewers’ ratings of psychosocial functioning, and psychiatric symptoms. Outcome measures also included data on utilization of mental health treatment and support services, and data on clients personal income, use of medical services and contact with the criminal justice system.

In a study of co-morbidity among alcoholics who sought treatment for substance abuse, Tomasson and Vaglum (1995) found that over 70% of pure alcoholics and over 90% of polysubstance abusers had a comorbid psychiatric diagnosis. The most prevalent disorders were affective disorders (33%), anxiety (65%), antisocial personality disorder (28%) and psychosexual dysfunction (20%).

In another study, patients with social phobia exhibited comparative high rates of problem alcohol use (Page & Andrews, 1996).



The subjects were SAMI clients who had been accepted into the centralized housing program. The comparison group consisted of SAMI clients in scattered housing.


Procedures, assessment tools and consent forms were reviewed and approved by the Institutional Review Board of Wright State University (Dayton, Ohio) before the study was implemented, for the purpose of protection of research subjects. Data were collected from July 1, 2000 through February 28, 2001. In addition, seven of the eight clients in residence in the centralized apartment units during October and November 2000, were extensively interviewed. These interviews were completed in the participants’ apartments on one or two occasions for each participant. Two of the three authors (PG and/or HR or AJ) were present during each interview.

Interview #1 (30-45 minutes) consisted of a semi-structured history and mental status exam, the Mini-Mental Status Exam, the Brief Psychiatric Rating Scale (BPRS), and the Addiction Severity Index (ASI). Interview #2 (15-30 minutes) consisted of a relatively unstructured follow-up interview. The BPRS and ASI were completed at this time also. Interview data was collected confidentially, and is reported anonymously in such a way that client identities can be protected.

Demographic and treatment unit counts in various categories for all persons who had participated in either the scattered housing program (n=14) or the centralized unit program (n=24) during the duration of the study were obtained. All data obtained from charts was recorded in group form, without identification of individual clients.

“Success rates” for scattered vs. centralized units for clients with chemical dependency were operationally defined as a “global assessment of the client’s performance” in the housing program, and included the following parameters: planned discharge (vs. eviction or suicide); attendance at treatment appointments (vs. non-attendance); negative urine screens (vs. positive screens); decrease of substance use (vs. no change or increase). No vocational or work parameter was included in this rating.

The global ratings were done by the Director of the Alcohol and Drug Treatment Program, on the basis of case outcomes reported by staff.


Client Demographics

Fifty-four percent of the 38 clients served in both programs were female. Fifty-four percent of clients had a family size of one, 13% had a family size of two, 25% had a family size of three, and 8% had a family size of 4. Seventy-nine percent of clients were single, and 13% were African American, and 87% were Caucasian. There were no statistically significant differences in demographic representation between the scattered versus centralized housing groups.

“Success Rates”

The “success rate” for clients in scattered housing (n=14) to be 70%. The success rate for clients in the centralized building (n=24) was 50% (chi square NS)..

Amounts of Alcohol/Drug Services and Mental Health Services Delivered

With respect to units of alcohol and drug services delivered (15 minutes per unit), the mean number of alcohol and drug treatment units for clients in scattered housing units was 50.16. The mean number of alcohol and drug treatment units for clients in centralized housing units was 35.57 (chi square NS). With respect to units of mental health services delivered (15 minutes per unit), the mean number of mental health units per client for clients in scattered housing units was 51.49. The mean number of mental health units for clients in centralized housing units was 45.59 (chi square NS).

Detailed Client Interviews of centralized housing unit SAMI clients (n=7)

The following material was extrapolated from the confidential client interviews. Seven out of eight centralized unit clients agreed to an initial interview. At the time of the first interview, four of the participants had been in the housing program for six months or less, and three had been in the program for up to one year (despite the original plan for the housing to last no more than six months). All seven participants were asked questions to elicit their current level of symptoms and were scored according to the criteria mused on the Brief Psychiatric Rating Scale. Six of the seven participants reported mild to severe symptoms of emotional withdrawal, six reported anxiety, five reported depression, four reported somatic symptoms, three reported either guilt, guardedness and/or hallucinations, and one each reported conceptual disorganization, tension, hostility and/or blunted affect. In addition, one individual reported also receiving services from the local board of mental retardation/developmental disabilities. Four out of seven clients reported past use (prior to program) of alcohol, three reported use of crack cocaine, two reported use of marijuana, and two reported use of other unspecified substances. Most of the seven individuals denied use of substances during the time that they were in the program. However, although only one person reported a recent relapse while a program participant, many other clients noted that there had been some tenants who relapsed during the admission the centralized housing program. On the basis of the interviews and chart reviews, 5 of the 7 clients received the diagnosis of alcohol abuse or dependence, and two of polysubstance abuse. Four of the seven clients had bipolar affective disorder, two had major depression, two had schizophrenia, and one had anorexia nervosa. Two participants reported a history of either drug use of alcohol use in their family, but the other five individuals denied any family history of drug or alchol abuse. Four out of the seven participants reported serious medical problems that complicated their lives. One person was being evaluated for a possible carcinoma. One individual was being evaluated for sleep apnea. Another person reported a congenital malformation that was currently causing symptoms. Another reported “sleepwalking” as a child, but not currently. However, he had been evaluated for possible epilepsy. The participants also noted having problems with legal consequences related to their alcohol and other drug abuse, which created a major barrier to securing permanent housing, since the local public housing authorities had rules barring persons with recent criminal offenses from their housing programs. Of the participants, three of the seven reported criminal offenses related to their drug use (cocaine trafficking and driving under the influence).


The Interface Between Housing Policy and Treatment Issues for the SAMI Client in Recovery

Simply collecting rural SAMI clients into a centralized apartment dwelling did not improve outcomes in this study. In fact, the trend was in the opposite direction. In retrospect, the main issue that seemed to contribute to this outcome was that structured treatment guidelines within the housing facility could not be created due to the political position of the funding agency, which insisted on “housing as housing”. This stance ultimately compromised the original goal of finding a more cost-effective means of treating the SAMI population than total on-site provision of services.

According to grant requirements, clients in centralized housing units were to participate in outpatient treatment programs, but no provisions were permitted to structure the milieu of the housing facility by using professional on-site staff or limited programs. Instead, a manager who was a recovering alcohol/drug client but not a case manager was hired. Unfortunately, when the manager was no longer abstinent, this had an increased effect on the relapse rate of other clients in the centralized housing program, due to the (expected) group bond that had formed between them since they all lived in the same location. Thus, without a more structured setting, the “ad hoc” milieu that spontaneously developed served to detract from treatment rather than to enhance treatment.

Also, since no case manager had been dedicated to the centralized housing program as a whole (although each client had his/her own case manager), there was less structured outpatient follow-up than needed, and clients did not access services that were available to them. As a result, the outpatient programs provided much less service intensity that originally planned. In addition, expectations for employment had not been set, and as a result no clients were referred to vocational programs and almost none were employed. The scattered housing program tended to receive more units of service for both alcohol/drug and mental health treatment (although this trend was not statistically significant) because case management was more actively involved in reminding clients of appointments, transporting them to the clinics and so forth.

Program Changes That Resulted From the Study

As a result to the study, changes in the centralized housing program were made. The lease was eliminated, and a formal behavioral “contract” initiated, so that clients who were not compliant with guidelines could be asked to leave more readily, and so that the program was less likely to be perceived as “permanent” housing. Written payment agreements with clients were established. The criteria for referral to the centralized SAMI housing program were clarified. A more structured model, with clear rules and consequences was adopted. The housing program became part of the treatment milieu, with a more appropriate manager chosen and a case manager dedicated to all clients in the centralized housing program. Some on site support services, including alcohol and drug support groups, were instituted but provided on an outreach basis by staff already employed at the outpatient agency. The bulk of treatment continued to be provided at the outpatient setting, to save on costs and staffing. A policy was set regarding the handling of relapses that recognized the fact that persons in a rural area may not have seen “examples” of severe alcohol-drug related debilitation, and may remain in denial longer than other populations. However, persons who have relapsed were not allowed to remain in centralized housing until they have regained sobriety.


Addington, J., & Duchak, V. (1997).Reasons for substance use in schizophrenia. Acta Psychiatrica Scandinavia, 96, 329-33

Carling, P.J. Housing and supports for persons with mental illness; emerging approaches to research and practice. (1993). Hospital and Community Psychiatry, 44, 439-449

Drug Abuse Treatment Outcome Study (DATOS) (1999). National Development and Research Institute, www.datos.org

Jerrell, J.M., & Ridgely, M.S. (1995). Evaluating changes in symptoms and functioning of dually diagnosed clients in specialixed treatment. Psychiatric Services, 46, 233-6

McClellan, A.T., Luborsky, L., O’Brien, C.P., & Woody, G.E. (1980). An improved evaluation instrument for substance abuse patients: the Addiction Severity Index. Journal of Nervous and Menal Diseases,168, 26-33

Overall, J.E. & Gorham, D.R. (1988). The brief psychiatric rating scale (BPRS): recent developments in ascertainment and scaling. Psychopharmacology Bulletin, 24, 97-99

Rhoades, H.M., & Overall J.E. (1988). The semistructured BPRS interview and rating guide. Psychopharmacology Bulletin, 24, 101-104

Serenity Plus Inc. d.b.a. Serenity House Plus (1999). Admission Agreement, Continued Care Agreement, Guidelines for New Residents, Policy, Procedure & Guidelines, Admission Check List

Siegel, H.A., Rapp, R.C., Li, L., Saha, P., & Kirk, K.D. (1997) The role of case management in retaining clients in substance abuse treatment: an exploratory analysis. Journal of Drug Issues, 27, 821-831

Tomasson, K., Vaglum, P. (1995). A nationwide representative sample of treatment-seeking alcoholics: a study of psychiatric comorbidity. Acta Psychiatrica Scandinavia, 92, 378-85

Westreich, I., Guidj, P., Galanter, M., & Baird, D. (1997). Differences between men and women in dual-diagnosis treatment. American Journal of Addictions, 6, 311-17