Read Educating%20the%20Public%20About%20Mental%20Illness%20and%20Homelessness.pdf text version

Educating the Public About Mental Illness and Homelessness: A Cautionary Note

George S Tolomiczenko, MPH, PhD1, Paula N Goering, PhD2, Janet F Durbin, MSc1

Objective: To determine whether the viewing of a video depicting the successful struggles of homeless persons with mental illness in finding and maintaining housing can have a positive impact on attitudes toward homeless persons with mental illness. Method: Five hundred and seventy-five high school students attending a brief educational session on mental illness participated in 1 of 3 comparison versions of the 2-hour program (control, video, video plus discussion). All completed an "Attitudes toward Homelessness and Mental Illness Questionnaire." Demographic and prior exposure variables were entered as covariates in between-group analyses of variance. Results: Females and subjects who had more prior encounters with homeless persons were found to have the most positive attitudes. After controlling for these effects, the video alone had a negative impact on attitudes relative to the other groups, while the video followed by a discussion with one of the people featured in it had a largely positive impact. Conclusions: The apparent immediacy and the evocative power of video presentations cannot substitute for direct contact for the purpose of promoting positive attitude change. The findings are consistent with prior research emphasizing the importance of direct interaction with members of stigmatized groups to reduce negative attitudes. Education programs trying to destigmatize mental illness and homelessness using videos should proceed with caution. (Can J Psychiatry 2001;46:253­257) Key words: homelessness, mental illness, attitudes, stigma, education program ecent Canadian and American studies have reported high prevalence rates of mental disorders among homeless persons in a number of urban settings across North America (1­3). Public perception often links the issues of homelessness and mental illness. While many attribute today's homeless ness to yes ter day's dein sti tu tion ali za tion of those suffering from mental illness, empirical data suggest that broader economic determinants (such as housing affordability and employment factors) and personal histories of poverty and lack of social support are by far the main driving forces behind the increasing number of homeless individuals and families across North America (4,5). At the same time, there has been a corresponding surge in public attention paid to homelessness. Media attention may, in turn, have induced "compassion fatigue" where an impatient public may demand more punitive measures aimed at curbing visible numbers of homeless persons (6). This reaction may have intensified during North America's recent run of economic prosperity, as


those who have fared well may find it harder to understand that, for some, economic conditions have worsened. An education campaign to inform the public and increase knowledge about how mental illness can affect a person's housing status or life while homeless would seem, at first glance, to be likely to have a positive impact on attitudes. Well-timed and designed, such a campaign could potentially inoculate recipients against compassion fatigue. Population survey research (7), public education evaluation research (8), and research reviews (9) show that positive impact is best made through direct contact with persons who experience symptoms of mental illness. This finding also applies to perceived dangerousness associated with mental illness; for example, among white respondents, lower levels of perceived dangerousness were associated with increased contact with persons suffering from mental illness (10). It seemed reasonable to expect, then, that a video produced in large part by formerly homeless persons with mental illness would be an effective tool to educate groups toward this end. Toronto's Hostel Outreach Program (HOP), jointly administered by Community Resources Consultants of Toronto and the Community Occupational Therapists Association, runs a case-management program for mentally ill homeless persons in metropolitan Toronto that was formally evaluated in a study published in 1993 (11). It was hypothesized that the program would improve residential stability, reduce psychiatric symptoms, improve social functioning, improve social networks, and increase use of appropriate services. Clients


Manuscript received April 2000, revised, and accepted June 2000. 1Assistant Professor, Department of Psychiatry, University of Toronto, Clarke Site, Centre for Addiction and Mental Health. 2 Professor, Department of Psychiatry, University of Toronto, Clarke Site, Centre for Addiction and Mental Health. Address for correspondence: Dr GS Tolomiczenko, Health Systems Research & Consulting Unit, Clarke Site, Centre for Addiction and Mental Health, 250 College Street, 4th Floor, Toronto, ON M5T 1R8 e-mail: [email protected]

Can J Psychiatry, Vol 46, April 2001


The Canadian Journal of Psychiatry

Vol 46, No 3

were assessed during the 9 months before and 9 months after program entry. At follow-up, significant improvements in residential stability and reductions in psychopathology were demonstrated. Improvements in social functioning and increases in social network size were significant. As a way of reaching a wider audience with these findings in a direct manner, funding to produce a video on the HOP was obtained through a donation from the Atkinson Charitable Foundation. While the primary goal of the video was to increase awareness of the effectiveness of the HOP program, it offered a powerful and moving portrayal of the challenges facing individuals doubly burdened by mental illness and homelessness. Entitled "A Fine Line," the video features caseworkers, clients, and psychiatrists describing the program's beginnings, services, and impact among treatmentresistant homeless persons who also have a mental illness (12). In the video, clients speak about the course of their illness, homelessness, and recovery. Clients--who were involved in the production of the video--conducted and taped interviews with staff members (case managers) and other professionals. Researchers and mental health workers saw the video as a potential tool to help destigmatize homeless persons who are also suffering from mental illness. This paper presents an evaluation of the video's impact on attitudes among nonprofessional groups. Intervention Each week during the school year, between 50 and 150 students from high schools across Toronto visit the Clarke Site of the Centre for Addiction and Mental Health for a public education program called "Beyond the Cuckoo's Nest." The program's chief aim is to destigmatize mental illness; it typically consists of segments that include a facility tour or video, dialogues with persons who have a history of mental illness or their family members or both, reviews of psychiatric terminology, group exercises to sensitize individuals regarding symptoms, and other activities. This served as the control condition (C). We then created 2 variations of this program. The video condition (V) substituted the HOP video for one routinely used in the program. The video plus discussion condition (V+D) used the HOP video and included a follow-up discussion with a client featured in the video. Hypothesis Compared with the unmodified education program, we expected that the HOP video, by focusing on mental illness among a homeless population, would have a positive impact on attitudes toward homelessness and that the video combined with a participant­discussant would potentiate this effect. More specifically, the HOP video would increase understanding of--and improve attitudes toward--homeless persons who suffer from mental illness.

Questionnaire We gathered data with a questionnaire that includes items from 2 previously developed questionnaires used to assess attitudes toward homelessness (13) and mental illness (14). This pool of items was reduced following preliminary administration of the items to a group of students enrolled in an urban planning seminar at a local university. Items that were ambiguous or did not contribute to item-total consistency within a scale were eliminated (Note 1). The questionnaire consists of items reflecting amount of exposure to homeless persons (unnumbered items); emotional responsiveness (items 1­3: higher average score associated with less empathy, = 0.62); attitude toward mental illness (items 4­14: higher average score associated with negative view of mental illness, = 0.64); dangerousness (items 15­18: higher average score associated with threatening view of homeless persons, = 0.56); aversion (items 19 ­ 2 3: higher average score associated with aversive reaction toward homeless persons, = 0.75); restrictions (items 24­27: higher average score associated with a belief that restrictions on homeless persons should be imposed, = 0.62); structural determinants (items 29­33: higher average score associated with belief that structural factors such as housing supply and government policy contribute to homelessness, = 0.72); disability (items 34­37: higher average score associated with belief that individual disability precipitates homelessness, = 0.70); and blame (items 38­39: higher average score associated with belief that homelessness is due to laziness and irresponsible behaviour, = 0.77). At best, the internal consistency of the scales was good ( > 0.70), though all but 1 reached acceptable levels of internal consistency (Note 2). Scale scores represent average item ratings from 1 to 4 (Note 3). Sample Five hundred and seventy-five students from 14 different high school programs completed questionnaires. The largest group was the video-only group (V, n = 214), followed by the video-plus-discussant group (V+D, n = 186), and the control group (C, n = 175). No exact numbers are available on the number and characteristics of students who did not complete the questionnaires. Overall, an estimated three-quarters of those attending the programs completed the questionnaires. Results Table 1 shows a clear effect of sex of respondent on several of the scales. In all cases, females scored favourably compared with males (more empathic, less of a stigmatizing view of mental illness, less sense of danger attached to homeless persons, and a better appreciation of the structural determinants of homelessness). Given these significant differences, sex was entered as a factor into the analyses of group differences.

April 2001

Educating the Public About Mental Illness and Homelessness: A Cautionary Note


Table 1. Attitude scores by sex

Scale Emotional responsiveness Mental illness Dangerousness Aversion Restrictions Structural determinants Disability Blame

aP bP cP

Female (n = 388) Mean (SD) 1.66 (0.48) 2.14 (0.33) 2.34 (0.47) 2.58 (0.50) 2.34 (0.52) 2.89 (0.67) 2.84 (0.68) 2.58 (0.86)

Male (n = 187) Mean (SD) 2.00 (0.54) 2.24 (0.36) 2.46 (0.46) 2.67 (0.56) 2.36 (0.60) 2.76 (0.58) 2.76 (0.58) 2.73 (0.77)

Combined (n = 575) Mean (SD) 1.77 (0.52) 2.17 (0.34) 2.38 (0.47) 2.61 (0.53) 2.34 (0.55) 2.85 (0.65) 2.81 (0.65) 2.63 (0.84)

t value ­7.67a ­3.40b ­2.88b ­1.93 ­0.51 2.37 c 1.33 ­1.90

< 0.001; < 0.01; < 0.05. Scale scores represent average item ratings from 1 to 4. See text for description of items.

Table 2. Attitude scores as a function of exposure

Scale Low level of exposure (n = 440) (2 or fewer times per week) Mean (SD) 1.79 (0.51) 2.17 (0.33) 2.38 (0.47) 2.63 (0.52) 2.36 (0.52) 2.80 (0.65) High level of exposure (n = 128) (3 or more times per week) Mean (SD) 1.71 (0.55) 2.16 (0.38) 2.37 (0.48) 2.51 (0.52) 2.25 (0.59) 3.05 (0.60) t value

Emotional responsiveness Mental illness Dangerousness Aversion Restrictions

1.50 0.27 0.28 2.20a 2.09a

Another variable that was significantly associated with several scales was the level of exposure that respondents had with homeless persons. The first 3 items of the questionnaire indicated that fewer than 10% of the respondents had not encountered any homeless people in the prior year. Since these 3 items were sig nifi cantly intercorrelated and the second (number of homeless persons seen in an average week) was most strongly correlated with the other 2, it was used to divide the sample into low- (less than 3 times per week) and high-level (3 or more times per week) subgroups. Significant differences between low- and highexposure subgroups were found on aversion (higher-exposure respondents were less averse), restrictions (higher-exposure respondents were less in favor of restrictions), and struc tural de ter mi nants (higherexposure respondents were more inclined to attribute homelessness to economic and governmental determinants). These differences justified also entering exposure as a factor into the analyses of group differences (Table 2).

Table 3 lists the results of analyses of between-group differences after excluding Disability 2.83 (0.65) 2.77 (0.64) 0.79 the variance explained by sex and exposure in hierarchical analyses of variance. The Blame 2.65 (0.83) 2.56 (0.83) 1.11 first 2 columns reiterate the finding that aP < 0.05; bP < 0.001. Scale scores represent average item ratings from 1 to 4. See text for description of items. there are significant differences in scale scores by sex and level of exposure. Significant between-group differences were found on 4 of the scales. Table 3. Hierarchial ANOVA analysesa These differences were not conVariables tested sistent with the primary hySex Exposure Group pothesis of the study, however. F value F value (Differences due to F value In particular, the video-only Scale (Sex differences) amount of exposure) (Pairwise group differences) condition showed significant b Emotional responsiveness 54.08 (Male > Female) 1.04 0.91 differences in a negative direcMental illness 10.00c (Male > Female) 0.02 4.43d (V > C > V+D) tion on all 4 of these scales. Dangerousness 7.01c (Male > Female) 0.01 5.21c (V> C > V+D) Com pared with the con trol Aversion 3.92d (Male > Female) 4.48d (High < Low) 2.10 group, the video-only group d d showed a more negative attitude Restrictions 0.98 4.40 (High < Low) 4.13 (V > V+D > C) t o w a r d m e n t a l ill n e s s , a Structural determinants 6.20d (Male < Female) 14.43b (High > Low) 0.51 stronger feeling of danger assob Disability 1.63 0.65 9.00 (V+D > V > C) ciated with homeless persons, Blame 3.87 1.05 0.12 endorsed more restrictions, and aTwo scales with significant interaction effects between sex and either level of prior exposure or group emphasized the sex were more likely to see individdifferences. On attitude toward mental illness scores, males with higher levels of exposure to homeless persons have lower scores ual disability as a key factor in and females with higher levels of exposure have higher scores than their respective low-exposure counterparts. Two significant interaction effects related to scores on the restrictions scale resulted from the video's stronger negative impact on males (group by contributing to homelessness. sex interaction) and the video groups' (V or V+D versus C) lower scores among high-exposure subjects contrasted with higher In the opposite direction and scores for V and V+D groups among low-exposure subjects (group by exposure interaction). bP < 0.001; cP < 0.01; dP < 0.05. consistent with the hypothesis,

Structural determinants ­3.93b


The Canadian Journal of Psychiatry

Vol 46, No 3

however, the video-plus-discussant group had lower scores (using adjusted values) than the control group on 2 scales. Average scores for the video-plus-discussant group reflected a view that was less stigmatizing toward those with mental illness and that saw homeless persons as less dangerous. Contrary to the hypothesis, those who saw the video--with or without discussion--were also more likely to endorse restrictions directed toward homeless persons. Finally, the highest group score for disability determinants of homelessness was among the video-plus-discussion group followed by the video-only group. Both video-condition groups again scored higher than the control group. Discussion The video documenting portraits of life on the street of a group of people with mental illness and their paths toward housing and mental stability had a powerful, though complex, impact on high school students. The simple hypothesis that this impact would positively affect attitudes toward homeless persons and people struggling with mental illness was not supported, however. Overall, the best predictors of positive attitudes were being female and having had more prior encounters with homeless persons. Our findings indicate that a video chronicling the successful housing struggles of people suffering with major psychiatric disorders seemed to intensify negative attitudes. This result is consistent with the negative results of a classic study in community-oriented mental health education conducted in Saskatchewan in the early 1950s. Following the Cummings' carefully designed and executed educational program, community attitudes toward mental illness showed no change on retesting (15). Not surprisingly, negative stereotyping serves to harden negative attidudes. A more recent study that looked at the impact of a television film on attitudes toward mental illness used film featuring a mentally ill killer. Viewers expressed more negative attitudes toward both mental illness and community care for those afflicted than did a control group of viewers who saw a similarly violent film that did not include mental illness. Even with a trailer specifically stating that violence is not characteristic among persons with psychiatric disorders, viewers of the film featuring the mentally ill killer showed the same shift toward negative attitudes (16). Without a member of the group highlighted in the video to share his or her experience, the images are likely to be interpreted as corroborating evidence for pre-existing negative attitudes. By their nature, personal accounts do not emphasize the larger economic and societal determinants of homelessness. This illustrates one of the limitations of using a video approach that does not specifically explore structural factors. It also suggests why the HOP video made no significant impact in this dimension of attitude change.

Clinical Implications · Effective destigmatization relies on direct interaction with negatively affected individuals.

· Video portrayals may have the unintended effect of hardening preconceived stereotypes.

Limitations · The study sample consisted of urban high school students. · A group comparison design was used rather than a "before" and "after" contrast design.

· Surveyed attitudes using self-report measures may differ from observed behaviours.

In a positive direction, the video combined with an audience discussion that included 1 of the video's subjects was found to decrease stigmatizing attitudes toward mental illness and dangerousness (Note 4). At least 1 prior study helps in interpreting the positive results. With regard to dangerousness, Arikan and colleagues (17) found that, among university students, those who felt treatment for mental illness was inefficacious believed sufferers to be more dangerous than did students, who were more positive about treatment results. Direct discussion involving a subject from the video may be the best means of convincing an audience that treatment works. Taken together, these findings point toward a pernicious aspect of stigma attached to mental illness; namely, a tendency to ratchet in a negative direction when emotionally charged material is not grounded or interpreted in the context of either direct discussion or a proximate relations (through a friend or a family member) with a person or persons living with mental illness. The findings also underline the importance of adjusting the form of a message to maximize impact in the intended direction among a particular audience. While the current study was limited by its group-comparison design (rather than a "before" and "after" design) and the use of selfreported attitudinal measures (rather than observed behavioural changes), the cautionary note it introduces should inform educational programs attempting to destigmatize mental illness and homelessness.

Acknowledgement The authors express their thanks to the organizers of the Beyond the Cuckoo's Nest program through which subjects were recruited and to the Atkinson Foundation for grant support.


1. The final version of the questionnaire is available on request from the authors. 2. Dangerousness is included in subsequent analyses for the sake of completeness even though, at = 0.56, its reliability is below a conservative level of acceptability. 3. Items 28 and 40 were omitted to improve the item-total reliability of their respective scales. 4. The finding that the V+D group scored higher on the disability scale is not surprising in light of the emphasis both discussants placed on how illness symptoms precipitated and maintained their periods of homelessness.

April 2001

Educating the Public About Mental Illness and Homelessness: A Cautionary Note



1. Koegel P, Sullivan G, Burnam A, Morton SC, Wenzel S. Utilization of mental health and substance abuse services among homeless adults in Los Angeles. Med Care 1999;37:306­17. 2. Tolomiczenko GS, Goering PN. Pathways into homelessness: broadening the perspective. Psychiatry Rounds 1998;2:1­6. 3. Culhane DP, Avery JM, Hadley TR. Prevalence of treated behavioral disorders among adult shel ter us ers: a lon gi tu di nal study. Am J Or thop sy chia try 1998;68:63­72. 4. Plumb J D. Homelessness: care, prevention, and public policy. Ann Intern Med 1997;126:973­5. 5. Shinn M, Weitzman BC, Stojanovic D, Knickman JR, Jimenez L, Duchon L, and others. Predictors of homelessness among families in New York City: from shelter request to housing stability. Am J Public Health 1998;88:1651­7. 6. Link BG, Schwartz S, Moore R, Phelan J, Struening E, Stueve A, and others. Public knowledge, attitudes, and beliefs about homeless people: evidence for compassion fatigue. American J Community Psychol 1995;23:533­55. 7. Link BG, Cullen F. Contact with the mentally ill and perceptions of how dangerous they are. J Health Soc Behav 1986;27:289­303.

8. Wolff G, Pathare S, Craig T, Leff J. Public education for community care: a new approach. Br J Psychiatry 1996;168:441­7. 9. Penn DL, Martin J. The stigma of severe mental illness: some potential solutions for a recalcitrant problem. Psychiatr Q 1998;69:235­ 47. 10. Whaley AL. Ethnic and racial differences in perceptions of dangerousness of persons with mental illness. Psychiatr Serv 1997;48:1328­30. 11. Wasylenki D, Goering PN, Lemire D, Lindsey S, Lancee W. The Hostel Outreach Program: assertive case management for homeless mentally ill persons. Hospital and Community Psychiatry 1993;44:848­53. 12. A fine line [film]. Cava B, producer and director. Toronto (ON): Consumer Resources and Consultants of Toronto; 1996. 13. Phelan J, Link BG, Stueve A, Moore R, Struening E, Colten ME. Education, social liberalism, and economic conservatism: attitudes toward homeless people. American Sociological Review 1995;60:126­40. 14. Borinstein AM. Public attitudes toward persons with mental illness. Health Aff 1992;11:186­96. 15. Cumming E, Cumming J. Closed ranks: an experiment in mental health education. Cambridge (MA): Harvard University Press; 1957. 16. Wahl OF, Lefkowits JY. Impact of a television film on attitudes toward mental illness. Am J Community Psychol 1989;17:521­8. 17. Arikan K, Uysal O, Cetin G. Public awareness of the effectiveness of psychiatric treatment may reduce stigma. Isr J Psychiatry Relat Sci 1999;36:95­9.

Résumé-- Éducation du public sur la maladie mentale et l'itinérance : mise en garde

Objectif : Déterminer si l'écoute d'un vidéo décrivant les démarches fructueuses de personnes sans-abri souffrant de maladie mentale en vue de trouver et de garder un logement peut avoir un effet favorable sur les attitudes envers les sans-abri souffrant de maladie mentale. Méthode : Cinq cent soixante-quinze élèves du secondaire assistant à une brève séance de sensibilisation à la maladie mentale ont participé à l'une des 3 versions comparatives de l'émission de 2 heures (contrôle, vidéo, vidéo et discussion). Tous ont rempli un questionnaire sur les attitudes à l'égard de l'itinérance et de la maladie mentale. Les variables démographiques et d'exposition antérieure ont été entrées à titre de covariables dans les analyses de variance entre groupes. Résultats : Les femmes et les sujets ayant davantage d'expérience préalable avec les sans-abri affichaient les attitudes les plus favorables. Après avoir tenu compte de ces effets, le vidéo à lui seul avait un effet défavorable sur les attitudes des autres groupes, alors que le vidéo suivi d'une discussion avec une personne qui y jouait avait un effet très favorable. Conclusions : L'apparente instantanéité et le pouvoir évocateur des présentations vidéo ne peuvent remplacer le contact direct pour promouvoir un changement d'attitude. Les résultats sont conformes aux recherches précédentes insistant sur l'importance de l'interaction directe avec les membres des groupes stigmatisés pour réduire les attitudes nuisibles. Les programmes d'éducation qui tentent d'éliminer les stigmates de la maladie mentale et de l'itinérance doivent faire preuve de prudence.



5 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Guidelines for the Use of Antiretroviral Agents
Microsoft Word - ECHO 2nd Progress Report_Final.doc
Kenneth Minkoff, MD