Papers by Julian M Somers
Supplemental Material for Bayesian sensitivity analysis for unmeasured confounding in causal medi... more Supplemental Material for Bayesian sensitivity analysis for unmeasured confounding in causal mediation analysis by Lawrence C McCandless and Julian M Somers in Statistical Methods in Medical Research
Underlying the daily lives of people with experiences of homelessness and mental illness is a com... more Underlying the daily lives of people with experiences of homelessness and mental illness is a complex interplay of individual and structural factors that perpetuate cycles of inequity. The introduction of novel methodological combinations within qualitative research has the potential to advance knowledge regarding the experience of health equity by such individuals and to clarify the relationship between these experiences and broader structural inequities. To explore the lived experience of inequity, we present a thematic analysis of narrative interviews in conjunction with timelines from 31 adults experiencing homelessness and mental illness. Use of these methods together enabled a novel and expanded appreciation for the varied ways in which differential access to the social determinants of health influences the trajectories and experiences of inequity for people who are homeless and mentally ill. The further utility of these methods for better understanding the experience of inequ...
Canadian Journal of Community Mental Health, 2020
During the period 2010–2011, when the At Home project was conducted, a questionnaire was sent to ... more During the period 2010–2011, when the At Home project was conducted, a questionnaire was sent to 420 non-governmental organization (NGO) key managers in six Canadian cities to enquire about their collaboration with public services and their perspective on the services for homeless people with serious mental illness (SMI). NGOs constituted a dense network of collaboration among themselves. With regard to public services, housing and shelters were two services that NGOs had frequent contact with, followed by the healthcare addiction sectors and, to a lesser extent, social service and the justice sectors. Education and employment were both located in the network periphery. In general, NGOs viewed housing availability and accessibility to health services as largely unsatisfactory. They called for better public support, coordination, and funding.
Wellbeing, Space and Society, 2020
Abstract We developed, tested, and operationalized a comprehensive objective measurement tool, th... more Abstract We developed, tested, and operationalized a comprehensive objective measurement tool, the Observer-rated Housing Quality Scale (OHQS), to assess housing unit, building, and neighbourhood characteristics in a large trial of Housing First (HF) in 5 Canadian cities. The OHQS provided an opportunity to explore the role of neighbourhood and its contribution to housing stability among formerly homeless individuals. A subset of 423 participants who had become housed in the second year of the trial were systematically selected. The psychometric properties of the neighbourhood quality (NQ) items were assessed within the OHQS. The association between NQ and housing stability was explored using ordinal logistic regression, adjusting for group, site, demographic, housing unit and building quality variables. Inter-rater reliability and internal consistency were very good for the NQ items and subscale. Although NQ was not predictive of housing stability after considering other residence and building features in the multivariable analysis, we assert further analysis is required to explore this relationship, especially for HF programs offering clients choice in housing and neighbourhood.
Journal of Community Safety and Well-Being, 2018
Co-occurring health and public safety concerns involving mental illness, substance use, and homel... more Co-occurring health and public safety concerns involving mental illness, substance use, and homelessness are increasingly prevalent challenges for policymakers in cities worldwide. The Hub model is a roundtable process where the combined resources of diverse agencies are used to mitigate urgent risk of crime, victimization, illness and death, by establishing immediate connections with appropriate services and supports. Initiated in Scotland, the model has been replicated in more than 60 communities across Canada since 2012. In November 2105, the Surrey Mobilization and Resiliency Table (SMART) became the first Hub in British Columbia. Little peer-reviewed research has examined the impact of Hub inter-ventions from a client perspective. We conducted semi-structured interviews with 16 SMART clients and analyzed their responses thematically. We also examined demographic- and intervention-related characteristics reported in the SMART database. Participants described positive experiences...
BMC Psychology, 2019
Background: Indigenous people are over represented among homeless populations worldwide and the p... more Background: Indigenous people are over represented among homeless populations worldwide and the prevalence of Indigenous homelessness appears to be increasing in Canadian cities. Violence against Indigenous women in Canada has been widely publicized but has not informed the planning of housing interventions. Despite historical policies leading to disenfranchisement of Indigenous rights in gender-specific ways, little is known about contemporary differences in need between homeless Indigenous men and women. This study investigated mental health, substance use and service use among Indigenous people who met criteria for homelessness and mental illness, and hypothesized that, compared to men, women would have significantly higher rates of trauma, suicidality, substance dependence, and experiences of violence. Methods: This study was conducted using baseline (pre-randomization) data from a multi-site trial. Inclusion in the current analyses was restricted to participants who self-reported Indigenous ethnicity, and combined eligible participants from Vancouver, BC and Winnipeg, MB. Logistic regression analyses were used to model the independent associations between gender and outcome variables. Results: In multivariable regression models among Indigenous participants (n = 439), female gender was predictive of meeting criteria for PTSD, multiple mental disorders, current high suicidality and current substance dependence. Female gender was also significantly associated with reported physical (AOR: 1.52, 95% CI = 1.10-2.23) and sexual (AOR: 6.31, 95% CI = 2.78-14.31) violence. Conclusions: Our analyses of Indigenous men and women who are homeless illustrate the distinct legacy of colonization on the experiences of Indigenous women. Our findings are consistent with the widely documented violence against Indigenous women in Canada. Housing policies and services are urgently needed that take Indigenous historical contexts, trauma and gender into account. Trial registration: This trial has been registered with the International Standard Randomized Control Trial Number Register and assigned ISRCTN42520374; ISRCTN57595077; ISRCTN66721740.
Suicide and Life-Threatening Behavior, 2019
Suicide is a leading cause of death, yet the accurate prediction of suicidal behavior is an elusi... more Suicide is a leading cause of death, yet the accurate prediction of suicidal behavior is an elusive target for clinicians and researchers. The current paper examines the predictive validity of the Mini Neuropsychiatric Interview (MINI) Suicidality subscale for suicide attempts (SAs) for a homeless population with mental illness. Methods: Two thousand two hundred and fifty-five homeless individuals with mental illness across five Canadian cities enrolled in the At Home/Chez Soi Housing First trial interviewed at baseline using the MINI Suicidality subscale with 2-year follow-up of self-reported SAs. Results: Two thousand two hundred and twenty-one participants were included in the analysis. High rates of mood and substance use disorders were present (56.5% and 67.4%, respectively). The mean MINI Suicidality subscale score was 7.71. Among 1,700 participants with follow-up data, 11.4% reported a SA over the 2-year study period. MINI Suicidality subscale scores were predictive of SAs (AUC ≥ 0.70) among those with and without a history of SAs, even among those with missing answers. A
Addiction (Abingdon, England), 2017
Evidence supports the effectiveness of Housing First (HF) programmes for people who are experienc... more Evidence supports the effectiveness of Housing First (HF) programmes for people who are experiencing homelessness and mental illness; however, questions remain about its use in people with comorbid substance use disorders (SUD). The aim of this project was to test whether SUD modifies the effectiveness of an HF intervention. Secondary analysis of data from a randomized controlled trial of HF versus treatment-as-usual (TAU) with 24-month follow-up, comparing those with and without SUD at trial entry. Vancouver, Toronto, Winnipeg, Moncton and Montreal, Canada. A total of 2154 participants recruited from 2009 to 2013 and randomized to HF versus TAU (67% male, mean age 40.8 ± 11.2, 25% ethno-cultural minority). All were homeless and had a mental disorder at baseline; 35% reported symptoms consistent with SUD. Housing paired with Intensive Case Management or Assertive Community Treatment. Primary outcomes were days housed and community functioning. Secondary outcomes were general and hea...
PloS one, 2017
No previous experimental trials have investigated Housing First (HF) in both scattered site (SHF)... more No previous experimental trials have investigated Housing First (HF) in both scattered site (SHF) and congregate (CHF) formats. We hypothesized that CHF and SHF would be associated with a greater percentage of time stably housed as well as superior health and psychosocial outcomes over 24 months compared to treatment as usual (TAU). Inclusion criteria were homelessness, mental illness, and high need for support. Participants were randomised to SHF, CHF, or TAU. SHF consisted of market rental apartments with support provided by Assertive Community Treatment (ACT). CHF consisted of a single building with supports equivalent to ACT. TAU included existing services and supports. Of 800 people screened, 297 were randomly assigned to CHF (107), SHF (90), or TAU (100). The percentage of time in stable housing over 24 months was 26.3% in TAU (reference; 95% confidence interval (CI) = 20.5, 32.0), compared to 74.3% in CHF (95% CI = 69.3, 79.3, p<0.001) and 74.5% in SHF (95% CI = 69.2, 79.7...
Schizophrenia Bulletin, 2016
Adherence to antipsychotic medication is a significant challenge among homeless patients. No expe... more Adherence to antipsychotic medication is a significant challenge among homeless patients. No experimental trials have investigated the impact of Housing First on adherence among patients with schizophrenia. We investigated whether Housing First in congregate and scattered-site configurations resulted in superior adherence compared to usual care. Adult participants (n = 165) met criteria for homelessness, schizophrenia, and initiation of antipsychotic pharmacotherapy prior to recruitment to an unblinded, 3-arm randomized controlled trial in Vancouver, Canada. Randomization arms were: congregate Housing First (CHF) with on-site supports (including physician and pharmacy services); scattered-site Housing First (SHF) with Assertive Community Treatment; or treatment as usual (TAU) consisting of existing services. Participants were followed for an average of 2.6 years. Adherence to antipsychotic medication was measured using the medication possession ratio (MPR), and 1-way ANOVA was used to compare outcomes between the 3 conditions. Data were drawn from comprehensive pharmacy records. Prior to randomization, mean MPR among participants was very low (0.44-0.48). Mean MPR in the follow-up period was significantly different between study arms (P < .001) and approached the guideline threshold of 0.80 in SHF. Compared to TAU, antipsychotic adherence was significantly higher in SHF but not in CHF. The results demonstrate that further implementation of SHF is indicated among homeless people with schizophrenia, and that urgent action is needed to address very low levels of antipsychotic adherence in this population (trial registration: ISRCTN57595077).
PLOS ONE, 2016
Background The prevalence of food insecurity and food insufficiency is high among homeless people... more Background The prevalence of food insecurity and food insufficiency is high among homeless people. We investigated the prevalence and correlates of food insecurity among a cohort of homeless adults with mental illness in Vancouver,
Emerging Themes in Epidemiology, 2015
Background: A subgroup of individuals becomes entrenched in a "revolving door" involving correcti... more Background: A subgroup of individuals becomes entrenched in a "revolving door" involving corrections, health, and social welfare services. Little research has investigated the numbers of people that are in frequent contact with multiple public agencies, the costs associated with these encounters, or the characteristics of the people concerned. The present study used linked administrative data to examine offenders who were also very frequent users of health and social services. We investigated the magnitude and distribution of costs attributable to different categories of service for those in the top 10 % of sentences to either community or custodial settings. We hypothesized that the members of these subgroups would be significantly more likely to have substance use and other mental disorders than other members of the offender population. Methods: Data were linked across agencies responsible for services to the entire population of British Columbia spanning justice, health, and income assistance. Individuals were eligible for inclusion in the study if they were sentenced at least once in the Vancouver Provincial Court between 2003 and 2012. We examined the subset of participants who fell within the top 10 % of sentences and at least two of the following service categories: community physician services; hospital days; pharmaceutical costs; or income assistance between 2007 and 2012. We examined two groups of offenders separately (those in the top ten percent sentenced to community supervision or to custody) due to differences in time at risk and availability to receive community-based services. Results: From more than 14,000 offenders sentenced in Vancouver's Downtown Eastside, very High Frequency service users associated with community (n = 216) and custody (n = 107) sentences incurred average attributable public service costs of $168,000 and $247,000 respectively over a 5-year period of observation. Health-related costs for both groups were over $80,000 per person, primarily associated with hospital admissions. Across both groups, 99 % had been diagnosed with at least one mental disorder and over 80 % had co-occurring substance use and another mental disorder. Conclusions: A subset of offenders with concurrent psychiatric disorders receives extremely high levels of service from health, social welfare, and justice sectors in close temporal succession. Members of this subpopulation require targeted supports in order to produce positive outcomes and prevent the perpetuation of a costly and ineffective revolving door.
Psychology, Public Policy, and Law, 2013
Abundant research confirms a high prevalence of substance use and mental disorders in correctiona... more Abundant research confirms a high prevalence of substance use and mental disorders in correctional samples. It is unclear, however, how these contribute to the risk of repeat offending. The present study examined offense trajectories within a Canadian Provincial offender population (N ϭ 31,014), and observed that offenders with non-substance-related mental disorders were at no greater risk of recidivism than those with no diagnosis. In contrast, odds of recidivism were significantly higher among those with substance use and/or co-occurring disorders. These findings add strength to the emerging conclusion that non-substance-related mental disorders are, as a group, less likely to predict recidivism than substance use disorders. Notably, nearly 50% of repeat offenders had a physician-diagnosed substance use disorder in the 5 years prior to their index offense. Results are discussed in relation to the necessity for targeted evidence-based partnerships between health and corrections sectors that are responsive to both public health and safety.
BMC Psychiatry, 2016
Background: Self-reported service use is an integral feature of interventional research with peop... more Background: Self-reported service use is an integral feature of interventional research with people who are homeless and mentally ill. The objective of this study was to investigate the accuracy of self-reported involvement with major categories of publicly funded services (health, justice, social welfare) within this sub-population. Methods: Measures were administered pre-randomization in two randomized controlled trials, using timeline follow back with calendar aids for Health, Social, and Justice Service Use, compared to linked administrative data. Variables examined were: psychiatric admissions (both extended stays of more than 6 months and two or more stays within 5 years); emergency department visits, general hospitalization and jail in the past 6 months; and income assistance in the past 1 month. Participants (n = 433) met criteria for homelessness and a least one mental illness. Results: Prevalence adjusted and bias adjusted kappa (PABAK) values ranged between moderate and almost perfect for extended psychiatric hospital separations (PABAK: 0.77; 95 % confidence interval (CI) = 0.71, 0.83), multiple psychiatric hospitalizations (PABAK = 0.50, 95 % CI = 0.41, 0.59), emergency department visits (PABAK: 0.77; 95 % CI = 0.71, 0.83), jail (PABAK: 0.74; 95 % CI = 0.68, 0.81), and income assistance (PABAK: 0.82; 95 % CI = 0.76, 0.87). Significant differences in under versus over reporting were also found. Conclusions: People who are homeless and mentally ill reliably reported their overall use of health, justice, and income assistance services. Evidence of under-reporting and over-reporting of certain variables has implications for specific research questions. ISRCTN registry: 57595077 (Vancouver at Home Study: Housing First plus Assertive Community Treatment versus congregate housing plus supports versus treatment as usual); and 66721740 (Vancouver at Home study: Housing First plus Intensive Case management versus treatment as usual).
International Journal of Social Psychiatry, 2014
Background: Incarceration and homelessness are closely related yet studied rarely. This article a... more Background: Incarceration and homelessness are closely related yet studied rarely. This article aimed to study the incarcerated homeless and identify specific vulnerabilities, which rendered them different from the nonincarcerated homeless. It also aimed to describe the homeless population and its significant involvement with the criminal justice and enforcement system. Methods: Data were derived from the British Columbia Health of the Homeless Study (BCHOHS), carried out in three cities in British Columbia, Canada: the large urban center Vancouver ( n = 250), Victoria ( n = 150) and Prince George ( n = 100). Measures included socio-demographic information, the Maudsley Addiction Profile (MAP), the Childhood Trauma Questionnaire (CTQ) and the Mini International Neuropsychiatric Interview (MINI) Plus. Results: Incarcerated homeless were more often male (66.6%), were in foster care (56.4%) and had greater substance use especially of crack cocaine (69.6%) and crystal methamphetamine (7...
BMC Psychiatry, 2021
Background Researchers have pointed out the paucity of research investigating long-term consequen... more Background Researchers have pointed out the paucity of research investigating long-term consequences of experiencing homelessness in childhood or youth. Limited research has indicated that the experience of homelessness in childhood or youth is associated with adverse adjustment-related consequences in adulthood. Housing First (HF) has acknowledged effectiveness in improving housing outcomes among adults experiencing homelessness and living with serious mental illness, although some HF clients struggle with maintaining housing. The current study was conducted to examine whether the experience of homelessness in childhood or youth increases the odds of poorer housing stability following entry into high-fidelity HF among adults experiencing serious mental illness and who were formerly homeless. Methods Data were drawn from the active intervention arms of a HF randomized controlled trial in Metro Vancouver, Canada. Participants ( n = 297) were referred to the study from service agenci...
Individuals randomized to abstinence-contingent housing (Oxford House) exhibited superior outcome... more Individuals randomized to abstinence-contingent housing (Oxford House) exhibited superior outcomes at 24 months compared to individuals assigned to usual care (outpatient treatment or self-help groups). Compared to usual care, individuals in the housing condition exhibited significantly lower substance use (31.3% vs. 64.8%), significantly higher monthly income ($989 vs. $440) and significantly lower incarceration rates (3% vs. 9%). 4. Implications for Drug and Alcohol-Free Housing Housing in the present study consisted of rented, multi-bedroom dwellings in lowcrime, residential neighbourhoods. Reductions in criminal activity may have been partially attributable to the structure and location of housing. The authors advise that housing may offer benefits to other subgroups of individuals who are attempting to maintain abstinence, such as ex-offenders, individuals with psychiatric disorders, and those who are homeless. The Oxford House model illustrates the value of peer-based abstinence-contingent housing. 5. Evaluation of paper (research methodology, level of confidence, etc.) The study involved a randomized controlled trial of 150 individuals, with over 90% follow up over 24 months. Collateral information was obtained in order to confirm selfreported abstinence rates at 24 months. The design and execution of the research was robust. Designing and implementing research with this type of peer-run program is difficult and the long term collaborative relationship between the researchers and the program was seen as critical for the successful implementation of the project. It would be challenging to replicate these findings without a similar collaborative research model.
International Journal of Criminal Justice Sciences, 2012
IntroductionThere is emerging consensus in correctional literature that a disproportionate amount... more IntroductionThere is emerging consensus in correctional literature that a disproportionate amount of crime (particularly property crime) is committed by a minority of offenders (Marlow, 2007; Mawby & Worrall, 2004; Millie & Erol, 2006; Vennard & Pearce, 2004). There is also general agreement among scholars that custody alone is a relatively ineffective method of reducing offending (Moore et al., 2006), and that 'getting tough on crime' has been unsuccessful (Andrews & Bonta, 2010). These realizations have spurred a variety of programs in several countries that emphasize inter-agency collaboration as a means of improving offender outcomes and increasing community safety (Mawby & Worrall, 2004). Illustrative programs are predicated on the emerging concept of therapeutic jurisprudence (Public Safety Canada and Alberta Solicitor General, 2010) and growing evidence that successful criminal justice policy requires a focus on evidence-based offender treatment and rehabilitation pra...
Schizophrenia Research
The current study investigated whether a previously reported beneficial effect of methadone maint... more The current study investigated whether a previously reported beneficial effect of methadone maintenance therapy (MMT) on antiretroviral adherence is also present in relation to antipsychotic treatment for schizophrenia. Administrative data were linked over a 17-year period for 1996 people who were dually diagnosed with schizophrenia and opioid dependence and, as an indicator of further marginalization, experienced at least one episode of correctional supervision in British Columbia. Adherence was estimated using the medication possession ratio (MPR ≥ 0.80), calculated in each 120-day period beginning with the first date of concurrent use of MMT and antipsychotic medication. Generalized Estimating Equations were used to estimate the association between independent and dependent variables. The probability of antipsychotic adherence doubled in periods that were preceded by a period of MMT adherence (AOR: P: 2.07; 95% CI: 1.90-2.26). Subgroup and sensitivity analyses yielded results similar to those derived through the primary analysis, examining: conviction history; length of follow-up; initiation of MMT prior to antipsychotic induction; excluding participants who died during the study period; and restricted to participants who received methadone exclusively as part of a MMT program. Despite a strong temporal association between MMT and antipsychotic adherence, overall MPRs for both prescriptions remained <0.50 throughout the study period. Antipsychotic adherence was more than twice as likely following periods of adherence to MMT among dually-diagnosed patients. Research is needed to identify the conditions responsible for MMT adherence, and to further clarify the relationship between opioid agonist treatment and antipsychotic pharmacotherapy in this vulnerable and under-studied population.
Uploads
Papers by Julian M Somers