I recently attended a conference on the PATH Housing First model. Nashville runs a pilot program of this model and during the discussions that ensued regarding the efficacy of the model, the agency charged with providing case managers to the participants in the housing projects ultimately concluded the same thing that this study did: "services may need to shift towards the provision of housing that most closely resembles that of the general population, for example independent scatter-site housing in the community."
I mention this because we all struggle mightily in trying to find the best solutions possible to address the most difficult of the "service resistant" homeless and slowly but surely we are gathering "best practice" techniques that seem to be working. It is a long, expensive and often difficult endeavor but it does appear we are gaining ground.
| Journal | Community Mental Health Journal |
| Publisher | : Springer Netherlands |
| Subject Collection | Behavioral Science |
| Saturday, March 31, 2007 |
Community integration of persons with psychiatric disabilities has been studied for several decades. While there is general consensus that services for individuals with psychiatric disabilities should foster community integration, the numerous and diverse service interventions, and a variety of research outcomes suggests that this concept is wanting for a clearly articulated conceptual framework. At its inception, community integration was narrowly defined as residential tenure outside of institutional settings, such as psychiatric hospitals or other treatment facilities (Rosenblatt & Mayer, 1974). The definition has subsequently been expanded to refer more generally to the extent to which an individual lives, participates, and socializes in his/her community (Wong & Solomon, 2002).
There has been a recent emphasis on this broader, more comprehensive view of integration as a multi-dimension
Researchers that have focused on investigating community integration find that persons with psychiatric disabilities report fairly low to moderate levels of community integration and that various client, residential facility, and community factors can influence the degree to which individuals are integrated. In a landmark study examining possible determinants of integration, Segal and Aviram (1978) found that social support as a community factor had the greatest influence on the level of participation in community activities for individuals with mental illness residing in board-and-care homes. Other studies have confirmed that greater community acceptance and lower levels of rejection are associated with increased integration (Nelson, Hall, Squire, & Walsch-Bowers, 1992; Segal & Aviram, 1978; Sherman, Frenkel, & Newman, 1986). City population density has been yet another community factor identified as a predictor of integration (Kruzich, 1985).
Characteristics
Studies of individual characteristics
Wong and Solomon’s (2002) conceptual model for studying the determinants of community integration of persons with psychiatric disabilities includes personal and program factors that potentially influence individuals’ levels of community integration. Personal factors are individual characteristics
Using a similar approach to community integration, Prince and Prince (2002) examined the impact of stigma on integration among clients of Assertive Community Treatment (ACT). Consumers’ greater psychosocial functioning and perceived community support were found to be related to both increased physical and psychological integration. Conversely, perceived stigma was most strongly and negatively associated with psychological integration, such that greater stigma was associated with a decreased sense of belonging in the community. While several studies have used this broader view of community integration, to our knowledge, only Prince and Prince have focused on investigating how various predictors influence the different dimensions of integration.
The study described here tests several components of Wong and Solomon’s (2002) model of community integration. It evaluates the three domains of community integration among a sample of persons with psychiatric disabilities who participated in a study in which they were assigned randomly to one of two groups. One of these settings (the experimental condition) was a program that implements a Housing First approach wherein consumers live in their own apartments (scatter-site) and support services are offered based on client choice. Apartments are located within normal residential buildings and are scattered throughout city neighborhoods. Services are provided flexibly by ACT teams and are located off-site (Tsemberis & Asmussen, 1999). Receiving and maintaining housing is not contingent on consumers’ sobriety or participation in psychiatric or substance use treatment. The other residential settings (services-as-us
This study was conducted during the last panel of an ongoing four-year project that examined various client outcomes including residential stability, substance use, mental health, and hospitalization
The analysis is driven by two sets of hypotheses derived from the aforementioned research evidence and clinical observations. The first focuses on identifying the components of integration, and the second on identifying possible determinants.
| 1. | Life satisfaction, perceived neighborhood cohesion, psychological well-being, self-actualizat |
| 2. | Organizational participation, activities of daily living (ADL), and participation in leisure activities would load onto a “physical” factor. |
| 3. | Satisfaction with social support and size of neighborhood social network would load onto a “social integration” factor. |
| 1. | Older age, higher psychopathology |
| 2. | Experimental status (normalized scatter-site housing), consumer choice, and utilization of psychiatric and substance use treatment services would be associated with greater community integration. |
| 3. | Stigma would be associated with decreased community integration. |
Two hundred and twenty five participants living in New York City were included in the original sample. Fifty-six percent of the sample (n = 126) were assigned to the services as usual condition (control) and typically resided in housing arrangements such as drop-in centers, SROs, community residences, and institutions throughout the study period. Participants assigned to the experimental condition (n = 99, 44%) were referred to a program that used a Housing First approach, providing participants with independent scatter-site apartments without prerequisites for treatment and sobriety. The control group was purposely over-sampled because these participants were expected to be more likely to become lost due to attrition. The sample was comprised of two subgroups: An original street sample of 157 participants who met the following eligibility criteria for the study: (1) spent 15 out of the last 30 days on the street or in other public places (not including shelters); (2) history of homelessness over the past 6 months; and (3) had an Axis I diagnosis of severe mental illness. The second group of 68 individuals was recruited from two state psychiatric hospitals. These individuals met the same entry criteria for homelessness and mental illness as the street sample prior to hospitalization
In all, 183 participants (81%) completed the final (48-month) interview and were included in these analyses. Eighty-two (44.8%) of the participants had been randomized into the experimental group (Housing First) and the remaining 101 (55.2%) had been randomized into the control group (services-as-us
After completing their baseline interviews, participants were interviewed every 6 months. Data collection for the larger 48-month project in which this sub-study was embedded began in December 1997 and ended in April 2003. During each interim 6-month period, 5-min monthly telephone calls were conducted to maintain contact with participants and establish their whereabouts. Participants were paid for all interviews. Six-month interviews were conducted in a variety of locations including the research office, the participant’s apartment/resid
Outcome variables
Five subscales were selected from the Wisconsin Quality of Life (Becker, Diamond, & Sainfort, 1993) as representative of components of community integration. These subscales were: (a) average life satisfaction, (b) psychological well-being, (c) ADL, (d) social and leisure activities, and (e) satisfaction with social support.
Average Life Satisfaction consisted of the mean of a 10-item Likert scale, with 1 = very dissatisfied and 7 = very satisfied. The scale is weighted by a measure of importance of each life domain. Internal consistency was high (n = 126, α = .82). Examples of items are: How satisfied or dissatisfied are you with your neighborhood? and how satisfied or dissatisfied are you with your housing?
ADL consisted of the weighted mean of a six-item measure of activities completed during the past week, such as going shopping, doing laundry, or going for a ride in a bus or car.
Social and Leisure Activities consisted of the mean of a 10-item measure of activities participated in during the past week such as going to a social group, a church, synagogue or mosque, playing a sport or going for a walk.
Psychological Well-Being consisted of the mean of a 10-item scale, with 1 = yes and 0 = no. Examples of items are: Have you felt depressed? Have you felt unhappy? and Have you felt that things went your way?
Satisfaction with Social Support consisted of the weighted mean of a 5-item Likert scale, with 1 = Very dissatisfied and 7 = Very satisfied. Internal consistency was moderate (n = 140, α = .54.). Examples of items are: How satisfied or dissatisfied are you with how you get along with your friends? and How satisfied or dissatisfied are you with your relationships with your family?
Five additional measures were also used as outcome variables representative of various aspects of community integration: (1) neighborhood cohesion, (2) self-actualizat
Neighborhood Cohesion (Bruckner, 1988) consisted of the mean of an 18-item Likert scale, with 1 = Strongly disagree and 5 = Strongly agree (n = 147, α = .88). Examples of items are: I feel like I belong to this neighborhood and I believe my neighbors would help me in an emergency.
Self-Actualizat
Lehman Quality of Life (1988). Each participant was asked to respond to the single question: How do you feel about your life in general? A score was obtained by taking the mean of this item which was rated twice, once at the beginning and once at the end of the interview. Scores ranged from 1 = Terrible to 7 = Delighted.
Organizational Participation. (Shinn, Personal Communication) Participants were asked to rate how often they attended each of four different types of organizations: religious/spiri
Network Members from Neighborhood: A modified version of a social network measure from Barrera (1981) was used. Participants responded to questions probing whether any members of the respondent’s social network fit certain criteria, either positive or negative. Items inquired whether there were persons to whom the participant could speak with about their problems, who might lend them money, or who were critical of the participant. The number of individuals who fit these criteria from the participants’ neighborhood were summed to yield a total score for the number of social network members from their neighborhood.
Predictor variables
Five “program domain” variables (Wong and Solomon, 2002) were chosen as predictors of community integration: (1) housing environment (assessed as program assignment), (2) consumer choice, (3) participation in substance use treatment, (4) participation in psychiatric treatment, (5) stigma. For the housing environment domain, program assignment was used as an indicator of normalization of housing, and perceived stigma as an indicator of degree of community support. For the domain of behavioral environment, perceived choice was used as a measure of independence, and utilization of mental health and substance use treatment services as a gauge for service availability.
The measure assessing consumer choice was modified from Srebnik, Livingston, Gordon, and King (1995) so that participants were asked to indicate their perceived level of choice for items assessing housing-related
Substance use treatment was measured using a modified shorter version of the Treatment Services Inventory (McLellan et al., 1992). In the interview, participants were asked if they received any substance use treatment during the past 2 weeks. Drug and alcohol treatment services included the average of a 7-item measure consisting of questions such as whether the participant had received treatment in a detox program or other program, consulted with a counselor to talk about substance problems, and attended AA, NA, or any other self-help group.
Psychiatric treatment was measured using a modified shorter version of the Treatment Services Inventory (McLellan et al., 1992). In the interview, participants were asked if they received any psychiatric treatment during the past 2 weeks. This included the average of a 7-item measure consisting of questions such as whether the participant had stayed overnight in a psychiatric hospital, attended a day treatment center, or made visits to a doctor or nurse to discuss emotional problems.
Stigma (Link, 1987) consisted of the mean of a 15-item Likert scale (n = 105, α = .75). with 1 = few and 4 = most and was used as a response to items such as: Members of my community look down on people who were once mental patients, People think less of a person who has been in a psychiatric hospital, and People feel it is all right to make fun of a mentally ill person.
Finally, a series of individual covariates drawn from the literature and selected for inclusion in the analysis were age, education, sex, level of psychopathology
After standardizing the outcome variables comprising community integration, an exploratory principal components factor analysis with Kaiser normalization was performed on the measures using data from the 183 participants. Principal components analysis was used for the initial factor extraction. Four factors obtained eigenvalues greater than 1.0, and these factors accounted for 19.92%, 14.65%, 12.04%, and 10.69% of the variance, respectively, totaling 57.3%. Three of these factors supported the hypothesized structure of psychological, physical, and social integration. An unexpected fourth factor, comprised of (1) ADL and (2) self-actualizat
An Oblimin rotation with Kaiser normalization was performed. Factor 1 included four of the five measures hypothesized to represent psychological integration (WQOL-I, Quality of Life, Psychological Well-being, Neighborhood Cohesion), with factors loading at .74, .66, .63 and .58, respectively. Factor 2 was comprised of two of the three measures hypothesized to represent physical integration (Organization Participation, Leisure Activities), with .85 and .83 factor loadings. Factor 3 was comprised of the two measures hypothesized to represent social integration (Social Support, Social Network Members from Neighborhood), with .73 and .69 factor loadings. The independence/se
For psychological integration, the negative bivariate association between psychological integration and baseline CSI and the positive bivariate association between psychological integration and membership in the hospital sample remained significant after all the program domain variables were entered. In addition, the non-significant
For physical integration, none of the personal history/demogra
For social integration, the positive bivariate association between social integration and baseline CSI and membership in the experimental condition remained significant after all the program domain variables were entered. In addition, substance use treatment remained marginally significant when all program domain variables were entered at the second step, and was negatively related to social integration.
For independence–se
This study explored community integration among individuals who have psychiatric disabilities and histories of homelessness. Using the Wong and Solomon model of community integration, we sought to identify both components of community integration and possible predictors. The factor analysis supported a comprehensive approach to the definition of community integration that includes psychological, physical, and social domains as posited by Wong and Solomon, but also suggested the existence of at least another factor that we refer to as independence/se
Past literature has suggested that normalized housing and services that emphasize independence foster greater community integration (Hull & Thompson, 1981; Kruzich, 1985; Segal & Aviram, 1978; Van Wel et al., 2003). As hypothesized in this study, choice and scatter-site housing were significant program predictors of psychological and social integration, respectively. Having been hospitalized in a psychiatric facility prior to housing enrollment was associated with greater subsequent psychological integration. This is an interesting and perhaps counter-intuiti
Similarly, participation in substance use treatment services had somewhat conflicting effects on integration. Utilization of substance use treatment was found to positively influence physical integration, but there was also a trend for it to negatively impact social integration. It is possible that this finding was driven by the fact that physical integration included participation in self-help organizations as an indicator. For many participants, these self-help groups pertained to substance use treatment, such as Narcotics Anonymous or Alcohol Anonymous. This suggests that individuals who attend substance use treatment services participate in more formal organizations and thereby have greater physical integration, but perhaps at the expense of informal socializing with other community and neighborhood members (i.e., low social integration). Another possibility is that engagement in substance abuse treatment results in consumers severing ties with those in their social network who are also substance users, thereby reducing the size of their social network. Finally, our data did not support findings from previous research that linked stigma to decreased community integration. However, it is possible that stigma may have been significant in the context of other factors that were not assessed.
These findings have several implications for facilitating the community integration of individuals with psychiatric disabilities. The first is for designing housing programs so as to maximize the potential for residents’ integration. For example, Aubry and Myner (1996) found that, as compared to community controls, persons with psychiatric disabilities experienced particularly low levels of social integration. Considering that our study found that a normalized residential arrangement was the only significant predictor of social integration, this would suggest that services may need to shift towards the provision of housing that most closely resembles that of the general population, for example independent scatter-site housing in the community. Additionally, housing agencies should encourage consumers to exercise choice regarding their lives, especially since this increased sense of autonomy leads to a greater sense of belonging and well-being. The Housing First model, with its emphasis on independent housing, consumer choice and empowerment, may therefore be particularly well suited for enhancing community integration.
Second, this analysis of the Wong and Solomon model demonstrates that individual predictors can influence different dimensions of integration and that the same predictor can have disparate effects within different domains. The findings regarding substance abuse treatment, for example, suggest that the same factor can differentially influence discrete aspects of integration and that there may be a need to adjust program variables to find an ideal balance.
The Wong and Solomon model provides an excellent foundation for community integration research but needs to be expanded. While the model includes personal and program variables, it does not sufficiently account for more macro-neighborh
The model may also become more elaborate when findings supporting the concept of person-environm
The study has some limitations that need to be highlighted. First, some important factors that might predict community integration are not included in the model and those factors could be equally or more important than those assessed. For example, the effects of neighborhood variables, such as neighborhood quality, building quality, crime rates and poverty were not assessed. Second, the results of the study could have been strengthened by a longitudinal approach which could have tracked changes in community integration over time. Finally, a triangulated approach assessing program factors, utilizing observational methods in addition to participant self-reports, would have provided an opportunity to corroborate the effects of external factors.
While this multi-dimension
Prior studies indicate that persons with psychiatric disabilities tend to experience only low to moderate levels of integration, thereby making ongoing research into the factors that facilitate community integration essential (Gerber et al., 2003). Existing research has mostly focused on the physical aspects of integration, but Gerber and colleagues (2003) demonstrated that the degree to which forensic outpatients were integrated into the community along psychological, social, and physical dimensions was also fairly low. Further, with the exception of Aubry and Myner (1996), few studies have conducted comparisons with community residents from the general population. Research on community integration can be greatly advanced by the inclusion of neighborhood controls and consumer perspectives on defining integration.
Studies of integration of individuals participating in housing programs should strive to include comparison groups of non-program neighbors. Without such neighborhood controls, investigators cannot determine which difficulties are specific to persons with psychiatric disabilities and which are common to all residents in general. Finally, there needs to be greater emphasis on direct consumer input. To our knowledge, no studies have explored the salience of the different domains of integration in consumers’ lives nor the respective importance that different individuals may ascribe to various aspects of integration. It is also imperative to explore the meaning of community integration from the perspective of consumers themselves, particularly through qualitative research designs. This will lead researchers to conceptualize integration in ways that are more relevant to the lives of persons with psychiatric disabilities. Such an understanding will further allow us to effectively tailor programs and services to meet consumers’ needs for community involvement and support.
References Aubry T., Myner J. (1996). Community integration and quality of life: A comparisons of persons with psychiatric disabilities in housing programs and community residents who are neighbors. Canadian Journal of Community Mental Health 15(1): 5–20 PubMed Aubry T. D., Teft B., Currie R. F. (1995). Public attitudes and intentions regarding tenants of community mental health residences who are neighbors. Community Mental Health Journal 31(1): 39–52 PubMed SpringerLink Barrera M. Jr. (1981). Social support in the adjustment of pregnant adolescents: Assessment issues In B. H. Gottlied (Ed.), Social networks and social support (pp. 69–96). Beverly Hills: Sage Becker M., Diamond R., Sainfort F. (1993). A new patient focused index for measuring quality of life in persons with severe and persistent mental illness. Quality of Life Research 2(4): 239–251 PubMed SpringerLink Bruckner J. C. (1988). The development of an instrument to measure neighborhood cohesion. American Journal of Community Psychology 16: 771–791 SpringerLink Caplan R. D., Harrison R. V. (1993). Person-environment fit theory: Some history, recent developments, and future directions. Journal of Social Issues 49: 253–276 Ciarolo J. A., Edwards D. W., Kiresuk T. J., Newman F. L., Brown T. R. (1981). Colorado symptom index Washington, DC: National Institute of Mental Health Gerber G. J., Prince P. N., Duffy S., McDougall L., Cooper J., Dowler S. (2003). Adjustment, integration, and quality of life among forensic patients receiving community outreach services. International Journal of Forensic Mental Health 2(2): 129–136 Greenwood R., Schaefer-McDaniel N., Winkel G., Tsemberis S. (2005). Decreasing psychiatric symptoms by increasing choice in services for adults with histories of homelessness. American Journal of Community Psychology 36(3/4): 223–238 PubMed SpringerLink Gulcur L., Stefancic A., Shinn M., Tsemberis S., Fischer S. (2003). Housing, hospitalization and cost outcomes for homeless individuals with psychiatric disabilities participating in continuum of care and housing first programmes. Journal of Community and Applied Social Psychology 13: 171–186 CrossRef Hull J. T., Thompson J. C. (1981). Factors which contribute to normalization in residential facilities for the mentally ill. Community Mental Health Journal 17(2): 107–113 PubMed SpringerLink Kennedy C. (1989). Community integration and well-being: Toward the goals of community care. Journal of Social Issues 45(3): 65–77 Kruzich J. M. (1985). Community integration of the mentally ill in residential facilities. American Journal of Community Psychology 13(5): 553–563 PubMed SpringerLink Lehman L. F. (1988). A quality of life interview for the chronically mentally ill Evaluation and Program Planning 11: 51–62 CrossRef Link B. G. (1987). Understanding labeling effects in the area of mental disorders: An assessment of the effects of expectations of rejection. American Sociological Review 52(1): 96–112 CrossRef Maslow A. (1971). The farther reaches of human nature. Oxford: Viking McLellan A. T., Alterman A. I., Woody G. E., et al. (1992). Treatment services review. Philadelphia, PA: University of Pennsylvania McMillan D. W., Chavis D. M. (1986). Sense of community: A definition and theory. Journal of Community Psychology 14: 6–23 CrossRef Moos R. H., Lemke S. (1996). Evaluating residential facilities: The multiphasic environmental assessment procedure. Thousand Oaks, CA: Sage Nagy M. P., Fisher G. A., Tessler R. C. (1988). Effects of facility characteristics on the social adjustment of mentally ill residents in board-and-care homes. Hospital and Community Psychiatry 39(12): 1281–1286 PubMed Nelson G., Hall G. B., Squire D., Walsh-Bowers R. T. (1992). Social network transactions of psychiatric patients. Social Science and Medicine 34(4): 433–445 PubMed CrossRef Perkins D. D., Florin P., Rich R. C., Wandersman A., Chavis D. M. (1990). Participation and the social and physical environment of residential blocks: Crime and community context. American Journal of Community Psychology 18(1): 83–115 SpringerLink Prince P. N., Prince C. R. (2002). Perceived stigma and community integration among clients of assertive community treatment teams. Psychiatric Rehabilitation Journal 25: 323–331 PubMed Rosenblatt A., Mayer J. (1974). Recidivism of mental patients: A review of past studies. American Journal of Orthopsychiatry 44: 697–706 PubMed Segal S. P., Aviram U. (1978). The mentally ill in community-based sheltered care: A study of community care and social integration. New York Wiley Segal S. P., Everett-Dille L. (1980). Coping styles and factors in male/female social integration. Acta Psychiatrica Scandinavica 61(1): 8–20 PubMed Segal S. P., Silverman C., Baumohl J. (1989). Seeking person-environment fit in community care placement. Journal of Social Issues 45(3): 49–64 Sherman S. R., Frenkel E. R., Newman E. S. (1986). Community participation of mentally ill adults in foster family care. Journal of Community Psychology 14: 120–133 CrossRef Srebnik D., Livingston J., Gordon L., et al. (1995). Housing choice and community success for individuals with serious and persistent mental illness. Community Mental Health Journal 31: 139–152 PubMed SpringerLink Timko C., Moos R. H. (1998). Outcomes of the treatment climate in psychiatric and substance abuse treatment programs. Journal of Clinical Psychology 54(8): 1137–1150 PubMed CrossRef Tsemberis S., Asmussen S. (1999). From streets to homes: Pathways to Housing consumer preference supported housing model. Alcoholism Treatment Quarterly 17: 113–131 CrossRef Tsemberis S., Gulcur L., Nakae M. (2004). Housing first, consumer choice, and harm reduction for individuals who are homeless with dual diagnoses: A 24 month clinical trial American Journal of Public Health 94(4): 651–656 PubMed Unger D. G., Wandersman A. (1985). The importance of neighbors: The social, cognitive, and affective components of neighboring. American Journal of Community Psychology 13: 139–163 SpringerLink Van Wel T., Felling A., Persoon J. (2003). The effect of psychiatric rehabilitation on the activity and participation level of clients with long-term psychiatric disabilities. Community Mental Health Journal 39(6): 535–546 PubMed SpringerLink Wong Y. I., Solomon P. L. (2002). Community integration of persons with psychiatric disabilities in supportive independent housing: A conceptual model and methodological considerations. Mental Health Services Research 4(2): 13–28
