BMJ Open. For peer review only - Journal: BMJ Open. Manuscript ID bmjopen

Transcription

1 How did a Housing First Intervention Improve Health and Social Outcomes among Homeless Adults with Mental Illness? Two Year Outcomes from a large Randomized Trial Journal: BMJ Open Manuscript ID bmjopen-0-00 Article Type: Research Date Submitted by the Author: 0-Dec-0 Complete List of Authors: O'Campo, Patricia; St.Michael's Hospital, Center for Research on Inner City Health; University of Toronto, Dalla Lana School of Public Health Stergiopoulos, Vicky; St. Michael's Hospital, Centre for Research on Inner City Health; University of Toronto, Department of Psychiatry Nir, Pam; Cota Levy, Matthew; University of Toronto, Department of Psychiatry Misir, Vachan; St. Michael's Hospital, Centre for Research on Inner City Health Chum, Antony; London School of Hygiene & Tropical Medicine, Social and Environmental Health Research Arbach, Bouchra; Cota Nisenbaum, Rosane; St. Michael's Hospital, Centre for Research on Inner City Health; University of Toronto, Dalla Lana School of Public Health To, Matthew; St. Michael's Hospital, Centre for Research on Inner City Health Hwang, Stephen W. ; St. Michael's Hospital, Centre for Research on Inner City Health; University of Toronto, Division of General Internal Medicine <b>primary Subject Heading</b>: Mental health Secondary Subject Heading: Public health Keywords: Homelessness, Mental Illness, Housing Stability, Housing First, Health services utilization BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

2 Page of BMJ Open Housing First Impact on Health and Social Outcomes How did a Housing First Intervention Improve Health and Social Outcomes among Homeless Adults with Mental Illness? Two Year Outcomes from a large Randomized Trial Patricia O'Campo a,b, Vicky Stergiopoulos a,c, Pam Nir e, Matthew Levy c, Vachan Misir a, Antony Chum f, Bouchra Arbach e, Rosane Nisenbaum a, b, Matthew J To a, Stephen W. Hwang a,d a Centre for Research on Inner City Health, St Michael's Hospital, Toronto, Ontario, Canada b Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada c Department of Psychiatry, University of Toronto, 0 College Street, th Floor, Toronto, Ontario MT R, Canada d Division of General Internal Medicine, Department of Medicine, University of Toronto, Toronto, Ontario, Canada e Cota, Toronto, Ontario, Canada f Social and Environmental Health Research, London School of Hygiene & Tropical Medicine, London, UK Corresponding author: Patricia O`Campo Centre for Research on Inner City Health Li Ka Shing Knowledge Institute, St. Michael s Hospital 0 Bond Street, Toronto, Ontario MB W, Canada o campop@smh.ca --0 Keywords: homelessness, RCT, mental illness, housing stability, Housing First, health services utilization BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

3 Page of Housing First Impact on Health and Social Outcomes Abstract OBJECTIVES: We studied the impact of a Housing First intervention on housing, crime, health care utilization and health outcomes among At Home/Chez Soi randomized trial participants in Toronto, a city with an extensive service network for social and health services for individuals who are experiencing homelessness and mental illness. METHODS: Participants identified as high needs were randomized to receive either the intervention which provided them with housing and supports by an Assertive Community Treatment team (HF+ACT) or treatment as usual (TAU). Participants (N=) had in-person interviews every six months for two years. RESULTS: The HF+ACT group spent more time stably housed compared to the treatment as usual (TAU) with the mean difference between the groups of.% (%CI:.%,.%, p<0.000). Accounting for baseline differences, HF+ACT showed significant improvements over TAU for community functioning, selected quality of life subscales, and arrests at some time points during follow up. No differences between HF+ACT and TAU over the follow-up were observed for health service utilization, community integration, and substance use. CONCLUSIONS: Housing First for individuals with high levels of need increased housing stability and selected health and justice outcomes over two years in a city with many social and health services. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

4 Page of BMJ Open Housing First Impact on Health and Social Outcomes Strengths and limitations of this study A Housing First intervention was assessed using a strong RCT design, matching the intensity of intervention to the participant s level of need. Retention rates over the two year period were high and we measured a wide range of health and social outcomes. Interviewers were not blinded to treatment status of participants. Not all outcome measures were highly sensitive to detecting changes over time. The intensity of observations may have facilitated change in both treatment and control groups BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

5 Page of Housing First Impact on Health and Social Outcomes Introduction Innovations in interventions serving individuals who experience long term homelessness are gaining ground []. The shift away from a reliance on a cycle of shelters and transitional housing toward providing permanent supportive housing has been documented across North America and Europe [-]. "Treatment first" models that seek to stabilize individuals with regard to mental illness and substance use prior to providing permanent housing are giving way to "housing first" models that are built on principles of consumer choice and mental health recovery while providing immediate access to housing and optional community supports. Housing First (HF) has been documented to improve a range of outcomes including housing stability, quality of life, vocational opportunities, and community integration over treatment first or usual care models [-] but less so for substance abuse outcomes []. The At Home/Chez Soi randomized trial was conducted in five cities across Canada and enrolled individuals with serious mental illness who were experiencing chronic homelessness in the largest trial to date of the Housing First model []. At Home/Chez Soi (AH/CS) tailored community supports to those with high needs versus moderate needs as assessed at baseline. Those with high needs were identified by severity of mental health conditions, and high levels of mental health care use and social problems experienced in the year prior to enrollment; participants randomized to receive the intervention were provided with housing and supports by an Assertive Community Treatment team. AH/CS was intentionally implemented in a variety of settings across Canada, with Toronto being a site where a large proportion of those enrolled were born outside of Canada or considered themselves to belong to an ethnoracial minority group. Moreover, Toronto has a broad network of mental health services --inpatient and outpatient services, assertive community treatment, court support services, crisis programs (e.g., mobile crisis team and the distress line), and ethno-racial focused agencies-- which also serve individuals who are homeless. The AH/CS intervention tailors services to level of need during initial assessment. The study is unique since it implements a Housing First model while providing services, training of AH/CS staff on the HF model and principles of mental health recovery, and fidelity assessments at points in time. Follow-up rates over time were high. While outcomes for the high needs portion of AH/CS for the full sample have been published, it is critical to examine site specific data for a variety of reasons. Findings from the overall comparison may differ from the site specific results because the TAU conditions vary across cities. Furthermore, the samples differed at baseline in each of the cities[]. Specifically, the Toronto sample differed from the five city At Home/Chez Soi sample on several characteristics. Toronto, compared to the five-city sample, enrolled a lower proportion of women, % versus %, respectively. Given that Toronto focused on those participants who identify as ethnoracial minorities, Toronto's high needs group was 0% ethnoracial compared to % for the five-city ACT sample. While Toronto, compared to the five-city sample of high needs participants, had a similar level of psychotic disorder ( % vs %, respectively) the Toronto sample had a lower proportion of substance-related problems, 0% versus %, respectively []. Each site's programs were tailored to local contexts while still being adherent to the basic principles of HF and ACT [, ]. Thus, an examination of site specific findings yields further insights into whether and how Housing First strategies are effective for health and social outcomes in diverse contexts. The Toronto site of AH/CS provides insights into the effectiveness of the intervention in well-resourced settings and ethnoracially diverse homeless populations. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

6 Page of BMJ Open Housing First Impact on Health and Social Outcomes Methods Sampling and Recruitment For inclusion in the study, at the time of enrollment participants had to be >= years of age, absolutely homeless or precariously housed, have a confirmed major mental disorder, and not be served by an Assertive Community Treatment (ACT) program[]. Participants were recruited from October 00 to July 0 using a targeted recruitment and referral strategy to ensure that the study sample approximated those experiencing homelessness in Toronto[]. Referrals came from shelters, drop-in centres, hospitals, outreach programs, and service providers and were reviewed by an intake coordinator for eligibility and assessment [0]. At baseline, participants were assessed for disability and severity of mental health problems and then stratified into a high or moderate needs group. Participants with a Multnomah Community Ability Scale (MCAS) score of less than, indicating at least moderate disability, a diagnosis of a psychotic or bipolar disorder and meeting one of the following criteria: being hospitalized for a mental illness at least twice in one year over the last five years, indication of co-morbid substance use, and recent arrest or incarceration were randomized into the high needs group. High needs participants were randomized after their baseline interview into treatment as usual (TAU) or in the Housing First plus Assertive Community Treatment (HF+ACT) arm of the trial. HF+ACT participants were provided with a rent allowance of $00. The ACT mental health team had a participant-tostaff ratio of 0:, available days/week; hours/day [0] and provided a number of relevant services which included case management, initial and ongoing assessment, psychiatric care, employment and housing assistance, family support and education, substance use services and intensive support to allow participants to live successfully in the community [0]. This study was approved by the Research Ethics Board of St. Michael s Hospital in Toronto and registered with the International Standard Randomized Control Trial Number Register (ISRCTN0). Data Collection Data were collected by in-person interviews every three months for a two year period and entered into a secure web-based database system. Participants were financially compensated when interviewed. Housing stability was the primary outcome of interest and was assessed using the Residential Timeline Followback (RTLFB) Questionnaires which captured participants month residential history (i.e., move in/out dates for each type of residence such as street, temporary housing, permanent housing, emergency/crisis and institution) at every interview point []. Among participants with at least one follow-up interview, the percentage of days stably housed was calculated as the total number of days stably housed divided by the total number of days for which any type of residence data were provided. The number of days spent in a hospital (psychiatric, general and psychiatric unit) was also calculated using the data obtained from the RTLFB questionnaire. Community Functioning was measured using the Multnomah Community Ability Scale (MCAS), a - item scale capturing degree of self-reported functional ability in the areas of health, adaptation, social skills and behavior. Higher scores indicate little to no impairment [, ]. The Colorado Symptom Index (CSI) was used to assess Mental Health symptom severity through questions asking how often participants experienced specific psychiatric symptoms at least every day to not at all (range BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

7 Page of Housing First Impact on Health and Social Outcomes -) generating an overall summary score ranging from to 0 with lower scores reflecting fewer symptoms[] []. Substance use severity was measured via the -item GAIN Substance Use Disorder Scale Short Screener (GAIN SS) which assessed the number of substance-related problems. Severity was derived by counting the number of problems in the past month with a range from 0- and higher values corresponding to a greater problem severity. Quality of life was measured using two different instruments. Quality of life was measured via the Euro Qol EQ- D VAS where participants rated their health from 0-00 with higher values reflecting better health. Previous studies have shown that the single-item quality of life visual analogue scale has a high level of agreement with a number of validated multi-item measures including the Health Utility Index[, ] and health perception scale of the short-form 0 health survey. Subjective and objective assessment of disease specific quality of life was measured using the Quality of Life Index (QoLI-0), a 0 item scale measuring participants' satisfaction on a point likert scale in a variety of areas including family, finance, leisure, living, safety, and social. Higher scores, ranging from -0, are indicative of better quality of life []. Community integration was measured using the Community Integration Scale (CIS) which assesses psychological integration based upon levels of satisfaction with home and neighbourhood ranging from -0, and physical integration from counts of social activities such as going to a concert, meeting people for coffee, participating in sports, etc on a scale of 0- with higher scores reflecting greater integration []. Health service and justice use were also measured using two items from the Health Service and Justice Service Use (HSJSU) Questionnaire: the number of emergency department visits and the number of arrests, in the last months []. Statistical Analyses Descriptive statistics were calculated as proportions for categorical variables and mean and standard deviation for continuous baseline variables among the HF + ACT and TAU groups. Data for each participant were included up until the end of the year follow up or until his or her point of death. The mean percentage of days stably housed over the -year period was compared for the HF + ACT and TAU group using a two sample t-test. For the number of days hospitalized over the -year period, a zero-inflated negative binomial model was used to account for the likelihood of excessive zeros and overdispersion (i.e. variance larger than the mean) among participants who had at least one day of hospitalization[] (PROC GENMOD in SAS (SAS Institute) version.). This model also included an offset for the number of years in the study so that estimated means could be interpreted as predicted number of days hospitalized per year. For secondary outcomes, which were measured longitudinally from baseline, mixed models were used for both continuous and count outcomes with a random intercept used to estimate subject-level variation. Similar to the previous models, the treatment arm was included, with the TAU group as the reference group and also a categorical time variable (0,,, and ) with baseline as the reference category.. Treatment differences specifically at each measurement period (0,,, and ) were examined through the use of mean differences for the continuous outcomes and rate ratios for the count models. To truly understand the longitudinal change for each treatment group taking into consideration the baseline differences, the change between arms over times were examined through the use of a treatment by time interaction. For continuous outcomes this value is interpreted as the difference in group mean changes from baseline at each time point. For count outcomes, change from baseline is estimated by the rate ratio in each BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

8 Page of BMJ Open Housing First Impact on Health and Social Outcomes arm (i.e., expected counts at each time point divided by expected counts at baseline), and the interaction represents the ratio of the rate ratios between the arms at each time point. Multiple imputation with chained equations (MICE) using STATA (StataCorp) version was used to impute missing data due to loss to follow-up, withdrawal, skipped interviews, nonresponse on specific items and lack of interviewer confidence in participant responses. Each of 0 imputed datasets was analyzed using PROC MIXED for continuous outcomes and PROC GLIMMIX for count outcomes, assuming an unstructured correlation matrix for the repeated measures. Results were combined across datasets using PROC MIANALYZE []. All tests were two-sided and statistical significance was set at p-value <0.0. Results Of the participants recruited into the study in Toronto, participants met criteria for the high needs group. Of these high needs participants, were randomly allocated to the HF+ACT group and 00 participants were randomly allocated to the TAU group. By months, (%) high needs participants were lost to follow-up, (%) died and (%) were withdrawn (See Figure ).Demographic and mental health characteristics of the participants at baseline are presented in Table. Over the -months period, participants in the HF+ACT group spent more time stably housed compared to the TAU (mean difference (%CI) =.% ( %CI:.,., p<0.000); group means (% CI):.% (% CI:.,.),.% (% CI:.,.) for HF+ACT and TAU, respectively. Participants in the HF+ACT group experienced significantly fewer days hospitalized in a psychiatric hospital compared with TAU group (mean=., %CI: (0., 0.);., %CI: (.,.0), respectively; Rate Ratio: 0., %CI: (0., 0.), p=0.00) (see Figure ). There were no significant differences observed between the HF + ACT and TAU group for the number of days spent in a psychiatric unit and in a general hospital. At each follow-up time point (see Figures ), participants in both arms showed improvements for almost all other secondary outcomes, compared with their baseline values. Measures of community functioning,--mcas and mental health status, CSI--showed steady improvements over the month follow-up for both arms. Quality of life measures--qol and EQD--also showed improvements over time with statistically significant differences observed for several domains at different follow-up time points (e.g., QOL Total, QOL Living, QOL Safety and EQD amongst others). Substance use and arrests also showed improvement for both arms of the trial. Only CIS physical did not show any improvements for either group over time. However, when changes were compared between groups, a slightly different picture emerged. The HF+ACT group showed greater improvements than the TAU over the two year follow up period for few outcomes. These outcomes for which treatment effects were observed over the month period accounting for baseline values are described here (see Table and Figure ). The mean change from baseline for the MCAS score was found to be significant at both months, with a treatment difference of.0 [% CI: (0.,.), p=0.0] and at months, with a difference of. [%CI: (0.,.), p=0.0]. These mean changes in the MCAS score over time were indicative of improvement over baseline for the HF+ACT group compared to the TAU group (See Table ). For the QoL Living subscale, a significant mean change from baseline was detected at months [0., %CI: (0.0,.), p=0.0] and a borderline difference for QoL Global at months [0., %CI: (-0.0,.), p=0.0] (See Table ). There was a rate reduction for arrests at months compared to baseline, with the HF+ACT group having a greater reduction [0., % CI: (0., 0.), p=0.0] (See Table ). BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

9 Page of Housing First Impact on Health and Social Outcomes For the EQ-D VAS, while there was no significant treatment difference from baseline, a significant difference was detected specifically at Months, with a mean difference of. [% CI: (0.,.), p=0.0] where the HF+ACT indicated a greater overall health score. Significant differences were detected for the Quality of life Total Score at every time point with the exception of the month interview [Baseline:. (p=0.00), Months:. (p=0.0), Months:. (p=0.0), Months:. (p=0.0) and Months:. (p=0.0)], all of which indicated a greater overall quality of life for the HF+ACT group. Similarly, significant differences at each time point were detected with the exception of the baseline interview [Baseline:. (p=0.0), Months:.0 (p=0.00), Months:. (p=0.0), Months:. (p=0.00) and Months:. (p=0.0)].for the Living subscale, significant treatment differences were detected at and months [ Months:., % CI: (0.,.), p=0.000, Months: 0., %CI: (0.,.), p=0.00]. A significant difference for the Finance scale was detected at study end [ Months:., %CI: (0.,.), p=0.0], for the Social subscale at baseline [., %CI: (0.,.), p=0.0] and Months [., %CI: (0.,.), p=0.00], and the Global subscale at Months [0., %CI: (0.,.), p=0.00]. Discussion For the primary outcome of Housing stability, the HF+ACT arm showed a large and significant difference of.% greater time stably housed compared the TAU arm (.% HF+ACT group vs..% TAU group. For secondary outcomes, participants in both the HF+ACT and TAU arms displayed improvements in almost all outcomes over the full month follow-up period. For several of the quality of life measures, there were significant differences between HF+ACT and TAU for several of the follow-up points. However, significant improvements over baseline values for the HF+ACT compared to the TAU participants were only observed for a handful of outcomes. Significant differences between HF+ACT and TAU for outcomes very early ( month point) in the follow-up period included those related to quality of life (QOL-Living) and functional ability (MCAS). While improvements continued for the two outcomes over the follow-up period, only for MCAS did the HF+ACT improve further over the TAU group at the end of the study ( month point) (see Figures). One outcome that showed significant improvements between HF+ACT and TAU toward the end of the study is arrests. Again, while arrests decreased over time for both HF+ACT and TAU, at the month point, arrests in HF+ACT decreased significantly more than in TAU (see Figure and Table ). While we had hypothesized that the HF+ACT group would show greater improvements in most or all our primary and secondary outcomes over the TAU group over the two year follow-up period, there are a number of possible reasons why we did not see improvements in more of the outcomes. First, we had a small sample size which may have worked against finding a true difference. In fact, for some additional outcomes some p-values were approaching the cut-off of p 0.0. However, analyzing our data using longitudinal methods helped us to overcome the problem of low statistical power. Longitudinal evaluations that enroll very high risk participants often experience results that reflect `regression to the mean` where participants naturally improve regardless of the treatments being administered. Given that we saw improvements in all participants over time for all but one of our outcomes, it is certainly possible that our findings reflect this phenomenon. Finally, given the service rich context, it is possible that many TAU participants had access to similar mental health and addictions services as HF participants. A major strength of the At Home/Chez Soi approach was the supports (e.g., training, staffing resources, ability to innovate and adapt to local contexts) provided to ensure that sites adhered to the HF model with fidelity assessments occurring throughout the project. Teams in all sites benefitted from the training, fidelity assessments and the local communities of practice created to ensure that providers in each site were adhering to and not deviating from the core models (Nelson et al 0). Yet, despite the high ratings for fidelity BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

10 Page of BMJ Open Housing First Impact on Health and Social Outcomes assessments, Toronto had other advantages that may have contributed to the overall improvements over time for all participants and the small differences observed between the HF+ACT and TAU groups. Toronto has a number of active housing programs with the largest one being Streets to Homes. It has been available prior to At Home/Chez Soi beginning in Feburary 00, where outreach workers house approximately 00 homeless people per year with % remaining housed after year []. In addition, more than 0 organizations provide various mental health or supportive housing services in the city []Thus, in Toronto, it is possible that the added benefit of the At Home/Chez Soi implementation of HF+ACT was small at best, given the rich array of preexisting ACT and housing first programs. One limitation to our study is that our interviewers could not be blinded to the treatment status of participants. Moreover, it is possible that given the intensity of the study activities and observations, even for those in the TAU group, many of whom were contacted monthly by study staff, that the Hawthorne effect, where participants positively change behaviours as a result of being part of a study, resulted in improvements in both groups. In summary, our findings suggest that even in a city with extensive mental health and housing services like Toronto, for a handful of outcomes, we can do better than providing (uncoordinated) services to high risk populations. This is supported by the few significant differences we saw in our data but, perhaps more importantly, by the improvements early in the program, right after participants are housed, for some of our outcomes in the HF+ACT group compared to the TAU group. On the other hand, our data also suggest that it may take more time for outcomes such as arrests to show improvements and that longer follow-up periods for two or more years might yield additional information about how HF+ACT impacts health and social outcomes for this population. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

11 Page 0 of Housing First Impact on Health and Social Outcomes Contribution statement Patricia O Campo, Vicky Stergiopoulos, Stephen W.Hwang, Pam Nir and Matthew Levy conceived of the research. Patricia O Campo Vachan Misir and Rosane Nisenbaum analyzed the data. Patricia O Campo, Vicky Stergiopoulos, Stephen W.Hwang, Rosane Nisenbaum, Pam Nir, Matthew Levy, Bouchra Arbach and Matthew J To wrote the paper. Competing interest All authors have completed the ICMJE uniform disclosure form at and declare: all authors had financial support from Mental Health Commission of Canada for the submitted work except for Antony Chum; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work." Funding This work was supported by Mental Health Commission of Canada Data sharing statement The At Home/Chez Soi project has a process by which interested investigators who would like to use the data for publication can make a formal request. The formal request is reviewed by a cross-site committee and as long as those particular analyses have not already been undertaken approval and data sharing can take place. 0 BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

12 Page of BMJ Open Housing First Impact on Health and Social Outcomes References. Doran K. M., Misa E. J., and Shah N. R., Housing as health care-new York's boundary-crossing experiment. New England Journal of Medicine, 0. (): p. -.. G., J., ParkinsonS., and Parsell C., Policy shift or program drift? Implementing Housing First in Australia. American Journal of Psychiatry, 0. (): p. -.. Nelson G., Goering P., and Tsemberis S., Housing for people with lived experience of mental health issues: Housing First as a strategy to improve quality of life. Community psychology and the socioeconomics of mental distress: International perspectives, 0: p Tsemberis S., Housing First: The Pathways model to end homelessness for people with mental illness and addiction manual. European Journal of Homelessness, 0. ().. Gilmer TP., et al., Effect of full-service partnerships on homelessness, use and costs of mental health services, and quality of life among adults with serious mental illness homelessness services in mentally ill adults. Arch Gen Psychiatry, 00. : p. -.. Gulcur L., et al., Housing, hospitalization, and cost outcomes for homeless individuals with psychiatric disabilities participating in continuum of care and housing first programmes. Journal of Community & Applied Social Psychiatry, 00. : p. -.. Leff HS., et al., Does one size fit all? What we can and can t learn from a meta-analysis of housing models for persons with mental illness. Psychiatric Services 00. 0(): p. -.. Nelson G., Aubry T., and Lafrance A., A review of the literature on the effectiveness of housing and support, assertive community treatment, and intensive case management interventions for persons with mental illness who have been homeless. American Journal of Orthopsychiatry, 00. (): p Patterson ML., et al., Housing First improves subjective quality of life among homeless adults with mental illness: -month findings from a randomized controlled trial in Vancouver, British Columbia. Soc Psychiatry Psychiatr Epidemiol, 0. : p Siegel C., et al., Tenant outcomes in supported housing and community residences in New York city. Psychiatr Serv, 00. : p. -.. Yanos PT., et al., Exploring the role of housing type, neighborhood characteristics, and lifestyle factors in the community integration of formerly homeless persons diagnosed with mental illness. J Ment Health, 00. : p Rog DJ., et al., Permanent supportive housing: assessing the evidence. Psychiatric Services, 0. (): p. -.. Palepu, A., et al., Housing First Improves Residential Stability in Homeless Adults With Concurrent Substance Dependence and Mental Disorders. American Journal of Public Health, 0. 0(S): p. e0- e.. Goering PN., et al., The At Home/Chez Soi trial protocol: a pragmatic, multi-site, randomised controlled trial of a Housing First intervention for homeless individuals with mental illness in five Canadian cities. BMJ Open, 0. ().. Goering, P., et al., National At Home/Chez Soi Final Report. 0, Mental Health Commission of Ca: Calgary, AB.. Aubrey T., et al., One-year outcomes of a randomized controlled trial of Housing First with ACT in five Canadian cities. Psychiatric Services. Psychiatric Services, 0. (): p. -.. Nelson G., et al., Implementation and fidelity evaluation of the Mental Health Commission of Canada's At Home/Chez Soi project: Cross-site report.. 0. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

13 Page of Housing First Impact on Health and Social Outcomes. O'Campo P., et al., Strategies to balance fidelity to housing first principles with local realities: lessons from a large urban centre. J Health Care Poor Underserved, 0. (): p. -.. City of Toronto Shelter Support and Housing Administration, 00 Street Needs Assessment Results. Toronto: City of Toronto Hwang S., et al., Ending Homelessness among people with mental illness: the At Home/Chez Soi randomized trial of a Housing First Intervention in Toronto. BMC Public Health, 0. (): p... Barker, S., et al., A community ability scale for chronically mentally ill consumers: Part II. Applications. Community Ment Health Journal,. 0(): p. -.. Dickerson, F.B., et al., An expanded version of the Multnomah Community Ability Scale: anchors and interview probes for the assessment of adults with serious mental illness. Community Ment Health Journal, 00. (): p. -.. Conrad, K.J., et al., Reliability and validity of a modified Colorado Symptom Index in a national homeless sample. Mental Health Services Research, 00. : p. -.. Bélanger A, B.J.-M., Guimond E, et and al., A head-to-head comparison of two generic health status measures in the household population: McMaster Health Utilities Index (Mark) and the EQ-D.. 00, Statistics Canada Report. Ottawa, Ontario. p. -.. de Boer AG., et al., Is a single-item visual analogue scale as valid, reliable and responsive as multi-item scales in measuring quality of life?. Qual Life Res, 00. : p Uttaro T. and A. Lehman, Graded response modeling of the Quality of Life Interview. Evaluation and Program Planning,. : p. -.. Falvo, N.H., Toronto s Streets to Homes Program. In: Hulchanski, J. David; Campsie, Philippa; and S.H. Chau, Stephen; Paradis, Emily (eds.) Finding Home: Policy Options for Addressing Homelessness in Canada (e-book), Chapter.. Toronto: Cities Centre, University of Toronto., Chapter. Homelessness, Toronto s Streets to Homes Program., in Finding Home: Policy Options for Addressing Homelessness in Canada, J.D. Hulchanski, et al., Editors. 00, Cities Centre, University of Toronto: Toronto.. The Access Point. The Toronto Mental Health and Addictions Access Point,. June 0]; Available from: BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

14 Page of BMJ Open Housing First Impact on Health and Social Outcomes Table. At Home/Chez Soi Toronto site, sample demographic, health and social characteristics at baseline for the total sample of High Needs participants, those receiving Assertive Community Treatment (HF+ACT) and those in the Treatment as Usual (TAU), Characteristics ACT (N=) TAU (N=00) Total (N=) Age years, mean (SD). (.0).0 (.) 0.0 (.) Gender, N (%) Female (.0) () (.) Male (.0) () (.) Other 0 (0) () (.0) Marital Status, N (%) Married/Partnered (.) () (.) Divorced/Separated/Widowed (0) () (.) Single/Never married (.) () (.) Country of Birth, N (%) Canada (.) 0 (0) (.) Other (.) 0 (0) (.) Ethnic or Cultural Identity, N (%) Aboriginal (.) () 0 (.0) Ethno-Racial (.) () 0 (.) White (.) () (.) Housing Status, N (%) Absolutely Homeless (.) () (.) Precariously Housed (.) () (.) Lifetime duration of homelessness years, mean (SD). (.). (.). (.) Education, N (%) Less than High School (.) (.) 0 (.) Completed High School (.) (.) (.) Some Post-Secondary School (.0) (.) (.) MCAS score, mean (SD). (.). (.). (.) MINI Diagnostic Categories, N (%) Depressive Episode (.) () (.) Manic or Hypomanic Episode (.) () 0 (0.) Post-Traumatic Stress Disorder (.) () (.) Panic Disorder (.) () (.) Mood Disorder with Psychotic Features (.) () (.) Psychotic Disorder (.) 0 (0) (.) Substance Use Related Problems (.) () 0 (0.) Suicidality Level No/Low (.) () (.) Moderate (.) () (.) High (.) 0 (0) (.) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

15 Page of Housing First Impact on Health and Social Outcomes Figure : At Home/Chez Soi Toronto site, participant flow through the study Moderate Needs (n=) Allocated to HF+ACT (n=) Withdrawn (n=) Death (n=) Lost to follow-up (n=) Analysed (n=) Assessed for eligibility (n=) Excluded (n=) Randomized (n=) High Needs (n=) Allocation Follow-Up Analysis Not meeting inclusion criteria (n=) Declined to participate (n=) Other reasons (n=) Allocated to treatment as usual (TAU), (n=00) Death (n= ) Lost to follow-up (n=) Analysed (n=00) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

16 Page of BMJ Open Housing First Impact on Health and Social Outcomes Figure. Differences between HF+ACT versus TAU for the High Needs groups in health care outcomes at months Toronto's At Home/Chez Soi study, on 0 September 0 by guest. Protected by copyright. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from

17 Page of Housing First Impact on Health and Social Outcomes Figure. Differences between HF+ACT versus TAU for the High Needs groups over a month period of follow-up for Toronto's At Home/Chez Soi study, ** Differences between HF+ACT and TAU significant at the p<0.0 level with baseline values taken into account Differences between HF+ACT and TAU significant at the p<0.0 level for that point in the follow-up BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

18 Page of BMJ Open Housing First Impact on Health and Social Outcomes Table. At Home/Chez Soi Toronto site treatment by time Interactions across months of follow-up INT vs. TAU Time point(s) vs. 0 vs. 0 vs. 0 vs. 0 Continuous Outcomes Mean CI p-value Mean CI p-value Mean CI p-value Mean CI p-value EQD* CSI QoLI0 Total Score Qoli0 Family Qoli0 Finance Qoli0 Leisure Qoli0 Living* Qoli0 Social Qoli0 Safety Qoli0 Global* MCAS* CIS Psychological Count Outcomes Rate CI p-value Rate CI p-value Rate CI p-value Rate CI p-value GAIN CIS Physical ER Visits Arrests* on 0 September 0 by guest. Protected by copyright. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from

19 Page of Housing First Impact on Health and Social Outcomes Figure. Toronto At Home/Chez Soi High Needs Group over months of follow up. Time by Treatment Interactions for outcomes that differed over the two year follow-up period on 0 September 0 by guest. Protected by copyright. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from

20 How did a Housing First Intervention Improve Health and Social Outcomes among Homeless Adults with Mental Illness in Toronto? Two Year Outcomes from a Randomized Trial Journal: BMJ Open Manuscript ID bmjopen-0-00.r Article Type: Research Date Submitted by the Author: 0-Jun-0 Complete List of Authors: O'Campo, Patricia; St.Michael's Hospital, Center for Research on Inner City Health; University of Toronto, Dalla Lana School of Public Health Stergiopoulos, Vicky; St. Michael's Hospital, Centre for Research on Inner City Health; University of Toronto, Department of Psychiatry Nir, Pam; Cota Levy, Matthew; University of Toronto, Department of Psychiatry Misir, Vachan; St. Michael's Hospital, Centre for Research on Inner City Health Chum, Antony; London School of Hygiene & Tropical Medicine, Social and Environmental Health Research Arbach, Bouchra; Cota Nisenbaum, Rosane; St. Michael's Hospital, Centre for Research on Inner City Health; University of Toronto, Dalla Lana School of Public Health To, Matthew; St. Michael's Hospital, Centre for Research on Inner City Health Hwang, Stephen W. ; St. Michael's Hospital, Centre for Research on Inner City Health; University of Toronto, Division of General Internal Medicine <b>primary Subject Heading</b>: Mental health Secondary Subject Heading: Public health, Health services research Keywords: Homelessness, Mental Illness, Housing Stability, Housing First, Health services utilization BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

21 Page of BMJ Open How did a Housing First Intervention Improve Health and Social Outcomes among Homeless Adults with Mental Illness in Toronto? Two Year Outcomes from a Randomized Trial Patricia O'Campo a,b, Vicky Stergiopoulos a,c, Pam Nir e, Matthew Levy c, Vachan Misir a, Antony Chum f, Bouchra Arbach e, Rosane Nisenbaum a, b, Matthew J To a, Stephen W. Hwang a,d a Centre for Research on Inner City Health, St Michael's Hospital, Toronto, Ontario, Canada b Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada c Department of Psychiatry, University of Toronto, 0 College Street, th Floor, Toronto, Ontario MT R, Canada d Division of General Internal Medicine, Department of Medicine, University of Toronto, Toronto, Ontario, Canada e Cota, Toronto, Ontario, Canada f Social and Environmental Health Research, London School of Hygiene & Tropical Medicine, London, UK Corresponding author: Patricia O`Campo Centre for Research on Inner City Health Li Ka Shing Knowledge Institute, St. Michael s Hospital 0 Bond Street, Toronto, Ontario MB W, Canada o`campop@smh.ca --0 Keywords: homelessness, RCT, mental illness, housing stability, Housing First, health services utilization BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

22 Page of Abstract OBJECTIVES: We studied the impact of a Housing First intervention on housing, contact with the justice system, health care utilization and health outcomes among At Home/Chez Soi randomized trial participants in Toronto, a city with an extensive service network for social and health services for individuals who are experiencing homelessness and mental illness. METHODS: Participants identified as high needs were randomized to receive either the intervention which provided them with housing and supports by an Assertive Community Treatment team (HF+ACT) or treatment as usual (TAU). Participants (N=) had in-person interviews every three months for two years. RESULTS: The HF+ACT group spent more time stably housed compared to the treatment as usual (TAU) with the mean difference between the groups of.% (%CI:.%,.%, p<0.000). Accounting for baseline differences, HF+ACT showed significant improvements over TAU for community functioning, selected quality of life subscales, and arrests at some time points during follow up. No differences between HF+ACT and TAU over the follow-up were observed for health service utilization, community integration, and substance use. CONCLUSIONS: Housing First for individuals with high levels of need increased housing stability and selected health and justice outcomes over two years in a city with many social and health services. Significant Outcomes The Housing First strategy resulted in immediate housing stability and observer rated mental health which remained high over the full follow-up period among those in the treatment arm. Participant functioning psychiatric hospitalizations and number of arrests improved among participants in the treatment compared to the control arm. In a well-resourced urban setting with myriad services to individuals who are homeless, those in the control arm were also able to achieve gains over the follow-up, though not as great as those in Housing First arm. Limitations Interviewers were not blinded to treatment status of participants. All but one outcome measure relied on self-reports. The intensity of observations may have facilitated change in both treatment and control groups BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

23 Page of BMJ Open Introduction Innovations in interventions serving individuals who experience long term homelessness are gaining ground []. The shift away from a reliance on a cycle of shelters and transitional housing toward providing permanent supportive housing has been documented across North America and Europe [-]. "Treatment first" models that seek to stabilize individuals with regard to mental illness and substance use prior to providing permanent housing are giving way to "housing first" models that are built on principles of consumer choice and mental health recovery while providing immediate access to housing and optional community supports. Housing First (HF) has been documented to improve a range of outcomes including housing stability, quality of life, vocational opportunities, and community integration over treatment first or usual care models [-] but less so for substance abuse outcomes []. The At Home/Chez Soi randomized trial was conducted in five cities across Canada and enrolled individuals with serious mental illness who were experiencing chronic homelessness in the largest trial to date of the Housing First model []. At Home/Chez Soi (AH/CS) consisted of scattered-site housing and off-site community supports promoting participant choice, personalized goals and fostering of resilience and empowerment. These supports were tailored to those who demonstrated high or moderate needs as assessed at baseline identified by severity of mental health conditions and levels of mental health care use and social problems experienced in the year prior to enrollment. Participants randomized to receive the intervention were provided with housing and supports while those randomized to treatment as usual (TAU) were not provided any active intervention or support from the trial. AH/CS was intentionally implemented in a variety of settings across Canada, with Toronto being a site where a large proportion of those enrolled were born outside of Canada or considered themselves to belong to an ethnoracial minority group. Moreover, Toronto has a broad network of mental health services --inpatient and outpatient services, assertive community treatment, court support services, crisis programs (e.g., mobile crisis team and the distress line), and ethno-racial focused agencies-- which serve individuals who are homeless and were available to those in TAU. The AH/CS intervention tailors services to level of need during initial assessment (high and moderate needs). Our Housing First intervention provided intensive training of AH/CS staff on the HF model and principles of mental health recovery, and implemented fidelity assessments at points in time. Follow-up rates over time were high. While results for selected outcomes for the high needs portion of AH/CS for the cross-site sample has been published [], examining site specific findings can yield important information about how the intervention works locally and can inform needed improvements to local services. Findings from the cross-site analysis may differ from the site specific results because the TAU conditions vary across cities. Of the five intervention cities, Toronto had one of the richest service settings at the start of the trial, all of which would have been available to the TAU group: ACT teams were operating in Toronto, some with a focus on serving individuals who are homeless; more than twenty Community Health Centres many with specialized services for individuals experiencing homelessness; over forty shelters that serve individuals living with homelessness with over 0% providing some type of health service; over 000 supportive housing units designed for individuals with persistent mental illness; and a city run program, Streets to Homes, that links individuals living with homelessness to temporary or permanent housing. Sites adapted core Housing First strategies to suit their own contexts and populations while still being adherent to the basic principles of HF and ACT [-]. Close to half of the population in Toronto was born outside of Canada, thus the Toronto-site focused on those participants who identify as ethnoracial minorities. Toronto's high needs group was 0% ethnoracial compared to % for the five-city ACT sample []. Prior large scale social interventions have noted the importance of including a focus on adaptation and context in understanding how and why programs are or are not successful [0]. There were also demographic differences at baseline in Toronto's sample compared to the full cross-site sample such as the proportion of women (% for Toronto versus % for the full sample), substance use problems (0% for Toronto versus % for the full sample) []. Thus, an examination of site specific findings yields further insights into whether and how Housing First strategies work in service rich settings and can inform changes at the local level to improve existing services. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

24 Page of Methods Sampling and Recruitment Participants for this study were from the Toronto site of the At Home/Chez Soi study. For inclusion in the study, at the time of enrollment participants had to be >= years of age, absolutely homeless or precariously housed, have a confirmed major mental disorder, and not be served by an Assertive Community Treatment (ACT) program[, ]. Participants were recruited from October 00 to July 0 using a targeted recruitment where local service providers followed a structured referral strategy and those referred were further screened for eligibility by the study team to ensure that the study sample approximated those experiencing homelessness in Toronto[, ] (see Figure ). Referrals came from shelters, drop-in centres, hospitals, outreach programs, and service providers and were reviewed by an intake coordinator for eligibility and assessment. A more detailed description of recruitment and exclusions at baseline can be found elsewhere []. At baseline, participants were assessed for disability and severity of mental health problems and then stratified into a high or moderate needs group. Participants with a Multnomah Community Ability Scale (MCAS) score of less than, indicating at least moderate disability, a diagnosis of a psychotic or bipolar disorder and meeting one of the following criteria: being hospitalized for a mental illness at least twice in one year over the last five years, indication of co-morbid substance use, and recent arrest or incarceration were randomized into the high needs group. High needs participants were randomized after their baseline interview into treatment as usual (TAU) or in the Housing First plus Assertive Community Treatment (HF+ACT) arm of the trial. HF+ACT participants were provided with a rent allowance of $00. The ACT mental health team had a participant-to-staff ratio of 0:, available days/week; hours/day [] and provided a number of relevant services which included case management, initial and ongoing assessment, psychiatric care, employment and housing assistance, family support and education, substance use services and intensive support to allow participants to live successfully in the community []. This study was approved by the Research Ethics Board of St. Michael s Hospital in Toronto and registered with the International Standard Randomized Control Trial Number Register (ISRCTN0). Data Collection Data were collected by in-person interviews every three months for a two year period and entered into a secure webbased database system. Participants were financially compensated when interviewed. For our primary, secondary and exploratory outcomes described below, data were collected and analyzed at baseline and,, and month time points. Housing stability, the proportion of days spent in long term housing while enrolled in the study, was the primary outcome of interest and was assessed using the Residential Timeline Follow-back (RTLFB) Questionnaires at each interview which captured participants month residential history (i.e., move in/out dates for each type of residence such as street, temporary housing, permanent housing, emergency/crisis and institution) []. Among participants with at least one follow-up interview, the percentage of days stably housed was calculated as the total number of days stably housed divided by the total number of days for which any type of residence data were provided by the participant. Two secondary outcomes were assessed, Quality of life was measured via the Euro Qol EQ-D visual analogue which enables respondents to rate their health on a vertical scale capturing the "worst imaginable health state" (score of 0) to "best imaginable health state" (score of 00)[]. Functional ability was assessed by trained interviewers at the end of the interview using the Multnomah Community Ability Scale (MCAS), a - item scale capturing degree of functional ability among individuals living with mental illness in the areas of health, adaptation, social skills and behavior. Higher scores indicate little to no impairment [, ]. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

25 Page of BMJ Open The remaining outcomes examined were considered exploratory and were examined to generate rather than test hypotheses about the impact of Housing First. Lehman Quality of Life Interview 0 captured disease-specific quality of life through subscales (family, finances, leisure, living situation, safety, social, and global). Higher scores, ranging from -0, are indicative of better quality of life []. The Colorado Symptom Index (CSI) was used to assess Mental Health symptom severity through questions asking how often participants experienced specific psychiatric symptoms at least every day to not at all (range -) generating an overall summary score ranging from to 0 with lower scores reflecting fewer symptoms[] []. The number of days spent in a hospital (psychiatric, general and psychiatric unit, separately) was calculated using the data obtained from the RTLFB questionnaire. The RTLFB assesses the days spent in each hospital type at each interview point and the outcome considered was the sum of days over all observation periods. []. Substance use severity was measured via the -item GAIN Substance Use Disorder Scale Short Screener (GAIN SS) which assesses the number of substance-related problems. Severity was derived by counting the number of problems in the past month with a range from 0- and higher values corresponding to a greater problem severity. Community integration was measured using the Community Integration Scale (CIS) which assesses psychological integration based upon levels of satisfaction with home and neighbourhood ranging from -0, and physical integration from counts of social activities such as going to a concert, meeting people for coffee, and participating in sports on a scale of 0- with higher scores reflecting greater integration []. Health service and justice use were also measured using two items from the Health Service and Justice Service Use (HSJSU) Questionnaire: the number of emergency department visits and the number of arrests, in the last months []. Statistical Power and Analyses Site specific sample sizes were set to 00 participants per group at enrollment and at per group at months to account for an attrition rate of %. These numbers were based upon the assumption of an α of 0.0, a β of 0.0 and a moderate effect size of 0. [, ] which was anticipated for the primary and secondary outcomes. Descriptive statistics were first generated for baseline variables among the HF + ACT and TAU groups. Data for each participant were included up until the end of the year follow up or until his or her point of death. Days stably housed over the -year period was examined using a two sample t-test which compares the average days stably housed for the HF + ACT and TAU group. For the number of days hospitalized over the -year period, a zero-inflated negative binomial model was used to account for the likelihood of excessive zeros and over dispersion (i.e. variance larger than the mean) among participants who had at least one day of hospitalization[] (PROC GENMOD in SAS (SAS Institute) version.). Some participants did not provide residence days for the full -years follow-up (due to missing data, attrition or death), and therefore this model also included an offset as the natural log of total reported residence years. The mean number of days hospitalized, among those admitted during the months period, rate ratios and % confidence intervals were also estimated. We generated figures to display the data for all secondary and exploratory outcomes for HF+ACT and TAU over month follow-up period. For secondary and exploratory outcomes, which were measured longitudinally from baseline, linear and generalized linear mixed models were used for continuous and count outcomes, respectively, with a random intercept to estimate subject-level variation. An indicator for the treatment arm was included in the models, with the TAU group as the reference group and also a categorical time variable (0,,, and ) with baseline as the reference category. By including the baseline outcomes in the model, changes between arms over time were examined through the use of a treatment by time interaction. For continuous outcomes this value may be interpreted as the difference in mean change from baseline, whereas for count outcomes these values correspond to the ratio of the average number of events between arms (a ratio between groups) and the baseline ratio (i.e. a ratio of ratios) at the,, and months measurement periods. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

26 Page of Multiple imputation with chained equations using the mi impute chained command in STATA (StataCorp) version was conducted to impute missing data due to loss to follow-up, withdrawal, skipped interviews, nonresponse on specific items and lack of interviewer confidence in participant responses. Forty imputed datasets were generated and analyzed using SAS PROC MIXED for continuous outcomes and PROC GLIMMIX for count outcomes, assuming an unstructured correlation matrix for the repeated measures. Results were combined across the 0 datasets using PROC MIANALYZE []. All tests were two-sided and statistical significance was set at p-value <0.0. Results Of the participants recruited into the study in Toronto, participants met criteria for the high needs group. Of these high needs participants, were randomly allocated to the HF+ACT group and 00 participants were randomly allocated to the TAU group. By months, (%) high needs participants were lost to follow-up, (%) died and (%) were withdrawn (See Figure ).Demographic and mental health characteristics of the participants at baseline are presented in Table. For our primary outcome, over the -months period, participants in the HF+ACT group spent more time stably housed compared to the TAU (mean difference =.% ( %CI:.,., p<0.000); group means:.% (% CI:.,.),.% (% CI:.,.) for HF+ACT and TAU, respectively. We also break this down by year in Table which illustrates the proportion of participants stably housed by year of follow-up. For the EQ-D quality of life, one of our secondary outcomes, there were improvements over the month period for both HF+ACT and TAU groups, but there was no significant treatment by time interaction (see Table ). The mean change from baseline for the MCAS score, another secondary outcome, was found to be significant for the HF+ACT group over the TAU group at both months, with a treatment difference of.0 [% CI: (0.,.), p=0.0] and at months, with a difference of. [%CI: (0.,.), p=0.0]. These mean changes in the MCAS score over time were indicative of improvement over baseline for the HF+ACT group compared to the TAU group (Table ). Results for our exploratory outcomes varied. Among those with at least one hospitalization, participants in the HF+ACT group experienced significantly fewer days in a psychiatric hospital compared with TAU group (HF+ACT mean=., %CI: (0., 0.); TAU mean=., %CI: (.,.0); Rate Ratio: 0., %CI: (0., 0.), p=0.00) (see bars in the middle of Figure, dark bars are the treatment group and light bars are TAU, vertical lines are % CIs). For stays in a psychiatric unit or in a general hospital, however, there were no significant differences observed between the HF + ACT and TAU(Figure ). When changes were compared between groups taking into account a time by treatment interaction, the HF+ACT group showed greater improvements than the TAU over the two year follow up period for just a few outcomes (see Table and Figures and ). There was a rate reduction for arrests at months compared to baseline, with the HF+ACT group having a greater reduction [0., % CI: (0., 0.), p=0.0] (See Table ).For the QoL Living subscale, a significant mean change from baseline was detected at months [0., %CI: (0.0,.), p=0.0] and for QoL Global at months [0., %CI: (-0.0,.), p=0.0] but for the latter not confirmed at the p<0.0 level (See Table ). Discussion For the primary outcome of housing stability, the HF+ACT arm showed a large and significant difference of.% greater time stably housed over the month follow up period compared the TAU arm (.% of the time stably housed HF+ACT group vs..% for the TAU group). This finding confirms what has been demonstrated in other RCTs and quasiexperiments of the Housing First approach in the USA [,, 0]and more recently multiple sites in Europe []. Despite variations in country and local contexts in terms of the generosity of the social safety net and rates of homelessness across these studies, similar rates of housing stability have been reported for those enrolled in a Housing First program: BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

27 Page of BMJ Open at or around 0% at one or two years of follow -up [, -]. With regard to our secondary outcomes, for quality of life as measured by EQ-D, both groups improved over the month follow up period with HF+ACT showing greater improvements but these differences did not meet the cutoff of p<0.0. For functional ability (MCAS), another secondary outcome, both groups showed improvements over time and significant differences between HF+ACT and TAU were observed very early ( months) and at the last follow-up points. Because it is assessed by trained observers, this measure compliments our other self-reported outcomes on mental health symptoms. For exploratory outcomes, only one outcome showed no improvement over the two year follow-up period, community integration (CIS) which reflects participation in activities such as going to restaurants, places of worship, libraries or volunteering. For both HF+ACT and TAU CIS levels remained flat over time. A previous study with a pre-post design in the USA reported similar findings of no change over a one year follow up in almost all of these community integration activities[]. For the remaining exploratory outcomes, both HF+ACT and TAU showed improvements the follow-up period, but the treatment by time interactions were not significant with the exception of three outcomes: psychiatric unit hospitalizations, arrests, and QOL-Living. Our study showed a treatment effect for days spent in psychiatric hospitals, but not in general hospitals or psychiatric units, though time spent in the latter two was very small over the months. Gulcur et al., (00), in a two year follow-up study examined psychiatric hospitalizations and among those recruited from psychiatric hospitals, Housing First reduced overall stays[]; other types of hospitalization were not examined in this study. While arrests were reduced among both HF+ACT and TAU in our study, arrests in HF+ACT decreased significantly more than in TAU over baseline levels at the month follow-up point. In one of the only studies of jail time among Housing First participants, investigators in the USA used a pre-post design to demonstrate significant reductions in arrests and days spent in jail attributed to the intervention[]. Clifasefi and colleagues who studied reasons for arrests before and after Housing First enrollment concluded that bookings were almost all misdemeanors and were due to issues of homelessness (e.g., public order crimes, theft, court order violation) which were therefore significantly reduced once participants were housed[]. We examined seven subscales (family, finances, leisure, living situation, safety, social and global) of the Lehman Quality of Life Interview and only for the Living subscale did we see significant improvements immediately after enrollment between HF+ACT and TAU accounting for treatment by time interactions. This difference might reflect the significant change in perceptions of living conditions upon receiving housing in the HF+ACT arm which would not have been the case for TAU as relatively few were housed within the first six months of the study, just over 0% for TAU and 0% for HF+ACT at the six month point []. A prior pre-post study measuring the Lehman QOL subscales found that at one year post enrollment significant improvements were seen for the living, family relations, financial situation and social subscales[]. While we did see improvements in our sample on all these same subscales of the QoL-0, both the HF+ACT and TAU groups increased at similar rates over time. We did not see greater improvements among the HF+ACT group for mental health as measured by the Colorado Symptom Inventory (CSI) compared to TAU as we anticipated. This is surprising given than the interviewer rated MCAS showed significant improvements over TAU for early and later time points and also psychiatric hospitalizations were significantly reduced in the HF+ACT group over the follow-up period. Past studies of Housing First have also reported no improvements in mental health symptoms over time (e.g.,[]). On the other hand, improvements in substance use (via the GAIN) were also not documented in our sample over the month period which often goes hand in hand with mental health symptoms. It may also be that CSI is less sensitive to detecting improvements in mental well-being than the MCAS. Overall, for our secondary and exploratory outcomes, our data supported fewer of the hypothesized changes than we originally anticipated would occur as a result of the HF+ACT. There are a number of possible reasons why we did not see more improvements in more of the outcomes. Longitudinal evaluations that enroll participants experiencing episodic crises (e.g., with housing, mental health, or substance use issues) often experience results that reflect `regression to the mean` where participants naturally improve regardless of the treatments being administered. Given BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

28 Page of that we saw improvements in outcomes for participants in both groups over time for all but one of our outcomes, it is certainly possible that our findings reflect this phenomenon. It is also possible that given the intensity of the study activities and observations, even for those in the TAU group, many of whom were contacted monthly by study staff, that the Hawthorne effect, where participants positively change behaviours as a result of being part of a study, resulted in improvements in both groups. Even if participants did not change their behaviours, providers in the service community may have provided more assistance to those in TAU. Given the service rich context in Toronto, including the availability of housing and housing supports, many TAU participants had access to housing and similar mental health and addictions services as HF participants resulting in minimal differences between the two groups over the follow-up period time. While it is beyond the scope of this paper to compare the intensity of service participation for HF+ACT and TAU as it requires additional data from across myriad sectors of the service system, we hope to investigate this further in the future. Sample size may have been an issue which we discuss further below. A major strength of the At Home/Chez Soi approach was the supports (e.g., training, staffing resources, ability to innovate and adapt to local contexts) provided to ensure that sites adhered to the HF model with fidelity assessments occurring at two time points in the project. Fidelity to Housing First program components has been associated with better housing and health outcomes in prior studies[]. Fidelity assessments in our study, conducted early to promote mid-course corrections and later to ascertain whether program fidelity was maintained, confirmed that the AH/CS program was highly adherent in all areas across the full follow-up period including housing choice, service philosophy, service array and program structure [, ]. The Toronto team benefitted from the training and the local communities of practice created to ensure that housing and support providers were adhering to and not deviating from the core models [, ]. Yet, despite the high ratings for fidelity assessments, Toronto had other advantages that may have contributed to the overall improvements over time for all participants and the small differences observed between the HF+ACT and TAU groups. Toronto has a number of active housing programs with the largest one being Streets to Homes. It has been available prior to At Home/Chez Soi beginning in February 00, where outreach workers house approximately 00 homeless people per year with % remaining housed after year []. In addition, more than 0 organizations provide various mental health or supportive housing services in the city []Thus, in Toronto, it is possible that the added benefit of the At Home/Chez Soi implementation of HF+ACT was small at best, given the rich array of preexisting ACT and housing first programs. One limitation to our study is that our interviewers could not be blinded to the treatment status of participants. This might have impacted our finding of more positive functional ability as measured by MCAS for example. Our sample size calculations were based upon a moderate effect size and some of our findings illustrated smaller impacts of the program. Thus, our sample size may have been insufficient to detect effect sizes smaller than a moderate effect. However, analyzing our data using longitudinal methods helped us to overcome the problem of low statistical power. Our findings have implications for services provided to individuals living with homelessness and severe mental illness in Toronto. The good news is that this service rich setting --including numerous ACT teams, housing and supports, shelters and health service agencies (e.g., CHCs)--enabled a significant proportion of the TAU group to improve in health and social outcomes over time. Yet, Housing First participants performed significantly better than TAU in the area of housing outcomes, psychiatric hospitalizations and mental health (as rated by independent observers). These findings suggest that there is room for improvement in the system of services for this population in Toronto. One area requiring more attention is systems integration of these services within sectors (e.g., health and housing) and across sectors to better serve clients facing multiple health and social challenges as well as to reduce duplication. HF+ACT had to overcome the siloed nature of these two sectors as part of the intervention [] which partially explained why HF+ACT participants were able to experience greater improvements than those in TAU. Thus, applying those lessons to the overall service environment in Toronto should benefit all clients being served by those systems. In summary, our findings suggest that even in a city with extensive mental health and housing services like Toronto, for a handful of outcomes, we can do better than providing (uncoordinated) services to high risk populations. This is supported by the few significant differences we saw in our data but, perhaps more importantly, by the improvements early in the program, right after participants are housed, for some of our outcomes in the HF+ACT group compared to BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

29 Page of BMJ Open the TAU group. On the other hand, our data also suggest that it may take more time for outcomes such as arrests to show improvements and that longer follow-up periods for two or more years might yield additional information about how HF+ACT impacts health and social outcomes for this population. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

30 Page 0 of Contribution statement Conceived of the research: PO VS SWH PN ML Analyzed the data: PO VM RN Wrote the paper: PO VS SWH RN PN, ML BA MT Competing interest There are no competing interests Funding This work was supported by Mental Health Commission of Canada Data sharing statement The At Home/Chez Soi project has a process by which interested investigators who would like to use the data for publication can make a formal request. The formal request is reviewed by a cross-site committee and as long as those particular analyses have not already been undertaken approval and data sharing can take place. 0 BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

31 Page of BMJ Open References. Doran K. M., Misa E. J., and Shah N. R., Housing as health care-new York's boundary-crossing experiment. New England Journal of Medicine, 0. (): p. -.. G., J., ParkinsonS., and Parsell C., Policy shift or program drift? Implementing Housing First in Australia. American Journal of Psychiatry, 0. (): p. -.. Nelson G., Goering P., and Tsemberis S., Housing for people with lived experience of mental health issues: Housing First as a strategy to improve quality of life. Community psychology and the socio-economics of mental distress: International perspectives, 0: p Tsemberis S., Housing First: The Pathways model to end homelessness for people with mental illness and addiction manual. European Journal of Homelessness, 0. ().. Gilmer TP., et al., Effect of full-service partnerships on homelessness, use and costs of mental health services, and quality of life among adults with serious mental illness homelessness services in mentally ill adults. Arch Gen Psychiatry, 00. : p. -.. Gulcur L., et al., Housing, hospitalization, and cost outcomes for homeless individuals with psychiatric disabilities participating in continuum of care and housing first programmes. Journal of Community & Applied Social Psychiatry, 00. : p. -.. Leff HS., et al., Does one size fit all? What we can and can t learn from a meta-analysis of housing models for persons with mental illness. Psychiatric Services 00. 0(): p. -.. Nelson G., Aubry T., and Lafrance A., A review of the literature on the effectiveness of housing and support, assertive community treatment, and intensive case management interventions for persons with mental illness who have been homeless. American Journal of Orthopsychiatry, 00. (): p Patterson ML., et al., Housing First improves subjective quality of life among homeless adults with mental illness: -month findings from a randomized controlled trial in Vancouver, British Columbia. Soc Psychiatry Psychiatr Epidemiol, 0. : p Siegel C., et al., Tenant outcomes in supported housing and community residences in New York city. Psychiatr Serv, 00. : p. -.. Yanos PT., et al., Exploring the role of housing type, neighborhood characteristics, and lifestyle factors in the community integration of formerly homeless persons diagnosed with mental illness. J Ment Health, 00. : p Rog DJ., et al., Permanent supportive housing: assessing the evidence. Psychiatric Services, 0. (): p. -.. Palepu, A., et al., Housing First Improves Residential Stability in Homeless Adults With Concurrent Substance Dependence and Mental Disorders. American Journal of Public Health, 0. 0(S): p. e0-e.. Goering PN., et al., The At Home/Chez Soi trial protocol: a pragmatic, multi-site, randomised controlled trial of a Housing First intervention for homeless individuals with mental illness in five Canadian cities. BMJ Open, 0. ().. Aubry T., et al., Multiple-City RCT of Housing First With Assertive Community Treatment for Homeless Canadians With Serious Mental Illness. Psychiatr Services, 0. (): p. -.. Nelson G., et al., Implementation and fidelity evaluation of the Mental Health Commission of Canada's At Home/Chez Soi project: Cross-site report O'Campo P., et al., Strategies to balance fidelity to housing first principles with local realities: lessons from a large urban centre. J Health Care Poor Underserved, 0. (): p. -.. Keller, C., et al., Initial implementation of Housing First in five Canadian cities: how do you make the shoe fit, when one size does not fit all?. Am J Psychiatr Rehabilitation, 0. : p... Goering, P., et al., National At Home/Chez Soi Final Report. 0, Mental Health Commission of Ca: Calgary, AB. 0. Weisz, A., Spouse Assault Replication Program: Studies of Effects of Arrest on Domestic Violence.. 00, National Resource Center on Domestic Violence/Pennsylvania Coalition Against Domestic Violence, Retrieved April 0, from: Harrisburg, PA: VAWnet. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

32 Page of Aubrey T., et al., One-year outcomes of a randomized controlled trial of Housing First with ACT in five Canadian cities. Psychiatric Services. Psychiatric Services, 0. (): p. -.. Hwang S., et al., Ending Homelessness among people with mental illness: the At Home/Chez Soi randomized trial of a Housing First Intervention in Toronto. BMC Public Health, 0. (): p... City of Toronto Shelter Support and Housing Administration, 00 Street Needs Assessment Results. Toronto: City of Toronto Tsemberis S., et al., Measuring homelessness and residential stability: The residential time-line follow-back inventory. Journal of Community Psychology, 00. : p. -.. Barker, S., et al., A community ability scale for chronically mentally ill consumers: Part II. Applications. Community Ment Health Journal,. 0(): p. -.. Dickerson, F.B., et al., An expanded version of the Multnomah Community Ability Scale: anchors and interview probes for the assessment of adults with serious mental illness. Community Ment Health Journal, 00. (): p. -.. Uttaro T. and A. Lehman, Graded response modeling of the Quality of Life Interview. Evaluation and Program Planning,. : p. -.. Conrad, K.J., et al., Reliability and validity of a modified Colorado Symptom Index in a national homeless sample. Mental Health Services Research, 00. : p. -.. Pearson C., A. Montgomery, and G. Locke, Housing Stability Among Homeless Individuals With Serious Mental Illness Participating In Housing First Programs. Journal of Community Psychology. (): p Stefancic, A. and S. Tsemberis, Housing First for long-term shelter dwellers with psychiatric disabilities in a suburban county: a four-year study of housing access and retention. J Prim Prev, 00. (-): p. -.. Busch-Geertsema, V., Housing First Europe Final Report. 0, European Union Programme for Employment and Social Solidarity Bremen/Brussels 0.. Henwood, B.F., et al., Quality of life after housing first for adults with serious mental illness who have experienced chronic homelessness. Psychiatry Res, 0. 0(-): p. -.. Clifasefi, S.L., D.K. Malone, and S.E. Collins, Exposure to project-based Housing First is associated with reduced jail time and bookings. Int J Drug Policy, 0. (): p. -.. Tsemberis, S., L. Gulcur, and M. Nakae, Tsemberis S, Gulcur L, Nakae M: Housing First, consumer choice, and harm reduction for homeless individuals with a dual diagnosis. American Journal of Public Health, 00. : p. -.. MacNaughton, E., et al., Implementing Housing First Across Sites and Over Time: Later Fidelity and Implementation Evaluation of a Pan-Canadian Multisite Housing First Program for Homeless People with Mental Illness. Am J Community Psychology, 0. : p. -.. Nelson, G., et al., Early implementation evaluation of a multi-site housing first intervention for homeless people with mental illness: A mixed methods approach. Evaluation and Program Planning, 0. : p. -.. Falvo, N.H., Toronto s Streets to Homes Program. In: Hulchanski, J. David; Campsie, Philippa; and S.H. Chau, Stephen; Paradis, Emily (eds.) Finding Home: Policy Options for Addressing Homelessness in Canada (e-book), Chapter.. Toronto: Cities Centre, University of Toronto., Chapter. Homelessness, Toronto s Streets to Homes Program., in Finding Home: Policy Options for Addressing Homelessness in Canada, J.D. Hulchanski, et al., Editors. 00, Cities Centre, University of Toronto: Toronto.. The Access Point. The Toronto Mental Health and Addictions Access Point,. June 0]; Available from: BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

33 Page of BMJ Open Table. At Home/Chez Soi Toronto site, sample demographic, health and social characteristics at baseline for the total sample of High Needs participants, those receiving Assertive Community Treatment (HF+ACT) and those in the Treatment as Usual (TAU), Characteristics ACT (N=) TAU (N=00) Total (N=) Age years, mean (SD). (.0).0 (.) 0.0 (.) Gender, N (%) Female (.0) () (.) Male (.0) () (.) Other 0 (0) () (.0) Marital Status, N (%) Married/Partnered (.) () (.) Divorced/Separated/Widowed (0) () (.) Single/Never married (.) () (.) Country of Birth, N (%) Canada (.) 0 (0) (.) Other (.) 0 (0) (.) Ethnic or Cultural Identity, N (%) Aboriginal (.) () 0 (.0) Ethno-Racial (.) () 0 (.) White (.) () (.) Housing Status, N (%) Absolutely Homeless (.) () (.) Precariously Housed (.) () (.) Lifetime duration of homelessness years, mean (SD). (.). (.). (.) Education, N (%) Less than High School (.) (.) 0 (.) Completed High School (.) (.) (.) Some Post-Secondary School (.0) (.) (.) MCAS score, mean (SD). (.). (.). (.) MINI Diagnostic Categories, N (%) Depressive Episode (.) () (.) Manic or Hypomanic Episode (.) () 0 (0.) Post-Traumatic Stress Disorder (.) () (.) Panic Disorder (.) () (.) Mood Disorder with Psychotic Features (.) () (.) Psychotic Disorder (.) 0 (0) (.) Substance Use Related Problems (.) () 0 (0.) Suicidality Level No/Low (.) () (.) Moderate (.) () (.) High (.) 0 (0) (.) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

34 Page of Table : Aggregated proportion (% CI) of participants stably housed by period of follow-up and trial arm. Period of follow-up HF+ACT TAU -months.% (.%,.%).% (.%,.%) -months.% (.0%,.%).0% (0.%,.%) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

35 Page of BMJ Open Table. At Home/Chez Soi Toronto site treatment by time Interactions across months of follow-up INT vs. TAU Time point(s) vs. 0 vs. 0 vs. 0 vs. 0 Continuous Outcomes Mean CI p-value Mean CI p-value Mean CI p-value Mean CI p-value EQD CSI QoLI0 Total Score Qoli0 Family Qoli0 Finance Qoli0 Leisure Qoli0 Living* Qoli0 Social Qoli0 Safety Qoli0 Global MCAS* CIS Psychological Count Outcomes Rate CI p-value Rate CI p-value Rate CI p-value Rate CI p-value GAIN CIS Physical ER Visits Arrests* on 0 September 0 by guest. Protected by copyright. BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from

36 Page of xmm (00 x 00 DPI) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

37 Page of BMJ Open x0mm (00 x 00 DPI) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

38 Page of xmm (00 x 00 DPI) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

39 Page of BMJ Open xmm (00 x 00 DPI) BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from on 0 September 0 by guest. Protected by copyright.

40 Page 0 of Section/Topic Title and abstract Introduction Background and objectives CONSORT 00 checklist of information to include when reporting a randomised trial* Item No Checklist item Reported on page No a Identification as a randomised trial in the title Page b Structured summary of trial design, methods, results, and conclusions (for specific guidance see CONSORT for abstracts) Page a Scientific background and explanation of rationale Page b Specific objectives or hypotheses Page Methods Trial design a Description of trial design (such as parallel, factorial) including allocation ratio Page - b Important changes to methods after trial commencement (such as eligibility criteria), with reasons n/a Participants a Eligibility criteria for participants Page b Settings and locations where the data were collected Page - Interventions The interventions for each group with sufficient details to allow replication, including how and when they were Page actually administered Outcomes a Completely defined pre-specified primary and secondary outcome measures, including how and when they Page were assessed b Any changes to trial outcomes after the trial commenced, with reasons n/a Sample size a How sample size was determined Page b When applicable, explanation of any interim analyses and stopping guidelines n/a Randomisation: Sequence a Method used to generate the random allocation sequence Page generation b Type of randomisation; details of any restriction (such as blocking and block size) Page Allocation concealment mechanism Mechanism used to implement the random allocation sequence (such as sequentially numbered containers), describing any steps taken to conceal the sequence until interventions were assigned Implementation 0 Who generated the random allocation sequence, who enrolled participants, and who assigned participants to Page interventions Blinding a If done, who was blinded after assignment to interventions (for example, participants, care providers, those n/a CONSORT 00 checklist Page on 0 September 0 by guest. Protected by copyright. Page BMJ Open: first published as 0./bmjopen-0-00 on September 0. Downloaded from