Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Utilization patterns and health impacts of supervised drug consumption facilities in Vancouver, Canada Kennedy, Mary Clare 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2019_may_kennedy_maryclare.pdf [ 1.94MB ]
Metadata
JSON: 24-1.0376545.json
JSON-LD: 24-1.0376545-ld.json
RDF/XML (Pretty): 24-1.0376545-rdf.xml
RDF/JSON: 24-1.0376545-rdf.json
Turtle: 24-1.0376545-turtle.txt
N-Triples: 24-1.0376545-rdf-ntriples.txt
Original Record: 24-1.0376545-source.json
Full Text
24-1.0376545-fulltext.txt
Citation
24-1.0376545.ris

Full Text

 UTILIZATION PATTERNS AND HEALTH IMPACTS OF  SUPERVISED DRUG CONSUMPTION FACILITIES IN VANCOUVER, CANADA   by   Mary Clare Kennedy   M.A., The University of Victoria, 2013 B.A. (Hons), The University of Victoria, 2009   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Population and Public Health)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    February 2019   © Mary Clare Kennedy, 2019           ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  Utilization patterns and health impacts of supervised drug consumption facilities in Vancouver, Canada  submitted by Mary Clare Kennedy in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Population and Public Health  Examining Committee: Dr. Thomas Kerr, Department of Medicine Co-supervisor Dr. Chris Richardson, School of Population and Public Health Co-supervisor  Dr. Ryan McNeil, Department of Medicine Supervisory Committee Member Dr. Gina Ogilvie, School of Population and Public Health University Examiner Dr. Grant Charles, School of Social Work University Examiner           iii Abstract Background: A growing body of research has demonstrated the role of supervised drug consumption facilities (SCFs) in mitigating harms associated with illicit drug use. However, scientific evidence concerning long-term SCF utilization patterns and health impacts among people who use drugs (PWUD) is still lacking. Further, little is known about novel approaches to SCF programming, including peer involvement as staff, and how this may influence service dynamics. This dissertation sought to: systematically review the literature on the health and community impacts of SCFs; longitudinally examine discontinued use of a SCF; assess the impact of SCF use on all-cause mortality and exposure to violence; and characterize peer involvement in low-threshold SCFs, including how this may shape service engagement and associated health outcomes among PWUD. Methods: Quantitative data were derived from two community-recruited prospective cohort studies of PWUD in Vancouver, Canada. These data were confidentially linked to an external vital statistics database. In addition, observational fieldwork and interview data were drawn from a rapid ethnographic study examining the implementation, operations and impacts of low-threshold SCFs in Vancouver. Multivariable regression analyses of cohort data were used to examine the association between various factors and discontinued SCF use, as well as to longitudinally assess the impact of SCF use on all-cause mortality and exposure to violence. Ethnographic data were analyzed thematically to characterize peer involvement in SCFs.   Results: Most SCF clients discontinued using this service during follow up, and injection cessation co-occurred with the majority of SCF use cessation events. Higher-risk subpopulations of PWUD were less likely to cease using the SCF while actively          iv injecting. Frequent SCF use was protective against all-cause mortality, as well as exposure to violence among men but not women. Peer involvement in SCFs was found to promote service engagement and the reduction of harms among PWUD. Conclusions: These findings support the inclusion of SCFs within the continuum of services for PWUD, particularly given that use of such services may reduce the risk of serious harms, including mortality and exposure to violence. The expansion of peer-run and women-only SCFs may afford opportunities to extend the reach and impact of this intervention.              v Lay Summary Supervised drug consumption facilities (SCFs), where people can use illicit drugs while supervised by trained staff, have increasingly been implemented in response to drug-related harms. This dissertation research sought to address gaps in scientific evidence by examining long-term SCF utilization patterns and health impacts among people who use drugs (PWUD) in Vancouver, Canada. Additionally, this work investigated how involving PWUD as staff at SCFs may influence service use and related outcomes among this population. The study found that three-quarters of SCF clients followed for an average of four years discontinued using this service, most often because they had stopped injecting drugs. Frequent SCF use was protective against mortality, as well as exposure to violence among men but not women. Involving PWUD as staff at SCFs was found to promote service engagement and the reduction of harms. These findings support the inclusion of SCFs within the continuum of services for PWUD.           vi Preface This statement certifies that all of the work presented henceforth was conceived, undertaken, and written by the author, Mary Clare Kennedy (MCK). All empirical research conducted for this dissertation was approved by the research ethics boards at the University of British Columbia/Providence Health Care (certificate H05-50233; H05-50234 and H17-00557). The co-authors of the manuscripts, including Dr. Thomas Kerr (TK), Dr. Chris Richardson (CR), Dr. Ryan McNeil (RM), Dr. Kanna Hayashi (KH), Dr. M-J Milloy (MJM), Dr. Evan Wood (EW), Dr. Jade Boyd (JB), Ms. Alex Collins (AC), Ms. Huiru Dong (HD), Dr. Mohammad Karamouzian (MK), Dr. David Klassen (DK), and Ms. Samara Mayer (SM) made contributions only as is commensurate with supervisory committee, collegial, or co-investigator duties. The principal investigators of the Vancouver Injection Drug Users Study (TK; KH), AIDS Care Cohort to evaluate Access to Survival Services (MJM), and An Ethno-epidemiological Study of the Implementation and Effectiveness of an Innovative and Comprehensive Response to the Opioid Epidemic (RM; TK), from which all empirical analyses were derived, had access to all of the data and as corresponding authors take full responsibility for the integrity of the results and the accuracy of the analyses. Relative contributions of the author, collaborators, and co-authors are described in detail below.  Chapters 1 and 7 are original, unpublished intellectual products of the author. With substantive guidance and input from co-supervisors (TK and CR) and supervisory committee member (RM), MCK searched and reviewed all of the literature presented, designed the research, synthesized the findings and prepared the first and final drafts of these chapters.  A version of Chapter 2 has been published and is reused here with kind          vii permission from Springer Science and Business Media: Kennedy MC, Karamouzian M, Kerr T. Public health and public order outcomes associated with supervised drug consumption facilities: A systematic review. Current HIV/AIDS Reports. 2017;14(5),161-183. MCK designed and led the systematic review with input from TK and MK. In collaboration with MK, MCK conducted the search strategy and selected eligible studies for inclusion. MCK drafted the manuscript and incorporated feedback from MK and TK. MK and TK contributed to the revision of the manuscript.  A version of Chapter 3 is currently under review for peer-reviewed publication: Kennedy MC, Klassen D, Dong H, Milloy MJ, Hayashi K, Kerr T. Discontinued use of a supervised injection facility among people who inject drugs in Vancouver, Canada: A prospective cohort study. MCK designed the study with guidance and input from TK. MCK conducted the statistical analyses and led the writing of the manuscript. DK, TK, HD, KH and MJM provided input to the draft and contributed to the revision of the manuscript.   A version of Chapter 4 is currently under review for peer-reviewed publication: Kennedy MC, Hayashi K, Milloy MJ, Wood E, Kerr T. The impact of supervised injection facility use on all-cause mortality among a cohort of people who inject drugs in Vancouver, Canada. MCK designed the study with guidance and input from TK. MCK conducted the statistical analyses and prepared the first and final drafts of the manuscript. TK, KH, MJM and EW provided input to the draft and contributed to the revision of the manuscript.  A version of Chapter 5 is currently under review for peer-reviewed publication: Kennedy MC, Hayashi K, Milloy MJ, Wood E, Kerr T. Supervised injection facility use and exposure to violence among a cohort of people who inject drugs in Vancouver,          viii Canada: A gender-based analysis. MCK designed the study with guidance and input from TK. MCK conducted the statistical analyses and prepared the first and final drafts of the manuscript. TK, KH, MJM and EW provided input to the draft and contributed to the revision of the manuscript.  A version of Chapter 6 has been published and is reused here with kind permission from Elsevier: Kennedy MC, Boyd J, Mayer S, Collins A, Kerr T, McNeil R. Peer worker involvement in low-threshold supervised consumption facilities in the context of an overdose epidemic in Vancouver, Canada. Social Science and Medicine. 2019;225,60-68. MCK collected data, led the data analysis, and prepared the first and final drafts of the manuscript. RM designed the rapid ethnographic study, oversaw the data collection process, and contributed to data analysis and the revision of the manuscript. JB provided strategic direction in the implementation of the study, and contributed to data analysis and the revision of the manuscript. SM and AC contributed to data collection, data analysis and the revision of the manuscript.            ix Table of Contents Abstract ........................................................................................................................................ iii Lay Summary ................................................................................................................................ v Preface ........................................................................................................................................... vi Table of Contents ........................................................................................................................ ix List of Tables ............................................................................................................................. xiv List of Figures ............................................................................................................................ xvi List of Abbreviations .............................................................................................................. xvii Acknowledgements ............................................................................................................... xviii Dedication ................................................................................................................................... xx Chapter 1: Introduction .............................................................................................................. 1 1.1 Health and social harms of illicit drug use ............................................................... 1 1.2 Supervised drug consumption facilities .................................................................... 2 1.3 Study justification ......................................................................................................... 7 1.4 Study setting and context ............................................................................................ 9 1.5 Conceptual framework .............................................................................................. 13 1.6 Study objectives .......................................................................................................... 16 1.7 Overview of study design and methods ................................................................. 18 1.7.1 Prospective cohort data ..................................................................................... 18 1.7.2 Ethnographic data .............................................................................................. 20 1.8 Summary ...................................................................................................................... 21 Chapter 2: Public health and public order outcomes associated with supervised drug consumption facilities: A systematic review ........................................................................ 24 2.1 Introduction ................................................................................................................. 24          x 2.2 Methods ........................................................................................................................ 25 2.2.1 Search strategy .................................................................................................... 25 2.2.2 Inclusion and exclusion criteria ........................................................................ 25 2.2.3 Study screening, data extraction and analysis ............................................... 25 2.2.4 Quality assessment ............................................................................................. 26 2.3 Results .......................................................................................................................... 26 2.3.1 Summary of included studies ........................................................................... 26 2.3.2 Objective 1: Reduce the harms associated with illicit drug use ................... 27 2.3.2.1 Overdose-related morbidity and mortality ................................................ 27 2.3.2.2 Drug-related risk behaviours ........................................................................ 28 2.3.2.3 Other health and social outcomes ................................................................ 29 2.3.3 Objective 2: Connect people who use drugs with addiction treatment and other health and social services ........................................................................................ 29 2.3.3.1 Addiction treatment ....................................................................................... 29 2.3.3.2 Other health and social services ................................................................... 30 2.3.4 Objective 3: Reduce the public order and safety problems associated with injection drug use ............................................................................................................... 31 2.3.4.1 Public drug use and publicly-discarded injection equipment ................. 31 2.3.4.2 Crime ................................................................................................................ 32 2.3.4.3 Cost effectiveness ............................................................................................ 32 2.4 Discussion .................................................................................................................... 33 2.4.1 Directions for future research ........................................................................... 38 2.4.2 Limitations ........................................................................................................... 41 2.5 Conclusions ................................................................................................................. 41          xi Chapter 3: Discontinued use of a supervised injection facility among people who inject drugs in Vancouver, Canada: A prospective cohort study ..................................... 58 3.1 Introduction ................................................................................................................. 58 3.2 Methods ........................................................................................................................ 59 3.2.1 Study sample ....................................................................................................... 59 3.2.2 Variable selection ................................................................................................ 59 3.2.3 Statistical analyses .............................................................................................. 60 3.3 Results .......................................................................................................................... 63 3.4 Discussion .................................................................................................................... 64 Chapter 4: The impact of supervised injection facility use on all-cause mortality among a cohort of people who inject drugs in Vancouver, Canada ................................ 72 4.1 Introduction ................................................................................................................. 72 4.2 Methods ........................................................................................................................ 73 4.2.1 Study sample ....................................................................................................... 73 4.2.2 Variable selection ................................................................................................ 74 4.2.3 Statistical analyses .............................................................................................. 75 4.3 Results .......................................................................................................................... 76 4.4 Discussion .................................................................................................................... 78 Chapter 5: Supervised injection facility use and exposure to violence among a cohort of people who inject drugs in Vancouver, Canada: A gender-based analysis .............. 89 5.1 Introduction ................................................................................................................. 89 5.2 Methods ........................................................................................................................ 93 5.2.1 Study sample ....................................................................................................... 93 5.2.2 Variable selection ................................................................................................ 93          xii 5.2.3 Statistical analyses .............................................................................................. 95 5.3 Results .......................................................................................................................... 97 5.4 Discussion .................................................................................................................. 100 Chapter 6: Peer worker involvement in low-threshold supervised drug consumption facilities in the context of an overdose epidemic in Vancouver, Canada ..................... 113 6.1 Introduction ............................................................................................................... 113 6.2 Methods ...................................................................................................................... 118 6.3 Results ........................................................................................................................ 119 6.3.1 OPS implementation and operations drew on existing community capacities ............................................................................................................................ 119 6.3.2 Peer workers fostering environments of comfort and safety at OPS ........ 121 6.3.3 Peer workers enabling harm reduction practices and other positive outcomes ............................................................................................................................ 124 6.3.4 Work-related benefits and challenges for peer workers ............................. 125 6.4 Discussion .................................................................................................................. 128 Chapter 7: Conclusion ............................................................................................................. 137 7.1 Summary of findings ................................................................................................ 137 7.2 Study strengths and unique contributions ........................................................... 141 7.3 Limitations ................................................................................................................. 143 7.4 Recommendations .................................................................................................... 145 7.5 Future research .......................................................................................................... 148 7.6 Conclusions ............................................................................................................... 150 References ................................................................................................................................. 152 Appendices ............................................................................................................................... 183          xiii Appendix A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist ............................................................................................................... 183 Appendix B Sample search strategy to search the Medline database via the OVID platform; May 01, 2017 ......................................................................................................... 186 Appendix C Quality assessment of included studies ...................................................... 187 C.1 Quality assessment of observational cohort and cross-sectional studies (NHBLI) .............................................................................................................................. 187 C.2 Quality assessment of before-after (pre-post) studies (NHBLI) .................... 189 C.3 Quality assessment of cost-effectiveness studies using the Joanna Briggs Institute’s Critical Appraisal Checklist for Economic Evaluations ........................... 191           xiv List of Tables Table 2.1 Population, interventions, comparisons, outcomes and study design (PICOS) criteria for study inclusion ........................................................................................................ 43 Table 3.1 Baseline characteristics of a cohort of 1336 clients of supervised injection facility (SIF) in Vancouver, Canada, stratified by reporting ceasing use of the SIF while actively injecting during follow up (2005-2016) ..................................................................... 70 Table 3.2 Bivariable and multivariable extended Cox regression analyses of factors associated with time to cessation of use of a supervised injection facility while actively injecting among 1336 people who inject drugs in Vancouver, Canada (2005- 2016) ........ 71 Table 4.1 Characteristics of 811 people who inject drugs in Vancouver, Canada, stratified by at least weekly supervised injection facility use at baseline (2006-2017) ..... 84 Table 4.2 Causes of death in a study of 811 people who inject drugs in Vancouver, Canada (2006-2017) ..................................................................................................................... 86 Table 4.3 Unadjusted and adjusted Cox regression analyses of factors associated with time to all-cause mortality among 811 people who inject drugs in Vancouver, Canada (2006- 2017) .................................................................................................................................. 87 Table 5.1 Baseline characteristics of 697 women who inject drugs in Vancouver, Canada, stratified by recent exposure to violence (2005-2016) .......................................................... 106 Table 5.2 Baseline characteristics of 1251 men who inject drugs in Vancouver, Canada, stratified by recent exposure to violence (2005-2016) .......................................................... 108 Table 5.3 Bivariable and multivariable generalized estimating equation analyses of factors associated with experiencing violence, stratified by gender, among 1930 people who inject drugs in Vancouver, Canada (2005-2016) .......................................................... 110          xv Table 5.4 Characteristics of 2039 violent incidents experienced by people who inject drugs in Vancouver, Canada, stratified by gender (2005-2014) ......................................... 112 Table 6.1 Service-level characteristics of four overdose prevention sites (OPS) in Vancouver, Canada (2017) ....................................................................................................... 135 Table 6.2 Characteristics of 72 participants in a rapid ethnographic study of overdose prevention sites in Vancouver, Canada (2017) ..................................................................... 136           xvi List of Figures Figure 1.1 Rhodes’ Risk Environment Framework ................................................................ 23 Figure 2.1 Flowchart of record screening and selection process ......................................... 44 Figure 4.1 Flowchart showing how the analytical sample (n = 811) was determined, Vancouver Injection Drug Users Study (VIDUS) and AIDS Care Cohort to Evaluate Exposure to Survival Services (ACCESS), Vancouver, Canada (2006-2017) ...................... 83           xvii List of Abbreviations ACCESS AIDS Care Cohort to Evaluate access to Survival Services AHR  Adjusted hazard ratio AIC   Akaike information criterion  AIDS  Acquired immunodeficiency syndrome  AHR  Adjusted hazard ratio AOR  Adjusted odds ratio BC  British Columbia BCCSU British Columbia Centre on Substance Use CI  Confidence interval GEE  Generalized estimating equations HCV  Hepatitis C virus HIV  Human immunodeficiency virus HR  Hazard ratio  ICD-10 International Classification of Diseases, 10th edition  IQR  Interquartile range MMT  Methadone maintenance therapy NHBLI National Heart, Blood and Lung Institute OPS  Overdose prevention site OR  Odds ratio PHN  Personal health number PWID  People who inject drugs PWUD People who use drugs RCT  Randomized controlled trial SCF  Supervised drug consumption facility SIF  Supervised injection facility VANDU Vancouver Area Network of Drug Users VIDUS  Vancouver Injection Drug Users Study           xviii Acknowledgements I extend a deeply enthusiastic and sincere thank you to my co-supervisors, Drs. Thomas Kerr and Chris Richardson, and my supervisory committee member, Dr. Ryan McNeil, for their mentorship and unwavering support over the course of my doctoral studies. I want to thank Dr. Thomas Kerr in particular for inspiring me to pursue this research, believing in my abilities, sharing his seemingly unlimited knowledge, and providing me with invaluable opportunities for career development over the years. I cannot thank him enough for his enduring support, encouragement and generosity. I am also very grateful to Dr. Chris Richardson for his thoughtful guidance, timely and encouraging advice, and for lending his epidemiological wisdom to my studies. Dr. Ryan McNeil has also been an exceptional mentor, and I offer my sincere thanks to him for his constant support, constructive insights, and demonstrated commitment to advancing my development as a researcher. I owe special appreciation to the participants of the Vancouver Injection Drug Users Study (VIDUS), the AIDS Care Cohort to Evaluate Access to Survival Services (ACCESS), and the Ethno-epidemiological Study of the Implementation and Effectiveness of an Innovative and Comprehensive Response to the Opioid Epidemic. Thank you for generously sharing your thoughts and experiences. This work would not have been possible without your contributions. I also offer my sincere gratitude to all of the study interviewers, field staff, study nurses, and other research and administrative staff at the British Columbia Centre on Substance Use (BCCSU) for their dedication and efforts. As well, I gratefully acknowledge that this research took place on the unceded traditional territories of the xʷməθkwəy ̓əm (Musqueam), Skwxwú7mesh (Squamish), and sel ̓íl ̓witulh (Tsleil-waututh) Nations.          xix I feel very privileged to have had the opportunity to benefit from the advice and encouragement of many mentors and colleagues at the University of British Columbia and the BCCSU. I would particularly like to thank Dr. Kanna Hayashi, Dr. M-J Milloy, and Dr. Evan Wood for providing me with inspiring mentorship, guidance, and enthusiastic support during my doctoral training. I would also like to acknowledge Dr. Jade Boyd, Ms. Alex Collins, Ms. Huiru Dong, Dr. Mohammad Karamouzian, Dr. David Klassen, and Ms. Samara Mayer, and thank them for their important contributions to this research. I express special thanks to my fellow trainees and colleagues Ms. Elena Argento, Ms. Brittany Barker, Dr. Tessa Cheng, Ms. Stephanie Lake, Dr. Lianping Ti, and Ms. Pauline Voon for their ongoing support and encouragement.  Funding to support my doctoral training and dissertation research was generously provided by the University of British Columbia, Mitacs Canada, and the Social Sciences and Humanities Research Council. I also thank the BCCSU for providing me with salary support during my doctoral program. Lastly, I would like to thank Dustin and my family and friends for their endless love, patience, support, and encouragement during my graduate studies. I offer you all a heartfelt thank you for everything that you have done for me.           xx Dedication For my parents, Maggie and Fergus Kennedy           1 Chapter 1: Introduction 1.1 Health and social harms of illicit drug use In Canada and internationally, illicit drug use remains a major public health challenge.1–12 In particular, accidental drug overdose is a key driver of morbidity and mortality among people who use illicit drugs (PWUD) in many settings worldwide, especially in North America, where overdose deaths fuelled by the proliferation of illicit synthetic opioids in drug markets continue to rise in numerous jurisdictions.9,10,12 These increases in overdose deaths appear to have contributed to recent declines in overall life expectancy at the national level in both the United States and Canada.13,14 In addition, injection drug use continues to be a major driving factor of the global HIV/AIDS and Hepatitis C (HCV) pandemics.7,8 Of the estimated 11.8 million people who inject drugs (PWID) worldwide, approximately 13% are infected with HIV, with injection drug use accounting for close to 30% of all new HIV infections outside of sub-Saharan Africa.7 Approximately half of the global PWID population has chronic HCV infection, with an estimated 28% of HCV-negative PWID acquiring HCV within a year of initiating injection drug use.7,8 Drug-using populations throughout the world, particularly PWID, also commonly experience other serious health-related harms, including exposure to violence, mental health concerns, liver disease, stroke, soft tissue infections (e.g., abscesses, cellulitis), sepsis, and endocarditis.1–6,11 The barriers that PWUD encounter in accessing health services have been found to further compromise their health and well-being.15–18  For example, factors such as stigma and past experiences of discrimination by healthcare professionals may delay PWUD from seeking medical treatment until conditions become more severe.15–18 This, in turn, may contribute to increases in need for          2 costly emergency department care and inpatient hospital admissions.15–18 In addition to direct harms to PWUD specifically, illicit drug use remains a source of community concerns in various settings. For example, public drug use, public disposal of used syringes and drug dealing pose problems in many communities, particularly in inner-city neighbourhoods.19–21 1.2 Supervised drug consumption facilities Although individually-focused educational and behavioural interventions remain the dominant public health response to drug-related harms in many settings, there is growing recognition of the limitations of such approaches.22–25 This, coupled with increasing scientific evidence of the critical role that social, structural and environmental conditions play in the production of drug-related risk and harms,22,26 has provided impetus for the development and implementation of safer environment interventions, which target contextual forces to foster risk reduction and promote health among PWUD.23 Among such safer environment interventions are supervised drug consumption facilities (SCFs), which have been implemented in a growing number of cities worldwide as part of efforts to mitigate the substantial health and social harms associated with illicit drug use.27–29 SCFs are regulated environments that provide secure and hygienic spaces in which PWUD can consume pre-obtained illicit drugs with sterile equipment under the supervision of health professionals or other trained staff.27 SCFs are also referred to as drug consumption rooms and include supervised injection facilities (SIFs), which accommodate PWID, and supervised inhalation rooms, which accommodate people who inhale drugs. Some SCFs also accommodate other modes of drug consumption, including oral and intranasal use.29,30 Within SCFs, clients are          3 typically provided with sterile drug use equipment, education on safer drug consumption methods, emergency intervention in the event of overdose, as well as referrals to co-located and external addiction treatment, health and social services.27  Since the first legally-sanctioned SCF opened in Berne, Switzerland in 1986,28 these facilities have increasingly been implemented and it is estimated that there are now more than 120 SCFs operating internationally, including in Canada, Australia, Mexico and ten countries in western Europe (Switzerland, The Netherlands, Germany, Spain, Norway, Luxembourg, Denmark, France, Belgium, and Portugal).28,29,31–34 In addition, plans are underway to establish SCFs in other settings, including in Ireland.28 Several cities in the United States are also currently considering implementing these facilities.35,36 Although SCFs vary in design and operational procedures, the objectives of SCFs are similar across sites.37,38 Specifically, SCFs aim to engage PWUD at elevated risk of drug-related harms, particularly those who would instead use drugs in unsanitary, insecure, isolated, or otherwise potentially unsafe environments, including public (e.g., alleyways) and semi-public settings (e.g., shooting galleries).37 Epidemiological and ethnographic research has illustrated how the physical and social contexts of such settings may promote drug-related risks and harms among PWUD.22,26,39–41 Of note, public drug use environments are often characterized by unsanitary conditions and lack basic amenities (e.g., sterile equipment, clean water, sterile surfaces to prepare injections) to foster the adoption of hygienic drug use practices.42 Further, individuals consuming drugs in public spaces are often subjected to contextual pressures, including policing and threat of predatory robbery and assault, that may exacerbate risk by          4 encouraging unsafe practices such as rushed injection.42 In providing a protected, monitored environment in which PWUD can use drugs with sterile equipment without fear of unwanted interference or criminal prosecution, SCFs seek to reshape the immediate physical environments and social contexts surrounding drug consumption to reduce the potential for drug-related harms, including fatal overdose and infectious disease transmission.23,37,43 As well, in recognizing that social and structural factors (e.g., stigma, long wait times) often constrain the ability of PWUD to access addiction treatment and other health and social resources, SCFs seek to mediate such barriers in providing non-judgmental environments in which PWID can be connected with such services in a non-coercive manner.37,44 Finally, in drawing PWUD into off-street environments to prepare and consume drugs, these facilities also often aim to reduce public order concerns, including public drug use and the public disposal of used syringes and other drug-related paraphernalia.37 This emphasis on potential improvements in public order has played a critical role in efforts to increase the public and political appeal of SCFs.45 However, it should be noted that this framing has also been the subject of considerable criticism, including for emphasizing the interests of community members rather than promoting client-centeredness, and positioning SCFs as vehicles for “purifying public spaces of ‘disorderly’ drug users,”45 thus contributing to the socio-spatial exclusion of PWUD.45  Over the past fifteen years, the body of evidence specific to SCFs has grown substantially.27 As described in detail in Chapter 2, numerous peer-reviewed studies have documented an array of short- and medium-term benefits of SCFs, as well as a lack of negative impacts. For example, SCFs have been found to decrease unsafe drug          5 use practices and overdose mortality,46–48 improve public order,20,21,49 and facilitate engagement with medical care and addiction treatment.2,50–53 Despite this evidence and the fact that SCFs have become a standard intervention for reducing drug-related harms and improving public order in many cities worldwide, these facilities remain highly controversial in some settings35,54–56 and efforts to establish SCFs are often met with considerable opposition from key stakeholders, including government officials, law enforcement authorities and members of the general public.35,54–56 For example, in recent years, public health and elected officials have developed proposals to implement SCFs in several major cities in the United States, including Seattle, San Francisco, Philadelphia, New York City, Baltimore and Denver.35,36 However, despite ongoing overdose epidemics in these settings, municipal-, state- and federal-level authorities have thus far prevented these efforts and a sanctioned SCF has yet to be established in the United States.35  Opposition to SCFs also persists in some settings where SCFs have already been established, including in Canada, as detailed in Section 1.4. Indeed, despite rigorous evidence demonstrating the successes of Canada’s first sanctioned SCF that opened in 2003, extensive efforts have been required to prevent the closure of this facility, and no further SCFs were established in the country until 2016.54 While additional SCFs have since been implemented in Canada, political opposition continues to constrain the operation of SCFs in some Canadian settings. For example, although multiple SCFs have opened in the Canadian province of Ontario since August 2017 in response to the overdose crisis, the provincial government implemented regulations in October 2018 stipulating that all existing and proposed SCFs would be required to undergo an          6 extensive application process in order to receive provincial approval and funding to operate.57,58 These new regulations also impose limits on the number of SCFs allowed to operate in Ontario and require that SCFs meet a number of criteria that could jeopardize the continued operation of existing sites and the establishment of new facilities in the province.57,58 Such criteria include requirements that SCFs provide a range of mandatory services in addition to supervised consumption spaces, and not operate near each other or certain organizations (e.g., postsecondary institutions, child care centres).57 In December 2018, the federal government announced that provincial support would no longer be required in applications for federal exemptions to establish SCFs in Ontario or elsewhere in Canada.59 Nonetheless, this situation reveals how this form of intervention is still often subjected to considerable political interference even in instances where there is an urgent public health need for SCFs and such services have already been successfully operating for some time. The continued opposition to SCFs appears to be largely rooted in objection based on ideological principles or emotion-based claims and, in some cases, inaccurate understanding of these interventions and the known benefits identified through scientific evaluations of such programming.37,60–62 However, SCF opponents, including high-ranking government officials, have also cited gaps in existing evidence pertaining to SCFs as a reason not to support the implementation of this intervention.61,63–65 For example, in emphasizing the need for evidence-based responses to the present overdose crisis, the Surgeon General of the United States recently stated that “high-quality scientific evidence… does not exist for supervised injection facilities.”63 Such comments          7 point to need for continued rigorous evaluation of SCFs to inform policy development concerning these facilities and improve comprehensive health services for PWUD.  1.3 Study justification Despite the rapidly growing body of scientific literature specific to SCFs, outstanding questions regarding the use and impacts of these facilities remain. Among the most important knowledge gaps related to evaluations of SCF effectiveness is that many existing analyses have been derived from short-term studies of PWUD with insufficient follow-up durations to assess long-term service engagement and impacts. For example, although previous studies have found that PWUD at heightened risk of drug-related harms are more likely to initiate SCF use,66–70 these have primarily been cross-sectional analyses with short study periods, and no known studies to date have characterized long-term patterns of SCF utilization, including discontinued use of these services. This is a notable gap in the current evidence regarding SCFs given that this has been identified as a priority research area to inform SCF service optimization,71 and that policymakers in various jurisdictions have repeatedly questioned whether SCFs may lead PWUD to prolong their drug use and perpetually use these services.27,72–74 For example, the Deputy Attorney General of the United States recently expressed opposition to SCFs, stating “injection sites normalize drug use and facilitate addiction.”72 Another area that warrants greater research attention is examination of potential long-term health outcomes associated with SCF use. For instance, although previous studies have documented reductions in overdose mortality rates in the immediate months after SCF establishment,48,75 the primary units of analysis for this work have          8 been aggregate measures. Studies have yet to determine individual-level estimates of the association between SCF use and mortality given that this requires prolonged follow up of a large number of PWUD. Evidence of the impact of SCF use on exposure to violence is also lacking, despite the fact that street-based drug scene violence remains a key driver of morbidity and mortality among PWUD and that few interventions exist that specifically target this health concern in this population.76 Given that SCFs continue to be subjected to higher standards of proof of effectiveness compared to other health interventions even in contexts with worsening overdose epidemics, generating high-quality data on potential long-term health impacts would not only make critical contributions to the body of literature specific to SCFs, but could also play an important role in informing evidence-based policy recommendations concerning these facilities.  It is also noteworthy that few studies have evaluated innovative and emerging models of SCF programming, thus missing opportunities to explore unique implementation and operation contexts across SCFs. For example, as described in Section 1.4, novel low-threshold peer-run SCFs, termed overdose prevention sites (OPS), have increasingly been implemented in Canada as part of the response to the ongoing overdose crisis.54,77 However, little is known about how involvement of peers (i.e., people who currently use or formerly used drugs) as staff may shape access to and impacts of this form of intervention. Gaining a stronger understanding of the role of this SCF operational feature in shaping service utilization and related health outcomes may provide important evidence to optimize the effectiveness of existing SCF programming and inform decision-making concerning the development and establishment of future SCFs.           9 1.4  Study setting and context  Vancouver, British Columbia (BC), Canada has long contended with an illicit drug use epidemic. The Downtown Eastside neighbourhood of Vancouver, a 40 to 50 block area that is home to an estimated 5,000 PWUD and is characterized by high levels of poverty and homelessness, remains the city’s epicentre of this epidemic and related health and social challenges.54,78–80 From the mid- to late-1990s, drug-related harms increased rapidly in this setting. Specifically, the city experienced an outbreak of HIV infection that peaked at an annual incidence of 19% among local PWID.81 In addition, hundreds of overdose deaths occurred in Vancouver during the same period of time, with the Downtown Eastside community being disproportionately affected.54,78 This prompted Vancouver’s health authority to declare a public health emergency in 1997, and a range of harm reduction and HIV care interventions were subsequently implemented and expanded in the city.54 In 2003, North America’s first legally-sanctioned SCF, Insite, was established in Vancouver’s Downtown Eastside and has remained in operation since then.54 The facility is typically open 18 hours per day and includes 13 spaces in which people can inject pre-obtained drugs under nurse supervision.54 Nurses at Insite also respond to on-site overdose events and address other health needs (e.g., provide treatment for injection-related infections), and addictions counsellors are available to refer clients to addiction treatment and other health and community services.54 Although Insite exclusively accommodates injection drug use at present, oral and intranasal modes of drug consumption may also be accommodated at the facility in the near future.29          10 As there was little in the way of high-quality empirical evidence documenting the effectiveness of SCFs when Insite was established, the facility acquired a federal exemption to operate under Section 56 of the Controlled Drugs and Substances Act under the condition that its effects be subject to rigorous scientific evaluation.54 As such, Insite subsequently became the focus of numerous research studies investigating the health and social impacts of the facility. This research, which has provided much of the scientific evidence on the impacts of SCFs to date, has demonstrated a wide array of health and community benefits of the facility,27 as detailed in Chapter 2.  Despite the large body of scientific evidence demonstrating Insite’s various positive impacts, the former federal Conservative government remained opposed to the facility during its near decade in office from 2006 to 2015, and routinely publicly expressed its intension to close the SCF.54 The federal government’s efforts to undermine Insite culminated in a legal case filed by the facility’s operator and two local PWID against the federal government that was ultimately heard by the Supreme Court of Canada.54 Many of Canada’s leading national health groups, including the Canadian Medical Association and the Canadian Nurses Association, intervened in support of Insite.54 In a landmark decision in September 2011, the Supreme Court unanimously ruled to uphold the facility’s exemption to operate, stating “Insite has been proven to save lives with no discernable negative impact on the public safety and health objectives of Canada.”82 The Supreme Court ruling appeared to set a precedent that would enable the expansion of federally-sanctioned SCFs in Vancouver and elsewhere in Canada. However, the federal Minister of Health continued to raise questions about the long-         11 term impacts of SCFs, including their impact on perpetuating high risk drug use,83 and the federal government eventually responded to the Supreme Court decision by introducing Bill C-2, the Respect for Communities Act, which became law in June 2015.54 In what many observers deemed as disregard of the Supreme Court’s ruling, Bill C-2 stipulated that legal exemptions may only be granted to SCFs in “exceptional circumstances”84 and outlined twenty-six requirements that needed to be fulfilled before the federal Minister of Health would consider granting such an exemption.84 Such requirements included a report of broad community consultation, as well as supporting letters from the municipal government, local police force, provincial public health authority, and provincial ministers of health and public safety.84 However, even if all requirements were met, there was no guarantee of approval by the federal government. This legislation ultimately served to impede the broader scale up of SCFs in Canada under the federal Conservative government, despite the fact that research indicated significant unmet need for SCFs in several Canadian cities, including in Vancouver, and that significant planning and feasibility work had already been undertaken in effort to establish these services in many municipalities across the country.54 In October 2015, the expansion of SCFs in Canada became more feasible with the election of a federal Liberal government, whose election platform had stated support for SCFs.85 Indeed, within three months of being elected, the Liberal government approved a second SCF, the Dr. Peter Centre, which had been operating in Vancouver without a federal exemption since 2002.54 Meanwhile, emerging overdose epidemics driven by the increasing presence of illicitly-manufactured fentanyl (a powerful synthetic opioid) and related analogues in illicit drug supplies were affecting a number of settings in          12 Canada.86 For instance, there were 993 overdose deaths in BC in 2016, an 89% increase from 2015,86 which led the provincial government to declare a public health emergency.54 These circumstances coupled with intensive lobbying by PWUD, activists, and health, legal and human rights experts ultimately prompted the federal government to replace Bill C-2 with new legislation, Bill C-37.54 This bill was implemented by the government with the intention of simplifying the process of applying for a federal exemption to operate a SCF by reducing the number of application criteria from twenty-six to five.87 Since Bill C-37 was implemented in May 2017, there has been a substantial expansion of SCFs in the country. Specifically, a total of twenty-eight SCFs have received federal approval and are presently operating in Canada, three of which are located in Vancouver. These three facilities include Insite, the Dr. Peter Centre and the Powell Street Getaway SCFs.29 In addition to federally-sanctioned SCFs, there has also been an expansion of provincially-sanctioned low-threshold SCFs, known as OPS, since December 2016 as part of the response to the overdose crisis in BC.54,77 Specifically, a total of twenty-four OPS are presently operating in the province, six of which are located in the city of Vancouver (five in the Downtown Eastside).88,89 Additional OPS are also presently operating in Alberta and Ontario.88 Like federally-sanctioned SCFs, OPS provide regulated environments in which PWUD can use drugs with sterile equipment and receive emergency intervention in the event of overdose. However, OPS differ from SCFs in that these usually have received provincial exemptions to operate as non-permanent services in response to the overdose crisis but have not received formal exemptions from federal drug laws.54,77 Further, OPS are primarily staffed by peers          13 rather than nurses, are typically integrated into existing health, housing and community services that serve drug-using populations rather than established as purpose-built sites, and do not usually offer the same extent of ancillary services, such as nursing care and referrals to addiction treatment, as federally-sanctioned SCFs.54,77 As well, OPS are considered to be lower threshold than federally-sanctioned SCFs given that these were implemented in the context of a public health emergency and sought to minimize barriers to service access, including by accommodating drug use practices (e.g., assisted injection, drug sharing, and in some cases, drug inhalation) that are not permitted at most federally-sanctioned SCFs under the parameters of exemptions to federal drug laws.54,77 1.5 Conceptual framework This research is informed by Rhodes’ Risk Environment Framework.22,26,39 Past research investigating health-related risk and harm among PWUD has primarily been informed by individually-focused theoretical frameworks that emphasize rational decision-making and conceptualize health outcomes as the result of individual action.22 However, there has been growing acknowledgement of the limitations of such approaches in informing public health interventions due to their inadequate conceptualization of the role of broader contextual forces in shaping individual agency to practice risk reduction and promote personal health.26 Accordingly, in recent years, there has been a shift away from individually-focused models towards broader social-ecological conceptual models, including Rhodes’ Risk Environment Framework, that focus on how aspects of the physical, social, and structural environment may influence the distribution of health and harm among drug-using populations.22,26,39           14 First applied to conceptualize the production of HIV acquisition risk,22 Rhodes’ Risk Environment Framework has recently been employed more broadly in the study of social-structural contexts affecting various drug-related risks and harms, including overdose and public injection, as well as health service utilization and other health-related outcomes among PWUD.23,39,42,43,90 Rhodes’ Risk Environment Framework conceptualizes the health of PWUD at the individual level as shaped by the interplay of physical, social, economic and policy factors that intersect at the macro-, meso-, and micro-environmental levels of influence.22,26,39 The micro-environment refers to dimensions of the immediate environments of PWUD such as social interactions and physical contexts surrounding drug use.39 The meso-environment includes factors such as the distribution of health services and local law enforcement practices, while the macro-environment includes structural inequities, laws, and public health, drug and economic policies.39  Within the Risk Environment Framework, SCFs are conceptualized as micro-environmental interventions that primarily aim to mitigate harm among PWUD by modifying aspects of the immediate social and physical contexts of illicit drug use that shape drug use practices and risk (see Figure 1.1).23,25,43 Further, SCFs may also mediate meso- and macro-environmental forces that contribute to harm among PWUD.25 Indeed, past research has illustrated how SCFs provide micro-environmental social and spatial conditions that reduce drug-related risks (e.g., rushed public injecting) and increase the capacity of PWUD to enact harm reduction, while also reducing exposure to meso-level risks (e.g., street-level policing) and macro-level forces (e.g., stigma; drug prohibition) that may constrain engagement with health services and promote harm among          15 PWUD.25,43,44,91 The present research seeks to extend on previous studies in drawing on the Risk Environment Framework to conceptualize how SIF use may modify such aspects of the risk environment of PWID to minimize vulnerability to all-cause mortality and exposure to violence among this population.  Past research has also utilized the Risk Environment Framework to identify social, structural and physical factors that may facilitate or constrain access to SCFs and other health and harm reduction services among PWUD.71,90,92,93 For example, studies undertaken in Canada and internationally have found that PWUD contending with homelessness and other structural vulnerabilities are more likely to engage with SCF services.66–70 However, certain regulatory and operating policies governing federally-sanctioned SCFs in Canada (e.g., rules prohibiting assisted injection within SCFs) may impede engagement with this form of health service among PWUD.44,92 As well, other programmatic issues (e.g., wait times) and geographic factors (e.g., longer travel distance) may constrain access to SCFs.71,94  This dissertation seeks to build on this research in employing Rhodes’ Risk Environment Framework to examine the longitudinal relationship between various socio-demographic, behavioural, social, structural and environmental factors and discontinued SIF use among active PWID. A range of contextual exposures will be examined, including those that have been found to shape uptake of SCFs and other harm reduction services (e.g., residency near services, homelessness).69,94 As well, the Risk Environment Framework will be applied to study peer involvement in low-threshold SCFs (i.e., OPS), including how this may modify social dimensions of service delivery to influence accessibility and impacts among clients. Identifying the role of          16 such factors in shaping SCF engagement and related outcomes may provide critical insights into how SCF operations can be tailored to maximize benefits among distinct groups of SCF clients.  1.6 Study objectives The primary aim of this dissertation is to generate evidence concerning engagement in and health outcomes associated with SCFs to inform policy and programmatic responses to drug-related harms. Specifically, this research seeks to address pressing gaps in the existing scientific literature by drawing on prospective cohort and ethnographic data to investigate SCF service utilization patterns and health impacts among PWUD in Vancouver, Canada. The analyses herein seek to longitudinally characterize discontinuation of SIF use, examine potential long-term impacts of SIF use on mortality and exposure to violence, as well as explore how the recent implementation of novel peer-run SCF models in this setting may shape service dynamics and related outcomes among PWUD. It is hoped this research will produce findings of national and international relevance to inform the ongoing development and optimization of SCF programming in Vancouver and elsewhere. The five specific objectives of this research are as follows: 1. To systematically review existing quantitative literature investigating the health and community impacts of SCFs. Although several previous reviews have summarized the evidence specific to SCFs, these have been characterized by some methodological weaknesses, including searching only a small number of online databases for potentially-relevant studies, providing limited detail with regard to study eligibility criteria, and failing to assess the methodological          17 quality of included studies.27,37,95 The systematic review undertaken for this dissertation seeks to address these limitations by conducting a comprehensive electronic search strategy, providing explicit study eligibility criteria, as well as synthesizing and evaluating the quality of existing quantitative research examining potential health and community outcomes associated with SCFs. Chapter 2 reports the findings of this systematic review, and identifies underexplored opportunities to inform future research specific to SCFs. Study objectives and hypotheses investigated for Chapters 3 to 6 were developed and refined based on the findings of this review. 2. To characterize SIF use cessation among SIF clients, including factors associated with discontinued use of a SIF during periods of active injection among this population. Using data from two prospective cohorts of PWID, Chapter 3 employs survival analyses for recurrent events to longitudinally examine individual and contextual factors associated with SIF use cessation during periods of active injection. This chapter also descriptively examines the prevalence of injection cessation events that co-occur with SIF use cessation, as well as self-reported reasons for discontinuing use of this service among former SIF clients.  3. To longitudinally assess the potential impact of SIF use on all-cause mortality among PWID. The study presented in Chapter 4 uses survival analyses to examine the individual-level association between frequent SIF use and all-cause mortality among a community-recruited prospective cohort of PWID. This          18 analysis tests the hypothesis that frequent SIF use is independently associated with decreased risk of all-cause mortality. 4. To examine the relationship between SIF use and exposure to violence among PWID. Chapter 5 uses generalized estimating equations (GEE) with logit link analyses to prospectively examine the gender-specific relationship between exclusively injecting at a SIF and exposure to physical or sexual violence among a community-recruited prospective cohort of PWID. This study tests the hypothesis that exclusive SIF use is independently associated with decreased odds of experiencing violence among both men and women PWID. 5. To characterize peer involvement as staff in low-threshold SCFs, including how this shapes service dynamics and impacts in the context of an overdose epidemic. Drawing on rapid ethnographic fieldwork undertaken in Vancouver, Chapter 6 seeks to investigate the role of peers in the implementation and delivery of low-threshold SCF services, and to identify specific impacts of peer-based SCF models in shaping service engagement and health-related outcomes among PWUD.  1.7 Overview of study design and methods 1.7.1 Prospective cohort data Data for Chapters 3 to 5 of this dissertation are derived from two long-running, community-recruited prospective cohort studies of PWUD operating in Vancouver, Canada: The Vancouver Injection Drug Users Study (VIDUS) and the AIDS Care Cohort to Evaluate access to Survival Services (ACCESS).96–98 As described previously,96–98 study participants have been recruited through self-referral, street outreach and word of          19 mouth since May 1996. VIDUS is a cohort of HIV-negative adult PWID who have injected illicit drugs at least once in the month prior to study enrolment. ACCESS is a cohort of HIV-positive adult drug users who have used illicit drugs (other than or in addition to cannabis) at least once in the month prior to enrolment. VIDUS participants who seroconvert to HIV following recruitment are transferred into the ACCESS study. All eligible participants provide written informed consent. The data collection and follow-up procedures for VIDUS and ACCESS are harmonized to allow for the merging of datasets and combined analyses. Specifically, at baseline and semi-annually thereafter, participants in both cohorts complete an interviewer-administered questionnaire and provide blood samples for serological analyses. The questionnaire elicits information about socio-demographic characteristics, drug use and other behavioural patterns, social-structural exposures, violent encounters, engagement with healthcare services, and experiences with the criminal justice system. Serological analyses include HIV and HCV testing, as well as HIV disease monitoring, as appropriate. At each study visit, participants are offered a nominal honorarium ($30 CAD).  For the analyses presented in Chapter 4, the cohort data were augmented by confidential linkages to the BC Vital Statistics Agency, the centralized mortality registry for the province. This linkage to the Vital Statistics database provided information on dates and recorded causes of death in accordance with the International Classification of Diseases, 10th edition (ICD-10).99 These linkages were conducted by using personal health numbers (PHNs), a unique and persistent identifier issued by the provincial government to all residents in BC for billing and health service tracking. Specifically,          20 the PHNs and dates of birth of cohort participants were provided to a data analyst at the BC Vital Statistics Agency, who extracted death records for deceased PHN- and date of birth-matched participants who died in BC between 2006 and 2017. The full names of deceased participants were removed from the received data file. A data analyst at the British Columbia Centre on Substance Use (BCCSU) then merged the cohort data with the death record data. The University of British Columbia/Providence Health Care Research Ethics Board has approved the VIDUS and ACCESS studies, including linkages to the BC Vital Statistics database. 1.7.2 Ethnographic data Data for Chapter 6 were derived from a rapid ethnographic study examining the implementation, operations, and impacts of OPS in the Downtown Eastside neighbourhood of Vancouver. Rapid ethnography harnesses researchers’ familiarity with the specific context under investigation to collect data through intensive ethnographic fieldwork, including observation and interviews, conducted within a short time frame.100 This approach has previously been employed in the study of complex public health emergencies.101  Between December 2016 and April 2017, two members of the research team (JB, RM) conducted approximately 185 hours of observational fieldwork at four provincially-sanctioned OPS operating in the Downtown Eastside. Fieldnotes were recorded in a research log following observation sessions and detailed the operational contexts of OPS, including peer worker interactions. In addition, in-depth qualitative interviews were conducted with 72 PWUD who were recruited from four OPS by the ethnographers or one of two peer researchers (i.e., team members with research training          21 who currently use or formerly used drugs). Individuals who were accessing services at a local OPS were eligible to participate in an interview. We sought to recruit a heterogeneous sample of participants based on socio-demographic characteristics (including gender, age and ancestry). We also aimed to recruit a sufficient number of participants from each of the four OPS in effort to understand dynamics operating within each individual site.  Eligible participants were interviewed either on site at OPS or at the nearby research office by the candidate (18 interviews in total) or one of five other research team members who had prior training and experience in qualitative interviewing (54 interviews in total). An interview topic guide that included questions on a range of topics related to the overdose epidemic and OPS services, including peer involvement at OPS, was used to facilitate discussion during interviews. Interviews ranged from approximately 45 to 60 minutes in duration, were audio recorded and transcribed verbatim. An online pseudonym generator was used to assign pseudonyms to participants. All participants provided informed consent prior to their interview and received a $30 CAD honourarium upon completion of an interview. The study was approved by the University of British Columbia/Providence Health Care Research Ethics Board. 1.8 Summary This dissertation consists of seven chapters. Chapter 1 provides an overview of health and social harms associated with illicit drug use and the role of SCFs in response to such harms. It also provides a rationale for this dissertation research, and presents information concerning study objectives, study context and setting, study design and          22 methods, and the conceptual framework guiding this work. Chapter 2 presents a systematic review of existing scientific and grey quantitative literature on the health and community impacts of SCFs. Chapter 3 longitudinally characterizes SIF use cessation, including individual and contextual factors associated with this outcome during periods of active injection. Chapters 4 and 5 provide epidemiological analyses of the impacts of SIF use on health-related harms. Specifically, Chapter 4 tests the hypothesis that frequent SIF use is independently associated with decreased risk of all-cause mortality. Chapter 5 tests the hypothesis that exclusive SIF use is independently associated with decreased odds of exposure to violence among men and women PWID, respectively. Chapter 6 presents a qualitative analysis that aims to characterize peer involvement in novel low-threshold SCFs, including how this influences service engagement and related health and social outcomes. Finally, Chapter 7 provides a summary of the primary findings of this research, and describes study limitations, policy implications, potential directions for future research, and key conclusions.          23 Figure 1.1 Rhodes’ Risk Environment Framework           24 Chapter 2: Public health and public order outcomes associated with supervised drug consumption facilities: A systematic review 2.1 Introduction In response to calls for novel interventions to address health and social harms stemming from illicit drug use, SCFs have been established in a number of cities in Canada and internationally.27,28 As described in Section 1.2, the primary objectives of SCFs are to attract higher-risk PWUD and to offer the following public health and public order benefits: (1) reduce the harms associated with illicit drug use, including fatal overdose and infectious disease transmission; (2) connect PWUD with addiction treatment and other health and social services; and (3) reduce public order and safety problems associated with illicit drug use (e.g., public drug use, publicly-discarded syringes).37,38  In recent years, an increasing number of scientific studies have evaluated the effectiveness of SCFs in attaining these goals. However, as noted in Section 1.6, previous reviews of this evidence have suffered from some notable methodological shortcomings, including employment of search strategies that were narrow in scope, application of broad study eligibility criteria that resulted in the inclusion of low-quality evidence, and/or lack of assessment of the quality of included evidence.27,37,95 Guided by the primary health and public order objectives of SCFs noted above, the purpose of the present study was to systematically review existing quantitative research on the health and community outcomes associated with SCFs. In addition, this study sought to identify underexplored opportunities to inform future research specific to SCFs.          25 2.2 Methods 2.2.1 Search strategy Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for systematic reviews (see Appendix A),102 we searched for SCF studies published in the following databases from inception to May 01, 2017: MEDLINE, EMBASE, Web of Science, PsychINFO, Google Scholar, and CINAHL. Search terms were combined using appropriate Boolean operators and included the subject heading terms or key words related to SCFs (see Appendix B for a detailed search strategy). In addition to electronic databases, we searched the reference lists of retrieved studies, relevant conference proceedings, and key journals in the area of addiction. We also conducted a comprehensive grey literature search (i.e., dissertations, reports). The search was not restricted to a specific language.  2.2.2 Inclusion and exclusion criteria The population, interventions, comparisons, outcomes, and study designs considered in the review are described in Table 2.1.  2.2.3 Study screening, data extraction and analysis Title and abstract screening were conducted to identify studies that potentially met our inclusion criteria. Full texts of all potentially eligible studies were retrieved by the candidate and independently assessed for eligibility by the candidate and MK. Disagreements between the authors were resolved through discussion. Extracted data on study-specific information were summarized narratively and in a structured table.           26 2.2.4 Quality assessment Quality assessment of cohort, cross-sectional and pre-post studies was conducted using the 14-item National Heart, Blood and Lung Institute (NHBLI) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies103 or the 12-item NHBLI Quality Assessment Tool for Before-After (Pre-Post) Studies,104 as appropriate. Quality assessment for cost-effectiveness studies was completed using the Joanna Briggs Institute’s Critical Appraisal Checklist for Economic Evaluations.105 2.3 Results As shown in Figure 2.1, database searching yielded 1476 records, and hand searching yielded an additional 85 records to account for a total of 1469 potentially eligible studies after duplicate removal. Of these, 1128 records were excluded through title and abstract screenings. Assessment of the full text of the remaining 341 records resulted in the exclusion of an additional 294 studies. In total, 47 studies published between 2003 and 2017 met the eligibility criteria and were included in the review. 2.3.1 Summary of included studies Of the 47 included studies, the majority (n = 28) were conducted in Vancouver, Canada, ten were conducted in Sydney, Australia, and the remaining studies were conducted in the following European countries: Germany (n = 4), Denmark (n = 2), Spain (n = 2) and the Netherlands (n = 1). Seventeen studies employed prospective cohort designs, while the remaining studies employed times series or pre-post ecological (n = 10), cross-sectional (n = 9), mathematical simulation (n = 8) or serial cross-sectional (n = 3) designs. Study quality scores are presented in Appendix C1-3. Overall, most studies had good methodological quality. Additional study-specific information          27 (including study location, design, participant characteristics, exposure(s), outcome(s), and main findings) is presented in Table 2.2.  2.3.2 Objective 1: Reduce the harms associated with illicit drug use 2.3.2.1 Overdose-related morbidity and mortality Of eight studies examining overdose-related outcomes,48,67,75,106–110 six suggested a protective effect of SCFs.48,75,107–110 For example, the establishment of Insite was associated with a 35% reduction in overdose deaths in the immediate vicinity of the SIF after the facility opened, compared to a 9% reduction in the rest of the city.48 An earlier simulation study found that Insite averts an estimated 1.9 to 11.7 overdose deaths per year.107 Similar findings have been observed in ecological and simulation studies conducted in Germany.75,109 Relatedly, the establishment of the SIF in Sydney, Australia was associated with declines in opioid poisoning emergency department presentations110 and ambulance attendances at opioid-related overdoses near the SIF.108 However, there were no statistically significant changes in the number of opioid-related deaths in the neighbourhood of the SIF compared to the rest of the state after the SIF opened, which the authors noted was likely due to insufficient statistical power resulting from the heroin shortage in Australia that contributed to an overall reduction in overdose mortality rates during the study period.110 Another Sydney study found that frequent SIF clients were more likely to experience an overdose within the SIF, likely due to their greater time spent at the facility.67 Finally, a study conducted in Vancouver examined the association between frequent SIF use and recent non-fatal overdose among PWID and produced null results.106           28 2.3.2.2 Drug-related risk behaviours Nine studies evaluated the relationship between SCFs and levels of drug use or drug-related behaviours that may increase risk of infectious disease transmission and other harms.46,47,111–117 Of these, four studies examined the relationship between SCF use and syringe sharing,47,111,112,114 three of which provided evidence of an inverse association.47,111,112 For example, a cross-sectional study of PWID in Vancouver found that regular SIF users were 70% less likely to report borrowing or lending used syringes, despite the fact that SIF users and non-users reported similar levels of syringe sharing prior to the establishment of the SIF in retrospective analyses.47 Two studies (conducted in Demark and Vancouver) demonstrated an association between SCF use and decreased likelihood of other types of unsafe injection behaviours, including syringe reuse, outdoor injection, and rushed injection, as well as an increased likelihood of safe behaviours such as using clean water for injecting, cooking or filtering drugs, and safely disposing syringes.46,113 Only one small German study with a short follow up period found no evidence of an association between SCF use and injection-related risks (e.g., public drug use; equipment sharing).114 This study also found that SIF use was not significantly associated with development of cutaneous injection-related infections, as was found in a prospective study conducted in Vancouver.114,117 With regards to drug use patterns, a study undertaken in Vancouver found no substantial changes in rates of relapse into injection drug use, ceasing injection, ceasing binge drug use, or participation in methadone maintenance therapy (MMT) after Insite opened among a prospective cohort of PWUD.115 As well, another prospective Vancouver study found that the proportion of Insite users who had recently initiated injection drug use was          29 substantially lower than the estimated background community-level rate of injection initiation.116 2.3.2.3 Other health and social outcomes Two prospective cohort studies from Vancouver examined health or social outcomes among PWUD other than overdose-related outcomes or drug-related behaviours.118,119 One of these found that SIF use was not significantly associated with employment in multivariable analyses.118 The other study found that both use of SIF services and time since recruitment from the SIF were independently and positively associated with consistent condom use among PWID with regular but not casual partners.119 2.3.3 Objective 2: Connect people who use drugs with addiction treatment and other health and social services 2.3.3.1 Addiction treatment Four studies provided evidence a positive association between SCF use and uptake of addiction treatment.50,51,53,120 For example, a prospective study of PWID in Vancouver found that at least weekly SIF use and contact with a SIF addictions counselor, respectively, were associated with more rapid entry into detoxification programmes.51 A follow-up study demonstrated that rates of entry into detoxification programmes among SIF users increased by more than 30% in the year after compared to the year before the SIF was established.53 Further, this study found that such enrolment in a detoxification programme was associated with earlier entry into MMT and other forms of addiction treatment, as well as subsequent declines in injections at the SIF.53 An additional prospective study in Vancouver found that at least weekly SIF use was          30 positively associated with enrolment in addiction treatment, which in turn was associated with an increased likelihood of injection cessation.50 Similarly, a prospective study of PWID in Sydney found that frequent SIF use was positively associated with referral to addiction treatment, although analyses with addiction treatment uptake as the outcome produced null results.120 In addition, a sole study examining barriers to treatment found that frequent SIF use was not significantly associated with inability to access addiction treatment among SIF users in Vancouver.121 2.3.3.2 Other health and social services Six studies examined the association between SCF use and utilization of health or social services other than addiction treatment.2,16,70,122–124 For instance, a recent multi-site cross-sectional study of SCF users in Denmark found that being advised to seek treatment for a medical condition by SCF staff was associated with an increased likelihood of receiving treatment.122 Additionally, two separate prospective cohort studies of SIF users in Vancouver found that those referred to hospital by Insite nurses were more likely to access the emergency department and receive hospital care, respectively, for cutaneous injection-related infections.2,16 Further, the latter study also found that such referrals were associated with shorter durations of hospitalization.16 Three studies (conducted in Canada, Germany, and Denmark) demonstrated links between SCF use and utilization of education on safer drug use practices at SCFs,70,122,124 while the German study also found an association between frequent SCF use and greater likelihood of accessing syringe exchange services, medical services and counseling at the SCF.70 Another study, conducted in three cities in the Netherlands,          31 found that SCF users had a higher level of awareness but a similar prevalence of uptake of a Hepatitis B vaccination programme compared to non-users.123  Two additional studies examined health-related outcomes associated with programmes offered within SCFs.125,126 A recent Vancouver study of a pilot drug checking programme offered within Insite found that SIF clients who checked their drugs and received a positive result for fentanyl were more likely to reduce their doses but not to dispose of their drugs compared to those receiving negative results.125 Another study found that the implementation of a smoking cessation organizational change intervention in the Sydney SIF was associated with an increased likelihood of receiving smoking cessation care among SIF clients.126 2.3.4 Objective 3: Reduce the public order and safety problems associated with injection drug use 2.3.4.1 Public drug use and publicly-discarded injection equipment Five studies have demonstrated the role of SCFs in addressing public disorder associated with illicit drug use.20,110,127–129 An ecological study employing a prospective data collection protocol found that the establishment of Insite in Vancouver was associated with reductions in the number of people injecting drugs in public, publicly-discarded syringes and injection-related litter, independent of changes in police presence and weather patterns.20 Similarly, there were observed declines in publicly-discarded syringes and public injection in the neighbourhood of the SIF in Sydney after the facility opened.110,128 There were also increases in the proportion of residents who agreed with positive statements regarding SIFs (including that these reduce public injection and public disposal of used syringes), although opinions were mixed among          32 business owners.129 Another study found that the opening of SCFs in Barcelona, Spain, was associated with a significant reduction in the number of publicly-discarded syringes collected by local services.127 2.3.4.2 Crime Six studies examined the association between SCF operations and drug-related crime.21,49,108,130–132 Of these, four were conducted in Sydney and found no changes in police-recorded thefts or robbery incidents, drug possession, drug dealing, or illicit drug offenses in the neighbourhood of the SIF after the facility was established.49,108,130,131 Similar results have been observed in Vancouver. For example, a before and after study of local crime statistics found no increases in incidents of drug trafficking or assaults/robbery in the neighbourhood of Insite after the facility opened.21 In addition, a prospective cohort study of PWID in Vancouver demonstrated that frequent SIF use was not associated with recent incarceration in multivariable analyses.132 2.3.4.3 Cost effectiveness A total of six studies have evaluated the cost-effectiveness of SCFs, all of which focused on SCFs in Vancouver.133–138 Five studies examined the economic impacts of Insite and found it to be cost-effective.133–137 For example, a simulation study estimated that the SIF provides an excess of $6 million CAD per year (due to averted overdose deaths and incident HIV cases) after considering the facility’s annual operating costs.137 Others have provided more conservative estimates, including a study estimating that the prevention of incident HIV cases and overdose deaths by the SIF provides an excess of $ 200,000 – 400,000 CAD per year.135 Additionally, a recent study of the cost-effectiveness of an unsanctioned peer-run supervised inhalation room found that the          33 facility saved an annual average of $1.8 million CAD due to the prevention of incident cases of HCV infection.138 2.4 Discussion In the present systematic review, we identified consistent, methodologically sound evidence to demonstrate the effectiveness of SCFs in achieving their primary health and public order objectives. Further, the available evidence does not support concerns regarding the potential negative consequences of establishing SCFs, including that these promote drug use or attract crime.   The prevention of drug-related overdose fatalities represents a significant public health challenge in many settings, particularly in North America, where opioid-related overdose deaths have reached epidemic levels and become a leading cause of accidental death in many areas in recent years.10,12,139 Given that early, rapid and well-equipped overdose intervention is available within SCFs,37 and that these facilities have been shown to attract PWUD who possess risk factors for overdose (e.g., homelessness, high-intensity drug use),37,66–70 the broader expansion of SCFs in settings contending with overdose epidemics may afford opportunities to mitigate overdose-related morbidity and mortality. Indeed, compelling ecological and simulation studies included in this review have demonstrated the contributions of SCFs to reductions in overdose-related deaths, emergency department presentations and ambulance attendances.48,75,107–110 It is also noteworthy that despite the millions of injections that have occurred within SCFs internationally over the past three decades, not a single overdose death has been observed within a SCF.27,37 In addition, although preventing non-fatal overdose is not a key objective of SCFs, frequent SCF use has not been found to increase non-fatal          34 overdose risk, which challenges the contention that these facilities promote riskier drug use practices (e.g., taking higher doses) associated with overdose.106 Although one report included in this review observed non-significant declines in opioid-related deaths in Sydney after the SCF was established, the authors note that this study was likely underpowered.110 It should be noted that the increasing adulteration of illicit drug supplies with fentanyl and other powerful opioids in North America in recent years10,12 could potentially limit the generalizability of historical studies concerning the relationship between SCFs and overdose-related harms to current circumstances in this setting. As such, future studies should continue to examine the role of SCFs in mitigating overdose mortality and overdose-related harms in this context of ongoing overdose epidemics.  As described elsewhere,37,109 methodological challenges have impeded efforts to examine the impact of SCFs on the incidence of infectious diseases such as HIV and HCV. However, the studies assessed herein indicate positive impacts of SCFs on reducing unsafe injection practices associated with infectious disease transmission among higher-risk PWUD. For example, several studies have demonstrated associations between SCF use and reductions in syringe sharing,47,111,112 with a previous meta-analysis of three studies undertaken in Canada and Spain providing a pooled estimate of a 70% decreased likelihood of syringe sharing among SCF users.140 Studies also suggest that SCFs contribute to declines in other unsafe injection practices such as reusing syringes, injecting outdoors, or rushing injections,46,113 as has been found in descriptive studies of SCFs that were ineligible for this review.75,94,141–151 In addition to the provision of sterile injection equipment on site, there are several other mechanisms          35 through which SCFs may reduce such behaviours. For example, SCFs often become a key source of sterile syringes for external use,43 which is notable given the well-documented impact of syringe exchange services in reducing risk of HIV and HCV transmission.152,153 Moreover, SCFs have been shown to increase access to safer injection education,70,122,124 and to decrease the need to rush injections due to fear of arrest.43 Collectively, these findings provide strong evidence to support the expansion of SCFs as an infectious disease prevention strategy.  While concerns persist that SCFs may increase illicit drug use and discourage PWUD from seeking addiction treatment,27,72–74 such concerns are not supported by existing evidence. Indeed, the establishment of SCFs has not been found to have significantly altered community drug use patterns such as rates of injection initiation, relapse or cessation.115,116 Further, several studies have demonstrated the role of SCFs in facilitating entry into addiction treatment programmes,50,51,53,120 and subsequent injection cessation and/or reduced injecting at SCFs.50,53 Thus, these facilities appear to support rather than undermine the goals of addiction treatment.  In addition to addiction treatment, the research assessed in this review also suggests that SCFs provide opportunities for PWUD to access co-located services, including counseling, and syringe exchange services,2,16,70,122,123,125,126 while also facilitating critical medical intervention for the treatment of complex conditions such as cutaneous injection-related infections.2,16,70,122 Similarly, descriptive studies not included in the present review have highlighted how SCFs may help to connect PWUD with other on-site services, including basic supportive services (e.g., food, personal care facilities), HIV testing, mental health care, and naloxone training and distribution programmes.38,154          36 Further, the integration of SCFs and other low-threshold services into existing HIV/AIDS healthcare programmes has been shown to improve access to and engagement with HIV treatment and care among PWUD.155–157 Recent qualitative work has provided insights into how SCFs foster a supportive and welcoming environment characterized by social acceptance and belonging in which PWUD feel comfortable engaging with SCF staff regarding health needs.157,158 Thus, although PWUD are known to commonly experience barriers in accessing conventional healthcare services,15,159 the available data suggest that SCFs may help to mitigate such barriers in mediating access to a range of internal and external health and social resources.  Studies assessed in this review also indicate that SCFs are largely successful in achieving their objective of reducing public disorder associated with illicit drug use through declines in public injection and discarded drug use-related paraphernalia.20,110,127,128 These findings are consistent with those observed in descriptive studies showing declines in self-reported public drug use among SCF users.75,110,147,150,151 Further, as has been found in descriptive studies undertaken in the Netherlands and Switzerland,145,160–162 the implementation of SCFs in Vancouver and Sydney did not appear to contribute to increases in drug dealing or drug-related crime.21,49,128,130,131 Additionally, there is some evidence from Sydney to suggest increasing public acceptance and support of these facilities over time, although support was somewhat inconsistent among business owners.129 This largely aligns with work conducted elsewhere suggesting mixed support in terms of public opinion of SCFs,38,163 but that this tends to increase with time.37,38,109 Finally, despite not being an explicit objective, economic evaluations undertaken in Vancouver indicate that SCFs also offer an          37 additional public benefit of reducing the burden of costs on the public healthcare system.133–138   Overall, high-quality scientific evidence derived from the observational and simulation studies included in this review demonstrates the effectiveness of SCFs in meeting their primary public health and order objectives. Although randomized controlled trials (RCTs) are typically defined as the ‘gold standard’ for yielding level-one evidence on the effectiveness of a given intervention, it should be noted that RCTs of SCFs have been deemed unethical due to a lack of clinical equipoise and therefore have not been conducted.37,164,165 However, reliance on hierarchies of evidence to guide public heath decision making has been contested in recent years.166–168 Indeed, there has been growing acknowledgment that, like observational studies, RCTs often suffer from notable methodological weaknesses, including limited external validity, and that while RCTs may provide evidence that effectively serves the needs of clinical medicine, this is not necessarily the case in the realm of public policy.166,168 This is particularly relevant to decisions concerning complex public health interventions, as evidence of effectiveness in ‘real world’ contexts and attention to considerations such as health equity and human rights may be of equal or greater relevance to public health goals than contolled study of intervention efficacy.168 Further, assigned level of evidence is not necessarily indicative of methodological quality, and therefore well-designed observational research can arguably provide a level of evidence that meets or exceeds that derived from RCTs.166–168 Thus, given that it will not be possible to obtain evidence from RCTs on SCFs, decisions regarding the implementation of these facilities should instead be informed by the best available evidence derived from scientifically-viable studies,          38 which clearly demonstrates the positive impacts of SCFs in improving public order and advancing the health and human rights of socially marginalized PWUD. 2.4.1 Directions for future research Although the available evidence suggests that SIFs improve the health of PWUD and reduce community concerns associated with illicit drug use, several important research opportunities remain unexplored. First, despite evidence of the short- and medium-term health impacts of SCFs, rigorous research on the long-term impacts of SCFs on the health of PWUD is lacking. For example, while previous work has found SCF establishment to contribute to reductions in population-level overdose mortality rates,48,75 it is not known if SCF use has an impact on all-cause mortality. An additional area of evaluation that has not received adequate attention is the impact of SCFs on exposure to violence. Although previous qualitative research indicates that SCF use may decrease exposure to violence associated with the consumption of drugs,77,92,169 no studies have quantitatively evaluated the potential role of SCFs in reducing overall levels of exposure to physical or sexual violence among PWUD.   There is also a need for research to evaluate SCF programming that aims to be more responsive to the needs of vulnerable and underserved subpopulations of PWUD. For example, with the exception of SCFs operating in Geneva and Barcelona,170 SCFs in most settings are legally prohibited from accommodating individuals who require manual assistance with injections, despite the fact that this subpopulation accounts for an estimated one-third of PWID,171 is comprised largely of women and people with disabilities,171 and is disproportionally vulnerable to an array of serious harms including overdose, HIV infection and violence.172–174 A qualitative evaluation of an unsanctioned,          39 peer-run SCF in Vancouver that offered manual assistance with injections found that the provision of this service in a regulated environment helped to reduce risk for the above-mentioned harms.92 Nonetheless, further research on the potential benefits of offering assisted injection within SCFs may help to strengthen the case for legal reforms to allow for the wider adoption of this practice. In addition, although SCFs have previously been shown to provide protection from street-based drug scene violence for some women PWUD,169 other women may avoid SCFs due to perceived threats of violence.175 In an effort to address such concerns, women-only SCFs have been implemented in several settings, including Hamburg, Germany and Vancouver, Canada.30,77 While research undertaken in Hamburg found that the overwhelming majority of women-only SCF clients felt safer and more comfortable using drugs and approaching staff at this SCF,30 studies should further explore the ability of this form of tailored service to engage and support the health of structurally vulnerable drug-using women.   An additional research opportunity is to evaluate the health and social impacts of supervised inhalation rooms, which accommodate people who inhale drugs. Although supervised inhalation rooms are presently operating in some European cities,30 and recent qualitative research indicates that these facilities have potential to promote safer smoking practices and reduce health-related harms,122,176,177 the health and community outcomes specific to supervised inhalation rooms have not been thoroughly evaluated. As supervised inhalation rooms remain underutilized in many settings,30,54 further inquiry in this area may provide critical information to inform decisions regarding the possible broader implementation of these facilities.           40 Another notable knowledge gap concerns the role and impacts of novel SCFs models, including those integrated into existing healthcare and social services. For instance, although there is evidence to suggest that a majority of PWID would use an in-hospital SCF,178 and SCFs have recently opened in hospitals in Edmonton, Canada and Paris, France,29,54 this type of SCF model has not yet been well evaluated. However, recent qualitative research conducted in Vancouver suggested that the provision of hospital-based SCFs could reduce instances of patients leaving hospital against medical advice, promote culturally safe care and prevent adverse outcomes associated with in-hospital drug use among PWUD.179 Future studies should also investigate if the benefits of stand-alone SCFs will extend to SCFs integrated into existing shelters, supportive housing, and community organizations that serve PWUD, as research on such integrated SCF services is lacking. A related recommendation is to further examine the uptake and potential outcomes associated with services co-located with SCFs, including on-site addiction treatment and low-threshold housing.180 As well, given the limited geographic coverage of fixed-site SCFs,48,69 studies should evaluate how the implementation of mobile SCFs might improve the responsiveness of SCF programming to the needs of PWUD, particularly those who reside in settings with geographically dispersed drug scenes or who experience social-structural barriers to attending fixed SCFs (e.g., sex workers working in remote locations; women who avoid SCFs due to previous experiences of violence).54,175,181 A final recommendation is the continued assessment of peer-run SCFs, which are prohibited in many settings despite emerging evidence of their ability to engage and reduce harms among PWUD who may encounter social-structural and programmatic          41 barriers in accessing SCFs operated by healthcare professionals.38,54,92,177 Specifically, future studies should seek to better characterize preferences for, engagement with, and outcomes associated with peer-run SCF models, as this may help to further elucidate the role of these facilities in complementing or extending the reach of conventional SCF programmes.  2.4.2 Limitations A number of limitations common to observational studies apply to many of the studies included in this review. First, it is possible that the findings of the studies assessed herein are affected by residual confounding. In addition, most studies relied on non-random samples of PWUD in resource-rich settings and therefore our findings may not be generalizable to other contexts. Further, as previous work has indicated that SCFs attract socially marginalized and higher-risk PWUD,37,66–70 observed measures of the health benefits of SCF use may be biased towards the null. Finally, a limitation of this review is that despite our comprehensive search strategy, it is possible that we neglected to include some relevant literature, particularly non-English literature, not indexed in the databases searched for this review. 2.5 Conclusions In summary, while SCFs remain under-utilized in many settings worldwide, high-quality scientific evidence suggests that these effectively achieve their primary public health and order objectives with a lack of adverse impacts, and therefore supports their role as part of a continuum of services for PWUD. However, further studies are needed to better understand the potential long-term health impacts of these facilities. In addition, future research should continue to investigate innovations in SCF          42 models and programming, including efforts to tailor SCFs to the needs of vulnerable subpopulations of PWUD, in order to optimize the effectiveness and extend the reach and coverage of this form of harm reduction intervention.                              43 Table 2.1 Population, interventions, comparisons, outcomes and study design (PICOS) criteria for study inclusion  Criteria Definition Population People who use or inject drugs and the broader communities in which supervised consumption facilities (SCFs) are located Interventionsa Use, establishment or operation of SCFs Comparisons No exposure to SCFs   Outcomesa All individual- or population-level health or social outcomes Study Designb Original quantitative studies that assessed associations between SCFs and outcome(s) of interest for statistical and/or a priori-defined clinical significance a Original quantitative research studies were included if they examined the relationship between any aspect of use, establishment or operation of SCFs (including any service provided within SCFs) and any individual- or population-level health or social outcome (with significance assessed through an appropriate statistical test; the estimation of a measure of association (such as an odds ratio or rate ratio) and 95% confidence intervals; or an a priori-defined effect size considered to be of clinical significance). Feasibility studies that considered potential outcomes associated with the hypothetical establishment of SCFs were excluded. Studies that examined SCF use as an outcome were excluded, as examining characteristics of SCF users was beyond the scope of the present study. We also excluded studies that examined outcomes associated with exposure to larger facilities with integrated SCFs (unless use or operation of the SCF specifically was examined). b Review articles, case reports, case series, commentaries, editorials, qualitative studies and descriptive studies (that did not assess statistical or a-priori defined clinical significance) were excluded. If separate records presented overlapping results, the publication with the most complete information was included.                           44 Figure 2.1 Flowchart of record screening and selection process*                                   *From 182: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta Analyses: The PRISMA Statement. PLoS Med 6(7): e1000097. doi:10.1371/journal.pmed1000097        Records identified through database searching (n = 1476) Screening Included Eligibility Identification Additional records identified through other sources (n = 85) Records after duplicates removed (n = 1469) Records screened (n = 1469) Records excluded (n = 1128) Full-text articles assessed for eligibility (n = 341) 294 full-text articles excluded: • SCF not exposure or SCF definition not met (n = 94) • Outcome(s) not eligible (n = 19) • Study design not eligible (n = 170) • Overlapping/ duplicate data (n = 11) Studies included in systematic review (n = 47)          45 Table 2.2 Summary of included studies examining health and community outcomes associated with supervised drug consumption facilities (SCFs), arranged chronologically (n = 47)  Authors Location  Study design Participants Exposure(s) Outcome(s) Main Findings Poschadel et al. 200375 Saarbrüken, Hannover, Hamburg, & Frankfurt, Germany Time series N/A Establishment of SCFs  Monthly police records of drug-related deaths After the establishment of SCFs, there were significant reductions in drug-related deaths in the four respective cities (all p < 0.05).  Zurhold et al. 200370 Hamburg, Germany   Cross-sectional 616 people who use illicit drugs (PWUD) who used the SCF; Mean age: 32.6 years; 20% Female Frequent (≥daily) SCF use; occasional (<daily to ≥weekly) SCF use; rare (<weekly) SCF use Self-reported utilization of other services since began visiting SCFs (yes vs. no)  Frequent SCF users were more likely to use syringe exchange services compared to occasional or rare visitors (59% vs. 54% and 44%, respectively; p < 0.05). The same was true of counselling services (corresponding percentages = 46% vs. 35% and 25%; p < 0.01), medical services (37% vs. 29% and 17%; p < 0.01) and education on safer use (9% vs. 3% and 3%; p < 0.05). Hedrich 2004†109 Germany Mathematical simulation  N/A  Operation of SCFs Estimated annual overdose fatalities prevented  An estimated >10 deaths are prevented by SCFs in Germany each year. van Beek et al. 200467 Sydney, Australia Cross-sectional (derived from prospective cohort) 3747 people who inject drugs (PWID) and used supervised injection facility (SIF) Frequent SIF use (top quartile of the visits’ frequency distribution during the study period (i.e., 11+ visits during 17 month study period, measured through SIF database)) Non-fatal overdose at SIF during the study period (yes vs. no), measured through SIF database In multivariable logistic regression analyses, frequent SIF use was positively associated with experiencing a non-fatal overdose within the SIF (Adjusted odds ratio [AOR] = 6.1; 95% CI: 4.3– 8.6).          46 Wood et al. 200420 Vancouver, Canada  Pre-post ecological  N/A  Establishment of SIF (6 weeks before vs. 12 weeks after SIF opened) Number of people injecting in public; publicly-discarded syringes; and injection-related litter in the 10 blocks surrounding the SIF (measured by researcher counts) The SIF opening was associated with reductions in the number of people injecting in public (mean daily #: 2.4 (95% confidence interval [CI]: 1.9 – 3.0) after vs. 4.3 (95% CI: 3.5 – 5.4) before SIF opening), publicly discarded syringes (mean daily #: 5.4 (95% CI: 4.7 – 6.3) after vs. 11.5 (95% CI: 10.0 – 13.2) before SIF opening), and injection-related litter (mean daily #: 310 (95% CI: 305 – 317) after vs. 601 (95% CI: 590 – 613) before SIF opening) (all p < 0.05). Freeman et al. 2005130 Sydney, Australia  Time series  N/A Establishment of SIF  Police-recorded trends in theft and robbery incidents; drug use and drug dealing (measured by proxy of drug-related and total loiterers, counted by hired personnel) The SIF opening did not contribute to significant changes in trends (increases/decreases) in theft incidents, robbery incidents or drug-related loitering at the front of SIF after it opened (all p > 0.05). There were slight increases in drug-related loitering at the back of the SIF and total loitering at both the back and front of the SIF after opening (all p < 0.05). Kerr et al. 200547 Vancouver, Canada Cross-sectional and retrospective analyses (derived from prospective cohort data) 431 PWID  Self-reported SIF use (all, most or some vs. few or no injections at SIF) in the previous 6 months Self-reported syringe sharing (borrowing or lending) in the previous 6 months (yes vs. no) In multivariable logistic regression analyses, SIF use was associated with reduced syringe sharing (AOR = 0.30; 95% CI: 0.11 – 0.82). The odds of syringe sharing between SIF users and non-users were similar prior to the SIF opening (p = 0.50), suggesting that the observed reduction in syringe sharing among SIF users was not due to the SIF selecting PWID at inherently lower risk of syringe sharing. Thein et al. 2005129 Sydney, Australia  Serial cross- sectional 515 and 540 residents; 209 and 207 business owners in the 2 Establishment and operation of SIF (17 months after vs. 7 months before) Support for SIFs; whether or not SIF reduces risk of HIV/ HCV; reduces publicly The level of support for the SIF significantly increased in the neighbourhood of established SIF (68 to 78%; p < 0.001) among residents. There was no significant change in support for the SIF among business owners (p > 0.20). There was an increase          47 respective study years (i.e., 2000 and 2002)  discarded syringes; show dangers of injecting; reduces public injection; encourages drug injection; attracts PWUD; encourages belief that heroin injection is legal; makes law enforcement difficult (all yes vs. no) in the proportion of residents who agreed that SIFs reduce risk of HIV/ HCV (87 to 92%; p = 0.0004) and reduce discarded syringes (80 to 82%; p = 0.01). There was an increase in the proportion of residents who disagreed that SIFS encourage illicit drug injection (62 to 73%; p < 0.001), or encourage belief that heroin injection is legal (44% to 52%; p = 0.006). Among business owners, there was an increase in the proportion who agreed that SIFs reduce public injection (67 to 72%; p = 0.01) and show the dangers of injecting drug use (47 to 51%; p < 0.001), and there was a decrease in the proportion who agreed that SIFs encourage people to think that heroin injection is legal (55 to 43%; p = 0.001). Wood et al. 2005112 Vancouver, Canada  Cross-sectional (derived from prospective cohort) 582 PWID who used SIF (479 HIV-negative and 103 HIV-positive at baseline); 30% Female Self-reported exclusive SIF use for injection drug use in the previous month (yes vs. no) Self-reported borrowing a used syringe in the previous 6 months among HIV-negative participants; Lending a used syringe in the previous 6 months among HIV-positive participants (both yes vs. no) In bivariable logistic regression analyses, exclusive SIF use was associated with decreased odds of syringe borrowing among HIV-negative participants (OR = 0.14; 95% CI: 0.00 – 0.78) but was not significantly associated with syringe lending among HIV-positive participants (OR = 0.94; 95% CI: 0.00 – 7.90). Kerr et al. 2006115 Vancouver, Canada Prospective cohort  871 PWID; Median age (IQR): 35.3 (28.6 – 41.3) years; 39% Female Establishment of SIF (year after the SIF’s opening vs. the year before) Self-reported relapse into injection among former users; stopping injecting; introduction /discontinuation of methadone  There were no substantial changes in rates of relapse into injection drug use (17% to 20%), stopping injected drug use (17% vs. 15%), starting methadone use (11% vs. 7%), or stopping methadone use (13% vs. 11%).           48 Wood et al. 2006a21 Vancouver, Canada  Pre-post ecological  N/A   Establishment of SIF (year before vs. year after SIF opened) Police-recorded drug trafficking and drug-related crime in neighbourhood of SIF There were no increases in the number of drug trafficking offenses (124 vs. 116; p = 0.803) or assaults/robbery offenses (174 vs. 180; p = 0.565). A decline was observed in vehicle break-ins/vehicle theft offenses (302 vs. 227; p = 0.001). Wood et al. 2006b51 Vancouver, Canada  Prospective cohort 1031 PWID who used the SIF* Regular SIF use (≥weekly vs. <weekly) in the previous 6 months; any contact with an addictions counselor at the SIF in the previous 6 months (both measured through the SIF database) Use of detoxification service (measured through database linkage) In multivariable Cox regression analyses, regular SIF use (Adjusted hazard ratio [AHR] = 1.72; 95% CI: 1.25 – 2.38) and contact with a SIF addictions counsellor (AHR = 1.98; 95% CI: 1.26 – 3.10) were associated with more rapid time to entry into a detoxification programme. Kerr et al. 2007116 Vancouver, Canada  Prospective cohort 1065 PWID who used the SIF* N/A Rate of initiation into injection drug use at the SIF (measured through self-report and subtracting age at first injection from current age) Among the entire population of SIF users (n = ~5000), the estimated number who may have initiated injection drug use inside the SIF since the SIF opened was 5 (95% CI: 2 – 12), which is comparatively lower than the expected rate of initiation into injection drug use among local street-involved youth during a similar follow-up period (100 initiations; 95% CI: 81 – 122).  NCHECR 2007†110 Sydney, Australia  Pre-post ecological  1652 opioid-related deaths; 1558 opioid poisoning presentations at emergency departments (EDs) Establishment of SIF (60 months after vs. 36 prior to opening) Opioid-related deaths (measured by the state health department); opioid poisoning presentations at two EDs (measured by ED records) There was a significant decrease from an average of 4 to 1 deaths per month in the immediate vicinity of the SIF after the SIF was established (p < 0.001), compared to a decrease from 27 to 8 deaths in the rest of the state (p < 0.001). This difference in rate changes was not statistically significant (p = 0.877). There was a significant decrease from an average of 11 to 7 opioid poisoning ED presentations (35% reduction) after the SIF establishment (p < 0.001).          49 Salmon et al. 2007128 Sydney, Australia  Serial cross-sectional 515, 540 and 316 residents; and 269, 207 and 210 business operators in the 3 respective study years (i.e., 2000, 2002 and 2005)  Establishment and operation of SIF  Witnessed public injection in last month; publicly-discarded syringes in last month; drugs offered for purchase in the last month (all yes vs. no) The proportions of residents who had witnessed public injecting in the last month were 33%, 28% and 19% in 2000, 2002 and 2005, respectively (p < 0.001), while the corresponding proportions for business operators were 38%, 32% and 28% (p = 0.03). The proportion of residents who had seen publicly-discarded syringes in the last month was 67%, 58% and 40% in 2000, 2002 and 2005, respectively (p < 0.001) while the corresponding proportions for business owners were 72%, 64% and 57% (p = 0.01). The proportion of residents who had been offered drugs for purchase in the last month was 28%, 29% and 26% in 2000, 2002 and 2005 (p = 0.80). The corresponding proportions for business owners were 33%, 34%, and 28% (p = 0.26).  Stoltz et al. 200746 Vancouver, Canada Cross-sectional (derived from prospective cohort) 760 PWID who used a SIF*  Consistent SIF use (≥25% of injections vs. <25%) in the previous 6 months Self-reported changes since SIF opening in: syringe reuse; rushed injecting; injecting outdoors; use of sterile water; cooking or filtering drugs; tying off; safer syringe disposal; easier finding vein; injecting in clean place (all yes vs. no) In multivariable logistic regression analyses, consistent SIF use was positively associated with a change in each injection behaviour: reuse syringes less often (AOR = 2.04; 95% CI: 1.38 – 3.01), less rushed during injection (AOR = 2.79; 95% CI: 2.03 – 3.85), less injecting outdoors (AOR = 2.70; 95% CI: 1.93 – 3.87), using clean water for injecting (AOR = 2.99; 95% CI: 2.13 – 4.18), cooking or filtering drugs prior to injecting (AOR = 2.76; 95% CI: 1.84 – 4.15), tying off prior to injection (AOR = 2.63; 95% CI: 1.58 – 4.37), safer disposal of syringes (AOR = 2.13; 95% CI: 1.47 – 3.09), easier finding of a vein (AOR = 2.66; 95% CI: 1.83 – 3.86) and injecting in a clean place (AOR = 2.85; 95% CI: 2.09 – 3.87). Wood et al. 200753 Vancouver, Canada  Prospective cohort 1031 PWID who use SIF* Establishment of SIF (year before vs. year after SIF opened) Enrolment in detoxification service; number of visits to the SIF in the month after detoxification In multivariable generalized estimated equations (GEE) with logit link analyses, there was a significant increase in uptake of detoxification services in the year after vs. the year before the SIF opened (AOR = 1.32; 95% CI: 1.11 – 1.58). In multivariable Cox regression analyses,          50 enrolment (both measured through database linkage) detoxification service use was associated with more rapid entry into MMT (AHR = 1.56; 95% CI: 1.04 – 2.34) and other forms of addiction treatment (AHR = 3.73; 95% CI: 2.57 – 5.39). Among those who enrolled in detoxification, the rate of SIF use declined in the month after enrollment compared to the rate of SIF use in the month prior to enrolment (19 vs. 24 visits; p = 0.002). Bayoumi and Zaric 2008134 Vancouver, Canada  Mathematical simulation  Estimated 3000 to 20,000 PWID infected with HIV and/or HCV SIF operation (simulation over 10 years) Cost-effectiveness of the SIF based on the prevention of incident HIV and HCV infections (with the SIF vs. without the SIF) An estimated 1191 incident HIV and 54 incident HCV cases were averted over 10 years, resulting in an estimated minimum net savings of $14 million CAD and 920 years of life gained over 10 years.  Kimber et al. 2008120 Sydney, Australia   Prospective cohort    3715 PWID who used SIF; 47%> 30 years; <40% Female  Frequent SIF use (top quartile of the visits’ frequency distribution during the 17 month study period (i.e., 12+ visits, measured through SIF database)) Addiction treatment referral (received at least one written referral during the study period); Addiction treatment uptake (use of referral card, yes vs. no). In multivariable Cox regression analyses, frequent SIF use was positively associated with drug treatment referral (AHR = 1.6; 95% CI: 1.2 – 2.2) but was not significantly associated with drug treatment referral uptake (AOR = 0.8; 95% CI: 0.4 – 2.0). Lloyd-Smith et al. 2008117 Vancouver, Canada Prospective cohort   1065 PWID who used SIF* Self-reported exclusive SIF use for injection drug use in the previous 6 months (yes vs. no) Current cutaneous injection-related infection (CIRI), measured visually by study nurse (yes vs. no) In multivariable generalized linear mixed-effects analyses, exclusive SIF use was not significantly associated with development of a CIRI (AOR = 0.58; 95% CI: 0.29– 1.19). Milloy et al. 2008a106 Vancouver, Canada Prospective cohort 1090 PWID and use SIF* Self-reported frequent SIF use in the previous 6 months (≥ 75% vs. Self-reported non-fatal overdose in the previous 6 In multivariable GEE analyses, frequent SIF use was not associated with recent non-fatal overdose (AOR: 1.01; 95% CI: 0.77 – 1.32). In bivariable GEE analyses, reporting that SIF use had changed          51 <75% of injections); Self-reporting that SIF use had resulted in a change in injection practices in the previous 6 months (yes vs. no). months (yes vs. no) injection practices was not associated with recent non-fatal overdose (AOR = 0.77; 95% CI: 0.53 – 1.11).  Milloy et al. 2008b107 Vancouver, Canada Mathematical simulation 453 potentially fatal overdose events out of 766,486 injections during the study period Operation of SIF Overdose deaths averted It was estimated that 1.9 to 11.7 overdose deaths were averted per year. Richardson et al. 2008118 Vancouver, Canada  Prospective cohort 1090 PWID who use SIF*   Self-reported SIF use (≥25% vs. <25% of injections) in the previous 6 months Self-reported employment (job with regular salary or temporary work) in the previous 6 months (yes vs. no) In multivariable GEE analyses, SIF use was not associated with employment (AOR = 1.05; 95% CI: 0.88 – 1.27). Wood et al. 2008124 Vancouver, Canada Prospective cohort 1087 PWID who use SIF* Self-reported frequent SIF use (≥75% vs. <75% of injections) in the previous 6 months Self-reported receipt of safer injection education at the SIF in the previous 6 months (yes vs. no) In multivariable GEE with logit link analyses, frequent SIF use was associated with an increased likelihood of receiving safer injection education at the SIF (AOR = 1.47; 95% CI: 1.22 – 1.77). Bravo et al. 2009111 Barcelona & Madrid, Spain  Cross-sectional (derived from prospective cohort)   249 people who inject heroin aged 30 years or younger (137 in Barcelona; 112 in Madrid); 76% Any use of at least one of five SIFs since last interview (mean=17.3 months [SD=5.7 Self-reporting of not borrowing used syringes; not sharing injection equipment since the last interview In multivariable logistic regression analyses, SIF use was associated with not borrowing used syringes (AOR = 3.3; 95% CI: 1.4 – 7.7). SIF use was not significantly associated with not sharing injection equipment (AOR = 1.1; 95% CI: 0.5 – 2.2).            52  > 25 years; 26% Female months]) (yes vs. no) Marshall et al. 2009119 Vancouver, Canada Prospective cohort  794 PWID who used the SIF and reported sexual activity during one or more interviews*   Time since recruitment from the SIF (measured in SIF database); Self-reported use of medical services at Insite (e.g., nurse consultation, HIV testing, referral to health services) in the previous 6 months (yes vs. no) Consistent condom use during vaginal and/or anal intercourse in the previous 6 months (always vs. usually, sometimes, occasionally, never)  In multivariable GEE with logit link analyses, use of SIF healthcare services was marginally associated with consistent condom use among those with regular partners (AOR = 1.27; 95% CI: 0.99 – 1.64) but not among those with casual partners (OR = 0.94; 95% CI: 0.71 – 1.26). Time since recruitment from the SIF was associated with consistent condom use among those with regular partners (AOR = 1.29; 95% CI: 1.06 – 1.55) but not those with casual partners (AOR = 1.15; 95% CI: 0.90 – 1.47). Milloy et al. 2009132 Vancouver, Canada Prospective cohort 902 PWID who used the SIF*   Self-reported frequent SIF use (≥75% vs. <75% of injections) in the previous 6 months Self-reported incarceration in the previous 6 months (yes vs. no) In multivariable GEE analyses, frequent SIF use was not associated with recent incarceration (AOR = 0.99; 95% CI: 0.79 – 1.23). Andresen and Boyd 2010137 Vancouver, Canada Mathematical simulation  Estimated 5000 PWID in population SIF operation Benefit cost-ratios for the SIF based on prevention of incident HIV infections and overdose deaths Mathematical modelling estimated that Insite prevents approximately 35 incident cases of HIV infection and 3 overdose deaths per year, providing an annual excess of $6 million with an average cost-benefit ratio of 5.12:1. Baars et al. 2010123 Rotterdam, Utrecht & South Limburg, Netherlands Cross-sectional  309 PWUD; Mean age (SD): 41.5 (7.4) years; 22% Female Self-reported SCF use at least once in the last 6 months (yes vs. no) Hepatitis B vaccination programme awareness (yes vs. no) and self-reported uptake (yes vs. no) In multivariable logistic regression analyses, SCF users were more likely to be aware of Hepatitis B vaccination programme than non-users (AOR = 1.86; 95% CI: 1.04 – 3.33), but SCF use was not associated with Hepatitis B vaccination uptake (p > 0.05; data not shown).           53 Fitzgerald et al. 2010†131 Sydney, Australia Time series N/A  Establishment and operation of SIF Police recorded trends of criminal incidents of robbery, property crime and illicit drug offenses (use or deal amphetamines narcotics and cocaine) Incidence of robbery and property offences declined in both the neighbourhood of SIF and the rest of Sydney between 1999 and 2010. Illicit drug offense incidents declined in the neighbourhood of the SIF between 1999 and 2003 and then remained stable until 2009. A similar pattern was observed in the rest of Sydney (drug arrests declined from 1999 to 2003, but with a slight upward trend from 2003 to 2010). Lloyd-Smith et al. 201016 Vancouver, Canada Prospective cohort 1083 PWID who use SIF* Referral to hospital by a SIF nurse (yes vs. no), measured by linkage to SIF database Hospitalization for CIRI (yes vs. no); duration of hospitalization (in days), both measured by linkages to local hospital inpatient database In multivariable Cox regression analyses, referral to hospital by SIF nurses was associated with increased likelihood of hospitalization for CIRI (AHR = 5.38; 95% CI: 3.39 – 8.55). Referral to hospital by SIF nurses was significantly and independently associated with shorter duration of stay in hospital (4 days [IQR: 2 – 7] vs. 12 days [IQR: 5 – 33]). Each referral by a SIF nurse would result in an estimated savings of $5,696 CAD [IQR: $2,136 – 18,512]. Milloy et al. 2010121 Vancouver, Canada  Prospective cohort 1083 PWID who use SIF*  Self-reported frequent SIF use (≥ 75% vs. <75% of injections) in the previous 6 months  Self-reported inability to access addiction treatment in the previous 6 months (yes vs. no) In bivariable GEE with logit link analyses, frequent SIF use was not significantly associated with trying but being unable to access addiction treatment  (OR = 1.08; 95% CI: 0.84 – 1.40). Pinkerton et al. 2010136 Vancouver, Canada  Mathematical simulation Estimated 5000 PWID Operation of SIF Annual number of HIV infections and associated costs If the SIF ceased operating, there would be an estimated increase from 179.3 to 262.8 annual incident HIV infections among local PWID, which would be associated with $17.6 million CAD in lifetime HIV-related healthcare costs. These savings from future hypothetical healthcare costs exceed the annual operating costs of the SIF (approximately $3 million CAD).           54 Salmon et al. 2010108 Sydney, Australia Pre-post ecological 20,409 ambulance attendees at opioid-related overdoses (1485 in the SIF neighbourhood) before and after the opening of the SIF Establishment and operation of SIF (36 months before vs. 60 months after the SIF opened) Average monthly ambulance attendances at suspected opioid-related overdoses in the vicinity of the SIF vs. the rest of the state (measured through ambulance service database) After the opening of the SIF, the average monthly ambulance attendances at suspected opioid-related overdoses declined significantly in the immediate vicinity of the SIF (by 68%) compared to 61% in the rest of the state during SIF operating hours (p = 0.002). During the SIF operating hours, this difference was more pronounced with an 80% decline in the immediate vicinity of the SIF compared to a 60% decline in the rest of the state (p < 0.001).  Scherbaum et al. 2010114 Essen, Germany  Prospective cohort 129 PWID who initiated use of the SIF or began attending the SIF again after 6+ weeks of non-attendance; Mean age (SD): 31 (6); 25% Female Changes over time (1, 2, 3 months after first use of SIF vs. first use of SIF) Outdoor drug use; use of non-sterile equipment; equipment sharing; injection-related abscesses  Compared to baseline, at 1 month follow-up of first use of the SIF, the proportion of 71 participants who reported outdoor drug use, use of non-sterile equipment and equipment sharing remained relatively stable at approximately 50%, 50% and 20%, respectively (all p > 0.30). At 1 month follow up compared to baseline, the proportion who had injection-related abscesses was similar (8.5% vs. 4.2%, p > 0.30). At 3 months follow-up of first use of the SIF, the proportion of 26 participants who used drugs outdoors, used non-sterile equipment, shared equipment and had abscesses were comparable to baseline (all p > 0.30; data not shown). Debeck et al. 201150 Vancouver, Canada Prospective cohort  1090 PWID and use SIF* Regular (≥weekly vs. <weekly) SIF use at baseline; Contact with addictions counselor (at least once before event or censor date) at the SIF (both measured through SIF database) Self-reported uptake of addiction treatment (all treatment modalities including residential treatment and methadone maintenance therapy); Self-In multivariable Cox regression analyses, regular SIF use (AHR = 1.33; 95% CI: 1.04 – 1.72) and having contact with the addiction counselor within the SIF (AHR = 1.54; 95% CI: 1.13 – 2.08) were independently and positively associated with initiation of addiction treatment. Enrolment in methadone maintenance therapy (AHR = 1.57; 95% CI: 1.02 – 2.40) and other addiction treatment (AHR = 1.85; 95% CI: 1.06 – 3.24) were positively associated with injection drug use cessation.          55 reported injection cessation for ≥6 months  Marshall et al. 201148 Vancouver, Canada  Pre-post ecological  290 decedents; Median age (IQR): 40 (32-48) years; 21% Female  Establishment of SIF (2 years after vs. 2 years prior to SIF opening)  Fatal overdose (measured by Coroner records) Fatal overdose decreased by 35.0% within 500 metres from the SIF from 253.8 to 165.1 deaths per 100,000 person-years (p = 0.048) in the two years after the opening of the SIF vs. the two years prior to the SIF opening, compared to a 9.3% reduction in fatal overdose from 7.6 to 6.9 per 100,000 person-years in the rest of the city (p = 0.490). These rate changes were significantly different (p = 0.049). Pinkerton et al. 2011135 Vancouver, Canada Mathematical simulation Estimated 5000 PWID Operation of SIF Annual number of HIV infections and associated costs The SIF prevents an estimated average of 5.6 infections year (90% CI: 4.0 – 7.6), reducing HIV incidence by an estimated 6–11% among local PWID and averting more than $1 million CAD in future HIV-related healthcare costs, and accounting for an estimated $200,000 – $400,000 CAD  in savings per year after considering the SIF’s operating costs. Andresen and Jozaghi 2012133 Vancouver, Canada Mathematical simulation  Estimated 5000 PWID in population SIF operation Cost-benefit ratios Mathematical modeling estimated that the SIF prevents 22 incident HIV infections per year, providing an average cost-benefit ratio of 3.09:1. Lloyd-Smith et al. 20122 Vancouver, Canada Prospective cohort 1083 PWID who use SIF* Referral to hospital by a SIF nurse (yes vs. no), measured by linkage to SIF database ED use for CIRI (yes vs. no), measured by linkage to local hospital ED database In multivariable Cox regression analyses, referral to hospital by SIF nurses was independently and positively associated with ED use for CIRI among females (AOR = 4.48; 95% CI: 2.76 – 7.30) and males (AOR = 2.97; 95% CI: 1.93 – 4.57). Donnelly and Mahoney 2013†49 Sydney, Australia   Time series   N/A  Establishment and operation of SIF Police-recorded trends in criminal incidents of robbery, theft and illicit drug offences  Incidents of robbery and theft incidents declined over time in neighbourhood of SIF since it was established (consistent with the rest of Sydney) (all p <0.001). Possession of illicit substances remained stable from May 2001 (when SIF opened) to 2008 but increased from 2009 onwards in both the neighbourhood of the SIF and in the rest of Sydney. A similar trend was documented with crime rates per 100,000 population. There were no          56 changes in trends of drug-related incidents occurring in the 50 m. of the SIF during the study period. Vecino et al. 2013127 Barcelona, Spain   Pre-post ecological  N/A   Establishment of SCFs (after vs. before) Monthly-averaged publicly discarded syringes (collected by local services) After the opening of two SCFs, there was a significant reduction in the average monthly number of publicly-discarded syringes (from 13.13 in 2004 to 3.19 in 2012). Jozaghi and Vancouver Area Network of Drug Users 2014138 Vancouver, Canada  Mathematical simulation  Estimated 4330 people who smoke crack cocaine Operation of an unsanctioned supervised inhalation room  Benefit-cost and cost-effectiveness ratios for the supervised inhalation room based on prevention of incident HCV infections The supervised inhalation room prevented an estimated 57 incident cases of HCV infection per year, providing average annual savings of $1.8 million CAD per year with an average benefit-cost ratio of 12.1:1 and a marginal cost-effectiveness ratio ranging from $1705 to 97,203 CAD. Kinnard et al. 2014113  Copenhagen, Denmark   Cross-sectional  41 PWID who used SIF; Median age (IQR): 37 (30; 43) years; 9.8% Female Opening of the SIF (after vs. before) Self-reported perceived changes in syringe disposal practices and injection-related risk  behaviours  In total, 24 participants (58.5%) reported changing syringe disposal practices (with 23 reporting change from not always to always disposing safely) after the SIF opening (p < 0.001). 75.6% reported reductions in injection risk behaviours after SIF opening (63.4% less rushed injecting; 56.1% fewer outdoor injections; 53.7% stopped syringe sharing; 43.9% cleaned injection sites more often).  Skelton et al. 2016126 Sydney, Australia Serial cross-sectional SIF staff and clients Smoking cessation care organizational change intervention at the SIF (after vs. before) Self-reported receipt of smoking cessation care at the SIF (among SIF users); self-reported smoking cessation care strategies (among SIF staff) In the post intervention period, more SIF users reported receiving smoking cessation care (p < 0.05), and more SIF staff reported providing verbal advice regarding smoking cessation, offer of free nicotine replacement therapy, referral to a physician and follow up to check on smoking cessation progress (all p < 0.01).          57 Toth et al. 2016122 Copenhagen, Aarhus and Odense, Denmark  Cross-sectional 154 PWUD who use at least one of five SCFs; 10% < 30 years; 25% Female   Self-reported receipt of education in hygienic injection practices at SCF; Self-reported referral to medical help by SCF staff  Self-reported use of SCF to access clean injection equipment (yes vs. no); Self-reported receipt of treatment for condition (yes vs. no) Those who had received education on hygienic injection practices at a SCF were more likely to access SCFs for clean injection equipment vs. those who had not received such education (68.8% vs. 25.9%; p = 0.024). Those advised to seek medical help by staff for a medical condition were more likely to receive treatment for the condition than who were not advised to seek treatment for a condition (51.3% vs. 25.7%; p = 0.003).  Lysyshyn et al. 2017125 Vancouver, Canada  Cross-sectional 472 drug checks for fentanyl at the SIF  Result of drug test for fentanyl at the SIF (positive vs. negative), measured using a test strip designed for urine testing Disposal of drugs; reduced dose of drugs; Experienced overdose at the SIF (all yes vs. no), all measured by SIF staff Receiving a positive fentanyl result was associated with increased drug dose reductions (37% vs. 8%; p < 0.05) but not disposals of drugs (9% vs. 8%; p > 0.05). A positive fentanyl result was also associated with overdosing at the SIF (9% vs. 2%; p < 0.05).          58 Chapter 3: Discontinued use of a supervised injection facility among people who inject drugs in Vancouver, Canada: A prospective cohort study 3.1  Introduction As detailed in Chapter 2, existing evidence demonstrates that SCFs effectively attain their primary objectives without producing unintended adverse consequences. Despite the established benefits of SCFs, these facilities remain controversial in some settings,27,37,55,56,183,184 and efforts to implement SCFs have been impeded by concerns regarding the potential adverse consequences of these programs, including that these may perpetually maintain people in active drug use.27,35,56,72–74,183,185 However, previous studies of SCF use and client characteristics have been limited to those with cross-sectional designs or short-term follow-up durations.66–70 As such, little is known about long-term SCF utilization patterns among PWID, including discontinued use of these services. Characterizing SCF use discontinuation may help to address concerns regarding the potential role of these facilities in perpetuating active injection drug use, and thus could provide important information to guide evidence-based policy recommendations concerning the potential implementation of future SCFs, including in settings where these facilities have not yet been established. Further, such information could inform the ongoing development of existing SCF programming in Canada and elsewhere, including how such services could be tailored to address the needs of various subpopulations of clients with distinct SCF utilization patterns. The present study was therefore undertaken to longitudinally characterize cessation of use of the          59 Insite SIF among a community-recruited prospective cohort of PWID in Vancouver, Canada. 3.2 Methods Data for these analyses were derived from the VIDUS and ACCESS studies, as described in Section 1.7.1.  3.2.1 Study sample All participants enrolled between December 1, 2005 and November 30, 2016 who reported using the Insite SIF in the previous six months at baseline or in a follow-up interview and had at least one subsequent follow-up interview were included in the present study. 3.2.2 Variable selection Given that SIFs are programmes designed for people engaged in active injection drug use,37 and given our interest in examining SIF use cessation independent of the influence of injection cessation, our outcome of interest was SIF use cessation during periods of active injection drug use. This outcome was ascertained by examining responses to the following two questions: “In the past six months, have you used a needle to chip, fix, or muscle even once?” and “Have you fixed at the Insite SIF in the last six months?” An event was defined as any instance of concurrently self-reporting “yes” to the first question and “no” to the second question. A range of individual and contextual factors that were selected based on Rhodes’ Risk Environment Framework22,23,39 and previous research examining use of SIFs and other harm reduction services among PWID69,71,94,122,186,187 were considered as explanatory variables: age (per year older), sex (male vs. female), ethnicity (white vs. others), residency in the          60 Downtown Eastside neighbourhood (yes vs. no), homelessness (yes vs. no), employment (i.e., regular, temporary, or self-employment; yes vs. no), HIV status (positive vs. negative serology), heroin injection (≥daily vs. <daily), cocaine injection (≥daily vs. <daily), crystal methamphetamine injection (≥daily vs. <daily), public injection (yes vs. no), binge injection (yes vs. no), require help injecting (yes vs. no), non-fatal overdose (yes vs. no), syringe sharing (yes vs. no), enrolment in MMT (yes vs. no), difficulty accessing addiction treatment (yes vs. no), exposure to violence (yes vs. no), sex work involvement (yes vs. no) and incarceration (yes vs. no). Difficulty accessing addiction treatment was defined in response to the question: “In the last six months, did you try to get into any treatment for your drug or alcohol use but were unable?” Unless otherwise indicated, all variables referred to experiences and activities in the previous six months and were treated as time-updated based on each semi-annual follow-up visit. 3.2.3 Statistical analyses First, Pearson’s Chi-squared test for categorical variables and the Mann-Whitney test for continuous variables were used to compare the baseline characteristics of those who did and did not report ceasing use of the SCF while actively injecting during follow up. The candidate then descriptively examined the number of SIF use cessation events that co-occurred with periods of injection cessation (defined as self-reporting not injecting drugs during the same six-month period in which SIF use cessation was reported). Next, the incidence density of SIF use cessation during periods of active injection and injection cessation, respectively, were calculated using the Poisson distribution.          61 During the study period, some participants reported more than one SIF use cessation event while actively injecting. Therefore, a recurrent event survival model was constructed to examine the relationship between explanatory variables and the (repeated) outcome of interest. This model incorporated information on all SIF use cessation events that occurred during periods of active injection drug use over the duration of the study period. A proportional rates-means model described by Lin et al.188 was used to account for correlation among the length of individuals’ repeated time with potential for SIF use cessation during periods of active injection. In this model, a counting process framework was specified to define time to repeated events, such that individuals were considered eligible for the outcome from time zero to the first event, from the first event to the second event, and so forth. For all participants, time zero was defined as the date of first reported use of the SIF. The date of each event was estimated as the mid-point of the previous six-month period given that the precise date of SIF use cessation was not known. During follow up, participants who did not cease using the SIF during periods of active injection were right censored at the date of their latest interview, the date of first report of injection drug use cessation, or November 30, 2016, whichever came first.  The candidate first computed crude hazard ratios (HRs) to estimate the bivariable association between each explanatory variable and repeated SIF use cessation while actively injecting. An a priori-defined model-building approach was then applied by fitting a multivariable model that included all variables significantly associated with SIF use cessation during periods of active injection at p < 0.20 in bivariable analyses. All remaining variables were then subjected to a backwards selection procedure based on          62 the Akaike information criterion (AIC) and Type III p-values.189 Each variable with the highest p-value was removed sequentially, with the final model including the set of variables associated with the lowest AIC.  Finally, as a sub-analysis, the candidate descriptively examined participants’ responses to the following question: “If you have fixed at Insite only once or have stopped fixing there, why have you not gone back?” As this question was only included in the study questionnaire between June 1, 2010 and November 30, 2012, this analysis was limited to participants who responded to this question at their first report of SIF use cessation during this time period. This question was asked of study participants who reported ceasing use of the SIF regardless of whether or not a cessation event occurred during periods of active injection. Response options, which were not read aloud to study participants, included: Don’t like the Downtown Eastside; already have a safe place in which to inject; prefer to inject at home; too far from where I score drugs; have to wait too long to inject; need help injecting; can’t split deals; entrance is too public; don’t want to have to have to register to use the site; too many police near the site; prefer to keep drug use private; prefer to inject alone; quit fixing; moved away; in jail; don’t want to inject with strangers; poor treatment by Insite staff; for heroin users only (interviewers explained that this is not true); just wanted to check it out; poor treatment by other users; banned/ barred; other (specify). Participants could provide more than one response. The candidate conducted all statistical analyses with SAS version 9.4 (SAS Institute Inc., Cary, NC), and all reported p-values are two-sided.          63 3.3 Results A total of 1336 individuals were eligible and included in the present analyses. Of these, the median age at baseline was 41 years (interquartile range [IQR] = 34 – 47), 467 (35.0%) were women, and 846 (63.6%) reported white ethnicity. Table 3.1 presents the baseline characteristics of participants, stratified by reporting at least one six-month period of SIF use cessation while actively injecting during follow up. During the eleven-year study period, participants contributed 6254 person-years of observation time (median = 50.1 person-months; IQR = 17.9 – 93.9 person-months). In total, 847 (63.4%) individuals reported 1663 six-month periods of SIF use cessation while actively injecting drugs, yielding an incidence density of 26.6 events per 100 person-years (95% confidence interval [CI]: 25.3 – 27.9). An additional 2282 SIF use cessation events occurred during periods of injection drug use cessation (incidence density: 36.5 (95% CI: 35.0 – 38.0) events per 100 person-years), accounting for 57.8% of the total 3945 SIF use cessation events that occurred among 1030 participants (77.1% of the study sample) during either periods of active or inactive injection drug use. Of the 847 (63.4%) individuals who reported at least one six-month period of SIF use cessation while actively injecting drugs, 398 (47.0%) reported a single SIF use cessation event, 223 (26.3%) reported two cessation events, 129 (15.2%) reported three cessation events, 59 (7.0%) reported four cessation events, 33 (3.9%) reported five cessation events, 4 (0.5%) reported six cessation events, and 1 (0.1%) reported seven cessation events, for a total of 1663 events.        The results of the bivariable and multivariable extended Cox regression analyses are shown in Table 3.2. In multivariable analyses, enrolment in MMT (Adjusted Hazard          64 Ratio [AHR] = 1.30; 95% CI: 1.11 – 1.53, p = 0.002) remained significantly and positively associated with SIF use cessation during periods of active injection. Factors that remained significantly and inversely associated with SIF use cessation during periods of active injection in adjusted analyses included: homelessness (AHR = 0.61; 95% CI: 0.51 – 0.73, p < 0.001), ≥daily heroin injection (AHR = 0.78; 95% CI: 0.65 – 0.94, p = 0.010), public injection (AHR = 0.78; 95% CI: 0.66 – 0.93, p = 0.005), binge injection (AHR = 0.77; 95% CI: 0.65 – 0.91, p = 0.003), and incarceration (AHR = 0.63; 95% CI: 0.50 – 0.81, p < 0.001). Difficulty accessing addiction treatment (AHR = 0.74; 95% CI: 0.53 – 1.02, p = 0.063) was marginally and inversely associated with SIF use cessation during periods of active injection in adjusted analyses. Of the 493 study participants who ceased using the SIF during periods of either active or inactive injection drug use and responded to the question regarding reasons for SIF use cessation, the most commonly reported reasons for doing so included: quit fixing (n = 150; 42.3%); prefer to inject at home (n = 109; 30.7%); already have a safe place in which to inject (n = 57; 16.1%); don’t like the Downtown Eastside (n = 14; 3.9%); and moved (n = 14; 3.9%). 3.4 Discussion In this study of a community-recruited cohort of 1336 PWID who used the Insite SIF in Vancouver, Canada and were subsequently followed for a median of 50 months, it was found that ceasing SIF use was common with approximately three-quarters (77%) of participants reporting at least one six-month period of SIF use discontinuation while either actively or inactively injecting drugs. The majority of the observed 3945 SIF use cessation events occurred during periods of injection cessation (58%), and the remaining          65 SIF use cessation events occurred during periods of active injection (42%). In multivariable recurrent event analyses, enrolment in MMT was positively associated with SIF use cessation during periods of active injection, while homelessness, high intensity heroin injection, binge injection, public injection, and incarceration were inversely associated with this outcome. Reporting having difficulty accessing addiction treatment was marginally and inversely associated with SIF use cessation during periods of active injection, and HIV seropositivity was marginally and positively associated with this outcome. The primary reported reasons for discontinuing use of the SIF included injection cessation, a preference for injecting at home, and already having a safe place in which to inject. Although previous cross-sectional studies have identified correlates of short-term SIF use,67–70,190 the present study addresses gaps in current knowledge of long-term SIF utilization patterns in that it is the first study, to our knowledge, to longitudinally characterize discontinuation of SIF use among a community-recruited prospective cohort of PWID. It was found that most PWID who used the Insite SIF discontinued using this service during follow up, and that injection drug use cessation co-occurred with the majority of SIF use cessations events. Furthermore, injection cessation was the most commonly reported reason for discontinuing use of this health service. Consistent with previous studies,50,53 these findings challenge the contention that SIFs inhibit injection cessation. The results of this study also demonstrate that PWID enrolled in MMT were more likely to discontinue SIF use while actively injecting drugs. This association is likely explained by a decreased need to use the SIF due to reductions in injection drug          66 use after initiating MMT, as the link between MMT and declines in opioid use has been well described.191,192 Moreover, previous studies of local PWID have found SIF use to be associated with increased uptake of addiction treatment, including MMT, and subsequent reductions in injection drug use and SIF use.50–53 Together with these findings and past research demonstrating the role of opioid agonist therapies in reducing various harms associated with opioid dependence among PWID,193–195 our findings highlight potential benefits of offering links to evidence-based addiction treatment modalities within SIFs. However, it should be noted that because SIF use cessation events examined in multivariable analyses occurred during periods of active injection, we cannot exclude the possibility that some of those on MMT who discontinued using the SIF may have instead transitioned to injecting drugs in other settings without necessarily decreasing their overall drug use. For example, users of both opioids and stimulants may have decreased their opioid use but continued using stimulants after enrolling in MMT, and then ceased injecting at Insite due to perceiving a reduced risk of overdose compared to when they were previously using opioids more frequently.196 Additional research, particularly qualitative research, may help to further elucidate potential explanations and mechanisms underlying the observed association between MMT enrolment and SIF use cessation. Further, our finding that difficulty accessing addiction treatment was marginally associated with decreased likelihood of SIF use cessation during periods of active injection highlights the need for the further development and evaluation of strategies to increase access to low-threshold addiction treatment options for SIF clients who may experience barriers to engaging with such services.          67 Results from our multivariable analyses also demonstrate that PWID with established markers of structural vulnerability and drug-related risk, including homelessness, high intensity heroin injection, binge injection, and public injection, were less likely to cease using the SIF while actively injecting drugs. Our finding that homeless PWID and public injectors were less likely to discontinue using the SIF largely corroborates with our descriptive findings indicating that some of the most common reasons for SIF use cessation were preferring to inject at home and already having a safe place in which to inject, as the alternative injection settings for marginally-housed PWID and those who inject in public may be particularly unsafe,23,42,197 thus contributing to their sustained engagement with SIF services. While previous studies have found that SIFs effectively attract their target population of PWID at elevated risk of health-related harms, including those who are homeless or engage in higher-risk drug use practices,66–70 the present study builds on this work in demonstrating that this health service also successfully retains higher-risk subpopulations of PWID during periods of active injection. This finding is encouraging given the numerous health benefits of SIF use reported in Chapter 2.  Our finding that recently incarcerated PWID were less likely to cease using the SIF during periods of active injection is somewhat surprising given that previous research conducted in Vancouver and other settings has demonstrated associations between incarceration and interruptions in the use of harm reduction and treatment services among PWID, including in periods post-release from custody, which, in turn, may increase risk of harms such as fatal overdose and HIV infection.198–201 In addition, a previous study that relied on longitudinal data found that frequent SIF use was not          68 significantly associated with recent incarceration among PWID in Vancouver.132 It is possible that the observed relationship between incarceration and decreased likelihood of SIF use cessation could be partially explained by the social and economic instability that often characterizes the immediate post-incarceration release period among PWID,200,202,203 which might contribute to increased reliance on SIFs due to a lack of stable access to alternative safe environments in which to inject. Alternatively, it could be that long-term SCF users were simply more likely to contend with exposures (e.g., socio-economic marginalization; involvement in prohibited income generation activities) that increased their vulnerability to criminalization.67–69,204 However, these interpretations were not investigated in the present study and therefore further empirical and qualitative examination of potential mechanisms underlying the association between incarceration and ongoing SIF use is warranted. We also observed that HIV seropositivity was marginally and positively associated with discontinued use of the Insite SIF during periods of active injection. This finding could be explained in part by HIV-positive PWID being more likely to transition to instead injecting drugs at the Dr. Peter Centre, an HIV care service organization in Vancouver that operates under an integrated services model and provides services, including two booths in which people can inject drugs under nurse supervision, to an estimated 425 people living with HIV/AIDS annually.155 However, this finding could also be due to HIV-positive PWID experiencing barriers to sustained engagement with the Insite SIF.155,190 Further research is needed to better understand potential explanations for this finding.          69 There are several limitations to the present study. First, this study relied on data from a non-random sample of SIF users, and therefore the findings cannot necessarily be generalized to PWID who use these facilities in Vancouver or other settings. However, the characteristics of the study sample are similar to those of previous studies of a random sample of people who used the Insite SIF.50,69 In addition, this study is susceptible to reporting biases, including social desirability bias. However, we have no reason to believe that the magnitude of such biases would differ between those who did and did not cease use of the SIF while actively injecting. Finally, as with all observational studies, it is possible that unmeasured or uncontrolled factors may explain the associations observed herein.  In summary, the present study demonstrates that periods of SIF use cessation were common among PWID in this setting and often co-occurred with injection drug use cessation. PWID enrolled in MMT were more likely to cease using the SIF during periods of active injection, while PWID with markers of structural vulnerability and drug-related risk were less likely to do so. These findings challenge the contention that SIFs may inhibit injection cessation, and suggest that PWID at heightened risk of drug-related harms are successfully retained in this form of health programming while actively injecting.                 70 Table 3.1 Baseline characteristics of a cohort of 1336 clients of supervised injection facility (SIF) in Vancouver, Canada, stratified by reporting ceasing use of the SIF while actively injecting during follow up (2005-2016)  Characteristic Total (%) (n = 1336) Cease SIF use while actively injecting during follow up p - value Yes (%) (n = 847) No (%) (n = 489) Age* (median, IQR†; per year older)    41 (34-48)    41 (34-47)    40 (33-48) 0.348 Sex (male vs. female) 868 (65.0) 552 (65.2) 316 (64.8) 0.878 Ethnicity (white vs. others) 846 (63.6) 530 (63.0) 316 (64.8) 0.508 Downtown Eastside residence*         (yes vs. no)  1035 (77.5)  662 (78.2)  373 (76.3)  0.429 Employment* (yes vs. no) 277 (20.7) 162 (19.1) 115 (23.5) 0.057 Homeless* (yes vs. no) 566 (42.6) 339 (40.2) 227 (46.6) 0.023 HIV status* (positive vs. negative serology)  447 (33.5)  294 (34.7)  153 (31.3)  0.202 Heroin injection* (≥daily vs. <daily) 512 (38.3) 298 (35.2) 214 (43.8) 0.002 Cocaine injection* (≥daily vs. <daily) 178 (13.4) 112 (13.3) 66 (13.5) 0.907 Crystal methamphetamine injection*  (≥daily vs. <daily)  103   (7.7)  51   (6.1)  52 (10.7)  0.002 Public injection* (yes vs. no) 690 (51.8) 413 (48.8) 277 (56.9) 0.005 Binge injection* (yes vs. no) 430 (32.2) 261 (30.9) 169 (34.6) 0.167 Require help injecting* (yes vs. no) 353 (26.6) 198 (23.5) 155 (31.8) <0.001 Non-fatal overdose* (yes vs. no) 136 (10.2) 69   (8.2) 67 (13.7) 0.001 Syringe sharing* (yes vs. no) 157 (11.8) 83   (9.9) 74 (15.2) 0.004 Experience violence* (yes vs. no) 368 (27.9) 231 (27.6) 137 (28.3) 0.792 Enrolment in MMT*‡ (yes vs. no) 613 (46.2) 406 (48.3) 207 (42.6) 0.043 Difficulty accessing addiction treatment* (yes vs. no)  87   (6.6)  55   (6.6)  32   (6.6)  0.985 Sex work involvement* (yes vs. no) 253 (19.0) 163 (19.3) 90 (18.5) 0.730 Incarceration* (yes vs. no) 263 (19.8) 160 (19.0) 103 (21.2) 0.325 * Refers to activities and experiences in the previous 6 months.  † IQR = interquartile range. ‡ MMT = methadone maintenance therapy.            71 Table 3.2 Bivariable and multivariable extended Cox regression analyses of factors associated with time to cessation of use of a supervised injection facility while actively injecting among 1336 people who inject drugs in Vancouver, Canada (2005- 2016) Characteristic Unadjusted  Adjusted Hazard Ratio (95% CI)† p - value  Hazard Ratio (95% CI)† p - value Age* (per year older) 1.00 (0.99 – 1.01)       0.583    Sex (male vs. female) 0.91 (0.76 – 1.09) 0.295    Ethnicity (white vs. others) 0.92 (0.77 – 1.09) 0.334    Downtown Eastside residence* (yes vs. no) 1.00 (0.85 – 1.18) 1.000    Employment* (yes vs. no) 0.94 (0.78 – 1.14) 0.534    Homeless* (yes vs. no) 0.47 (0.40 – 0.57) <0.001  0.61 (0.51 – 0.73)  <0.001 HIV status* (positive vs. negative serology) 1.29 (1.08 – 1.54) 0.005  1.17 (0.98 – 1.40) 0.089 Heroin injection* (≥daily vs. <daily) 0.58 (0.48 – 0.69) <0.001  0.78 (0.65 – 0.94) 0.010 Cocaine injection* (≥daily vs. <daily) 0.73 (0.57 – 0.93) 0.010    Crystal methamphetamine injection* (≥daily vs. <daily) 0.93 (0.69 – 1.25) 0.621    Public injection* (yes vs. no) 0.56 (0.48 – 0.66) <0.001  0.78 (0.66 – 0.93)           0.005 Binge injection* (yes vs. no) 0.69 (0.58 – 0.81) <0.001  0.77 (0.65 – 0.91) 0.003 Require help injecting* (yes vs. no) 0.82 (0.67 – 1.00) 0.047    Non-fatal overdose* (yes vs. no) 0.61 (0.45 – 0.82) 0.001  0.74 (0.53 – 1.03) 0.074 Syringe sharing* (yes vs. no) 0.79 (0.55 – 1.12) 0.186    Experience violence* (yes vs. no) 0.79 (0.65 – 0.95) 0.015    Enrolment in MMT*‡ (yes vs. no) 1.43 (1.22 – 1.67) <0.001  1.30 (1.11 – 1.53)            0.002 Difficulty accessing addiction treatment* (yes vs. no) 0.62 (0.46 – 0.84) 0.002  0.63 (0.50 – 0.81) 0.063 Sex work involvement* (yes vs. no) 1.06 (0.85 – 1.32) 0.593    Incarceration* (yes vs. no) 0.52 (0.42 – 0.66) <0.001  0.63 (0.50 – 0.81)          <0.001 †CI = confidence interval. *Refers to activities or experiences in the previous 6 months. ‡ MMT = methadone maintenance therapy.          72 Chapter 4: The impact of supervised injection facility use on all-cause mortality among a cohort of people who inject drugs in Vancouver, Canada 4.1  Introduction PWID are known to be at heightened risk of premature mortality.9 A 2013 systematic review and meta-analysis of 67 cohort studies estimated that PWID worldwide have a crude all-cause mortality rate of 2.35 deaths per 100 person years, a rate 13.0 to 16.4 times that of the general population.9 Globally, the leading causes of death among PWID are accidental drug overdose and HIV-related disease,9 and in the United States and Canada in particular, overdose deaths have increased dramatically in recent years to become a leading cause of accidental death at the population level.10,12 As noted in Section 1.1, this rise in overdose deaths has contributed to a recent decline in overall life expectancy in the United States, and preliminary reports indicate that a similar decline has likely occurred at the national level in Canada.13,14 As well, previous studies undertaken in diverse settings have found that other underlying causes of death, including suicide, liver-related conditions, and other non-accidental causes (e.g., circulatory and respiratory infections or diseases), are also common among PWID.205–209 As described in Section 1.2, SIFs have been implemented in many municipalities in Canada and internationally as a strategy to reduce drug-related harms, including premature mortality.37 Chapter 2 found that existing scientific evidence suggests that SIFs contribute to reductions in overdose-related mortality.48,75,107,109 For example, a geospatial analysis of death records reported that the establishment of the Insite SIF in Vancouver was associated with a 35% population-level decrease in the fatal overdose          73 rate in the surrounding vicinity of the SIF, compared to a 9% decrease in the rest of the city.48 Further, an earlier simulation study estimated that between 8 and 51 overdose events that occurred at Insite during its first four years of operation would have been fatal if they had instead occurred in the surrounding community.107 Although these analyses indicate a protective role of SIFs against overdose mortality, the potential impact of SIF use on all-cause mortality is unknown. Information concerning the relationship between SIF use and mortality may be of public health importance given that evidence-based interventions to mitigate premature death among PWID are urgently needed at present, and that many settings in Canada and elsewhere are currently debating the merits of implementing SIFs as a strategy to address drug-related harms.29,35,54 The present study was therefore undertaken to examine the association between frequent SIF use and all-cause mortality among a community-recruited cohort of PWID in Vancouver, Canada between 2006 and 2017.   4.2 Methods Data for these analyses were derived from the VIDUS and ACCESS studies, as described in Section 1.7.1.  4.2.1 Study sample The present analyses were restricted to participants who completed at least one baseline or follow-up interview in which they reported past six month injection drug use between December 1, 2006 and June 30, 2017, the time period during which all variables of interest were available. As has been demonstrated in research on needle exchange use,210 existing literature indicates that SIF users are more likely to contend with structural vulnerabilities and engage in higher risk drug-related behaviours than          74 non-users, and therefore may have an inherently greater risk of death.66,68,69 Thus, analyses were further restricted to participants who reported past six month SIF use in ≥50% of available follow up visits in effort to avoid potential bias due to lack of comparability of explanatory variable groups when estimating the association between frequent SIF use and mortality.211,212 4.2.2 Variable selection The primary outcome for this analysis was all-cause mortality. As described in Section 1.7.1, this variable and specific underlying causes of death were ascertained through confidential record linkages with the BC Vital Statistics Agency using government-issued PHNs. The Vital Statistics Agency database recorded causes of death in accordance with the International Classification of Diseases, Tenth Revision (ICD-10)99 codes used in medical records. To avoid potential bias due to long durations between study visits and death,206 individuals who died more than 24 months after their last recorded follow-up visit were censored on the date of their last follow-up. Consistent with previous studies of PWID,9,206–208 causes of death were classified into the following eight categories: HIV-related, overdose, liver-related, homicide, suicide, other accidental, other non-accidental, and ill-defined/unknown causes. The primary exposure of interest was frequent SIF use. This was defined in response to the question, “In the last six months, how often have you used Insite to inject?” (December 2006 to November 2016) or “In the last six months, how often have you used SIFs to inject?” (December 2016 to June 2017), with responses classified as ≥once a week vs. <once a week.           75 To examine the independent association between frequent SIF use and time to all-cause mortality, the following were assessed as potential confounding variables on the basis of Rhodes’ Risk Environment framework22,26,39 and previous literature concerning mortality and SIF use among PWID:9,66,68,69,206,208 age (per year older), sex (male vs. female), ethnicity (white vs. non-white), HIV status (positive vs. negative serological test); HCV status (positive vs. negative serological test), and heavy alcohol use (average of >3 alcoholic drinks per occasion or >7 drinks per week in the previous 6 months for women, and an average of >4 alcoholic drinks per occasion or >14 drinks in total per week in the previous 6 months for men).213 Other potential confounders examined included Downtown Eastside residence, unstable housing, binge injection, public injection, non-fatal overdose, enrolment in any form of addiction treatment, experienced violence, incarceration, and involvement in sex work (all yes vs. no). Finally, the candidate assessed as confounders frequent use of injection heroin, injection cocaine, injection crystal methamphetamine, non-injection crack cocaine, injection or non-injection prescription opioids, and cannabis (all ≥daily vs. <daily). Unless otherwise indicated, all variables refer to activities and experiences that occurred in the six-month period preceding the date of the interview, and were treated as time-updated based on each semi-annual follow-up visit. 4.2.3 Statistical analyses First, the crude mortality rates and 95% CIs for all-cause mortality and each specific cause of death were calculated using the Poisson distribution. Second, Pearson’s Chi-squared test for categorical variables and the Mann-Whitney test for continuous variables were used to compare baseline characteristics of those who reported frequent          76 SIF use at baseline with those who did not. Third, bivariable extended Cox regression analyses were used to examine the association between each explanatory variable and time to all-cause mortality. An a priori-defined statistical protocol was applied to estimate the independent association between frequent SIF use and time to all-cause mortality. First, the candidate fit a multivariable model that included all variables significantly associated with all-cause mortality at p < 0.10 in bivariable analyses. Next, the secondary explanatory variable corresponding to the smallest relative change in the frequent SIF use coefficient was removed. The candidate continued this iterative process until the minimum change of the value of the coefficient for frequent SIF use exceeded 5%. Lastly, age, sex and unstable housing were forced into the model to account for the established associations between these variables and the primary explanatory and outcome variables of interest.66,68,69,206,214 All statistical analyses were conducted with SAS version 9.4 (SAS Institute Inc., Cary, NC). All reported p-values are two-sided. 4.3 Results Between December 2006 and June 2017, 2139 participants were recruited into the VIDUS and ACCESS cohorts. As shown in Figure 4.1, 1328 individuals were excluded from the present study because they either did not report past six month injection drug use in any interviews during follow up (n = 268) or did not report past six month SIF use in at least 50% of available interviews (n = 1060). Compared with participants in the analytic sample (n = 811), those excluded were more likely to: be older, be HIV seropositive and report heavy alcohol use (all p < 0.05). Additionally, participants excluded from the analytic sample were less likely than those included to: reside in the          77 Downtown Eastside, be unstably housed, be HCV seropositive, inject heroin ≥daily, inject cocaine ≥daily, inject methamphetamine ≥daily, use prescription opioids ≥daily, use crack cocaine ≥daily, inject in public, binge inject, have had a recent non-fatal overdose, have recently experienced violence, have recently engaged in sex work, and have been recently incarcerated (all p < 0.05). The 811 PWID included in present study were followed for a median of 72 months (IQR: 24 – 123) and collectively contributed a total of 4928.1 person-years of observation. At baseline, 278 (34.5%) participants were women and the median age was 39 years (IQR: 33 – 46). Table 4.1 reports the baseline characteristics of the study participants stratified by frequent SIF use. As shown, at baseline, persons who reported frequent SIF use were more likely than those who did not to: be younger (median age = 38 vs. 40 years), reside in the Downtown Eastside (85.1% vs. 74.7%), be unstably housed (85.8% vs. 77.6%), inject heroin ≥daily (52.8% vs. 27.9%), inject cocaine ≥daily (17.3% vs. 6.2%), inject methamphetamine ≥daily (12.8% vs. 8.0%), use prescription opioids ≥daily (14.2% vs. 5.7%), inject in public (65.5% vs. 49.7%), have had a recent non-fatal overdose (14.0% vs. 9.0%), and been recently incarcerated (32.0% vs. 17.0%). Those who reported frequent SIF use at baseline were less likely to be HIV seropositive (25.5% vs. 35.9%) and to be enrolled in addiction treatment (48.9% vs. 58.2%) at baseline (all p < 0.05).  A total of 112 participants (13.8%) died during the 10.5-year study period, corresponding to a crude mortality rate of 22.7 (95% CI: 18.7 – 27.4) deaths per 1,000 person-years. The underlying causes of death are presented in Table 4.2. The leading observed causes of death were: other non-accidental (n = 30; 26.8%), ill-defined or          78 unknown causes (n = 27; 24.1%), overdose (n = 19; 17.0%) and HIV-related causes (n = 15; 13.4%). Table 4.3 presents the crude and adjusted HRs for the association between frequent SIF use and other covariates with all-cause mortality. In bivariable extended Cox regression analyses, frequent SIF use (HR = 0.64; 95% CI = 0.43 – 0.97) was significantly and inversely associated with time to all-cause mortality. In the final multivariable Cox regression model, frequent SIF use (AHR = 0.63; 95% CI = 0.42 – 0.95) remained significantly associated with decreased risk of all-cause mortality after adjusting for age, sex, unstable housing, ≥daily heroin injection, and enrolment in addiction treatment. 4.4 Discussion In this prospective 10.5-year study of a community-recruited cohort of more than 800 PWID in Vancouver, Canada, we observed a high burden of death, with an estimated crude mortality rate of 22.7 deaths per 1,000 person-years. It was found that frequent SIF use predicted a lower risk of all-cause mortality, independent of potential confounders including socio-demographic factors, enrolment in addiction treatment and high intensity use of injection heroin. The primary causes of death were other non-accidental, unknown factors, overdose and HIV-related causes.  Existing literature indicates that SIFs avert numerous overdose deaths per year.107,109 Moreover, past research relying on aggregate data has demonstrated the role of SIFs in reducing local population-based rates of fatal overdose.48,75 However, to our knowledge, this is the first study to prospectively identify an individual-level          79 association between frequent SIF use and decreased risk of all-cause mortality among a community-recruited cohort of PWID. There are likely multiple explanations for the protective association between frequent SIF use and death observed in the present study. For instance, SIF use has been associated with positive changes in various injecting practices, including declines in syringe sharing, syringe reuse, outdoor injecting and rushed injecting, thereby reducing the risk of acquiring HIV and other viral or bacterial infections that may contribute to premature mortality.46,47,113 In addition, the provision of rapid, well-equipped emergency response in the event of overdose within SIFs (e.g., oxygen and naloxone administration), has served to prevent the occurrence of on-site overdose deaths. Indeed, no overdose deaths have ever occurred within any SIF in operation in Canada or internationally to date.27 Use of SIFs has also been found to facilitate access to addiction treatment and other health and social resources, which may help to prevent mortality related to diverse causes among PWID.2,16,50–53 However, interpretations of the underlying explanations for the statistical association between frequent SIF use and reduced risk of all-cause mortality cannot be confirmed based on the present analyses and therefore further investigation of these issues is warranted. In particular, future studies should seek to determine individual-level estimates of the impact of SIF use on specific causes of death, as well as discern any mediating factors underlying these potential associations. This is especially important given that almost a quarter of the deaths included in the present study were listed in the Vital Statistics Agency database as due to ill-defined or unknown causes, and therefore important questions remain          80 about the pathways and mechanisms explaining the observed protective relationship between SIF use and mortality among PWID in this setting.  Together with the findings of previous research,48,75,107 our findings underscore the need for continued efforts to enhance access to SIFs as a strategy to reduce mortality among PWID. In particular, given that SIFs have limited geographic coverage and that PWID have been found to often encounter long wait times in accessing SIF services in this setting, the broader expansion of SIFs may serve to improve service accessibility and thereby reduce the potential for mortality and other harms among this population.54,71,94,215 The recent scale up of SIFs in Vancouver provides an opportunity for future research to examine these issues, including potential impacts of this expansion on service utilization patterns and related health and social outcomes among PWID. As well, further efforts should be undertaken to mitigate other barriers to engagement with SIFs. For example, increasing SIF operating hours may promote more frequent use of this service, and amending SIF regulations that have been shown to constrain access to SIFs (e.g., rules prohibiting the provision of manual assistance with injections) may help to engage vulnerable and underserved populations of PWID.69,71,172  This study has a number of limitations. First, the VIDUS and ACCESS cohorts are community-recruited, non-randomized samples of PWID, and therefore the findings may not be generalizable to PWID in Vancouver or elsewhere. Moreover, the present study further restricted the cohorts to PWID who reported recent SIF use in at least half of their available study visits. While we believe that this served to enhance the internal validity of the study by reducing the potential for biased measures of association,211,212 this approach likely further reduced the generalizability of our findings. In addition,          81 this study relied on self-reported information and is therefore susceptible to reporting biases including social desirability bias. However, it is noteworthy that the primary outcome of mortality was based on objective measures derived from linkages to an external administrative database. An additional limitation is that deaths that occurred outside of BC were not recorded in the Vital Statistics Agency database, and thus the mortality rates reported herein may be underestimated. However, past research has found that migration to settings outside of BC is low among PWID in Vancouver.216 As previously noted, a further limitation is that just under a quarter of all deaths observed in the present study were listed in the Vital Statistics Agency database as being due to ill-defined or unknown causes, which complicates interpretations of the observed protective association between SIF use and mortality. The observed excess of deaths of unknown causes is likely largely explained by delays in updating causes of death in the database in recent years as a result of a backlog in post-death toxicology testing due to the present overdose crisis.217 Indeed, 42.9% of deaths of ill-defined or unknown causes observed in the present study occurred within the last three years of the study period, and we received confirmation via email correspondence with the Vital Statistics Agency that recent increases in deaths in the province had contributed to delays in updating the database (M. Spearman, email communication, April 19, 2016). As such, the true prevalence of overdose-related deaths may have been underestimated in the present study, as may have been deaths of other specific causes. However, given that the primary aim of this study was to examine the independent association between SIF use and all-cause mortality (rather than distinct causes of death), we believe that improvements in statistical power resulting from including recent deaths in our          82 analyses offset potential benefits concerning interpretations if we had instead restricted the study period to reduce the number of deaths of unknown causes. A final limitation is that the observed relationship between frequent SIF use and decreased risk of mortality might be influenced by unmeasured confounding. However, previous studies have found that frequent SIF users tend to exhibit characteristics and contend with exposures associated with greater risk of mortality,66,68,69 and thus the direction of any bias in our estimate of the association would likely be towards the null. Moreover, we sought to reduce the potential for this bias by restricting our study sample based on SIF utilization patterns and by adjusting multivariable analyses for key confounding factors.   In conclusion, this study of a cohort of PWID in Vancouver, Canada reports a previously unidentified independent association between frequent SIF use and decreased risk of all-cause mortality. This relationship warrants further investigation. In particular, future studies should seek to examine the individual-level association between SIF use and distinct causes of death among PWID. Nonetheless, the findings of the present study suggest that efforts to scale up access to SIFs may serve to reduce preventable deaths among this population.                    83 Figure 4.1 Flowchart showing how the analytical sample (n = 811) was determined, Vancouver Injection Drug Users Study (VIDUS) and AIDS Care Cohort to Evaluate Exposure to Survival Services (ACCESS), Vancouver, Canada (2006-2017)                      2139 individuals enrolled in the VIDUS and ACCESS cohorts between December 1, 2006 and June 30, 2017 1328 individuals excluded:  268 individuals did not report past 6 month injection drug use at least once during the study period  1060 individuals did not report past 6 month supervised injection facility use in ≥50% of available study visits   811 individuals included in the study          84 Table 4.1 Characteristics of 811 people who inject drugs in Vancouver, Canada, stratified by at least weekly supervised injection facility use at baseline (2006-2017) Characteristic Total (%) (n = 811) ≥Weekly SIF use*‡ p - value Yes (%) (n = 424) No (%) (n = 387) Age*      Median [IQR]† 39 [33-46] 38 [32-45]  40 [33-47] 0.004 Sex      Male 529 (65.5)    277 (65.7) 252 (65.5) 0.956 Female 278 (34.5) 145 (34.4) 133 (34.6)  Ethnicity      White 526 (65.2)  276 (65.4)  250 (64.9) 0.889  Non-white 281 (34.8) 146 (34.6) 135 (35.1)  Downtown Eastside residence*  Yes 650 (80.2) 361 (85.1) 289 (74.7) <0.001 No 161 (19.9) 63 (14.9) 98 (25.3)  Unstable housing* Yes 663 (81.8) 363 (85.8) 300 (77.6) 0.002 No 147 (18.2) 60 (14.2) 87 (22.5)  HIV seropositive* Yes 247 (30.5) 108  (25.5) 139 (35.9) 0.001 No 564 (69.5) 316  (74.5) 248 (64.1)  HCV seropositive*     Yes 688 (85.0) 369 (87.2) 319 (82.6) 0.069 No 121 (15.0) 54 (12.8) 67 (17.4)  Heroin injection*      ≥Daily  332 (40.9) 224 (52.8) 108 (27.9) <0.001 <Daily 479 (59.1) 200 (47.2) 279 (72.1)  Cocaine injection*      ≥Daily  97 (12.0) 73 (17.3) 24   (6.2) <0.001 <Daily 713 (88.0) 350 (82.7) 363 (93.8)  Crystal methamphetamine injection*      ≥Daily  85 (10.5) 54 (12.8)  31   (8.0) 0.028 <Daily 724 (89.5) 368 (87.2) 356 (92.0)  Non-injection crack cocaine use*       ≥Daily  311 (38.4) 173 (40.9) 138 (35.7) 0.126       <Daily 499 (61.6) 250 (59.1) 249 (64.3)  Prescription opioid use*      ≥Daily  82 (10.1) 60 (14.2) 22   (5.7) <0.001      <Daily 729 (89.9) 364 (86.0) 365 (94.3)           85 Cannabis use*       ≥Daily  176 (21.8) 86 (20.3) 90 (23.4) 0.287 <Daily 633 (78.2) 338 (79.7) 295 (76.6)  Heavy alcohol use*^            Yes 97 (12.0) 51 (12.0) 46 (11.9) 0.950        No 714 (88.0) 373 (88.0) 341 (88.1)  Public injection*     Yes 469 (58.0) 277 (65.5) 192 (49.7) <0.001 No 340 (42.0) 146 (34.5) 194 (50.3)  Binge injection*     Yes 259 (32.0) 137 (32.3) 122 (31.7) 0.850 No 550 (68.0) 287 (68.0) 263 (68.3)  Non-fatal overdose*            Yes 94 (11.6) 589 (14.0) 35   (9.0) 0.031        No 716 (88.4) 364 (86.1) 352 (90.1)  Enrolled in addiction treatment*            Yes 429 (53.4) 206 (48.9) 223 (58.2) 0.006        No 375 (46.6) 215 (51.1) 160 (41.8)  Exposure to violence*            Yes 238 (29.5) 136 (32.4) 102 (26.4) 0.064        No 568 (70.5) 284 (67.6) 284 (73.6)  Sex work involvement*     Yes 149 (18.4) 77 (18.3) 72 (18.7) 0.882 No 659 (81.6) 345 (81.8) 314 (81.4)  Incarceration*     Yes 201 (24.9) 135 (32.0) 66 (17.0) <0.001 No 608 (75.2) 287 (68.0) 321 (83.0)  Note: Column percentages may not necessarily sum to 100% due to missing data or rounding error.  ‡ SIF = supervised injection facility. *Refers to the 6 months prior to a baseline interview. †IQR = interquartile range. ^Average of >3 alcoholic drinks on at least 1 day per week or >7 drinks in total per week (women), or >4 alcoholic drinks on at least 1 day per week or >14 drinks in total per week (men).               86  Table 4.2 Causes of death in a study of 811 people who inject drugs in Vancouver, Canada (2006-2017) Cause of death  n = 811  n % Rate* 95% CI‡ All causes 112 100.0 22.7 (18.7 – 27.4) HIV-related 15  13.4 3.0 (1.7 – 5.0) Overdose 19 17.0 3.9 (2.3 – 6.0) Liver-related 11 9.8 2.2 (1.1 – 4.0) Suicide 3 2.7 0.6 (0.1 – 1.8) Homicide 2 1.8 0.4 (0.1 – 1.5) Other accidental 5 4.5 1.0 (0.3 – 2.4) Substance-related 4 3.6   Other causes 1 0.9   Other non-accidental 30 26.8 6.1 (4.1 – 8.7) Neoplasms 10 8.9   Circulatory disease 8 7.1   Respiratory disease 6 5.4   Other causes 6 5.4   Ill-defined or unknown 27 24.1 5.5 (3.6 – 8.0) *Per 1,000 person-years. ‡ CI = confidence interval.              87 Table 4.3 Unadjusted and adjusted Cox regression analyses of factors associated with time to all-cause mortality among 811 people who inject drugs in Vancouver, Canada (2006- 2017) Characteristic Unadjusted  Adjusted Hazard Ratio (95% CI)‡ p - value  Hazard Ratio (95% CI)‡ p - value Age*          (per year older) 1.03 (1.01 – 1.06) 0.008  1.03 (1.00 – 1.05) 0.050 Sex           (male vs. female) 1.23 (0.82 – 1.86) 0.323  1.06 (0.69 – 1.63) 0.795 Ethnicity           (white vs. non-white) 0.81 (0.55 – 1.18) 0.269    Downtown Eastside residence*           (yes vs. no) 0.85 (0.57 – 1.26) 0.406    Unstable housing*           (yes vs. no) 1.45 (0.89 – 2.37) 0.133  1.57 (0.96 – 2.54) 0.070 HIV seropositive*           (yes vs. no) 3.23 (2.20 – 4.73) <0.001    HCV seropositive*           (yes vs. no) 1.33 (0.58 – 3.04) 0.501    ≥Weekly supervised injection facility use*           (yes vs. no) 0.64 (0.43 – 0.97) 0.035  0.63 (0.42 – 0.95) 0.026 ≥Daily heroin injection*           (yes vs. no) 0.69 (0.45 – 1.05) 0.081  0.77 (0.49 – 1.20) 0.249 ≥Daily cocaine injection*           (yes vs. no) 1.14 (0.65 – 2.02) 0.648    ≥Daily crystal methamphetamine injection*           (yes vs. no) 0.60 (0.26 – 1.39) 0.234             88 ≥Daily non-injection crack cocaine use*           (yes vs. no) 0.88 (0.57 – 1.36) 0.564    ≥Daily prescription opioid use*           (yes vs. no) 0.90 (0.45 – 1.78) 0.758    ≥Daily cannabis use*           (yes vs. no) 0.92 (0.55 – 1.54) 0.759    Heavy alcohol use*†           (yes vs. no) 0.87 (0.47 – 1.59) 0.643    Public injection*           (yes vs. no) 0.87 (0.60 – 1.28) 0.491    Binge injection*           (yes vs. no) 0.96 (0.65 – 1.43) 0.847    Non-fatal overdose*           (yes vs. no) 0.79 (0.40 – 1.56) 0.496    Enrolled in addiction treatment*           (yes vs. no) 0.58 (0.40 – 0.85) 0.005  0.56 (0.38 – 0.82) 0.003 Exposure to violence*           (yes vs. no) 0.75 (0.44 – 1.29) 0.294    Sex work involvement*           (yes vs. no) 0.93 (0.51 – 1.70) 0.801    Incarceration*           (yes vs. no) 0.64 (0.34 – 1.21) 0.168    ‡ CI = confidence interval. * Refers to previous six months and treated as time-updated.  † Average of >3 alcoholic drinks on at least 1 day per week or >7 drinks in total per week (women), or >4 alcoholic drinks on at least 1 day per week or >14 drinks in total per week (men).           89 Chapter 5: Supervised injection facility use and exposure to violence among a cohort of people who inject drugs in Vancouver, Canada: A gender-based analysis 5.1 Introduction Exposure to physical and sexual violence among drug-using populations remains a public health problem in diverse settings worldwide.6,218–222 Studies have documented particularly heightened rates of exposure to violence among marginalized populations of drug-using men and women, including PWID.6,218–222 For example, a community-based study of women who use drugs in San Francisco found that 26% had experienced partner violence and 28% had experienced non-partner violence in the previous six months.219 Similarly, another study found that 22% of a community-recruited cohort of PWID in Vancouver had experienced physical or sexual violence in the previous six months.76 In addition to physical injury, such violent encounters may have further adverse consequences for the health and well-being of PWID.223–225 Indeed, exposure to violence has been associated with an increased likelihood of mental health problems, including post-traumatic stress-disorder, mood and anxiety disorders, and suicidal ideation.223,224 Moreover, PWID who have experienced violence have been found to be more likely to engage in drug use practices associated with elevated risk of overdose and infectious disease transmission,225 and are also known to avoid healthcare services near locations where they have experienced violence.175 Over the past decade, there has been growing research interest in the role of contextual determinants of exposure to violence among PWID.6,77,169,175,219,226,227 Accompanying this shift has been the increasing application of Rhodes’ Risk          90 Environment Framework as a heuristic for guiding such investigations.22,41 As described in Section 1.5, Rhodes’ Risk Environment Framework has primarily been applied in studies of HIV risk.22,41 However, this framework has been extended to conceptualize how experiences of violence at the individual level among PWID are shaped by the interplay between various social, structural and environmental forces operating at the macro-, meso- and micro-levels of environmental influence.6,77,169,175,219,226 For instance, studies have illustrated how street-based drug scenes (i.e., inner-city areas characterized by high concentrations of PWUD and drug market activity), are key risk environments that contribute to the production of violence among PWID,42,169,175 with many violent encounters related to the purchasing of drugs and active involvement in informal income generating activities, including drug dealing, sex work and other street-based economic activities, within these environments.76,181,228,229 Further, the preparation and injection of drugs in particular public spaces within street-based drug scenes (e.g., alleyways, public washrooms) has been found to increase susceptibility to violence from ‘street predators’ and police.23,42,175 Such threats of violence are exacerbated among people who require manual assistance with injecting, due in part to regulations prohibiting the provision of assisted injections within health services in many settings, which may drive these individuals to inject in particularly unsafe social and physical contexts.23,92 In addition, studies have demonstrated how social-structural exposures that disproportionately affect urban drug-using populations, including housing instability and street-based law enforcement, contribute to violence among PWID.6,77,226,230,231          91 Rhodes’ Risk Environment Framework is also useful for conceptualizing how gendered structures and social relations may shape susceptibility to violence and exploitation within street-based drug scenes.77,169,175,219,232,233 For example, qualitative and ethnographic studies have highlighted how women are often rendered vulnerable to violence as a result of their subordination within street-based drug economy hierarchies.175,234 Gendered power relations operating within street-based drug scenes also contribute to the frequent “grinding” (i.e., attempts to obtain drugs and money through begging or coercion) of women by men due in large part to women’s perceived ability to generate resources through sex work.92,169 Further, violence against women is common in injection drug-using partnerships, with physical and emotional abuse often ensuing in conflicts over control of income and resources generated by women.169,175,235 While less research attention has focused on exposure to violence among men, quantitative studies of PWID have demonstrated that correlates of received violence, as well as types and perpetrators of assault, may differ between men and women.220,226 For example, men have been found to experience higher levels of assault from police officers and strangers than women.220,226 Increased recognition of the role of contextual and gendered forces in shaping vulnerability to violence among PWID has drawn attention to the need for interventions that seek to address social, structural and environmental determinants of risk.23,25,43 In providing safer alternative environments to the street-based drug scene for the consumption of drugs, SIFs could potentially reduce exposure to street-based drug scene violence among PWID. However, as noted in Chapter 2, the potential relationship between SIF use and violence has not been thoroughly evaluated.           92 Several qualitative and ethnographic studies have provided evidence indicating that SIFs may offer protection from some forms of violence among certain subpopulations of PWID, including women and people who require help injecting.77,91,92,169 For example, an evaluation of an unsanctioned SIF in Vancouver that offered manual assistance with injections found that the provision of this service allowed individuals who required help injecting to escape the violence and exploitation associated with assisted injection within the street-based drug scene.92 Similarly, studies of sanctioned SIFs in this setting have illustrated how these facilities may serve to temporarily protect women PWID from violence during drug preparation and consumption, and might also help to mitigate grinding and inter-partner violence by enabling women to exercise greater control over resources and the injecting process.77,91,169 However, recent ethnographic work has also described how, despite operating as gender-neutral programmes, these facilities remain ‘male-dominated environments’ in which women PWID are routinely subjected to harassment from men, which may deter women’s access.77 As such, questions remain about the potential role of SIFs in providing protection from violence among both women and men PWID. We therefore undertook the present study to longitudinally examine the gender-specific relationship between exclusively injecting at the Insite SIF and exposure to physical or sexual violence among a community-recruited cohort of PWID in Vancouver, Canada. This study may provide useful information to guide the development and implementation of evidence-based violence prevention strategies that are aligned with the specific needs of men and women PWID.          93 5.2 Methods Data for these analyses were derived from the VIDUS and ACCESS studies, as described in Section 1.7.1.  5.2.1 Study sample The present analyses were restricted to participants who completed at least one study interview between December 1, 2005 and November 30, 2016, the time period during which all variables of interest were available. We further restricted the inclusion of data from participants to observations in which participants reported injecting drugs in the six months prior to each interview, as only active injectors were eligible to use the SIF. 5.2.2 Variable selection The primary outcome for this analysis was response to the question: “Have you been attacked, assaulted (including sexual assault) or suffered any kind of violence in the last six months?” (December 2005 to June 2014) (yes vs. no). Post June 2014, this was defined in response to the questions “In the last 6 months, have you been physically attacked or suffered any kind of physical violence, including torture or punishment related to a drug debt?” and “In the last 6 months, have you been forced to have sex or perform a sexual act against your will, or experienced any kind of sexual assault?” (yes to either question vs. no to both questions). The primary explanatory variable of interest was exclusive SIF use, defined in response to the question, “In the last six months, what proportion of injections did you do at Insite?” (all vs. most, some, a few or none). Given that SIFs have been found to engage PWID who are more likely to inject drugs in public and other potentially unsafe settings,66,68,69 this measure of SIF use was employed in          94 effort to rule out the possibility that estimates of the association between SIF use and violence may be biased due to SIF users being more likely to experience violence while consuming drugs in such settings. This variable was measured longitudinally at each follow-up and included in the analysis as a time-updated measure. To estimate the independent association between exclusive SIF use and exposure to violence, a range of socio-demographic, social-structural and behavioural variables were assessed as potential confounders. The selection of these variables was informed by Rhodes’ Risk Environment framework,22,41 and previous studies investigating SIF use and/or violence among drug-using populations.76,77,92,169,219,226,227,236,237 These variables included: age (per year older); ethnicity (white vs. non-white); relationship status (legally married/common law/regular partner vs. other); injection heroin use (≥daily vs. <daily); injection cocaine use (≥daily vs. <daily); injection crystal methamphetamine use (≥daily vs. <daily); non-injection crack cocaine use (≥daily vs. <daily); childhood emotional abuse (moderate/severe vs. low/none); and calendar year of interview (per year increase). Other variables considered included: Downtown Eastside residence; employment (regular job/temporary job/self-employed); homelessness; HIV seropositivity; binge injection; heavy alcohol use; enrolment in MMT; sex work involvement; drug dealing; and incarceration (all yes vs. no); As with previous studies investigating violence among PWID,6,238 childhood emotional abuse was defined as reporting a score of ≥13 (moderate/severe) vs. <13 (low/none) on the emotional abuse subscale of the Childhood Trauma Questionnaire.239 All variables were treated as time-updated and refer to the six-month period preceding the date of interview unless otherwise indicated.           95 5.2.3 Statistical analyses As a first step, the candidate examined characteristics associated with recent exposure to violence at baseline, stratified by self-reported current gender identity (women (trans inclusive) vs. men (trans inclusive)). Participants with non-binary gender identities (n = 4; 0.2%) were included in the ‘women’ category given that we had insufficient statistical power to conduct meaningful analyses among this specific subgroup. While we recognize the limitations of such classification,240–243 we believe that this was the most appropriate approach for the present study given that a women-only SIF (trans-inclusive) that also accommodates individuals with non-binary gender identities recently opened in this setting,77 and thus our findings may have implications for the future development of such programming. Continuous variables were compared using the Wilcoxon Rank-Sum test. Categorical variables were compared using Pearson’s Chi-squared test or Fisher’s exact test when expected cell counts were less than or equal to five. Next, GEE for binary outcomes with a logit link function and exchangeable working correlation structure was used to estimate unadjusted odds ratios (OR) for the association between experiencing violence and each explanatory variable, stratified by gender. GEE for the analysis of correlated data was used given that the factors potentially associated with experiencing violence during follow-up were time-dependent measures. GEE analyses allowed for consideration of factors associated with experiencing violence over the full length of the study period, with standard errors calculated using an exchangeable correlation structure adjusted for multiple observations for each individual.244 This method has been used successfully in previous prospective cohort studies of violence among PWID.6,76,226,238          96  To estimate the gender-specific independent association between exclusive SIF use and exposure to violence, the candidate constructed two multivariable models using an a-priori variable selection process described previously.245 In this process, a full model was fit, for women and men, respectively, including all explanatory variables that were significant at the level of p < 0.10 in bivariable GEE analyses, noting the value of the coefficient associated with SIF use in each model. In a stepwise manner, the candidate fit a series of reduced models that excluded each of the secondary explanatory variables. The secondary explanatory variable corresponding to the smallest relative change in the SIF use coefficient was then removed from further consideration. The candidate continued this iterative process until the minimum change of the value of the SIF use coefficient from the full model exceeded 5%. Remaining variables were considered confounders in the multivariable model. This modelling approach has previously been used to estimate the independent relationship between a primary explanatory variable and violence among PWID.6,238  The candidate also recognized that defining participants with non-binary gender identities as women in this analysis could potentially bias the women-specific estimate of the association between SIF use and violence. Thus, as a sensitivity analysis, the above-noted modelling building procedure for women was replicated but with individuals with non-binary gender identities excluded from this group. Finally, as a subanalysis, the candidate descriptively assessed the characteristics of violent incidents experienced by participants, including perpetrators and types of violence, stratified by gender. Specifically, responses were analyzed for the following questions that were asked of participants who reported experiencing physical or sexual          97 violence in the previous six months: (1) “Who has attacked you?” and (2) “What type of attack was it?” Response options for the former question included: stranger; dealer; police; husband/wife; boyfriend/girlfriend; partner; sex work client; sex worker; friend; regular sex partner; casual sex partner; security guard; acquaintance; don’t know; and, other (specify). Response options for the latter question included: beating; sexual assault/rape; attacked with weapons; strangled; attacked or threatened with a gun; robbery; and, other (specify). Participants could provide more than one response for each of these questions, and response options were not read aloud to participants. These questions were only included in the study questionnaire from December 2005 to December 2014, and thus analyses of participant responses were restricted to this time period. The characteristics of violent encounters between women and men were compared using Pearson’s Chi-squared test or Fisher’s exact test when expected cell counts were less than or equal to five. All statistical analyses were conducted with SAS version 9.4 (SAS Institute Inc., Cary, DC, USA) and all p-values are two-sided. 5.3 Results Between December 2005 to December 2016, 1930 participants completed at least one study interview in which they reported recently injecting drugs and were therefore included in the present study. At baseline, the median age of the study sample was 41.2 years (IQR = 34.1 – 47.7), 679 (35.2%) were women, 1242 (64.5%) were white, and 491 (25.4%) reported experiencing violence in the previous six months. Participants contributed a total of 9290.5 person-years of follow-up, with a median follow-up duration of 44.1 months (IQR: 11.7 – 106.3) per participant among women and 52.2 months (IQR: 11.8 – 105.4) per participant among men. Of the total 14351 observations          98 included in the study, 2451 (17.1%) included at least one reported incident of exposure to violence in the previous six-month period, including 782 (16.2%) among women and 1669 (17.8%) among men. In total, 353 (52.0%) women and 694 (55.5%) men reported at least one six-month period in which they experienced violence during the study period. Among participants who reported experiencing at least one incident of violence during follow up, the median number of six-month periods in which they experienced violence was 1 (IQR = 1 – 3) for women and 2 (IQR = 1 – 3) for men. A total of 104 (15.3%) women and 197 (15.8%) men reported at least one six-month period of exclusive SIF use. Table 5.1 and 5.2 present the baseline characteristics of the study sample, stratified by recent exposure to violence, among women and men, respectively. Table 5.3 presents the results of the crude and adjusted longitudinal estimates of the odds of experiencing violence, stratified by gender. As shown, in unadjusted analyses, exclusive SIF use was associated with decreased odds of experiencing violence (OR = 0.68; 95% 95% CI: 0.51 – 0.92, p = 0.014) among men. Exclusive SIF use was not significantly associated with experiencing violence (OR = 0.82; 95% CI: 0.53 – 1.29, p = 0.398) among women in unadjusted analyses. In the final multivariable models, exclusive SIF use remained independently associated with reduced odds of experiencing violence among men (adjusted odds ratio [AOR] = 0.64; 95% CI: 0.46 – 0.89, p = 0.009), after adjustment for age, Downtown Eastside residency, homelessness, heavy alcohol use, drug dealing, incarceration and calendar year of interview. Among women, exclusive SIF use was not significantly associated with experiencing violence (AOR = 0.97; 95% CI: 0.57 – 1.65, p = 0.918), after adjustment for age, Downtown Eastside residency, homelessness, HIV seropositivity,          99 ≥daily heroin injection, ≥daily crystal methamphetamine injection, ≥daily crack cocaine use, enrolment in MMT, requiring manual assistance with injecting, heavy alcohol use, sex work involvement, drug dealing, incarceration, childhood emotional abuse and calendar year of interview. The multivariable model fit for women that excluded individuals with non-binary gender identities produced nearly identical results: exclusive SIF use was not significantly associated with exposure to violence after adjusting for the same set of variables as the primary model (AOR = 0.98; 95% CI: 0.58 – 1.66, p = 0.937; data not shown). Table 4 presents the characteristics of the 2039 incidents of violence experienced by participants between 2005 to 2014, stratified by gender. As shown, strangers and acquaintances were the most commonly reported perpetrators of violence among both men and women. Women were significantly more likely than men to report that perpetrators were current or former partners (14.9% vs. 4.2% of violent incidents), sex partners (1.9% vs. 0.1%), sex work clients (6.1% vs. 0.1%), sex workers (2.3% vs. 0.1%), and family members (1.8% vs. 0.6%). Men were more likely than women to report that perpetrators of violence were strangers (44.2% vs. 33.4%), police officers (14.1% vs. 5.5%), and drug dealers (10.1% vs. 5.2%) (all p < 0.05). The most commonly reported types of violence experienced by both men and women were beatings and assaults with weapons. Women were more likely than men to report having been sexually assaulted (8.7% vs. 0.7%), strangled (1.6% vs. 0.5%) and experienced other forms of violence (7.6% vs. 4.2%), while men were more likely than women to report having been assaulted with weapons (27.5% vs. 17.5%) (all p < 0.05).          100 5.4 Discussion The present study found that exposure to violence was common among a community-recruited cohort of 1930 PWID in Vancouver, Canada, with 52% of women and 56% of men reporting experiencing at least one incident of physical or sexual violence over a median follow-up duration of four years. In longitudinal multivariable analyses that adjusted for a range of socio-demographic characteristics, drug use practices, and social-structural exposures, exclusive use of the Insite SIF for injections was independently associated with decreased odds of experiencing violence among men. However, we did not observe a significant association between exclusive SIF use and exposure to violence among women in bivariable or multivariable analyses. Strangers and acquaintances were the most common perpetrators of violence among both women and men. However, women were more likely than men to report that perpetrators were current or former partners, sex partners, sex work clients, sex workers and family members, while men were more likely than women to report that perpetrators were strangers, police officers and drug dealers. Violent incidents among both men and women most often involved beatings and attacks with weapons, although men were more likely than women to report having been attacked with weapons, while women were more likely to report having been sexually assaulted, strangled, or experienced other forms of violence.  The present study is the first, to our knowledge, to quantitatively examine the association between SIF use and exposure to violence among a community-recruited prospective cohort of PWID. Our finding that exclusive SIF use was protective against exposure to violence among men but not women is perhaps understandable given that          101 men were more likely than women to experience violence from strangers, drug dealers and police officers in this study. As predatory and police violence often occur when PWID are preparing and consuming drugs in public settings within the local street-based drug scene,42,77,92,169 exclusive use of an alternative, off-street, monitored environment for the consumption of drugs likely provided men with protection from such violent encounters. In light of the recent expansion of SIFs in Vancouver and elsewhere in Canada,29,54,88,89 these findings are encouraging in highlighting how such facilities may reshape the social and environmental contexts of injection drug use to mitigate the risk of violence among men PWID. As such, these findings build on past studies demonstrating the various health and social benefits of SIFs25,27 and provide further evidence to support the inclusion of these facilities within the continuum of services for this population.  Although past qualitative research undertaken in this setting has illustrated how SIFs may offer protection from violence against women around the time of injection,77,92,169 this study found that exclusive use of the Insite SIF for injections was not significantly associated with reduced exposure to violence overall among women. While the underlying explanations for this finding cannot be determined based on the analyses presented herein, this finding is likely largely explained by gender dynamics of violence among PWID. Specifically, violence associated with the consumption of drugs may account for a relatively smaller proportion of violent encounters among women compared to men given the pervasiveness of violence in other contexts of their lives.77,226 Indeed, in the present study, women were more likely than men to experience violence from sex workers and clients, and sex work involvement was associated with          102 increased exposure to violence among women, as has been well documented in previous studies.181,219,226 Further, perpetrators of violence against women in this study were also more likely to be current or former partners, sex partners, and family members, and past research has demonstrated that violent encounters with these individuals often occur outside of injection-related contexts.169,175,225,226 As SIFs specifically aim to offer protection against violence during drug preparation and consumption, such forms of violence are likely beyond the scope of violence that SIFs would be expected to address.   Our finding of a lack of a protective association between exclusive SIF use and exposure to violence among women might also be partially explained by the fact that men sometimes harass women within local mixed-gender SIFs.77 Although SIF staff have been found to intervene to stop such behaviours from escalating to violence,77 it is possible that violent encounters could have subsequently ensued once women had exited the SIF.169 As well, women have been found to inject drugs within and leave SIFs more quickly than men,77 thus limiting the time during which they may have been provided with greater protection from violence. Further, the Insite SIF is located in a geographical area in the Downtown Eastside characterized by particularly high levels of violence.77 Indeed, previous studies have demonstrated that women often avoid injecting at Insite due to the threat of violence in the surrounding neighbourhood of the facility.175,181 Thus, women who inject exclusively at Insite might have an elevated risk of violence due to their greater exposure time within this specific area of the local drug scene, thereby offsetting potential safety benefits stemming from injecting within the SIF.          103 These findings underscore the need for social-structural interventions and supports that are more responsive to the specific needs of women PWID in relation to violence prevention. First, given that women have been found to experience barriers in accessing local mixed-gender SIFs as well as harassment from men within these facilities,77,175 the implementation of women-only SIFs may help to foster service engagement and better support the health and safety of women PWID.30,77 The integration of anti-violence programming within these services may further extend the impact of such programming in mitigating violence, including violence that occurs outside of injection-related contexts. The recent establishment of a women-only SIF in Vancouver, which also offers violence prevention resources and counselling, provides an opportunity to further investigate these issues.77 Future research should also seek to examine if the recent scale up of SIFs in the city has helped to extend service coverage and reduce instances of violence and other harms among local PWID, including among women who avoid the area surrounding Insite due to the potential for violence. As well, the findings of this study corroborate with past research indicating that requiring help with injecting increases the risk of exposure to violence,92,226,246 and thus reinforce the need for policy reforms that allow SIFs to accommodate assisted injection in effort to enhance service access and reduce violence and other harms among this highly-vulnerable subpopulation.77,92 Until recently, federal regulations had prohibited assisted injection within all federally-sanctioned SIFs in Canada.77,92 However, beginning in July 2018, Health Canada has approved several sites to allow assisted injecting on a trial basis,247 and thus the continued evaluation of such practices may provide critical evidence to inform the potential scale up of assisted injection within SIFs.77           104 Finally, as SIFs alone are unlikely to provide sufficient protection from violence against women, interventions that seek to alter social-structural determinants of risk in the broader risk environment of women PWID are needed to better protect their safety and well-being.25,219,226 In particular, given that homelessness and sex work involvement were associated with an increased likelihood of experiencing violence among women in the present study and previous studies,175,181,219,226 efforts to increase access to social-structural supports, including social housing and regulated safer sex work environments, should be included as part of broader strategies to minimize exposure to violence among women PWID in this setting.25,226 Additionally, these findings support calls for legal reforms, including decriminalization of sex work, to enhance occupational health and safety among PWID involved in sex work.248  This study has several limitations. First, the findings are based on a non-random sample of PWID, and thus may not be generalizable to PWID in Vancouver or elsewhere. Second, the study design limited our ability to determine a temporal relationship between explanatory variables and the outcome. As such, we cannot exclude the possibility that the association between SIF use and reduced exposure to violence observed among men PWID in the present study may be due to those who have experienced violence being subsequently less likely to use the SIF. For example, given that Insite is located in an area characterized by high levels of violence,77 men PWID who have experienced violence near the facility could potentially avoid the area of the SIF after experiencing such violent encounters. Third, our reliance on self-reported data may have resulted in reporting biases, including socially desirable reporting. Fourth, although a range of socio-demographic, behavioural, and social-         105 structural variables were assessed as potential confounders, our findings could be affected by residual confounding. A final limitation is that we lacked sufficient statistical power to examine violent encounters among PWID with nonconforming gender identities. Further studies should be undertaken to determine the potential role of SIFs in shaping exposure to violence among these individuals and how such services could be adapted to better address their needs. In summary, this study found that more than half of women and men in a cohort of PWID in Vancouver, Canada experienced violence over a median of four years of follow up. Exclusive SIF use was independently associated with decreased odds of exposure to violence among men, but was not significantly associated with this outcome among women. These findings suggest that the recent scale up of SIFs in this setting may afford opportunities to reduce exposure to violence among men. However, further efforts are needed to tailor SIF services to the needs of women and to increase access to broader social-structural interventions in order to more comprehensively protect the health and safety of women PWID in this setting.                           106 Table 5.1 Baseline characteristics of 697 women who inject drugs in Vancouver, Canada, stratified by recent exposure to violence (2005-2016) Characteristic Yes n (%) n = 162 No n (%) n = 501 Odds Ratio (95% CI)‡ p - value Age (per year older)*     Median (IQR)†   38 (31–44)   38 (31–45) 0.99 (0.97 – 1.01) 0.301 Ethnicity     White 90 (55.9) 242 (48.4) 1.35 (0.95 – 1.92) 0.098 Non-white 71 (44.1) 258 (51.6)   Downtown Eastside residence*     Yes 135 (83.3) 368 (73.5) 1.81 (1.14 – 2.86) 0.011 No 27 (16.7) 133 (26.6)   Employment*     Yes 25 (15.4) 62 (12.4) 1.29 (0.78 – 2.14) 0.317 No 137 (84.6) 439 (87.6)   In a relationship*     Yes 61 (37.9) 181 (36.6) 1.06 (0.73 – 1.53) 0.763 No 100 (62.1) 314 (63.4)   Homeless*     Yes 78 (48.2) 166 (33.3) 1.86 (1.30 – 2.66) <0.001 No 84 (51.9) 332 (66.7)   HIV seropositive*     Yes 45 (27.8) 210 (41.9) 0.53 (0.36 – 0.79) 0.001 No 117 (72.2) 291 (58.1)   Heroin injection*     ≥Daily 77 (47.5) 177 (35.3) 1.66 (1.16 – 2.37) 0.006 <Daily 85 (52.5) 324 (64.7)   Cocaine injection*     ≥Daily 17 (10.5) 59 (11.8) 0.88 (0.50 – 1.56) 0.656 <Daily 145 (89.5) 442 (88.2)   Crystal methamphetamine injection*     ≥Daily 21 (13.0) 25   (5.0) 2.86 (1.55 – 5.26) <0.001 <Daily 140 (87.0) 476 (95.0)   Crack cocaine use*     ≥Daily 85 (52.5) 231 (46.1) 1.29 (0.91 – 1.84) 0.159 <Daily 77 (47.5) 270 (53.9)   Binge injecting*     Yes 56 (34.8) 139 (28.0) 1.37 (0.94 – 2.01) 0.101 No 105 (65.2) 358 (72.0)            107 Require help injecting*     Yes 76 (46.9) 154 (31.1) 1.96 (1.37 – 2.82) <0.001 No 86 (53.1) 342 (69.0)   Exclusive supervised injection facility use*     Yes 4   (2.5)  16   (3.2) 0.77 (0.25 – 2.34) 0.795 No 157 (97.5) 484 (96.8)   Heavy alcohol use*     Yes 29 (18.1) 76 (15.2) 1.24 (0.77 – 1.98) 0.374 No 131 (81.9) 425 (84.8)   Enrolment in MMT*^     Yes 79 (48.8) 262 (52.3) 0.87 (0.61 – 1.24) 0.435 No 83 (51.2) 239 (47.7)   Sex work involvement*     Yes 82 (51.6) 199 (40.0) 1.60 (1.12 – 2.29) 0.010 No 77 (48.4) 299 (60.0)   Drug dealing*     Yes 68 (42.2) 151 (30.2) 1.69 (1.17 – 2.44) 0.005 No 93 (58.0) 349 (69.8)   Incarceration*     Yes 25 (15.4) 71 (14.2) 1.10 (0.67 – 1.81) 0.699 No 137 (84.6) 429 (85.8)   Childhood emotional abuse     Yes 96 (68.1) 244 (55.5) 1.71 (1.15 – 2.56) 0.009 No 45 (31.9) 196 (44.6)   ‡ = confidence interval. * Refers to activities or experiences in the 6-month period prior to a baseline interview. † IQR = interquartile range. ^ MMT = Methadone maintenance therapy. Note: not all cells add up to n = 679 due to missing values.                         108  Table 5.2 Baseline characteristics of 1251 men who inject drugs in Vancouver, Canada, stratified by recent exposure to violence (2005-2016) Characteristic Yes n (%) n = 323 No n (%) n = 921 Odds Ratio (95% CI)‡ p - value Age (per year older)*     Median (IQR)†   42 (33–47)   44 (36–50) 0.97 (0.95 – 0.98) <0.001 Ethnicity     White 238 (73.7) 662 (72.0) 1.09 (0.82 – 1.45) 0.568 Non-white 85 (26.3) 257 (28.0)   Downtown Eastside residence*     Yes 250 (77.4) 619 (67.2) 1.67 (1.24 – 2.24) <0.001 No 73 (22.6) 302 (32.8)   Employment*     Yes 77 (23.9) 262 (28.5) 0.79 (0.59 – 1.06) 0.116 No 245 (76.1) 659 (71.6)   In a relationship*     Yes 91 (28.4) 205 (22.5) 1.37 (1.03 – 1.83) 0.032 No 229 (71.6) 707 (77.5)   Homeless*     Yes 154 (47.8) 327 (35.6) 1.66 (1.28 – 2.15) <0.001 No 168 (52.2) 592 (64.4)   HIV seropositive*     Yes 111 (34.4) 374 (40.6) 0.77 (0.59 – 1.00) 0.048 No 212 (65.5) 547 (59.4)   Heroin injection*     ≥Daily 97 (30.0) 233 (25.3) 1.27 (0.96 – 1.68) 0.098 <Daily 226 (70.0) 688 (74.7)   Cocaine injection*     ≥Daily 22   (6.8) 99 (10.8) 0.61 (0.38 – 0.99) 0.043 <Daily 300 (93.2) 822 (89.3)   Crystal methamphetamine injection*     ≥Daily 26   (8.1) 51   (5.5) 1.49 (0.91 – 2.44) 0.110 <Daily 297 (92.0) 869 (94.5)   Crack cocaine use*     ≥Daily 123 (38.2) 286 (31.1) 1.37 (1.05 – 1.78) 0.020 <Daily 199 (61.8) 633 (68.9)   Binge injecting*     Yes 97 (30.1) 241 (26.4) 1.20 (0.91 – 1.59) 0.197 No 225 (69.9) 672 (73.6)            109 Require help injecting*     Yes 102 (31.7) 211 (23.2) 1.54 (1.16 – 2.04) 0.003 No 220 (68.3) 700 (76.8)   Exclusive supervised injection facility use*     Yes 11   (3.4) 52   (5.7) 0.58 (0.30 – 1.13) 0.111 No 312 (96.6) 860 (94.3)   Heavy alcohol use*     Yes 47 (14.6) 85   (9.2) 1.67 (1.14 – 2.45) 0.008 No 276 (85.5) 835 (90.8)   Enrolment in MMT*^     Yes 127 (39.6) 368 (40.0) 0.98 (0.76 – 1.28) 0.902 No 194 (60.4) 553 (60.0)   Sex work involvement*     Yes 16   (5.0) 37   (4.0) 1.24 (0.68 – 2.27) 0.479 No 307 (95.0) 882 (96.0)   Drug dealing*     Yes 166 (51.4) 288 (31.3) 2.32 (1.79 – 3.01) <0.001 No 157 (48.6) 632 (68.7)   Incarceration*     Yes 85 (26.3) 150 (16.3) 1.83 (1.35 – 2.48) <0.001 No 238 (73.7) 768 (83.7)   Childhood emotional abuse     Yes 154 (51.9) 358 (42.1) 1.48 (1.14 – 1.93) 0.004 No 143 (48.2) 492 (57.9)   ‡ = confidence interval. * Refers to activities or experiences in the 6-month period prior to a baseline interview. † IQR = interquartile range. ^ MMT = Methadone maintenance therapy. Note: not all cells add up to n = 1251 due to missing values.               110 Table 5.3 Bivariable and multivariable generalized estimating equation analyses of factors associated with experiencing violence, stratified by gender, among 1930 people who inject drugs in Vancouver, Canada (2005-2016) Characteristic Women (n = 679)  Men (n = 1251) Unadjusted OR†  (95% CI)‡ Adjusted OR†  (95% CI)‡  Unadjusted OR†  (95% CI)‡ Adjusted OR† (95% CI)‡ Exclusive supervised injection     facility use* (yes vs. no)  0.82 (0.53 – 1.29)  0.97 (0.57 – 1.65)   0.68 (0.51 – 0.92)  0.64 (0.46 – 0.89) Age* (per year older) 0.98 (0.96 – 0.99) 1.01 (1.00 – 1.03)  0.95 (0.94 – 0.96) 0.97 (0.96 – 0.98) Ethnicity (white vs. non-white) 1.17 (0.93 – 1.47)   1.18 (0.98 – 1.42)  Downtown Eastside residence*         (yes vs. no)  1.41 (1.14 – 1.75)  1.09 (0.86 – 1.40)   1.50 (1.30 – 1.73)  1.27 (1.10 – 1.47) Employment* (yes vs. no) 1.13 (0.88 – 1.44)   0.88 (0.77 – 1.00)  In a relationship* (yes vs. no) 1.06 (0.89 – 1.25)   1.15 (0.99 – 1.33)  Homeless* (yes vs. no) 1.98 (1.67 – 2.36) 1.74 (1.41 – 2.15)  1.87 (1.66 – 2.10) 1.29 (1.14 – 1.48) HIV seropositive* (yes vs. no) 0.63 (0.50 – 0.79) 0.64 (0.49 – 0.84)  0.76 (0.64 – 0.89)  Heroin injection* (≥ daily vs. <daily) 1.49 (1.25 – 1.77) 1.13 (0.91 – 1.39)  1.26 (1.09 – 1.45)  Cocaine injection* (≥ daily vs. <daily) 1.20 (0.96 – 1.51)   0.91 (0.75 – 1.11)  Crystal methamphetamine injection*          (≥ daily vs. <daily)  1.45 (1.07 – 1.97)  1.51 (1.05 – 2.15)   1.56 (1.28 – 1.92)  Crack cocaine use* (≥ daily vs. <daily) 1.48 (1.25 – 1.74) 1.10 (0.87 – 1.39)  1.61 (1.40 – 1.85)  Binge injection* (yes vs. no) 1.31 (1.11 – 1.54)   1.07 (0.95 – 1.19)     Require help injecting* (yes vs. no) 2.01 (1.68 – 2.40) 1.73 (1.42 – 2.11)  1.49 (1.28 – 1.73)  Heavy alcohol use* (yes vs. no) 1.29 (1.03 – 1.62) 1.20 (0.93 – 1.56)  1.59 (1.33 – 1.90) 1.66 (1.38 – 1.99) Enrolment in methadone maintenance therapy * (yes vs. no)  0.69 (0.57 – 0.82)  0.86 (0.70 – 1.07)   0.87 (0.76 – 0.99)  Sex work involvement* (yes vs. no) 1.62 (1.36 – 1.93) 1.35 (1.10 – 1.65)  1.65 (1.14 – 2.39)  Drug dealing* (yes vs. no) 1.58 (1.34 – 1.86) 1.26 (1.03 – 1.54)  2.32 (2.05 – 2.62) 1.86 (1.63 – 2.12) Incarceration* (yes vs. no) 1.76 (1.38 – 2.25) 1.36 (1.02 – 1.81)  2.28 (1.95 – 2.66) 1.57 (1.33 – 1.85)          111 Childhood emotional abuse              (yes vs. no)  1.99 (1.55 – 2.59)  2.12 (1.63 – 2.76)   1.51 (1.28 – 1.80)  Calendar year of interview*              (per year increase)  0.94 (0.92 – 0.96)  0.97 (0.94 – 1.00)   0.90 (0.89 – 0.92)  0.95 (0.92 – 0.97) † Odds ratio ‡ 95% confidence interval *Refers to the 6-month period prior to the interview.          112 Table 5.4 Characteristics of 2039 violent incidents experienced by people who inject drugs in Vancouver, Canada, stratified by gender (2005-2014) Characteristic Number of violent incidents (%)*† p - value Women  (n = 619) Men (n = 1420) Perpetrator of violence    Stranger 207 (33.4) 627 (44.2) <0.001 Acquaintance 154 (24.9) 329 (23.2) 0.404 Police             34 (5.5) 205 (14.4) <0.001 Current or former partner‡      92 (14.9)        60 (4.2) <0.001 Drug dealer        32 (5.2) 143 (10.1) <0.001 Sex partner£         12 (1.9)          2 (0.1) <0.001 Sex worker        14 (2.3)          1 (0.1) <0.001 Sex work client        38 (6.1)          1 (0.1) <0.001 Friend        33 (5.3)        72 (5.1) 0.807 Family        11 (1.8)          8 (0.6) 0.008 Unknown        15 (2.4)        55 (3.9) 0.098 Other        31 (5.0)        93 (6.6) 0.181 Type of violence    Beating 446 (72.1) 1079 (76.0) 0.060 Robbery 64 (10.3) 170 (12.0) 0.288 Attacked with gun          4 (0.7)        18 (1.3) 0.212 Attacked with other weapons 108 (17.5) 391 (27.5) <0.001 Sexual assault       54 (8.7)        10 (0.7) <0.001 Strangled        10 (1.6)          7 (0.5) 0.010 Other        47 (7.6)        59 (4.2) 0.001 * Refers to total number of reported incidents of received violence. † Total percentages may exceed 100 as participants could select multiple response options. ‡ Refers to the pre-defined response categories of ‘husband/wife’, ‘boyfriend/girlfriend’ and ‘partner’ as well as open-ended responses referring to a current or former partner.  £ Refers to regular or casual sex partner.             113 Chapter 6: Peer worker involvement in low-threshold supervised drug consumption facilities in the context of an overdose epidemic in Vancouver, Canada 6.1  Introduction As noted in Section 1.1, communities across North America are presently contending with epidemics of overdose death, driven largely by the adulteration of drug supplies with illicitly-manufactured fentanyl and related analogues.10,12 In 2017, there were an estimated 72,000 overdose deaths in the United States, almost 30,000 of which involved the use of non-methadone synthetic opioids such as fentanyl and its analogues.10 In Canada, there were almost 4,000 opioid-related deaths in 2017, 72% of which involved fentanyl or fentanyl analogues.12  In effort to address this ongoing public health crisis, a number of overdose response interventions have been implemented or scaled up in North America in recent years.249,250 For example, as described in Section 1.2, there has been a substantial expansion of SCFs in Canada since 2016.29,54 Momentum is also growing for the establishment of SCFs in several jurisdictions in the U.S., including San Francisco, Seattle, New York City, and Philadelphia,35,251,252 and an unsanctioned SCF is presently operating in an undisclosed urban location in the U.S.251 While a large body of evidence has demonstrated the role of SCFs in reducing various drug-related risks and harms, including fatal overdose, the systematic review presented in Chapter 2 found that most research in this area has focused on SCFs operated by health professionals. Comparatively less research attention has been paid to SCF programming delivered by peers. This dearth of evidence is due in large part to          114 regulations that have prohibited the legal establishment of peer-run SCFs in many settings worldwide.38,253,254 For example, in Canada, federally-sanctioned SCFs are governed by regulations under Section 56 of the Controlled Drugs and Substances Act stipulating that only licensed healthcare professionals may supervise drug consumption.253 Thus, at SCFs such as Insite in Vancouver, peer involvement has been largely limited to the role of peer support workers, who provide support counselling and information on local services (e.g., treatment and harm reduction programmes) to clients in an on-site “chill out” room outside of the area where the supervision of injections occurs.253 While social-structural obstacles to engagement of peer workers in SCFs persist, numerous studies have illustrated the value of peer-based harm reduction initiatives, including syringe exchange, safer assisted injecting and overdose response interventions, in engaging and reducing harms among higher-risk subpopulations of PWUD.246,255–259 For example, an evaluation of a peer-based overdose response training and naloxone distribution programme in Chicago found that this was an effective strategy to reduce overdose-related harms, particularly given that this training (including naloxone administration) was subsequently utilized in response to an overdose event by the majority of those trained.259 Similarly, an evaluation of a peer-run outreach-based syringe exchange programme in Vancouver found that this service effectively reached subgroups of PWUD who often face barriers to accessing conventional syringe exchange programmes, and that users of this service were less likely to reuse syringes compared to non-users.258 In addition, studies undertaken in various settings have demonstrated how peer-led outreach and support services may          115 promote utilization of addiction treatment,260,261 as well as infectious disease prevention, testing and counselling services among PWUD.262–265 There is also increasing evidence of the feasibility, acceptability and positive health impacts of peer-run SCFs.77,92,176,266 For example, in a previous feasibility study, some subpopulations of PWUD in Vancouver indicated a strong preference for peer-run SCF models over facilities operated by health professionals.266 Further, evaluations of unsanctioned peer-run SCFs in Vancouver have found these to be well utilized by PWUD, including structurally vulnerable drug-using populations.92,176 Indeed, such unsanctioned SCFs have been found to effectively engage individuals who inhale drugs and those who require manual assistance with injections, subpopulations who often encounter barriers in accessing local federally-sanctioned SCFs due to federal regulations that prohibit drug inhalation and assisted injection in existing federally-approved SCFs in Vancouver.92,176 Peer-based SCFs have also been found to play a role in mitigating drug-related risks and harms among these subpopulations of PWUD.92,176 For example, a qualitative study of people who smoke crack cocaine in Vancouver illustrated how an unsanctioned peer-run safer smoking room reshaped social-environmental contexts of drug use to minimize exposure to drug scene violence and reduce potential for unsafe smoking practices (e.g., crack pipe sharing) and related harms.176 However, little is known about peer engagement in SCF programming, including how this may shape service dynamics and related health and social outcomes, such as overdose-related risks and harms, among PWUD.  The Canadian province of BC has experienced particularly heightened rates of overdose death in recent years. Between 2014 and 2017, the annual fatal overdose rate in          116 BC increased from 7.9 to 30.9 per 100,000 population, prompting the provincial government to declare a public health emergency in April 2016, as described in Section 1.4.86 In response, various harm reduction strategies have been implemented or expanded in BC, particularly in Vancouver’s Downtown Eastside.54,249,267,268 Much of the response to the overdose epidemic in this neighbourhood has been spearheaded by peers involved in collective action initiatives and harm reduction programming delivered as part of the public health system and established peer-based drug user organizations in the Downtown Eastside.54,77,255,267,269 For example, peers have led the development and implementation of interventions such as naloxone training, distribution and response initiatives, as well as unsanctioned SCFs, including “pop-up” SCF tents that were erected in the Downtown Eastside beginning in September 2016.54,77,255 Such efforts stem from a long history of peer-led grass-roots collective action initiatives in the city.54 For example, local activists and PWUD opened and operated an unsanctioned SCF, known as the “327 Carrall Street” SCF, for approximately six months beginning in April 2003 when delays prevented the opening of Insite.270 Additionally, the Vancouver Area Network of Drug Users (VANDU), a peer-led organization that is internationally recognized for its advocacy efforts, public education, and implementation of novel peer-run interventions, was actively involved in efforts to sanction Insite, and has also played a critical role in innovating and extending the coverage of local supervised consumption programming since the establishment of the facility.54,92,176 Of particular importance, VANDU operated in its offices the peer-run          117 unsanctioned SCFs described above where individuals could get manual assistance with injections and inhale drugs.92,176 As detailed in Section 1.4, the BC Minister of Health issued a ministerial order to support the immediate implementation of low-threshold SCFs, known as overdose prevention sites (OPS), in BC in December 2016.54 Of note, several unsanctioned peer-run SCFs were subsequently sanctioned as OPS upon the enactment of this provincial order, and six of these facilities are presently operating in Vancouver.54,77 A notable distinction of OPS in comparison to conventional SCFs in Canada is that OPS are primarily staffed by peer workers and volunteers who manage service operations, supervise drug consumption, and provide emergency response in the event of overdose.54,77 These peer workers receive training in first aid and overdose response (including naloxone administration), as needed, and typically work shifts of four to five hours in duration. Information on service-level characteristics, including staffing models, of OPS in the Downtown Eastside is shown in Table 6.1. In the present study, we sought to extend existing research by drawing on rapid ethnographic fieldwork undertaken in Vancouver to characterize involvement of peer workers in OPS programming, including how this shapes dynamics of service engagement and related health and social outcomes among PWUD in the context of an overdose epidemic. As described in Section 1.5, this study was guided by Rhodes’ Risk Environment Framework.22,26,39 Given that SCFs and OPS are increasingly being established in communities across Canada as part of the response to the overdose epidemic,29,54 this study may provide important evidence to inform the ongoing development and optimization of such services.          118 6.2 Methods Data were drawn from a rapid ethnographic study examining the implementation, operations, and impacts of OPS in Vancouver’s Downtown Eastside neighbourhood, as described in Section 1.7.2.  Of the 72 PWUD interviewed as part of this study, 40 were women and 64 were presently unstably housed (i.e., homeless, having no fixed address or living in a single room occupancy hotel or shelter). Approximately one in six interviewed PWUD reported past or current involvement in the implementation or delivery of OPS services in Vancouver, and team members (JB, RM) spoke extensively with peer workers during ethnographic fieldwork. Characteristics of interviewed participants are presented in Table 6.2.  Data were analyzed thematically using deductive and inductive methods.271 Specifically, we developed a preliminary coding framework that drew on a priori categories extracted from the interview topic guide and our ethnographic fieldnotes, such as perspectives on peer workers, facilitators to accessing OPS, and barriers to accessing OPS. We also included emerging themes identified by research team members at regular meetings held during our fieldwork, such as peer roles in OPS implementation and operations, and challenges for peer workers. Data were imported into NVivo qualitative data analysis software program to facilitate data management and were coded thematically by multiple team members using an inductive and iterative process. Our team met regularly during the coding process to further refine the coding framework until the final thematic categories were established, and also          119 solicited feedback from OPS workers and community members in the Downtown Eastside during presentations. 6.3 Results 6.3.1 OPS implementation and operations drew on existing community capacities Our findings illustrate that peer involvement in OPS implementation and operations was in many ways an extension of the roles that peers were already undertaking in the community in response to the overdose epidemic. For example, most interviewed participants who worked at OPS and peer workers encountered during ethnographic fieldwork had previously received training in overdose response, including naloxone administration, through local organizations that serve drug-using populations. In addition, community capacity to open and operate OPS was largely developed through the active participation of peers engaged in drug user advocacy programming and collective action initiatives in the Downtown Eastside, as well as through peer-based harm reduction programs delivered as part of the public health system. Specifically, participants working at OPS commonly reported past or current participation in interventions such as unsanctioned SCFs, overdose response alley patrols and naloxone training and distribution initiatives. As explained by Kevin, a 51 year-old white man:  I took the [overdose response] training at [drug user organization] and I was one – I’m one of the supervisors at [an OPS] now and I was with [drug user organization] when we were patrolling the alleys too. As a result of their overdose response training and active involvement within existing peer-run organizations and harm reduction programming in the Downtown Eastside, peers were equipped with the critical competencies needed to rapidly          120 implement and operate OPS in the community. Moreover, during our ethnographic fieldwork, it was apparent that the experiences of peers within local drug user organizations and peer-based initiatives were crucial in positioning these individuals to occupy leadership roles in this critical part of the overdose response. For example, Julie, a 51-year old Indigenous woman with a longstanding history of involvement in peer-led initiatives, described her participation in the implementation of an unsanctioned SCF in the Downtown Eastside that was later sanctioned as an OPS:  Because so many people were ODing in our markets, and in the alley behind our markets. So then we did something about it... I’m part of the beginning of that opening up. In addition, peers, most of whom were actively engaged in leadership roles in existing local peer-based programming for PWUD, were largely tasked with the responsibility of establishing new OPS after the BC Health Minister issued the ministerial order to immediately open these services. For example, Abigail, a 60 year-old Indigenous woman, described how she and other peers involved in a local drug user organization acted quickly to fulfill the request of the local health authority to develop and open an on-site OPS with limited notice:  It was a great big thing, so like we were talking for hours, you know…Then we had extra meetings just to see what we can do, how we’re going to do it… We had 24 hours to open up the room, and that was that. As noted above, OPS were not subject to the same structural constraints as federally-sanctioned SCFs, including federal regulations that prohibit peers from working in drug consumption rooms, and were therefore able to provide greater opportunities for the direct involvement of peers in core service operations and delivery. As Mark, a 53 year-old Black man, articulated:           121 There’s no people who are from the community working in the injection room [at Insite]… There’s no people from the community in management there, right? There’s no people from the community in there, period, like except in the chill room pouring coffee. And that alone shows me the disdain that you have towards people in this community, right? So just on that – just on that very premise, that’s night and day to me, right? Where you come here [to an OPS] and it’s peer-run, you have people from the community in every level of the hierarchy here. Although women peer workers were often involved in high level operations of OPS, we observed in our ethnographic fieldwork that individuals who led the day-to-day service operations for OPS were more often men. These circumstances increased the potential for reproducing some gender dynamics that adversely impacted women, which we have examined elsewhere.77 6.3.2 Peer workers fostering environments of comfort and safety at OPS Involvement of people with lived experience as workers at OPS was commonly described by participants as providing a “good sense of community” that fostered a safe environment for PWUD characterized by comfort and inclusivity. Indeed, many OPS client participants emphasized during interviews and ethnographic fieldwork that peer workers were empathic to their life circumstances and needs due to their shared lived experience related to drug use and broader structural vulnerabilities (e.g., poverty, housing instability):  Because they’re addicts like us, you know, from the same lifestyle, you know. And they’ve been in our shoes many times over and they’ve been walking down the road we’re walking down right now. (Jeremy, Indigenous man, age 42) A few interviewed OPS clients reported feeling less confident in the ability of peer workers to effectively respond to overdose events at OPS in comparison to healthcare professionals given that peer workers typically had less education and training in this area:           122 A street person only knows so much. They’re not very educated when it comes to, you know, bringing a person back. Like did they get the training? No. That could cost a life. (Leslie, 56 year-old white woman) However, it is notable that peer workers successfully responded to all overdoses observed during our fieldwork despite challenges in identifying fentanyl-related overdoses,272 and no fatal overdoses have occurred at OPS.86 Moreover, most interviewed OPS clients emphasized that peer workers were well trained in overdose response and responded effectively to overdoses occurring at OPS: As soon as somebody [overdoses] – this person needs help – and they’re on their feet, ready, and already beside the person, which I like… Insite does that too but I see more reaction in the trailer [OPS]. These people on the street, they look after each other. They do care… It’s saved a lot of lives. (Paul, 55 year-old Indigenous man) In addition, many participants emphasized that peer workers had unique and relevant experiential knowledge, including drug-related expertise, that was critical to providing appropriate and effective services for PWUD. Dean, a 53 year-old white man, explained:  Because why send in people that don’t – haven’t been there, and don’t have knowledge of it? Like if you’ve never done dope, and how can you sit there and tell someone like me, that’s been a heroin addict for 30, 40 years, what it does to me and what it doesn’t? This expertise of peer workers was often described in participant interviews as fostering feelings of safety among OPS clients, particularly in relation to overdose response: “I feel very safe… because they’re users themselves, and so they know what to do in an emergency” (Emily, Indigenous woman, age 25). Participant accounts indicated that the expertise and lived experience of peer workers also enhanced feelings of comfort among PWUD at OPS because “the level of          123 trust [was] already there.” Specifically, interviewed participants characterized interactions with peer workers as generally more equitable and less intimidating and stigmatizing than interactions with non-peer staff at other organizations given that peers were more understanding of their social positions and experiences. These dynamics served to promote open communication and the development of more intimate personal relationships between OPS clients with peer workers, particularly in comparison to interactions with healthcare professionals. Jacob, a 23 year-old white man, explained:  I would open up more to the [staff] at [an OPS], and more to the people that are either previous users or current users, than somebody who’s like a professional doctor who maybe never has used in their life, because their opinions and just outlooks on things, in my opinion, aren’t really right. Because if you haven’t experienced it, you’ve never been a part of it, you can’t – you’ll never fully understand it.  Similarly, Leslie, a 56 year-old white woman who had accessed services at several OPS, noted: “Sometimes a person’s more comfortable telling a street worker, as opposed to telling a nurse, because they’ll feel intimated or nervous about saying it [to a nurse].” Moreover, characterizations of interactions with peer workers suggested that communication was further enhanced because these individuals often contended with structural vulnerabilities, such as poverty and criminalization, experienced by many OPS clients and were therefore able to relate to common stressors and provide relevant advice and support stemming from shared lived experience:  You know that they’ve been in that same position because they – when you’re talking to them [peer workers] about something that’s happened in your life, and you’re really messed up over, and they tell you exactly what they went through at times, and it’s so much exactly like what you went through. Anybody that didn’t go through it would never be able to say those words right at that time, you know,          124 and be able to tell you in such detail of what they were fighting with just to keep themselves alive. (Jeremy, Indigenous man, age 42) 6.3.3 Peer workers enabling harm reduction practices and other positive outcomes Our analysis highlights how peer involvement as staff at OPS facilitated engagement with OPS services and workers in several key ways that supported the adoption of harm reduction practices and promoted positive health and social outcomes among PWUD. First, many OPS clients who were interviewed and encountered during our ethnographic fieldwork linked their feelings of greater comfort and safety at OPS to being less rushed when using these services in comparison to SCFs operated by healthcare professionals: Well it’s more safe [at an OPS]… You’re not rushed, you’re more than welcome to chill out and hang out in the room, and do your thing, and play your music… Because the peers are there, I think people are more relaxed. (Brad, white man, age 47)  Participant accounts also illustrated how appreciation of peer expertise fostered the enactment of harm reduction practices, including overdose prevention strategies, among OPS clients. For example, some participants discussed how they reduced their doses at the advice of peer workers:  That guy [a peer worker] saved my life twice, or stopped me from overdosing twice like that, right? Because I had heroin, and I was going to use what I normally use, but he’d used it before, so he said, “[Matthew], don’t do that,” right? “Just maybe do half of that.” (Matthew, Indigenous man, age 44)   Further, interviewed participants highlighted how the experiential knowledge of peer workers promoted communication concerning addiction treatment and other health needs among OPS clients: I’m not going to talk to some stranger about wanting to go into detox or wanting rehab or something, right? But somebody here [at an OPS], if I’m feeling like I          125 needed rehab or detox I’m saying, “Hey, man, have you ever been through that? What’s it like?” Right? You can open up, right, where you can’t – you don’t want to do that with strangers or other people that don’t even know what you’re going through, right? So I’m 100 percent, it has to be peer-run. (Matthew, Indigenous man, age 44) In addition to supporting transitions to treatment, other participants indicated that the past experiences of peer workers provided these individuals with unique knowledge that supported reductions in drug use and facilitated engagement with other health, harm reduction and social services among OPS clients. Thus, in some cases, peer OPS workers appeared to be assuming responsibilities beyond overdose response that encouraged broader-spectrum health benefits among clients. 6.3.4 Work-related benefits and challenges for peer workers The structural context of OPS programming depended on peer worker involvement and thus allowed for formal recognition of the strengths and capacities of peer workers given that these individuals were critical to the work of implementing and operating these sites as part of the province-wide public health response to the overdose epidemic. This task shifting,273 which allowed peers to assume roles and responsibilities within the formal health system that were previously restricted to health professionals at federally-sanctioned SCFs, was described by some participants as providing more meaningful and rewarding workforce inclusion than what is typically available to PWUD in the community. As expressed by Michael, a 52 year-old Indigenous man:  You know, helping people helped me, right? You know, listening to them, it opened a lot in me, for sure, and made me want to go back to work and want to not just stay on methadone. And I’ve never done this sort of thing in my life, like worked at a place like this. In addition, others emphasized how working at an OPS provided unique opportunities for PWUD to expand their skillsets and increase their employability, with many          126 discussions during ethnographic fieldwork centring around these new possibilities for employment as a part of the overdose response. Speaking of her friend who worked at an OPS, Laura, a 52 year-old white woman, explained:  She started doing some volunteering [at an OPS], and she started to like build up a résumé so to speak… and she was saying how she actually felt that it improved her quality of life.  These descriptions highlight how meaningfully involving peer workers in the delivery of services at OPS may have served to mitigate the tokenization that PWUD sometimes report in relation to harm reduction programming.274 However, it should be noted that, although some peer workers had salaried positions, most were hired as volunteers at OPS who were provided with small stipends amounting to less than the provincially-mandated minimum wage. Steven, a 65-year old Indigenous man, described how peer workers at OPS typically received minimal financial compensation for their work in comparison to non-peer staff engaged in similar work in Vancouver: “It’s cheaper to get people like us on the street… one person from the city that you hire, you could hire three of us per hour.” Thus, despite the central role of peers in OPS service delivery, these inequities in compensation practices fostered perceptions that their expertise, time and efforts were not valued.  As with the broader local PWUD population, many peer workers who were interviewed and encountered during ethnographic fieldwork revealed that they had lost at least one friend or family member to overdose death, and also routinely encountered overdose events while working at OPS and in the broader community. As such, participants often experienced considerable trauma (i.e., “experiences that cause          127 intense physical and psychological stress reactions”275 and grief due to the emotional toll of the overdose epidemic and a lack of adequate supports:  When you see your friends go down or you come across your friends and they’re dead, like it’s – it really really gets to you after a while… Everybody down here is broken some sort of fuckin’ scale, 1 to 10, right?… At the end of the day, we’re all hurting and we all lean on each other. (Samantha, white woman, age 27) This grief associated with high rates of overdose mortality among participants’ peer groups was often described in participant interviews and during ethnographic fieldwork as compounded by structural vulnerabilities, such as poverty and criminalization, as well as other stressors, including elevated risk of overdose and pronounced withdrawal symptoms due to the proliferation of fentanyl and other powerful opioids in illicit drug supplies and lack of legal access to unadulterated drugs:  There are a ton of stressors right. I mean people are losing their friends and there are so many, but a big one is, on top of all that, they’re dope sick constantly. They’re constantly dope sick because they never know when it’s coming. You could never know when it’s coming, so on top of having all of these horrible things happening, your friends dying, your friends going down while you have friends dying, having to worry about your own life while your friends are dying… It’s just bonkers. It affects every aspect of your life, and that’s just crazy, because it affects the staff [at OPS]. It affects everybody. (Amy, white woman, age 47)  Participants who worked at OPS often described how the grief and trauma they experienced as a result of routine exposure to overdose events and the significant loss to overdose death in the community contributed to burnout in regards to their roles as peer workers. Specifically, many interviewed OPS workers described how these circumstances resulted in feelings of emotional exhaustion and disconnection from their work,276,277 and some OPS workers encountered during our fieldwork reduced their          128 shifts or left peer positions altogether due to these challenges. Kevin, a 51 year-old white man who worked at an OPS, explained:  I’ve spent so many years sitting in rooms watching people like this… It’s really hard for my post-traumatic stress, sitting there watching that go on [at an OPS]… so that’s why I haven’t been putting in as many hours there.    However, most peer workers were not provided with employee benefits and supports (e.g., health benefits, counselling, stress leave) that are typically afforded to salaried non-peer employees in similar positions at other local organizations. As such, participant responses highlighted the need for interventions to address gaps in social and emotional supports for peer workers and other local PWUD experiencing trauma, grief and other adverse psychosocial responses elicited as a result of the overdose epidemic:  I think that’s one thing they should think about setting up, is some place to go to talk about this, right? Because a lot of people just want to, you know, explain their feelings and stuff, which is – you know, there’s no better therapy than talking, right? (Michael, Indigenous man, age 52)  6.4 Discussion In summary, the findings of this study illustrate how the structural context of OPS implementation and operations depended on peer worker involvement and thus allowed for formal recognition of peer capacities that were developed through roles that peers were already undertaking in existing programming for PWUD in the broader community. We found that peer involvement as staff at OPS enhanced feelings of comfort and safety among OPS clients, and thereby facilitated their engagement with OPS services. These dynamics and appreciation of peer worker expertise among PWUD promoted client communication with peer staff in ways that fostered the enactment of          129 harm reduction practices and encouraged health and social benefits. Moreover, the central involvement of peer workers in the implementation and operation of OPS provided peers with rewarding workforce inclusion that allowed these individuals to expand their skillsets and increase their employability. However, peer workers typically received limited financial compensation and work-related benefits. Further, our findings revealed that many PWUD experienced considerable trauma and grief due to the emotional toll of the overdose epidemic and lack of adequate social and emotional supports, which contributed to burnout among peer workers.  Previous research has illustrated how task shifting, the systematic redistribution of healthcare tasks from specialized health professionals to individuals with less training such as lay workers, may help to reduce health inequities by improving access to care for underserved populations.273,278,279 To date, this research has primarily demonstrated how task shifting may help to overcome health human resource shortages, particularly in response to HIV/AIDS epidemics in resource-limited contexts.278,279 The present study builds on this work in identifying the use of task shifting to peer workers as a successful novel approach for facilitating the rapid implementation and delivery of low-threshold SCFs in the context of an overdose epidemic, thereby strengthening emergency response capacity in a timely manner. Furthermore, similar to previous studies of other peer-run harm reduction initiatives,92,176,246,255–259,266 our findings illustrate how this task shifting approach functioned to enhance the effectiveness of this form of harm reduction programming, including by improving service engagement, reducing potential for overdose-related          130 harms, and promoting uptake of addiction treatment and other health services among PWUD.  Amidst the ongoing overdose epidemic, these findings support the expansion of formalized peer worker involvement in SCF and OPS programming as a feasible and effective public health strategy to mitigate overdose and related harms. However, given that Vancouver has a well-established system of drug user organizations and peer-based programming that was harnessed to support this response, this approach may be less feasible in settings where peers are not already actively engaged in the delivery of such services. Thus, our findings underscore the need for adequate funding to support drug user organizations and other peer-based initiatives,280,281 as these appear to be critical in ensuring community capacity for successful peer-driven responses to the overdose epidemic. In addition, future research should seek to explore the role of various factors, including a lack of established drug user organizations, in influencing peer involvement in SCF programming and related outcomes in diverse settings.  Given that the peer-run SCFs studied in the present study were sanctioned to operate under a provincial ministerial order, future studies should also continue to examine how implementing and operating peer-run SCFs with a lack of formal legal sanction may shape service dynamics and the effectiveness of this form of intervention.92,176,251 This is particularly important in light of the findings of a recent evaluation of an unsanctioned SCF operating within a community-based organization in an urban area in the United States.251 This study found that the perceived illegality of the facility and resulting fears of potential legal repercussions among operators and staff shaped operational processes in ways that hindered client recruitment, contributed          131 to a lack of diversity in terms of the socio-demographic characteristics of clients, and impeded the ability of SCF staff to connect clients with other health and social services.251 Thus, further inquiry in this area, including examination of potential strategies to mitigate such issues, may provide important information to inform the optimization of unsanctioned SCFs as well as efforts to establish sanctioned SCFs.  Consistent with previous studies demonstrating the non-material benefits of peer work positions in drug user organizations and harm reduction programming, including empowerment and enhanced authority,273,280,281 our findings illustrate how involving peers as workers in OPS programming provided formal recognition of the unique expertise of PWUD while also enabling these individuals to enhance their skillsets and employability. As such, this approach served to mitigate the tokenization that some peers have described in relation to their engagement in other harm reduction interventions.274 However, our findings revealed gaps in existing supports available to peer workers. Of note, many peers received inadequate financial compensation for their work at OPS, which has been identified in the task shifting literature as a key factor that perpetuates health workforce inequities, contributes to the stigmatization and disempowerment of workers, and increases worker attrition, thereby compromising the long-term sustainability and effectiveness of task shifting approaches.273,278,279,281  In addition, this study found that many peers experienced trauma, grief and burnout in their roles as OPS workers due to their routine exposure to overdose events and the significant losses to overdose death in the community. Concurrently, peer workers often contended with various structural vulnerabilities, including poverty and housing instability, and other stressors, including pronounced withdrawal symptoms          132 and heightened risk of overdose due to the proliferation of illicitly-manufactured fentanyl and related analogues in illicit drug supplies. Despite these issues and their central role in OPS service delivery, peer workers were not provided with work-related employee benefits and supports, including health benefits, counselling and stress leave, that are typically afforded to non-peer employees engaged in similar work in this setting.  Thus, although task shifting the frontline response at OPS to peer workers appears to be a feasible and effective response to the overdose epidemic, efforts are needed to ensure that peer workers receive adequate financial, social and emotional supports in order to avoid perpetuating workforce power imbalances, imposing unfair burdens on peer workers, and compromising the sustainability of this approach over the long term.273,279,281,282 In particular, strategies such as offering fair and standardized compensation to peer workers would help to address concerns about their potential devaluation and exploitation, and would also ease financial constraints among these individuals to support their sustained retention as workers in OPS programming, thereby improving the quality and sustainability of services provided.273,278–281 As well, greater funding and resources should be dedicated to expanding formalized trauma- and grief-related supportive services for local PWUD, and to providing peer workers with work-related benefits to support their well-being and mitigate potential harms or burdens they may be experiencing due to their critical role in the frontline response to the overdose epidemic.273,281 This study has several limitations. First, our findings are specific to OPS clients and workers recruited from four OPS in Vancouver’s Downtown Eastside and are not          133 representative of these subpopulations or the broader drug-using population in the Downtown Eastside. Moreover, the Downtown Eastside is distinct from other communities in many ways, including being characterized by high levels of drug use and poverty, a long history of drug user activism, and a high concentration of harm reduction programming.54,78 Thus, our findings are not generalizable to PWUD in other settings. Finally, although our findings highlight benefits and challenges related to peer involvement as staff at OPS among local PWUD, further research is needed to fully determine the medium- and long-term impact of peer-run OPS on such outcomes. In particular, future studies should be conducted to determine if the previously-observed health impacts of SCFs operated by health professionals25,27 may extend to those operated by peers. This could include quantitatively assessing potential impacts on overdose-related harms, drug use practices associated with infectious disease transmission, and uptake of addiction treatment.  In conclusion, this study demonstrates that peer worker involvement in OPS was a feasible approach to support the rapid implementation and effective delivery of such programming in this setting. Peer worker engagement at OPS contributed to improved engagement with OPS services, fostered the enactment of harm reduction practices, and promoted health and social benefits among PWUD. This approach also provided rewarding workforce inclusion for PWUD, although many peer workers were inadequately compensated and experienced considerable grief, trauma, and burnout in their roles as OPS workers due to the overdose epidemic and a lack of formalized supports. These findings support the expansion of peer involvement as staff in SCF programming as a strategy to reduce overdose-related morbidity and mortality and          134 other harms among PWUD, particularly in locales with established peer-based programming for PWUD. However, further efforts are needed to ensure that peer workers receive adequate financial, social and emotional supports in order to promote the sustainability of this approach.                                                  135 Table 6.1 Service-level characteristics of four overdose prevention sites (OPS) in Vancouver, Canada (2017)  OPS #1 OPS #2 OPS #3 OPS #4 Location Within a drug user organization in the Downtown Eastside Adjacent to an alley in the Downtown Eastside Within a non-profit housing building in the Downtown Eastside Within a women-only organization in the Downtown Eastside Capacity • ~10 clients • ~22 clients • 7 clients • 15 clients Staffing composition • Peer site operators • Trained peer staff  • Non-peer site operators • Trained peer staff • Non-peer site operators • Trained peer staff • Non-peer site operators • Trained peer staff Services offered* • Supervision of drug consumption • Naloxone administration • Harm reduction supplies and education • Take home naloxone training  • Supervision of drug consumption • Naloxone administration • Oxygen administration • Harm reduction supplies and education  • Supervision of drug consumption • Naloxone administration • Oxygen administration • Harm reduction supplies and education  • Supervision of drug consumption • Naloxone administration • Oxygen administration • Harm reduction supplies and education • Medical care • Referrals to addiction treatment, health and community services Typical hours of operation† 10 AM – 10 PM 10 AM – 10 PM 12 PM – 10 PM 6 AM – 12 PM; 6 PM – 12 AM *Indicates the services that were available during the study period, but which have since been expanded in some OPS. †Indicates typical hours of operation during the study period, but which have since changed for some OPS.                136 Table 6.2 Characteristics of 72 participants in a rapid ethnographic study of overdose prevention sites in Vancouver, Canada (2017)  Characteristic   n (%) Age (years) Median (Interquartile range)  44 (34-53) Gender Men Women Transgender, two-spirit, or non-binary  29 (40%) 40 (56%)   3   (4%) Ancestrya White Indigenous Other  32 (44%) 33 (48%)   3   (7%) Employedb 10 (14%) Unstably housedc 64 (86%) Homeless in the previous year 47 (65%) Incarcerated in the previous year 27 (38%) Substance use in the previous 30 days      Powder cocaine      Crack cocaine      Crystal methamphetamine      Heroin      Fentanyl      Other opioids      Other substances    8 (11%)   3   (4%)   8 (11%)                60 (69%)   2   (3%)   2   (3%)   8 (11%) Overdose events in the previous year  1 overdose 2 overdoses 3 or more overdoses  14 (20%) 10 (14%) 19 (26%) a Participants could select more than one response option. b Defined as currently having full- or part-time employment. c Defined as currently living in single room occupancy hotel, shelter, homeless or having no fixed address.          137 Chapter 7: Conclusion 7.1 Summary of findings This dissertation sought to address gaps in knowledge, policy and health services for PWUD by longitudinally investigating discontinuation of SIF use, assessing the long-term impact of SIF use on all-cause mortality and exposure to violence, and characterizing peer involvement in low-threshold SCFs, including how this may shape aspects of service engagement and related outcomes, among PWUD in Vancouver, Canada. Chapter 2 presented the findings of a systematic review of existing evidence concerning the health and community impacts of SCFs. This review described how consistent evidence of strong methodological quality indicates that these facilities effectively achieve their primary public health and order aims with a lack of negative effects. As such, the review recommended that SCFs be included as part of comprehensive strategies to reduce harm and promote the health of PWID. However, this review also identified important knowledge gaps concerning the long-term impact of these facilities on health-related outcomes, including all-cause mortality and exposure to violence. Further, this review highlighted the need for further evaluation of novel SCF models, including peer-run models, and how these may modify the effectiveness of this form of programming, including by influencing service coverage and impacts.  The quantitative study presented in Chapter 3 drew on Rhodes’ Risk Environment Framework to examine the association between discontinued SIF use and various individual and contextual factors among a community-recruited prospective cohort of PWID who used the Insite SIF. The analyses demonstrated that approximately          138 three-quarters of participants (77%) ceased using the SIF over a median follow-up duration of approximately four years, with the majority of cessation events (58%) occurring during periods of injection cessation. In multivariable analyses, enrolment in MMT was positively associated with SIF use cessation during periods of active injection, while homelessness, high intensity heroin injection, binge injection, public injection and incarceration were inversely associated with this outcome. The most common reasons for discontinuing use of the SIF were: injection cessation, a preference for injecting at home, and already having a safe place in which to inject. These findings indicate that this health service successfully retains PWID at elevated risk of drug-related harms and that most SIF clients do not inject drugs or use SIFs perpetually.  Chapters 4 and 5 presented epidemiological analyses of the longitudinal association between SIF use and health outcomes among a prospective cohort of PWID. Chapter 4 longitudinally examined the relationship between SIF use and all-cause mortality. It was hypothesized that frequent SIF use would be associated with decreased risk of mortality. We observed a high burden of death in this study with 112 (14%) of participants dying over a median follow-up duration of six years, yielding a crude mortality rate of 22.7 deaths per 1,000 person years. The leading observed causes of death were other non-accidental, unknown causes, overdose and HIV-related causes. In multivariable analyses, frequent SIF use predicted a lower risk of all-cause mortality, independent of socio-demographic, social-structural, and behavioural factors. These findings suggest that, in promoting the enactment of harm reduction strategies, providing emergency response in the event of overdose, and facilitating uptake of          139 various health and social services,25,27 SIFs may reshape the risk environment of PWID to mitigate premature mortality among this population. Chapter 5 sought to longitudinally examine the gender-specific relationship between SIF use and experiencing physical or sexual violence among PWID. It was hypothesized that SIF use would be associated with a decreased likelihood of exposure to violence among both men and women. This study found that more than half of men and women PWID in the study experienced at least one incident of violence over a median follow-up duration of four years. In multivariable analyses, exclusive use of a SIF for injections was independently associated with decreased odds of experiencing violence among men, but was not significantly associated with this outcome among women. These findings highlight how, in providing a protected and monitored space in which to inject drugs, SIFs may reshape the social and physical contexts of injection drug use to reduce exposure to violence among men.91,92 However, SIF use alone may not adequately protect women from violence given that violent encounters that occur outside of drug use contexts appear to be more common among women than men. Indeed, in this study, men were more likely to report experiencing violence from strangers and police, while women were more likely to report experiencing violence from intimate partners, family members and people involved in the sex industry. Moreover, the SIF examined in this study is located in a geographic area characterized by particularly elevated levels of violence, and women have also been found to experience harassment when accessing local SIFs,77,175 which might also partially explain the observed lack of a protective association between exclusive SIF use and violence among women.           140 Chapter 6 drew on data from a rapid ethnographic study of low-threshold SCFs to characterize peer involvement as staff within these facilities, including how this shapes service engagement and related health and social outcomes among PWUD. This study demonstrated that the implementation and operation of low-threshold SCFs in this setting relied on peer involvement and thus allowed for formal recognition of capacities that peer workers had developed through active involvement in local peer-based programming. Findings also illustrated how peer involvement at these SCFs enhanced feelings of comfort and safety among PWUD, and thereby facilitated engagement with these services. These dynamics promoted client communication with peer staff in ways that fostered the enactment of harm reduction practices and other positive health-related outcomes. Moreover, involving peer workers in the implementation and operation of OPS provided peers with meaningful workforce inclusion that allowed these individuals to enhance their skillsets and employability. However, many peer workers received minimal financial compensation and experienced trauma and grief due to the emotional toll of the overdose crisis and a lack of adequate supports, which contributed to staff burnout. These findings suggest that involving peers as staff in low-threshold SCFs is a feasible approach to support the rapid implementation and effective delivery of such programming, particularly in settings where peers are actively involved in the delivery of existing programming for PWID. However, the findings also highlight existing gaps in formalized supports for peer workers that may compromise the sustainability of this approach over the long term.          141 7.2 Study strengths and unique contributions This dissertation has some notable strengths and makes several key contributions to the literature specific to SCFs. The systematic review of quantitative research on health and community outcomes associated with SCF use addressed the methodological shortcomings of previous reviews27,37,95 by employing a comprehensive search strategy that resulted in the inclusion of relevant research, including non-English literature, that had been overlooked in past reviews. Further, this review applied strict study inclusion criteria to reduce the likelihood of including low-quality evidence and, in contrast with previous reviews, also formally assessed the methodological and reporting quality of included studies using established tools in effort to maximize the validity of conclusions drawn and provide informed recommendations. Finally, this review identified gaps in the scientific literature to guide future research specific to SCFs, including the data-driven studies conducted for the present dissertation research.  Another unique contribution of this work is the inclusion of the first study to longitudinally characterize discontinuation of SIF use. While previous studies have examined the characteristics of SIF clients, these have primarily relied on cross-sectional designs with short-term durations of study follow up.66–70 In drawing on over eleven years of prospective cohort data and employing longitudinal statistical modelling methods with time-updated measures, this research addresses critical gaps in evidence concerning long-term SIF utilization patterns. Specifically, this research extends on past studies to demonstrate that SIFs not only attract,66–70 but are also more likely to retain their target population of PWID at elevated risk of harms when these individuals are actively injecting. This study also highlighted how treatment exposure may reduce the          142 need for SIF services, which has important implications for the development of SIF programming. Further, in identifying injection cessation as the most common explanation for SIF use cessation, this research builds on previous studies50–53,120 to challenge the argument that SIFs may encourage ongoing drug use.  Although previous population-based studies have demonstrated reductions in overdose mortality with the establishment of SIFs,48,75 these have primarily relied on aggregate rather than individual-level data, and no studies have examined the impact of SIF use on all-cause mortality. Moreover, existing research investigating the impact of SIFs on overdose mortality has been limited to analyses with relatively short study periods.48,75 As such, in drawing on data derived from a long-running cohort of community-recruited PWID, this dissertation makes an important contribution to existing literature in that it includes the first study to longitudinally identify an independent individual-level association between frequent SIF use and reduced risk of mortality from all causes. Some notable strengths of this study are that mortality was objectively measured through linkages to an external vital statistics database, and self-reported explanatory variable measures were time-updated based on semi-annual follow up data. As well, multivariable analyses for this study were adjusted for an array of key confounding factors, including socio-demographic, behavioural and social-structural variables.  This dissertation makes a further unique contribution to the literature in that it includes the first known study to quantitatively examine the association between SIF use and exposure to violence. Employing longitudinal methods that accounted for within and between individual correlation, and adjusting for a range of time-updated          143 potential confounders strengthened the study and the inferences drawn. Conducting a gender-based analysis was a further strength in that this allowed for the identification of the moderating influence of gender, with a significant inverse association between exclusive SIF use and violence observed among men, but a non-significant association observed among women. This approach highlighted the importance of considering potential gender differences when developing SIF programming and other violence prevention initiatives for PWID.   Lastly, in characterizing peer involvement in the implementation and operation of low-threshold SCFs, this dissertation addresses pressing gaps in current evidence concerning peer-run SCF models. Specifically, this research allowed for the generation of unique insights concerning the benefits and challenges associated with task shifting SCF roles from healthcare professionals to peers in the context of an overdose epidemic, which has important implications for the ongoing development and optimization of SCF programming. Employing rapid ethnographic methods contributed to an in-depth understanding of this phenomenon in a timely manner, providing evidence that may be useful in informing policy and programmatic responses to the present overdose crisis and potentially other complex public health problems related to illicit drug use. 7.3 Limitations This research has several important limitations that should be noted. First, many of the explanatory and outcome variables of interest examined in the quantitative analyses presented herein relied on self-reported information. As such, findings involving such measures may be biased by reporting biases, including socially desirable reporting and deficiencies in recall. However, previous studies have demonstrated the          144 validity of self-reported data among PWID.283 Further, a number of techniques were employed to minimize the potential for such biases, including having interviewers reassure participant anonymity and confidentiality of data, as well as ask potentially sensitive questions regarding stigmatized behaviours or sensitive topics near the end of interviews. Another notable limitation is that the VIDUS and ACCESS cohorts are non-random samples and thus the findings of this research may not be generalizable to PWID in Vancouver or elsewhere. However, it should be noted that no population-level registries of PWID exist, and therefore it is not possible to construct probability registry-based samples of PWID. Moreover, a variety of techniques were undertaken to maximize the representativeness of these cohorts to the broader population of PWID in Vancouver, including by recruiting eligible participants through close collaboration with local service agencies, snowball sampling, word-of-mouth, and street-based outreach undertaken in various neighbourhoods where PWID are known to congregate.284 It is also worth noting that the socio-demographic profile of PWID in the  VIDUS and ACCESS cohorts is similar to other samples of PWID in BC.285 Nonetheless, the cohorts have some unique characteristics, including high levels of polysubstance use, that may limit the generalizability of the findings of this dissertation to populations of PWID in other settings.286–288 A similar limitation is that interview data collected as part of the ethnographic component of this work is specific to staff and clients of local low-threshold SCFs who participated in this study, and are not representative of these groups or the broader population of PWUD in Vancouver. Further, Vancouver’s Downtown Eastside differs from other geographic settings in many ways, including in that it has long been an epicentre of illicit drug use and related harms in North America,          145 and has a well-developed system of harm reduction programming.54,78 Such characteristics further limit the generalizability of findings to other contexts. An additional limitation is that while the epidemiological analyses conducted for this dissertation employed multivariable regression techniques and other strategies in effort to control for potential confounders, it is possible that the findings observed herein may have been affected by unmeasured or uncontrolled confounding. Finally, although Rhodes’ Risk Environment framework22,26,39 was applied in this dissertation research in effort to conceptualize how social-structural and environmental conditions may shape health-related outcomes, including SCF use, mortality and exposure to violence, the quantitative analyses presented herein would not have captured how potential heterogeneity within strata of assessed explanatory variables may have influenced such outcomes or modified associations of interest.     7.4 Recommendations Although specific recommendations are detailed in Chapters 2 to 6, this section summarizes several key policy recommendations derived from this collective work. First, although SCFs remain underutilized in many settings worldwide, the findings of this dissertation underscore the need to scale up access to this form of health intervention as part of broader public health strategies to reduce drug-related harms. This is especially important given that there is presently a pressing need for evidence-based interventions to mitigate morbidity and mortality among PWUD, particularly in Canada and the United States, where overdose epidemics remain a major public health challenge.10,12 This recommendation is supported by the findings presented in Chapters 2 to 6, which demonstrate the effectiveness of SCFs in attracting and retaining their          146 target population of PWUD at heightened risk of drug-related harms, as well as in improving public order and mitigating the risk of serious harms, including mortality, exposure to violence, and overdose-related harms among PWUD. Further, as concerns regarding the potential negative consequences of SCFs have made these facilities difficult to implement in many settings, it should be noted the findings of Chapters 2, 3 and 6 indicate that such concerns, including that SCFs perpetuate ongoing injection drug use and undermine access to addiction treatment, are not supported by existing scientific evidence. In light of this strong evidence base, public health advocates and elected officials should continue to advocate for the inclusion of SCFs within the continuum of health and harm reduction services for PWUD.  In addition to the broader expansion of SCFs, efforts to better align SCF programming with the needs of vulnerable and underserved populations should be considered a public health priority. In particular, there remains an urgent need to amend federal regulations prohibiting assisted injection within SCFs in Canada given that Chapter 5 found that individuals who require help injecting were more likely to experience violence, which is consistent with past work indicating that this subpopulation of PWID is at particularly elevated risk of health harms.172–174 Moreover, previous studies have found that providing manual assistance with injections to these individuals within a regulated environment may help to mitigate such harms.92 As noted in Chapter 5, Health Canada recently initiated a trial to evaluate an assisted injection programme in several SCFs, including in Toronto, Ottawa, Montreal and Lethbridge.247 However, further legislative changes are needed to allow the scale up and optimization of this service in settings across Canada.          147 The findings of this dissertation also support the expansion of novel approaches to SCF programming, including peer-run and women-only SCFs, as a strategy to extend the reach and impacts of this health intervention. Of note, the findings of Chapter 6 align with previous studies of peer-run harm reduction services,92,176,246,255–259 in demonstrating that formally involving peer workers in SCF implementation and service delivery may afford opportunities to enhance the effectiveness of this intervention, including by promoting SCF service engagement and the reduction of harms among PWUD. Further, these findings suggest that this approach may support the rapid implementation of SCFs in response to drug-related public health emergencies. However, additional funding and resources should be dedicated to supporting drug user organizations and other peer-based initiatives in settings seeking to adopt this strategy, as the findings of Chapter 6 suggest that these are critical in ensuring community capacity for the successful implementation and operation of peer-run SCFs. As well, efforts are needed to ensure that peer workers at SCFs are fairly compensated and have access to necessary social and emotional resources in order to better support their well-being and the long-term sustainability of this approach.   The findings presented in Chapter 5 also reinforce previous calls for the implementation of women-only SCF programming in effort to maximize opportunities to engage and reduce harm among women PWID.77 Specifically, these findings indicate that exclusive use of a mixed-gender SCF was not significantly associated with reduced exposure to violence among this group, which could be due in part to gendered power relations that foster SCF environments in which the needs of men are privileged over those of women, and women are potentially vulnerable to harassment from men.77          148 Further, past work has also revealed that such gender dynamics operating within mixed-gender SCFs may hinder women’s engagement with such services.77 Thus, efforts to tailor SCF programming to the unique needs of women may serve to better support their health and safety. Indeed, an evaluation of a women-only SCF in Hamburg, Germany found that 80% of women accessing this facility reported feeling more comfortable and safe within this environment compared to mixed-gender SCFs, while 90% reported enhanced feelings of trust towards staff.30  7.5 Future research The studies presented in this dissertation suggest important new directions for future research seeking to advance understanding of SCF utilization patterns and health impacts among PWUD. First, there is need for further research to explore potential underlying explanations for the findings of the quantitative analyses presented herein. In particular, given that a substantial of the proportion of deaths examined for Chapter 4 were due to unknown causes, future studies should seek to determine potential individual-level associations between SCF use and specific causes of death (e.g., overdose mortality; HIV-related mortality), as well as identify possible explanatory mechanisms that may mediate such associations. Additionally, the finding of this study that exclusive SCF use was significantly associated with lower overall exposure to violence among men but not women suggests a need to better understand potential underlying explanations for this effect modification. For example, future studies should seek to more fully determine the role of SCFs in shaping exposure to specific forms of violence (e.g., inter-partner violence; police violence; violence during drug consumption) among men and women, respectively.          149 As previous research specific to SCFs has primarily examined service uptake and impacts related to the operation and use of individual facilities, an additional research opportunity is to investigate the potential health and social impacts of scaling up SCFs within specific geographic settings. As described previously, there has an expansion of SCFs in Vancouver and other communities across Canada in recent years,29,88,89 and thus further examination of potential outcomes related to this scale up in diverse settings would provide unique and important evidence for policymakers considering the merits of implementing or expanding access to such services. As discussed in Chapters 2, 3 and 5, such research should investigate potential impacts of SCF expansion on service uptake and retention, including among PWUD who have previously been found to encounter geographic or other barriers to accessing this form of programming (e.g., women who have avoided SCFs due to the threat of violence). A further priority research topic is to examine potential impacts of this service scale up on relevant health-related outcomes, including mortality and exposure to violence.  The findings of the present work also highlight the need for the continued evaluation of innovations in SCF models and service delivery. For example, the systematic review conducted for Chapter 2 identified gaps in evidence specific to supervised inhalation rooms, which remain underutilized in Canada despite successfully operating in European settings for many years.30,122 Accordingly, efforts to formally evaluate the effectiveness of supervised inhalation rooms would be useful in informing the development of evidence-based recommendations concerning the potential scale up of such services. This is especially important given that interventions specifically targeting the health and social harms of drug inhalation are presently          150 lacking in many settings worldwide.176 Further, the findings of Chapter 2 suggest that further evaluation of novel SCF models, including mobile SCFs and SCFs integrated into existing healthcare and community services, is warranted given that previous studies examining the effectiveness of SCFs have predominantly focused on fixed-site, stand-alone SCFs. As such, the extent to which previous findings concerning utilization and outcomes of SCFs may hold in evaluations of mobile and integrated SCF models is not known. An additional research opportunity discussed in Chapter 5 is to further evaluate women-only SCF models, including how these may extend service coverage and impacts among underserved subpopulations of women PWID. Future studies should also explore how SCF services could be modified to improve their responsiveness to the specific needs of individuals with nonconforming gender identities given that data in this area is lacking. Finally, the findings of Chapter 6 point to the need to assess potential medium- and long-term outcomes associated with peer involvement in SCF service delivery among both clients and staff. As well, given the unique characteristics of the present study setting,54,78 additional studies are needed to identify potential benefits and challenges related to the implementation and operation of peer-run SCFs in other settings.  7.6 Conclusions This dissertation research sought to generate scientific evidence concerning SCF service engagement and health impacts to inform policy and programmatic responses to the pressing public health challenges associated with illicit drug use. This research addresses important knowledge gaps concerning long-term SCF utilization patterns in demonstrating that the overwhelming majority of SCF clients followed over time          151 discontinued using this service, most often during periods of injection cessation. However, higher-risk subpopulations of PWID were more likely to be retained in this form of programming while actively injecting, which is reassuring given the opportunities afforded by this intervention to minimize harm and support the health of PWID.25,27 Specifically, past research has identified various short- and medium-term positive health impacts of SCFs,25,27 and the present study builds on this work in demonstrating some notable long-term health benefits, including decreased risk of all-cause mortality, as well as reduced overall exposure to violence among men PWID. Furthermore, this dissertation research highlights the value of involving peers as staff in SCFs, including how this may help to extend the reach and health impacts of this intervention. Collectively, these findings support the inclusion of SCFs as part of the continuum of services for PWUD, and also offer critical insights into how these services could be modified to further enhance their effectiveness in reducing the outstanding harms associated with illicit drug use.           152 References  1.  Tookes H, Diaz C, Li H, Khalid R, Doblecki-Lewis S. A cost analysis of hospitalizations for infections related to injection drug use at a county safety-net hospital in Miami, Florida. PLoS One. 2015 Jun 15;10(6). 2.  Lloyd-Smith E, Tyndall M, Zhang R, Grafstein E, Sheps S, Wood E, et al. Determinants of cutaneous injection-related infections among injection drug users at an emergency department. The Open Infectious Diseases Journal. 2012 Jan;6.  3.  Kaushik KS, Kapila K, Praharaj AK. Shooting up: The interface of microbial infections and drug abuse. Journal of Medical Microbiology. 2011;60(4):408–22.  4.  Darke S, Torok M, Kaye S, Ross J, McKetin R. Comparative rates of violent crime among regular methamphetamine and opioid users: Offending and victimization. Addiction. 2010 May 1;105(5):916–9.  5.  Gibson A, Randall D, Degenhardt L. The increasing mortality burden of liver disease among opioid-dependent people: Cohort study. Addiction. 2011 Dec 1;106(12):2186–92.  6.  Kennedy MC, McNeil R, Milloy M-J, Dong H, Kerr T, Hayashi K. Residential eviction and exposure to violence among people who inject drugs in Vancouver, Canada. International Journal of Drug Policy. 2017 Mar 1;41:59–64.  7.  Joint United Nations Programme on HIV/AIDS (UNAIDS). Harm reduction saves lives [Internet]. Geneva, Switzerland; 2017 [cited 2018 Nov 10]. Available from: http://www.unaids.org/sites/default/files/media_asset/harm-reduction-saves-lives_en.pdf 8.  Hagan H, Pouget ER, Des Jarlais DC, Lelutiu-Weinberger C. Meta-regression of Hepatitis C virus infection in relation to time since onset of illicit drug injection: The influence of time and place. American Journal of Epidemiology. 2008 Nov 15;168(10):1099–109.  9.  Mathers BM, Degenhardt L, Bucello C, Lemon J, Wiessing L, Hickman M. Mortality among people who inject drugs: A systematic review and meta-analysis. Bulletin of the World Health Organization. 2013 Feb 1;91(2):102–23.  10.  National Institute on Drug Abuse. Overdose death rates [Internet]. 2018 [cited 2018 Aug 30]. Available from: https://www.drugabuse.gov/related-topics/trends-statistics/overdose-death-rates          153 11.  Lemstra M, Rogers M, Thompson A, Moraros J, Buckingham R. Risk indicators of depressive symptomatology among injection drug users and increased HIV risk behaviour. Canadian Journal of Psychiatry. 2011 Jun 1;56(6):358–66.  12.  Health Canada. Overview of national data on opioid-related harms and deaths [Internet]. 2018 [cited 2018 Sep 1]. Available from: https://www.canada.ca/en/health-canada/services/substance-use/problematic-prescription-drug-use/opioids/data-surveillance-research/harms-deaths.html 13.  Murphy S, Xu J, Kochanek K, Arias E. Mortality in the United States, 2017- NCHS Data Briefs No. 328 - [Internet]. Centers for Disease Control and Prevention; 2018 Nov [cited 2018 Dec 10]. Available from: https://www.cdc.gov/nchs/products/databriefs/db328.htm 14.  Public Health Agency of Canada. The Chief Public Health Officer’s report on the state of public health in Canada 2018: Preventing problematic substance use in youth [Internet]. 2018 Oct [cited 2018 Dec 10]. Available from: https://www.canada.ca/en/public-health/corporate/publications/chief-public-health-officer-reports-state-public-health-canada/2018-preventing-problematic-substance-use-youth.html#a1.2 15.  Fairbairn N, Milloy M-J, Zhang R, Lai C, Grafstein E, Kerr T, et al. Emergency department utilization among a cohort of HIV-positive injecting drug users in a Canadian Setting. The Journal of Emergency Medicine. 2012 Aug;43(2):236–43.  16.  Lloyd-Smith E, Wood E, Zhang R, Tyndall MW, Sheps S, Montaner JS, et al. Determinants of hospitalization for a cutaneous injection-related infection among injection drug users: A cohort study. BMC Public Health. 2010;10(1):327.  17.  Nelson PK, Mathers BM, Cowie B, Hagan H, Des Jarlais D, Horyniak D, et al. Global epidemiology of hepatitis B and hepatitis C in people who inject drugs: Results of systematic reviews. The Lancet. 2011 Aug 13;378(9791):571–83.  18.  van Boekel LC, Brouwers EPM, van Weeghel J, Garretsen HFL. Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: Systematic review. Drug and Alcohol Dependence. 2013 Jul 1;131(1):23–35.  19.  Freeman K, Jones CG, Weatherburn DJ, Rutter S, Spooner CJ, Donnelly N. The impact of the Sydney Medically Supervised Injecting Centre (MSIC) on crime. Drug and Alcohol Review. 2005 Mar 1;24(2):173–84.           154 20.  Wood E, Kerr T, Small W, Li K, Marsh DC, Montaner JSG, et al. Changes in public order after the opening of a medically supervised safer injecting facility for illicit injection drug users. Canadian Medical Association Journal. 2004 Sep 28;171(7):731–4.  21.  Wood E, Tyndall MW, Lai C, Montaner JS, Kerr T. Impact of a medically supervised safer injecting facility on drug dealing and other drug-related crime. Substance Abuse Treatment, Prevention, and Policy. 2006;1:13.  22.  Rhodes T. The ‘risk environment’: A framework for understanding and reducing drug-related harm. International Journal of Drug Policy. 2002 Jun;13(2):85–94.  23.  Rhodes T, Kimber J, Small W, Fitzgerald J, Kerr T, Hickman M, et al. Public injecting and the need for ‘safer environment interventions’ in the reduction of drug-related harm. Addiction. 2006 Oct;101(10):1384–93.  24.  Dietze P, Jolley D, Fry CL, Bammer G, Moore D. When is a little knowledge dangerous?: Circumstances of recent heroin overdose and links to knowledge of overdose risk factors. Drug and Alcohol Dependence. 2006 Oct 1;84(3):223–30.  25.  McNeil R, Small W. ‘Safer environment interventions’: A qualitative synthesis of the experiences and perceptions of people who inject drugs. Social Science & Medicine. 2014 Apr;106:151–8.  26.  Rhodes T, Singer M, Bourgois P, Friedman SR, Strathdee SA. The social structural production of HIV risk among injecting drug users. Social Science & Medicine. 2005 Sep;61(5):1026–44.  27.  Potier C, Laprévote V, Dubois-Arber F, Cottencin O, Rolland B. Supervised injection services: What has been demonstrated? A systematic literature review. Drug and Alcohol Dependence. 2014 Dec 1;145:48–68.  28.  European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). Drug consumption rooms: An overview of provision and evidence [Internet]. 2018 [cited 2018 Jun 28]. Available from: http://www.emcdda.europa.eu/topics/pods/drug-consumption-rooms 29.  Health Canada. Supervised consumption sites: Status of applications [Internet]. 2018 [cited 2018 Dec 1]. Available from: https://www.canada.ca/en/health-canada/services/substance-abuse/supervised-consumption-sites/status-application.html          155 30.  Schäeffer D, Stöever H, Weichert L. Drug consumption rooms in Europe: Models, best practices and challenges [Internet]. Amsterdam: European Harm Reduction Network; 2014 [cited 2018 Jan 9]. Available from: http://www.eurohrn.eu/images/stories/pdf/publications/dcr_europe.pdf 31.  Correlation Network, De Regenboog Groep Amsterdam, International Network of Drug Consumption Rooms. DCRs: Geography of safe spaces [Internet]. 2018 [cited 2018 Oct 20]. Available from: https://drive.google.com/file/d/1oIVCpKGUFdChAcoOEJi4s8fYpmGUcfPI/view?fbclid=IwAR05TwgsE-vwupPL7cHVm-MAvEbq57BuDrI3HPjHP8jic2Qeht6-8r2dnUA 32.  Romero C. Mexicali offers safe spot for heroin addicts. El Universal [Internet]. 2018 Nov 23 [cited 2018 Nov 30]; Available from: https://www.eluniversal.com.mx/english/mexicali-offers-safe-spot-for-heroin-addicts 33.  Daphne Bramham. Portugal’s goal is setting drug users on a path to abstinence. Vancouver Sun. 2018 Sep 14 [cited 2019 Feb 19]; Available from: https://vancouversun.com/news/national/portugals-goal-is-setting-setting-drug-users-on-a-path-to-abstinence 34.  La première salle belge de consommation à moindre risque pour toxicomanes ouvre ce mercredi à Liège. TodayInLiege [Internet]. 2018 Sep 3 [cited 2018 Dec 18]; Available from: https://www.todayinliege.be/la-premiere-salle-belge-de-consommation-a-moindre-risque-pour-toxicomanes-ouvre-ce-mercredi-a-liege/ 35.  Kennedy MC, Kerr T. Overdose prevention in the United States: A call for supervised injection sites. American Journal of Public Health. 2017 Jan 1;107(1):42–3.  36.  Highleyman L. U.S. cities consider supervised injection sites. Medpage Today [Internet]. 2018 May 31 [cited 2018 Nov 18]; Available from: https://www.medpagetoday.com/reading-room/aga/lower-gi/73190 37.  Hedrich D, Kerr T, Dubois-Arber F. Drug consumption facilities in Europe and beyond. In: Rhodes T, Hedrich D, editors. Harm reduction: Evidence, impacts, and challenges. Lisbon: European Monitoring Centre for Drugs and Drug Addiction; 2010. p. 306–31.           156 38.  Woods S. Drug consumption rooms in Europe: Organizational overview [Internet]. Peacey J, Geise M, editors. European Harm Reduction Network. Regenboog Groep, Amsterdam; 2014 [cited 2017 Jun 14]. Available from: http://www.drugconsumptionroom-international.org/images/pdf/dcr_in_europe.pdf 39.  Rhodes T. Risk environments and drug harms: A social science for harm reduction approach. International Journal of Drug Policy. 2009 May;20(3):193–201.  40.  Strathdee SA, Hallett TB, Bobrova N, Rhodes T, Booth R, Abdool R, et al. HIV and risk environment for injecting drug users: The past, present, and future. The Lancet. 2010 Jul 24;376(9737):268–84.  41.  Rhodes T, Wagner K, Strathdee SA, Shannon K, Davidson P, Bourgois P. Structural violence and structural vulnerability within the risk environment: Theoretical and methodological perspectives for a social epidemiology of HIV risk among injection drug users and sex workers. In: O’Campo P, Dunn JR, editors. Rethinking social epidemiology. Springer Netherlands; 2012  p. 205–30.  42.  Small W, Rhodes T, Wood E, Kerr T. Public injection settings in Vancouver: Physical environment, social context and risk. International Journal of Drug Policy. 2007 Jan 1;18(1):27–36.  43.  Kerr T, Small W, Moore D, Wood E. A micro-environmental intervention to reduce the harms associated with drug-related overdose: Evidence from the evaluation of Vancouver’s safer injection facility. International Journal of Drug Policy. 2007 Jan;18(1):37–45.  44.  Small W, Van Borek N, Fairbairn N, Wood E, Kerr T. Access to health and social services for IDU: The impact of a medically supervised injection facility. Drug and Alcohol Review. 2009 Jul 1;28(4):341–6.  45.  Fischer B, Turnbull S, Poland B, Haydon E. Drug use, risk and urban order: Examining supervised injection sites (SISs) as ‘governmentality.’ International Journal of Drug Policy. 2004 Dec 1;15(5):357–65.  46.  Stoltz J-A, Wood E, Small W, Li K, Tyndall M, Montaner J, et al. Changes in injecting practices associated with the use of a medically supervised safer injection facility. Journal of Public Health. 2007 Mar 1;29(1):35–9.           157 47.  Kerr T, Tyndall M, Li K, Montaner J, Wood E. Safer injection facility use and syringe sharing in injection drug users. The Lancet. 2005 Jul 29;366(9482):316–8.  48.  Marshall BD, Milloy M-J, Wood E, Montaner JS, Kerr T. Reduction in overdose mortality after the opening of North America’s first medically supervised safer injecting facility: A retrospective population-based study. The Lancet. 2011 Apr 29;377(9775):1429–37.  49.  Donnelly N, Mahoney N. Trends in property and illicit drug crime around the Medically Supervised Injecting Centre in Kings Cross: 2012 update [Internet]. 2013 [cited 2017 May 27]. Available from: http://www.bocsar.nsw.gov.au/Documents/BB/bb90.pdf 50.  DeBeck K, Kerr T, Bird L, Zhang R, Marsh D, Tyndall M, et al. Injection drug use cessation and use of North America’s first medically supervised safer injecting facility. Drug and Alcohol Dependence. 2011 Jan 15;113(2–3):172–6.  51.  Wood E, Tyndall MW, Zhang R, Stoltz J-A, Lai C, Montaner JSG, et al. Attendance at supervised injecting facilities and use of detoxification services.  New England Journal of Medicine. 2006 Jun 8;354(23):2512–4.  52.  Gaddis A, Kennedy MC, Nosova E, Milloy M-J, Hayashi K, Wood E, et al. Use of on-site detoxification services co-located with a supervised injection facility. Journal of Substance Abuse Treatment. 2017 Nov;82:1–6.  53.  Wood E, Tyndall MW, Zhang R, Montaner JSG, Kerr T. Rate of detoxification service use and its impact among a cohort of supervised injecting facility users. Addiction. 2007 Jun 1;102(6):916–9.  54.  Kerr T, Mitra S, Kennedy MC, McNeil R. Supervised injection facilities in Canada: Past, present, and future. Harm Reduction Journal. 2017;14:28.  55.  McGinty EE, Barry CL, Stone EM, Niederdeppe J, Kennedy-Hendricks A, Linden S, et al. Public support for safe consumption sites and syringe services programs to combat the opioid epidemic. Preventive Medicine. 2018 Jun 1;111:73–7.  56.  Jauffret-Roustide M, Pedrono G, Beltzer N. Supervised consumption rooms: The French Paradox. International Journal of Drug Policy. 2013 Nov;24(6):628–30.  57.  Ontario Ministry of Health and Long-Term Care. Consumption and treatment services: Application guide [Internet]. 2018 Oct [cited 2018 Nov 10]. Available from:          158 http://health.gov.on.ca/en/pro/programs/opioids/docs/CTS_application_guide_en.pdf 58.  Pagliaro J. Supervised injection sites face new roadblocks in Toronto with provincial guidelines. The Star [Internet]. 2018 Nov 6 [cited 2018 Nov 11]; Available from: https://www.thestar.com/news/city_hall/2018/11/06/supervised-injection-sites-face-new-roadblocks-in-toronto-with-provincial-guidelines.html 59.  Weeks C. Ottawa bypasses Ontario’s rules on supervised drug consumption sites. The Globe and Mail; Toronto, Ont. 2018 Dec 14;A.1.  60.  Zlotorzynska M, Wood E, Montaner JS, Kerr T. Supervised injection sites: Prejudice should not trump evidence of benefit. Canadian Medical Association Journal. 2013 Oct 15;185(15):1303–4.  61.  Faher M. State injection site report relied on retracted research. VTDigger [Internet]. 2018 Oct 19 [cited 2018 Dec 6]; Available from: https://vtdigger.org/2018/10/19/state-injection-site-report-relied-retracted-research/ 62.  Barry CL, Sherman SG, Stone E, Kennedy-Hendricks A, Niederdeppe J, Linden S, et al. Arguments supporting and opposing legalization of safe consumption sites in the U.S. International Journal of Drug Policy. 2019 Jan;63:18–22.  63.  Joseph A. Jerome Adams reflects on his first year as U.S. surgeon general. STAT [Internet]. 2018 Sep 20 [cited 2018 Nov 18]; Available from: https://www.statnews.com/2018/09/20/surgeon-general-jerome-adams-year-one/ 64.  Collier R. Clement unleashes barrage at safe injection site supporters. Canadian Medical Association Journal. 2009 Dec 8;181(12):E316–E316.  65.  Shapiro J. Do safe-injection sites offer proven benefits to those facing addiction? Billy Penn [Internet]. 2018 Jan 30 [cited 2018 Dec 19]; Available from: https://billypenn.com/2018/04/05/do-safe-injection-sites-offer-proven-benefits-to-those-facing-addiction/ 66.  Kimber J, MacDonald M, van Beek I, Kaldor J, Weatherburn D, Lapsley H, et al. The Sydney Medically Supervised Injecting Centre: Client characteristics and predictors of frequent attendance during the first 12 months of operation. Journal of Drug Issues. 2003 Jul 1;33(3):639–48.           159 67.  van Beek I, Kimber J, Dakin A, Gilmour S. The Sydney Medically Supervised Injecting Centre: Reducing harm associated with heroin overdose. Critical Public Health. 2004 Dec 1;14(4):391–406.  68.  Wood E, Tyndall MW, Li K, Lloyd-Smith E, Small W, Montaner JSG, et al. Do supervised injecting facilities attract higher-risk injection drug users? American Journal of Preventive Medicine. 2005 Aug;29(2):126–30.  69.  Wood E, Tyndall MW, Qui Z, Zhang R, Montaner JSG, Kerr T. Service uptake and characteristics of injection drug users utilizing North America’s first medically supervised safer injecting facility. American Journal of Public Health. 2006 May 1;96(5):770–3.  70.  Zurhold H, Degkwitz P, Verthein U, Haasen C. Drug consumption rooms in Hamburg, Germany: Evaluation of the effects on harm reduction and the reduction of public nuisance.  Journal of Drug Issues. 2003 Jul 1;33(3):663–88.  71.  Small W, Shoveller J, Moore D, Tyndall M, Wood E, Kerr T. Injection drug users’ access to a supervised injection facility in Vancouver, Canada: The influence of operating policies and local drug culture. Qualitative Health Research. 2011 Jun;21(6):743–56.  72.  Rosenstein RJ. Fight drug abuse, don’t subsidize it. The New York Times [Internet]. 2018 Aug 27 [cited 2018 Nov 16]; Available from: https://www.nytimes.com/2018/08/27/opinion/opioids-heroin-injection-sites.html 73.  Highleyman L. Supervised injection sites reduce drug-related harm. Medpage Today [Internet]. 2018 Oct 23 [cited 2018 Nov 18]; Available from: https://www.medpagetoday.com/hivaids/hivaids/75871 74.  The Canadian Press. Ontario to review safe injection, overdose prevention sites, health minister says. CBC News [Internet]. 2018 Jul 24 [cited 2018 Nov 18]; Available from: https://www.cbc.ca/news/canada/toronto/ontario-safe-injection-sites-limbo-1.4760002 75.  Poschadel S, Höger R, Schnitzler J, Schreckenberger J. Evaluation der arbeit der drogenkonsumräume in der bundesrepublik Deutschland: Endbericht im auftrag des bundesministeriums für Gesundheit, Das bundesministerium für Gesundheit und soziale sicherung (Schriftenreihe Bd 149). Baden-Baden: Nomos- Verlags-Gesellschaft; 2003.           160 76.  Richardson LA, Long C, DeBeck K, Nguyen P, Milloy M-JS, Wood E, et al. Socioeconomic marginalisation in the structural production of vulnerability to violence among people who use illicit drugs. Journal of Epidemiology & Community Health. 2015 Jul 1;69(7):686–92.  77.  Boyd J, Collins AB, Mayer S, Maher L, Kerr T, McNeil R. Gendered violence and overdose prevention sites: A rapid ethnographic study during an overdose epidemic in Vancouver, Canada. Addiction. 2018;12:2261-70 78.  Wood E, Kerr T. What do you do when you hit rock bottom? Responding to drugs in the city of Vancouver. International Journal of Drug Policy. 2006 Mar 1;17(2):55–60.  79.  Himsworth CG, Bidulka J, Parsons KL, Feng AYT, Tang P, Jardine CM, et al. Ecology of leptospira interrogans in Norway rats (rattus norvegicus) in an inner-city neighborhood of Vancouver, Canada. PLOS Neglected Tropical Diseases. 2013 Jun 20;7(6):e2270.  80.  Campbell L, Boyd N, Culbert L. A thousand dreams: Vancouver’s Downtown Eastside and the fight for its future. Greystone Books; 2009.  81.  Strathdee SA, Patrick DM, Currie SL, Cornelisse PGA, Rekart ML, Montaner JSG, et al. Needle exchange is not enough: Lessons from the Vancouver injecting drug use study. AIDS. 1997 Aug;11(8):F59.  82.  Canada (Attorney General) v. PHS Community Services Society, 2011 SCC 44. 2011.  83.  Blanchfield M. Supreme Court OKs Insite Safe Injection Site In Unanimous Ruling. Huffington Post Canada [Internet]. 2011 Sep 30 [cited 2018 Nov 29]; Available from: http://www.huffingtonpost.ca/2011/09/30/supreme-court-insite-unanimous-ruling_n_988733.html 84.  Butler M, Phillips K. Bill C-2: The Respect for Communities Act [Internet]. Ottawa, ON: Parliament of Canada; [cited 2016 Mar 1]. Available from: http://www.parl.gc.ca/HousePublications/Publication.aspx?Language=E&Mode=1&DocId=8056955&File=24#1 85.  Valiante G. Supervised injection sites: Quebec expects federal approval. CBC [Internet]. 2015 Nov 15 [cited 2018 Nov 11]; Available from:          161 https://www.cbc.ca/news/canada/montreal/safe-injection-sites-supervised-trudeau-canada-government-quebec-1.3319725 86.  BC Coroners Service. Illicit drug overdose deaths in BC (January 1, 2008 - November 30, 2018) [Internet]. Burnaby, BC: Office of the Chief Coroner; 2018 [cited 2018 Dec 12]. Available from: https://www2.gov.bc.ca/assets/gov/birth-adoption-death-marriage-and-divorce/deaths/coroners-service/statistical/illicit-drug.pdf 87.  Parliament of Canada. Bill C-37: An Act to amend the Controlled Drugs and Substances Act and to make related amendments to other Acts [Internet]. 2017 [cited 2018 Nov 11]. Available from: https://www.parl.ca/DocumentViewer/en/42-1/bill/C-37/royal-assent 88.  Pivot Legal Society. Canada’s supervised consumption and overdose prevention sites [Internet]. Pivot Legal Society. 2018 [cited 2018 Aug 21]. Available from: http://www.pivotlegal.org/scs_map 89.  Vancouver Coastal Health. Overdose prevention & response [Internet]. 2018 [cited 2018 Aug 21]. Available from: http://www.vch.ca/public-health/harm-reduction/overdose-prevention-response 90.  McNeil R, Small W, Wood E, Kerr T. Hospitals as a ‘risk environment’: An ethno-epidemiological study of voluntary and involuntary discharge from hospital against medical advice among people who inject drugs. Social Science & Medicine. 2014 Mar;105:59–66.  91.  Small W, Moore D, Shoveller J, Wood E, Kerr T. Perceptions of risk and safety within injection settings: Injection drug users’ reasons for attending a supervised injecting facility in Vancouver, Canada. Health, Risk & Society. 2012 Jun;14(4):307–24.  92.  McNeil R, Small W, Lampkin H, Shannon K, Kerr T. “People knew they could come here to get help”: An ethnographic study of assisted injection practices at a peer-run ‘unsanctioned’ supervised drug consumption room in a Canadian setting. AIDS & Behavior. 2014;18(3):473–85.  93.  Milloy M-JS, Marshall BDL, Kerr T, Buxton J, Rhodes T, Montaner J, et al. Social and structural factors associated with HIV disease progression among illicit drug users: A systematic review. AIDS. 2012 Jun;26(9):1049–63.           162 94.  Petrar S, Kerr T, Tyndall MW, Zhang R, Montaner JSG, Wood E. Injection drug users’ perceptions regarding use of a medically supervised safer injecting facility. Addictive Behaviors. 2007 May;32(5):1088–93.  95.  Rapid Response Service. What is the effectiveness of supervised injection services? [Internet]. Ontario HIV Treatment Network; 2014 [cited 2017 Jun 8]. Available from: http://www.ohtn.on.ca/Pages/Knowledge-Exchange/Rapid-Responses/Documents/RR83-Supervised-Injection-Effectiveness.pdf 96.  Strathdee SA, Palepu A, Cornelisse PGA, Yip B, O’Shaughnessy MV, Montaner JSG, et al. Barriers to use of free antiretroviral therapy in injection drug users. Journal of the American Medical Association.  1998 Aug 12;280(6):547–9.  97.  Wood E, Tyndall MW, Spittal PM, Li K, Kerr T, Hogg RS, et al. Unsafe injection practices in a cohort of injection drug users in Vancouver: Could safer injecting rooms help? Canadian Medical Association Journal. 2001 Aug 21;165(4):405–10.  98.  Ti L, Kerr T, Wood E, Werb D, DeBeck K, Graham D, et al. Drug situation in Vancouver. [Internet]. Urban Health Research Initiative of the British Columbia Centre for Excellence in HIV/AIDS. 2013 [cited 2018 Dec 7]. Available from: http://cfenet.ubc.ca/sites/default/files/uploads/news/releases/war_on_drugs_failing_to_limit_drug_use.pdf 99.  World Health Organization. International statistical classification of diseases and health related problems (ICD-10). 2004.  100.  Handwerker PW. Quick ethnography: A guide to rapid multi-method research. Rowman Altamira; 2001. 311 p.  101.  Johnson GA, Vindrola-Padros C. Rapid qualitative research methods during complex health emergencies: A systematic review of the literature. Social Science & Medicine. 2017 Sep 1;189:63–75.  102.  Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: Explanation and elaboration. British Medical Journal.  2009 Jul 21;339:b2700.  103.  National Heart, Blood and Lung Institute. Quality assessment tool for observational cohort and cross-sectional studies. [Internet]. 2014 [cited 2017 May          163 14]. Available from: http://www.nhlbi.nih.gov/health-pro/guidelines/in-develop/ cardiovascular-risk-reduction/tools/cohort 104.  National Heart, Blood and Lung Institute. Quality assessment tool for before-after (pre-post) studies. [Internet]. 2014 [cited 2017 May 14]. Available from: https://www.nhlbi.nih.gov/health-pro/guidelines/in-develop/cardiovascular-risk-reduction/tools/before-after 105.  Joanna Briggs Institute. Joanna Briggs Institute Critical Appraisal Checklist for Economic Evaluations. [Internet]. 2016 [cited 2017 May 14]. Available from: http://joannabriggs.org/research/critical-appraisal-tools.html 106.  Milloy M-JS, Kerr T, Mathias R, Zhang R, Montaner JS, Tyndall M, et al. Non-fatal overdose among a cohort of active injection drug users recruited from a supervised injection facility. The American Journal of Drug and Alcohol Abuse. 2008 Jan 1;34(4):499–509.  107.  Milloy M-J, Kerr T, Tyndall M, Montaner J, Wood E. Estimated drug overdose deaths averted by North America’s first medically-supervised safer injection facility. PLoS ONE. 2008;3(10):e3351 108.  Salmon AM, Van Beek I, Amin J, Kaldor J, Maher L. The impact of a supervised injecting facility on ambulance call-outs in Sydney, Australia. Addiction. 2010 Apr 1;105(4):676–83.  109.  Hedrich D. European report on drug consumption rooms [Internet]. European Monitoring Centre for Drugs and Drug Addiction, Lisbon; 2004 [cited 2017 May 20]. Available from: http://www.emcdda.europa.eu/html.cfm/index54125EN.html 110.  NCHECR. Sydney Medically Supervised Injecting Centre evaluation report no. 4: evaluation of service operation and overdose-related events [Internet]. Sydney: National Centre in HIV Epidemiology and Clinical Research, University of New South Wales; 2007 [cited 2017 Jun 7]. Available from: https://kirby.unsw.edu.au/report/sydney-medically-supervised-injecting-centre-msic-evaluation-report-4 111.  Bravo MJ, Royuela L, De la Fuente L, Brugal MT, Barrio G, Domingo-Salvany A, et al. Use of supervised injection facilities and injection risk behaviours among young drug injectors. Addiction. 2009 Apr 1;104(4):614–9.           164 112.  Wood E, Tyndall MW, Stoltz J-A, Small W, Lloyd-Smith E, Zhang R. Factors associated with syringe sharing among users of a medically supervised safer injecting facility. American Journal of Infectious Diseases. 2005 Jan 1;1(1):50–4.  113.  Kinnard EN, Howe CJ, Kerr T, Skjødt Hass V, Marshall BD. Self-reported changes in drug use behaviors and syringe disposal methods following the opening of a supervised injecting facility in Copenhagen, Denmark. Harm Reduction Journal. 2014;11:29.  114.  Scherbaum N, Specka M, Schifano F, Bombeck J, Marrziniak B. Longitudinal observation of a sample of German drug consumption facility clients. Substance Use & Misuse. 2010 Jan 1;45(1–2):176–89.  115.  Kerr T, Stoltz J-A, Tyndall M, Li K, Zhang R, Montaner J, et al. Impact of a medically supervised safer injection facility on community drug use patterns: A before and after study. British Medical Journal. 2006;332(7535):220–2.  116.  Kerr T, Tyndall M., Zhang R, Lai C, Montaner J., Wood E. C Circumstances of first injection among illicit drug users accessing a medically supervised safer injection facility. American Journal of Public Health. 2007;97(7).  117.  Lloyd-Smith E, Wood E, Zhang R, Tyndall MW, Montaner JS, Kerr T. Risk factors for developing a cutaneous injection-related infection among injection drug users: A cohort study. BMC Public Health. 2008;8(1):405.  118.  Richardson L, Wood E, Zhang R, Montaner J, Tyndall M, Kerr T. Employment among users of a medically supervised safer injection facility. The American Journal of Drug and Alcohol Abuse. 2008 Jan 1;34(5):519–25.  119.  Marshall BDL, Wood E, Zhang R, Tyndall MW, Montaner JSG, Kerr T. Condom use among injection drug users accessing a supervised injecting facility. Sexually Transmitted Infections. 2009 Apr 1;85(2):121–6.  120.  Kimber J, Kimber J, Mattick RP, Kimber J, Mattick RP, Kaldor J, et al. Process and predictors of drug treatment referral and referral uptake at the Sydney Medically Supervised Injecting Centre. Drug and Alcohol Review. 2008 Jan 1;27(6):602–12.  121.  Milloy M-JS, Kerr T, Zhang R, Tyndall M, Montaner J, Wood E. Inability to access addiction treatment and risk of HIV infection among injection drug users recruited from a supervised injection facility. Journal of Public Health. 2010 Sep 1;32(3):342–9.           165 122.  Toth EC, Tegner J, Lauridsen S, Kappel N. A cross-sectional national survey assessing self-reported drug intake behavior, contact with the primary sector and drug treatment among service users of Danish drug consumption rooms. Harm Reduction Journal. 2016;13:27.  123.  Baars JE, Boon BJF, Garretsen HFL, van de Mheen D. The reach of a free hepatitis B vaccination programme: Results of a Dutch study among drug users. International Journal of Drug Policy. 2010 May;21(3):247–50.  124.  Wood RA, Wood E, Lai C, Tyndall MW, Montaner JSG, Kerr T. Nurse-delivered safer injection education among a cohort of injection drug users: Evidence from the evaluation of Vancouver’s supervised injection facility. International Journal of Drug Policy. 2008 Jun;19(3):183–8.  125.  Lysyshn M, Dohoo C, Forsting S, Kerr T, McNeil R. Evaluation of a fentanyl drug checking program for clients of a supervised injection site, Vancouver, Canada. In: 25th Harm Reduction International Conference Montreal, May 14-17 Abstract 188. 2017. Available from: https://www.hri.global/abstracts/abstrct/188/print 126.  Skelton E, Bonevski B, Tzelepis F, Shakeshaft A, Guillaumier A, Wood W, et al. Addressing tobacco smoking in a medically supervised injecting center with an organizational change intervention: An acceptability study. Asia-Pacific Journal of Clinical Oncology. 2016;12(Suppl. 6):13.  127.  Vecino C, Villalbi J., Guitart A, Espelt A, Bartroli M, Castellano Y, et al. Apertura de espacios de consumo higienico y actuaciones policiales en zonas con fuerte trafico de drogas. Evaluacion mediante el recuento de las jeringas abandonadas en el espacio publico [Safe injection rooms and police crackdowns in areas with heavy drug dealing. Evaluation by counting discarded syringes collected from the public space]. Adicciones. 2013;25(4):333–8.  128.  Salmon A., Thein H., Kimber J, Kaldor J., Maher L. Five years on: What are the community perceptions of drug-related public amenity following the establishment of the Sydney Medically Supervised Injecting Centre? International Journal of Drug Policy. 2007;18(1):46–53.  129.  Thein H., Kimber J, Maher L, MacDonald M, Kaldor J. Public opinion towards supervised injecting centres and the Sydney Medically Supervised Injecting Centre. International Journal of Drug Policy. 2005;16(4):275–80.           166 130.  Freeman K, Jones C., Weatherburn D., Rutter S, Spooner C., Donnelly N. The impact of the Sydney Medically Supervised Injecting Centre (MSIC) on crime. Drug and Alcohol Review. 2005;24(2):173–84.  131.  Fitzgerald J, Burgess M, Snowball L. Trends in property and illicit drug crime around the Medically Supervised Injecting Centre in Kings Cross: An update [Internet]. 2010 [cited 2017 May 27]. Available from: http://www.bocsar.nsw.gov.au/Documents/BB/bb51.pdf 132.  Milloy M, Wood E, Tyndall M, Lai C, Montaner J, Kerr T. Recent incarceration and use of a supervised injection facility in Vancouver, Canada. Addiction Research & Theory. 2009 Jan 1;17(5):538–45.  133.  Andresen MA, Jozaghi E. The point of diminishing returns: An examination of expanding Vancouver’s Insite. Urban Studies. 2012 Dec 1;49(16):3531–44.  134.  Bayoumi AM, Zaric GS. The cost-effectiveness of Vancouver’s supervised injection facility. Canadian Medical Association Journal. 2008 Nov 18;179(11):1143–51.  135.  Pinkerton SD. How many HIV infections are prevented by Vancouver Canada’s supervised injection facility? International Journal of Drug Policy. 2011 May;22(3):179–83.  136.  Pinkerton S. Is Vancouver Canada’s supervised injection facility cost-saving? Addiction. 2010;105(8):1429–36.  137.  Andresen M., Boyd N. A cost-benefit and cost-effectiveness analysis of Vancouver’s supervised injection facility. International Journal of Drug Policy. 2010;21(1):70–6.  138.  Jozaghi E, Vancouver Area Network of Drug Users. A cost-benefit/cost-effectiveness analysis of an unsanctioned supervised smoking facility in the Downtown Eastside of Vancouver, Canada. Harm Reduction Journal. 2014;11(1):30.  139.  Center for Disease Control and Prevention, National Vital Statistics System, Mortality File. Number and age-adjusted rates of drug-poisoning deaths involving opioid analgesics and heroin: United States, 2000–2014 [Internet]. Atlanta, GA: Center for Disease Control and prevention; 2015 [cited 2018 Jan 8]. Available from:          167 http://www.cdc.gov/nchs/data/health_policy/AADR_drug_poisoning_involving_OA_Heroin_US_2000-2014.pdf 140.  Milloy M-J, Wood E. Emerging role of supervised injecting facilities in human immodeficiency virus prevention. Addiction. 2009 Apr 1;104(4):620–1.  141.  Minder Nejedly M, Bürki CM. Monitoring HIV risk behaviours in a street agency with injection room in Switzerland. [Berne]: Medical Faculty of the University in Berne; 1999.  142.  Reyes Fuentes V. 15 Jahre Fixerraum Bern. Auswirkungen auf soziale und medizinische Aspekte bei Drogenabhängigen. [Berne]: Medical Faculty of the University of Berne; 2003.  143.  Ronco S, Spuler G, Coda P, Schöpfer R. Evaluation der Gassenzimmer I, II und III in Basel. Sozial- und Präventivmedizin. 1996;41(Supplement 1):S58–S68.  144.  Ronco C, Spuhler G, Kaiser R. Evaluation des “Aufenthalts- und Betreuungsraums für Drogenabhängige” in Luzern [Evaluation of a stay and care center for drug addicts in Lucerne]. Sozial- und Präventivmedizin. 1996;41(Supplement 1):S45–S57.  145.  Benninghoff F, Solai S, Huissoud T, Dubois-Arber F. Evaluation de Quai 9 ‘Espace d’acceuil et d’injection’ à Genéve: période 12/2001–12/2000, Raisons de santé 103,. Lausanne: Institut universitaire de médecine sociale et préventive; 2003.  146.  Benninghoff F, Geense R, Dubois-Arber F. Resultats de l’étude ‘La clientèle des structures à bas seuil d’accessibilité en Suisse. Lausanne: Institut universitaire de médecine sociale et préventive; 2001.  147.  Benninghoff F, Dubois-Arber F. Résultats de l’étude de la clientèle du Cactus BIEL/BIENNE 2001. Lausanne: Institut universitaire de médecine sociale et préventive; 2002.  148.  Solai S, Benninghoff B, Meystre-Agustoni G, Jeannin A, Dubois-Arber F. Evaluation de l’espace d’accueil et d’injection “Quai 9” à Genève: deuxième phase 2003. Raisons de santé. 2004;102, Institut universitaire de médecine sociale et préventive, Lausanne.  149.  Jacob J, Rottman J, Stöver H. Entstehung und Praxis eines Gesundheitsraumangebotes für Drogenkonsumierende. Abschlußbericht der          168 einjährigen Evaluation des ‘drop-in Fixpunkt’ in Hannover. Schriftenreihe Sucht- und Drogenforschung. 1999;2, BIS-Verlag, Universität Oldenburg, Oldenburg.  150.  van der Poel A, Barendregt C, van de Mheen D. Drug consumption rooms in Rotterdam: An explorative description. European Addiction Research. 2003;9:94–100.  151.  Zurhold H, Kreuzfeld N, Degkwitz P, Verthein U. Drogenkonsumräume. Gesundheitsförderung und Minderung öffentlicher Belastungen in europäischen Grossstädten, Lambertus, Freiburg. 2001;  152.  Aspinall EJ, Nambiar D, Goldberg DJ, Hickman M, Weir A, Van Velzen E, et al. Are needle and syringe programmes associated with a reduction in HIV transmission among people who inject drugs: A systematic review and meta-analysis. International Jorunal of Epidemiology. 2014 Feb 1;43(1):235–48.  153.  Abdul-Quader AS, Feelemyer J, Modi S, Stein ES, Briceno A, Semaan S, et al. Effectiveness of structural-level needle/syringe programs to reduce HCV and HIV infection among people who inject drugs: A systematic review. AIDS & Behavior.  2013 Nov 1;17(9):2878–92.  154.  Wood W. Distributing “take home” naloxone via Sydney medically supervised injecting centre: Where to from here? Paper 187. Drug and Alcohol Review; 2015;34:67. 155.  Collins A, Parashar S, Hogg R, Fernando S, Worthington C, McDougall P, et al. Integrated HIV care and service engagement among people living with HIV who use drugs in a setting with a community-wide treatment as prevention initiative: A qualitative study in Vancouver, Canada. Journal of the International AIDS Society. 2017;20(1):21407. 156.  Ti L, Dong H, Kerr T, Turje R, Parashar S, Min J, et al. The effect of engagement in an HIV/AIDS integrated health programme on plasma HIV-1 RNA suppression among HIV-positive people who use illicit drugs: A marginal structural modelling analysis. HIV Medicine. 2017;18(8):580-6. 157.  McNeil R, Dilley LB, Guirguis-Younger M, Hwang SW, Small W. Impact of supervised drug consumption services on access to and engagement with care at a palliative and supportive care facility for people living with HIV/AIDS: A qualitative study. Journal of the International AIDS Society. 2014;17(1):18855.          169 158.  Kappel N, Toth E, Tegner J, Lauridsen S. A qualitative study of how Danish drug consumption rooms influence health and well-being among people who use drugs. Harm Reduction Journal. 2016;13:20.  159.  Meyer JP, Althoff AL, Altice FL. Optimizing care for HIV-infected people who use drugs: Evidence-based approaches to overcoming healthcare disparities. Clinical Infectious Diseases. 2013 Nov 1;57(9):1309–17.  160.  Linssen. Gebruiksruimten. Een systematisch overzicht van de voorziening en de effecten ervan. 2003;  161.  Meijer G, de Jong A, Koeter M, Bielman B. Gebruik van de straat. Evaluatie gebruiksruimnte Binnenstad-Zuid Groningen. Groningen-Rotterdam: INTRAVAL; 2001.  162.  Spreyermann C, Willen C. Evaluationsbericht Öffnung der Kontakt- und Anlaufstellen für risikoärmere Konsumformen. Evaluation der Inhalationsräume der Kontakt- und Anlaufstellen Selnau und Seilergraben der Ambulanten Drogenhilfe Zürich [Internet]. 2003 [cited 2017 May 27]. Available from: http://www.indro-online.de/zuerich.pdf 163.  BC Centre for Excellence in HIV/AIDS. Evaluation of the supervised injection site: year one summary [Internet]. BC Centre for Excellence in HIV/AIDS, Vancouver; 2004. Available from: http://www.salledeconsommation.fr/_media/bc-centre-for-excellence-in-hiv-aids-evaluation-of-the-supervised-injection-site-year-one-summary-september-17-2004.pdf 164.  Christie T, Wood E, Schechter MT, O’Shaughnessy MV. A comparison of the new federal guidelines regulating supervised injection site research in Canada and the Tri-Council Policy Statement on Ethical Conduct for Research Involving Human Subjects. International Journal of Drug Policy. 2004 Feb 1;15(1):66–73.  165.  Maher APL, Maher APL, Salmon A, Maher APL, Salmon A. Supervised injecting facilities: How much evidence is enough? Drug and Alcohol Review. 2007 Jan 1;26(4):351–3.  166.  Worrall J. Evidence: Philosophy of science meets medicine. Journal of Evaluation in Clinical Practice. 2010 Apr 1;16(2):356–62.  167.  Burns PB, Rohrich RJ, Chung KC. The levels of evidence and their role in evidence-based medicine. Plastic Reconstruction Surgery.  2011 Jul;128(1):305–10.           170 168.  Parkhurst JO, Abeysinghe S. What constitutes “good” evidence for public health and social policy-making? From hierarchies to appropriateness.  Social Epistemology. 2016 Nov 1;30(5–6):665–79.  169.  Fairbairn N, Small W, Shannon K, Wood E, Kerr T. Seeking refuge from violence in street-based drug scenes: Women’s experiences in North America’s first supervised injection facility. Social Science & Medicine. 2008 Sep;67(5):817–23.  170.  Solai S, Dubois-Arber F, Benninghoff F, Benaroyo L. Ethical reflections emerging during the activity of a low threshold facility with supervised drug consumption room in Geneva, Switzerland. International Journal of Drug Policy. 2006 Jan 1;17(1):17–22.  171.  Wood E, Spittal PM, Kerr T, Small W, Tyndall MW, O’Shaughnessy MV, et al. Requiring help injecting as a risk factor for HIV Infection in the Vancouver epidemic: Implications for HIV prevention. Canadian Journal of Public Health / Revue Canadienne de Sante’e Publique. 2003;94(5):355–9.  172.  Fairbairn N, Small W, Van Borek N, Wood E, Kerr T. Social structural factors that shape assisted injecting practices among injection drug users in Vancouver, Canada: A qualitative study. Harm Reduction Journal. 2010;7:20.  173.  O’Connell JM, Kerr T, Li K, Tyndall MW, Hogg RS, Montaner JS, et al. Requiring help injecting independently predicts incident HIV infection among injection drug users. Journal of Acquired Immunodeficiency Syndrome. 2005 Sep 1;40(1):83–8.  174.  Kerr T, Fairbairn N, Tyndall M, Marsh D, Li K, Montaner J, et al. Predictors of non-fatal overdose among a cohort of polysubstance-using injection drug users. Drug & Alcohol Dependence.  2007 Feb 23;87(1):39–45.  175.  McNeil R, Shannon K, Shaver L, Kerr T, Small W. Negotiating place and gendered violence in Canada’s largest open drug scene. International Journal of Drug Policy. 2014 May;25(3):608–15.  176.  McNeil R, Kerr T, Lampkin H, Small W. “We need somewhere to smoke crack”: An ethnographic study of an unsanctioned safer smoking room in Vancouver, Canada. International Journal of Drug Policy. 2015 Jul;26(7):645–52.  177.  Jozaghi E, Lampkin H, Andresen M. Peer-engagement and its role in reducing the risky behavior among crack and methamphetamine smokers of the Downtown          171 Eastside community of Vancouver, Canada. Harm Reduction Journal.  2016;13(1):19.  178.  Ti L, Buxton J, Harrison S, Dobrer S, Montaner J, Wood E, et al. Willingness to access an in-hospital supervised injection facility among hospitalized people who use illicit drugs. Journal of Hospital Medicine. 2015 May 1;10(5):301–6.  179.  McNeil R, Kerr T, Pauly B, Wood E, Small W. Advancing patient-centered care for structurally vulnerable drug-using populations: A qualitative study of the perspectives of people who use drugs regarding the potential integration of harm reduction interventions into hospitals. Addiction. 2016;11(4):685–94.  180.  Golden RE, Collins CB, Cunningham SD, Newman EN, Card JJ. Overview of structural interventions to decrease injection drug use risk. In: Best evidence structural interventions for HIV prevention. Springer New York; 2013. p. 41–121.  181.  Shannon K, Rusch M, Shoveller J, Alexson D, Gibson K, Tyndall MW. Mapping violence and policing as an environmental–structural barrier to health service and syringe availability among substance-using women in street-level sex work. International Journal of Drug Policy. 2008 Apr 1;19(2):140–7.  182.  Moher D, Liberati A, Tetzlaff J, Altman DG, Group TP. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLOS Medicine. 2009 Jul 21;6(7):e1000097.  183.  Merali F. PCs “playing politics with people’s lives” on injection sites, drug policy expert warns. CBC News [Internet]. 2018 Aug 4 [cited 2018 Oct 20]; Available from: https://www.cbc.ca/news/canada/toronto/supervised-injection-sites-waiting-1.4771143 184.  Strike C, Watson TM, Kolla G, Penn R, Bayoumi AM. Ambivalence about supervised injection facilities among community stakeholders. Harm Reduction Journal. 2015;12(1):26. 185.  Devine J. Harm reduction programs like safe injection sites don’t beat addiction. They just sustain it. National Post. 2017 Mar 2 [cited 2018 Oct 5]; Available from: https://nationalpost.com/opinion/jeremy-devine-harm-reduction-programs-like-safe-injection-sites-dont-beat-addiction-they-just-sustain-it 186.  Archibald C, Ofner M, Strathdee S, Patrick D, Sutherland D, Rekart M, et al. Factors associated with frequent needle exchange program attendance in injection          172 drug users in Vancouver, Canada.  Journal of Acquired Immunodeficiency Syndrome. 1998;17(2):160–6.  187.  Bennett AS, Bell A, Tomedi L, Hulsey EG, Kral AH. Characteristics of an overdose prevention, response, and naloxone distribution program in Pittsburgh and Allegheny County, Pennsylvania. Journal of Urban Health.  2011 Dec 1;88(6):1020–30.  188.  Lin DY, Wei LJ, Yang I, Ying Z. Semiparametric regression for the mean and rate functions of recurrent events. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2000 Nov;62(4):711–30.  189.  Lima VD, Harrigan R, Murray M, Moore DM, Wood E, Hogg RS, et al. Differential impact of adherence on long-term treatment response among naive HIV-infected individuals. AIDS. 2008 Nov 12;22(17):2371–80.  190.  Reddon H, Wood E, Tyndall M, Lai C, Hogg R, Montaner J, et al. Use of North America’s first medically supervised safer injecting facility among HIV-positive injection drug users.  AIDS Education and Prevention. 2011 Oct;23(5):412–22.  191.  Gowing L, Farrell MF, Bornemann R, Sullivan LE, Ali R. Oral substitution treatment of injecting opioid users for prevention of HIV infection. Cochrane Database of Systematic Reviews. 2011 Aug 10;(8):CD004145.  192.  Mattick RP, Breen C, Kimber J, Davoli M. Methadone maintenance therapy versus no opioid replacement therapy for opioid dependence. Cochrane Database of Systematic Reviews. 2009 Jul 8;(3):CD002209.  193.  MacArthur GJ, Minozzi S, Martin N, Vickerman P, Deren S, Bruneau J, et al. Opiate substitution treatment and HIV transmission in people who inject drugs: A systematic review and meta-analysis. British Medical Journal. 2012 Oct 4;345:e5945.  194.  Tsui JI, Evans JL, Lum PJ, Hahn JA, Page K. Association of opioid agonist therapy with lower incidence of hepatitis C virus infection in young adult injection drug users. Journal of the American Medical Association Internal Medicine.  2014 Dec 1;174(12):1974–81.  195.  Sordo L, Barrio G, Bravo MJ, Indave BI, Degenhardt L, Wiessing L, et al. Mortality risk during and after opioid substitution treatment: Systematic review and meta-analysis of cohort studies. British Medical Journal. 2017 Apr 26;357:j1550.           173 196.  Rowe C, Santos G-M, Behar E, Coffin PO. Correlates of overdose risk perception among illicit opioid users. Drug and Alcohol Dependence. 2016 Feb 1;159:234–9.  197.  DeBeck K, Small W, Wood E, Li K, Montaner J, Kerr T. Public injecting among a cohort of injecting drug users in Vancouver, Canada. Journal of Epidemiology & Community Health. 2009 Jan 1;63(1):81–6.  198.  Degenhardt L, Larney S, Kimber J, Gisev N, Farrell M, Dobbins T, et al. The impact of opioid substitution therapy on mortality post-release from prison: Retrospective data linkage study.  Addiction. 2014 Aug 1;109(8):1306–17.  199.  Lo A, Kerr T, Hayashi K, Milloy M-J, Nosova E, Liu Y, et al. Factors associated with methadone maintenance therapy discontinuation among people who inject drugs. Journal of Substance Abuse Treatment. 2018 Nov 1;94:41–6.  200.  Milloy M, Buxton J, Wood E, Li K, Montaner JS, Kerr T. Elevated HIV risk behaviour among recently incarcerated injection drug users in a Canadian setting: A longitudinal analysis. BMC Public Health. 2009 May 27;9(1):156.  201.  Small W, Wood E, Betteridge G, Montaner J, Kerr T. The impact of incarceration upon adherence to HIV treatment among HIV-positive injection drug users: A qualitative study. AIDS Care. 2009 Jun;21(6):708–14.  202.  Binswanger IA, Stern MF, Deyo RA, Heagerty PJ, Cheadle A, Elmore JG, et al. Release from prison — A high risk of death for former inmates. New England Journal of Medicine. 2007 Jan 11;356(2):157–65.  203.  Walker S, Higgs P, Stoové M, Wilson M. Narratives of young men with injecting drug use histories leaving adult prison. International Journal of Offender Therapy and Comparative Criminology. 2018 Sep 1;62(12):3681–707.  204.  Milloy M, Wood E, Small W, Tyndall M, Lai C, Montaner J, et al. Incarceration experiences in a cohort of active injection drug users. Drug and Alcohol Review. 2008 Nov;27(6):693–9.  205.  Darke S, Degenhardt L, Mattick R. Mortality amongst illicit drug users: Epidemiology, causes and intervention. Cambridge University Press; 2006.  206.  Hayashi K, Dong H, Marshall BDL, Milloy M-J, Montaner JSG, Wood E, et al. Sex-based differences in rates, causes, and predictors of death among injection drug users in Vancouver, Canada.  American Journal of Epidemiology. 2016 Mar 15;183(6):544–52.           174 207.  Copeland L, Budd J, Robertson JR, Elton RA. Changing patterns in causes of death in a cohort of injecting drug users, 1980-2001.  Archives of Internal Medicine. 2004 Jun 14;164(11):1214.  208.  Miller CL, Kerr T, Strathdee SA, Li K, Wood E. Factors associated with premature mortality among young injection drug users in Vancouver. Harm Reduction Journal. 2007 Jan 4;4(1):1.  209.  Vanichseni S, Martin M, Suntharasamai P, Sangkum U, Mock PA, Gvetadze RJ, et al. High mortality among non-HIV-infected people who inject drugs in Bangkok, Thailand, 2005-2012. American Journal of Public Health. 2015 Jun;105(6):1136–41.  210.  Wood E, Lloyd-Smith E, Li K, Strathdee SA, Small W, Tyndall MW, et al. Frequent needle exchange use and HIV incidence in Vancouver, Canada. The American Journal of Medicine. 2007 Feb 1;120(2):172–9.  211.  Pourhoseingholi MA, Baghestani AR, Vahedi M. How to control confounding effects by statistical analysis. Gastroenterology and Hepatology from Bed to Bench. 2012;5(2):79–83.  212.  Rothman KJ, Gallacher JE, Hatch EE. Why representativeness should be avoided. International Journal of Epidemiology. 2013 Aug 1;42(4):1012–4.  213.  National Institute on Alcohol Abuse and Alcoholism. Rethinking drinking: Alcohol and your health. NIH Publication No. 13-3770. [Internet]. 2016 [cited 2018 Aug 24]. Available from: https://pubs.niaaa.nih.gov/publications/RethinkingDrinking/Rethinking_Drinking.pdf 214.  Zivanovic R, Milloy M, Hayashi K, Dong H, Sutherland C, Kerr T, et al. Impact of unstable housing on all-cause mortality among persons who inject drugs. BMC Public Health. 2015;15:106.  215.  Wood E, Tyndall MW, Montaner JS, Kerr T. Summary of findings from the evaluation of a pilot medically supervised safer injecting facility. Canadian Medical Association Journal. 2006 Nov 21;175(11):1399–404.  216.  Rachlis BS, Hogg RS, Wood E, Li K, Kerr T. Factors associated with geographic migration among a cohort of injection drug users. Health & Place. 2008 Sep;14(3):536–43.           175 217.  Wilson D. Toxicology backlog delaying drug overdose test results. CBC [Internet]. 2017 Mar 14 [cited 2018 Aug 23]; Available from: https://www.cbc.ca/news/canada/british-columbia/drugs-overdose-fentanyl-toxicology-backlog-1.4024621 218.  Degenhardt L, Hall W. Extent of illicit drug use and dependence, and their contribution to the global burden of disease. The Lancet. 2012 Jan 13;379(9810):55–70.  219.  Lorvick J, Lutnick A, Wenger LD, Bourgois P, Cheng H, Kral AH. Non-partner violence against women who use drugs in San Francisco.  Violence Against Women. 2014 Oct 5;20(11):1285-98.  220.  Kutsa O, Marcus R, J Bojko M, Zelenev A, Mazhnaya A, Dvoriak S, et al. Factors associated with physical and sexual violence by police among people who inject drugs in Ukraine: Implications for retention on opioid agonist therapy. Journal of the International AIDS Society. 2016;19:20897. 221.  Hayashi K, Ti L, Csete J, Kaplan K, Suwannawong P, Wood E, et al. Reports of police beating and associated harms among people who inject drugs in Bangkok, Thailand: A serial cross-sectional study. BMC Public Health. 2013;13(1):733. 222.  Williams JE, Dangerfield DT, Kral AH, Wenger LD, Bluthenthal RN. Correlates of sexual coercion among people who inject drugs (PWID) in Los Angeles and San Francisco, CA. Journal of Urban Health. 2018:1-8. 223.  Afifi TO, MacMillan H, Cox BJ, Asmundson GJG, Stein MB, Sareen J. Mental health correlates of intimate partner violence in marital relationships in a nationally representative sample of males and females. Journal of Interpersonal Violence. 2009 Aug;24(8):1398–417.  224.  Hedtke KA, Ruggiero KJ, Fitzgerald MM, Zinzow HM, Saunders BE, Resnick HS, et al. A longitudinal investigation of interpersonal violence in relation to mental health and substance use. Journal of Consulting and Clinical Psychology. 2008 Aug;76(4):633–47.  225.  Braitstein P, Li K, Tyndall M, Spittal P, O’Shaughnessy MV, Schilder A, et al. Sexual violence among a cohort of injection drug users. Social Science & Medicine. 2003 Aug;57(3):561–9.           176 226.  Marshall BDL, Fairbairn N, Li K, Wood E, Kerr T. Physical violence among a prospective cohort of injection drug users: A gender-focused approach. Drug and Alcohol Dependence. 2008 Oct 1;97(3):237–46.  227.  Jessell L, Mateu-Gelabert P, Guarino H, Vakharia SP, Syckes C, Goodbody E, et al. Sexual violence in the context of drug use among young adult opioid users in New York City. Journal of Interpersonal Violence. 2017 Oct 1;32(19):2929–54.   228.  Erickson PG. Drugs, violence and public health: What does the harm reduction approach have to offer? Vancouver: Fraser Institute. 2001. 229.  Romero-Daza N, Weeks M, Singer M. “Nobody gives a damn if I live or die”: Violence, drugs, and street-level prostitution in inner-city Hartford, Connecticut. Medical Anthropology. 2003 Jul;22(3):233–59.  230.  Werb D, Rowell G, Guyatt G, Kerr T, Montaner J, Wood E. Effect of drug law enforcement on drug market violence: A systematic review. International Journal of Drug Policy. 2011 Mar;22(2):87–94.  231.  Aitken C, Moore D, Higgs P, Kelsall J, Kerger M. The impact of a police crackdown on a street drug scene: Evidence from the street. International Journal of Drug Policy. 2002 Sep 1;13(3):193–202.  232.  Epele ME. Gender, violence and HIV: Women’s survival in the streets. Culture, Medicine & Psychiatry. 2002;26(1):33–54. 233.  Bungay V, Johnson JL, Varcoe C, Boyd S. Women’s health and use of crack cocaine in context: Structural and ‘everyday’ violence. International Journal of Drug Policy. 2010 Jul;21(4):321–9.  234.  Maher L, Hudson SL. Women in the drug economy: A metasynthesis of the qualitative literature. Journal of Drug Issues. 2007 Oct 1;37(4):805–26.   235.  Bourgois P, Prince B, Moss A. The everyday violence of Hepatitis C among young women who inject drugs in San Francisco. Human Organization. 2004 Sep 1;63(3):253–64.  236.  Wenzel SL, Leake BD, Gelberg L. Risk factors for major violence among homeless women. Journal of Interpersonal Violence. 2001 Aug 1;16(8):739–52.           177 237.  Wenzel SL, Tucker JS, Elliott MN, Marshall GN, Williamson SL. Physical violence against impoverished women: A longitudinal analysis of risk and protective factors. Women’s Health Issues. 2004 Sep;14(5):144–54.  238.  Lake S, Wood E, Dong H, Dobrer S, Montaner J, Kerr T. The impact of childhood emotional abuse on violence among people who inject drugs. Drug and Alcohol Review. 2015 Jan 1;34(1):4–9.  239.  Bernstein D, Fink L. Childhood Trauma Questionnaire: A retrospective self-report manual. San Antonio, TX: Psychological Corp; 1998.  240.  Bauer GR, Hammond R, Travers R, Kaay M, Hohenadel KM, Boyce M. “I don’t think this is theoretical; This is our lives”: How erasure impacts health care for transgender people. Journal of the Association of Nurses in AIDS Care. 2009 Sep;20(5):348–61.  241.  Bauer GR, Braimoh J, Scheim AI, Dharma C. Transgender-inclusive measures of sex/gender for population surveys: Mixed-methods evaluation and recommendations. PLOS ONE. 2017 May 25;12(5):e0178043.  242.  Meerwijk EL, Sevelius JM. Transgender population size in the United States: A Meta-regression of population-based probability samples. American Journal of Public Health. 2017 Feb;107(2):e1–8. 243.  Fraser G. Evaluating inclusive gender identity measures for use in quantitative psychological research. Psychology & Sexuality. 2018 Oct 2;9(4):343–57.  244.  Lee J-H, Herzog TA, Meade CD, Webb MS, Brandon TH. The use of GEE for analyzing longitudinal binomial data: A primer using data from a tobacco intervention. Addictive Behaviors. 2007 Jan;32(1):187–93.  245.  Maldonado G, Greenland S. Simulation study of confounder-selection strategies. American Journal of Epidemiology. 1993 Dec 1;138(11):923–36.  246.  Small W, Wood E, Tobin D, Rikley J, Lapushinsky D, Kerr T. The Injection Support Team: A peer-driven program to address unsafe injecting in a Canadian setting. Substance Use & Misuse. 2012 Mar 20;47(5):491–501.  247.  Lupick T. Assisted-injection trial means Vancouver’s most-marginalized drug users could soon access Insite. The Georgia Straight [Internet]. 2018 Jul 17 [cited 2018 Dec 12]; Available from: https://www.straight.com/news/1103766/assisted-injection-trial-means-vancouvers-most-marginalized-drug-users-could-soon          178 248.  Platt L, Grenfell P, Meiksin R, Elmes J, Sherman SG, Sanders T, et al. Associations between sex work laws and sex workers’ health: A systematic review and meta-analysis of quantitative and qualitative studies. PLOS Medicine. 2018 Dec 11;15(12):e1002680.  249.  Wood E. Strategies for reducing opioid-overdose deaths — Lessons from Canada.  New England Journal of Medicine. 2018 Apr 26;378(17):1565–7.  250.  Nadelmann E, LaSalle L. Two steps forward, one step back: Current harm reduction policy and politics in the United States. Harm Reduction Journal. 2017 Jun 12;14(1):37.  251.  Davidson PJ, Lopez AM, Kral AH. Using drugs in un/safe spaces: Impact of perceived illegality on an underground supervised injecting facility in the United States. International Journal of Drug Policy. 2018 Mar 1;53:37–44.  252.  Winfield Cunningham P. The Health 202: Supervised injection facilities are illegal in the United States. These cities want to open them anyway. Washington Post [Internet]. 2018 Apr 30 [cited 2018 Oct 1]; Available from: https://www.washingtonpost.com/news/powerpost/paloma/the-health-202/2018/04/30/the-health-202-supervised-injection-facilities-are-illegal-in-the-united-states-these-cities-want-to-open-them-anyway/5ae5dbc630fb043711926901/ 253.  BC Centre on Substance Use. Supervised consumption services operational guidance [Internet]. Vancouver, BC; 2018 [cited 2018 Aug 26]. Available from: http://www.bccsu.ca/wp-content/uploads/2017/07/BC-SCS-Operational-Guidance.pdf 254.  Belackova V, Salmon AM, Schatz E, Jauncey M. Online census of drug consumption rooms (DCRs) as a setting to address HCV: Current practice and future capacity. Amsterdam, Sydney: International Network of Drug Consumption Rooms, Correlation Network, Uniting Medically Supervised Injecting Centre; 2017.  255.  Bardwell G, Kerr T, Boyd J, McNeil R. Characterizing peer roles in an overdose crisis: Preferences for peer workers in overdose response programs in emergency shelters. Drug and Alcohol Dependence. 2018 Sep 1;190:6–8.  256.  Ti L, Tzemis D, Buxton JA. Engaging people who use drugs in policy and program development: A review of the literature. Substance Abuse Treatment, Prevention, and Policy. 2012 Nov 24;7(1):47.           179 257.  Greer AM, Luchenski SA, Amlani AA, Lacroix K, Burmeister C, Buxton JA. Peer engagement in harm reduction strategies and services: A critical case study and evaluation framework from British Columbia, Canada. BMC Public Health. 2016 May 27;16(1):452.  258.  Hayashi K, Wood E, Wiebe L, Qi J, Kerr T. An external evaluation of a peer-run outreach-based syringe exchange in Vancouver, Canada. International Journal of Drug Policy. 2010 Sep 1;21(5):418–21.  259.  Sherman SG, Gann DS, Scott G, Carlberg S, Bigg D, Heimer R. A qualitative study of overdose responses among Chicago IDUs. Harm Reduction Journal. 2008 Jan 24;5(1):2.  260.  Bassuk EL, Hanson J, Greene RN, Richard M, Laudet A. Peer-delivered recovery support services for addictions in the United States: A systematic review. Journal of Substance Abuse Treatment. 2016 Apr 1;63:1–9.  261.  Deering KN, Kerr T, Tyndall MW, Montaner JSG, Gibson K, Irons L, et al. A peer-led mobile outreach program and increased utilization of detoxification and residential drug treatment among female sex workers who use drugs in a Canadian setting. Drug and Alcohol Dependence. 2011 Jan;113(1):46–54.  262.  Broadhead RS, Heckathorn DD, Altice FL, van Hulst Y, Carbone M, Friedland GH, et al. Increasing drug users’ adherence to HIV treatment: Results of a peer-driven intervention feasibility study. Social Science & Medicine. 2002 Jul 1;55(2):235–46.  263.  Deering KN, Shannon K, Sinclair H, Parsad D, Gilbert E, Tyndall MW. Piloting a peer-driven intervention model to increase access and adherence to antiretroviral therapy and HIV Care among street-entrenched HIV-positive women in Vancouver. AIDS Patient Care and STDs. 2009 Jul 10;23(8):603–9.  264.  Needle RH, Burrows D, Friedman SR, Dorabjee J, Touzé G, Badrieva L, et al. Effectiveness of community-based outreach in preventing HIV/AIDS among injecting drug users. International Journal of Drug Policy. 2005 Dec 1;16:45–57.  265.  Ti L, Hayashi K, Kaplan K, Suwannawong P, Wood E, Montaner J, et al. Willingness to access peer-delivered HIV testing and counseling among people who inject drugs in Bangkok, Thailand. Journal of Community Health.  2013 Jun 1;38(3):427–33.           180 266.  Kerr T, Wood E, Palepu A, Wilson D, Schechter MT, Tyndall MW. Responding to an explosive HIV epidemic driven by frequent cocaine injection: Is there a role for safe injecting facilities? Journal of Drug Issues. 2003 Jul 1;33(3):579–608.  267.  BC Ministry of Health and Addictions. Responding to BC’s illegal drug overdose epidemic: Progress update [Internet]. 2018 [cited 2018 Aug 28]. Available from: https://www2.gov.bc.ca/assets/gov/overdose-awareness/final_responding_to_bcs_opioid_overdose_epidemic_-_progress_updatejan_2018.pdf 268.  Bardwell G, Boyd J, Kerr T, McNeil R. Negotiating space and drug use in emergency shelters with peer witness injection programs within the context of an overdose crisis: A qualitative study. Health & Place. 2018 Sep 1;53:86–93.  269.  Jozaghi E, Greer AM, Lampkin H, Buxton JA. Activism and scientific research: 20 years of community action by the Vancouver area network of drug users. Substance Abuse Treatment, Prevention, and Policy. 2018 May 22;13(1):18.  270.  Kerr T, Oleson M, Tyndall MW, Montaner J, Wood E. A description of a peer-run supervised injection site for injection drug users. Journal of Urban Health. 2005 Jun 1;82(2):267–75.  271.  Bradley EH, Curry LA, Devers KJ. Qualitative data analysis for health services research: Developing taxonomy, themes, and theory. Health Services Research. 2007;42(4):1758–72.  272.  Mayer S, Boyd J, Collins A, Kennedy MC, Fairbairn N, McNeil R. Characterizing fentanyl-related overdoses and implications for overdose response: Findings from a rapid ethnographic study in Vancouver, Canada. Drug and Alcohol Dependence. 2018 Dec;193:69–74.  273.  Buchman DZ, Orkin AM, Strike C, Upshur REG. Overdose education and naloxone distribution programmes and the ethics of task shifting. Public Health Ethics. 2018 Jul 1;11(2):151–64.  274.  Greer AM, Amlani A, Burmeister C, Scott A, Newman C, Lampkin H, et al. Peer engagement barriers and enablers: Insights from people who use drugs in British Columbia, Canada. Canadian Journal of Public Health. 2019:1-9. 275.  Center for Substance Abuse Treatment (U.S). Trauma-Informed Care in Behavioral Health Services. [Internet]. Rockville, MD: Substance Abuse and          181 Mental Health Services Administration (US). 2014. p. xix.  Available from: http://www.ncbi.nlm.nih.gov/books/NBK207192/ 276.  Ben-Porat A, Itzhaky H. Burnout among trauma social workers: The contribution of personal and environmental resources. Journal of Social Work. 2015 Nov 1;15(6):606–20.  277.  Maslach C, Leiter MP. Early predictors of job burnout and engagement. Journal of Applied Psychology. 2008 May;93(3):498–512.  278.  Mijovic H, McKnight J, English M. What does the literature tell us about health workers’ experiences of task-shifting projects in sub-Saharan Africa? A systematic, qualitative review. Journal of Clinical Nursing. 2016 Aug 1;25(15–16):2083–100.  279.  Mundeva H, Snyder J, Ngilangwa DP, Kaida A. Ethics of task shifting in the health workforce: Exploring the role of community health workers in HIV service delivery in low- and middle-income countries. BMC Medical Ethics. 2018 Dec;19(1):71. 280.  Bardwell G, Anderson S, Richardson L, Bird L, Lampkin H, Small W, et al. The perspectives of structurally vulnerable people who use drugs on volunteer stipends and work experiences provided through a drug user organization: Opportunities and limitations. International Journal of Drug Policy. 2018 May 1;55:40–6.  281.  Greer AM, Pauly B, Scott A, Martin R, Burmeister C, Buxton J. Paying people who use illicit substances or ‘peers’ participating in community-based work: A narrative review of the literature. Drugs: Education, Prevention and Policy. 2018 Sep 6;3:1–13. 282.  Shearer D, Fleming T, Fowler A, Boyd J, McNeil R. Naloxone distribution, trauma, and supporting community-based overdose responders. International Journal of Drug Policy. 2018 Dec 5. 283.  Darke S. Self-report among injecting drug users: A review. Drug and Alcohol Dependence. 1998 Aug;51(3):253–63.  284.  Tyndall MW, Craib KJP, Currie S, Li K, O’Shaughnessy MV, Schechter MT. Impact of HIV infection on mortality in a cohort of injection drug users. Journal of Acquired Immunodeficiency Syndrome. 2001 Dec;28(4):351–7.           182 285.  Wallace B, Kennedy MC, Kerr T, Pauly B. Factors associated with nonfatal overdose during a public health emergency. Substance Use & Misuse. 2019 Jan 2;54(1):39–45.  286.  Fairbairn N, Kerr T, Buxton JA, Li K, Montaner JS, Wood E. Increasing use and associated harms of crystal methamphetamine injection in a Canadian setting. Drug and Alcohol Dependence. 2007 May;88(2–3):313–6.  287.  Werb D, Debeck K, Kerr T, Li K, Montaner J, Wood E. Modelling crack cocaine use trends over 10 years in a Canadian setting. Drug and Alcohol Review. 2010 Jan 28;29(3):271–7.  288.  Hayashi K, Milloy M-J, Lysyshyn M, DeBeck K, Nosova E, Wood E, et al. Substance use patterns associated with recent exposure to fentanyl among people who inject drugs in Vancouver, Canada: A cross-sectional urine toxicology screening study. Drug and Alcohol Dependence. 2018 Feb 1;183:1–6.            183 Appendices Appendix A  Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist* Section/topic  # Checklist item  Reported on page #  TITLE  Title  1 Identify the report as a systematic review, meta-analysis, or both.  24 ABSTRACT  Structured summary  2 Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.  iii INTRODUCTION  Rationale  3 Describe the rationale for the review in the context of what is already known.  24 Objectives  4 Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS).  24 METHODS  Protocol and registration  5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number.  N/A Eligibility criteria  6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale.  25; Table 2.1 Information sources  7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.  25 Search  8 Present full electronic search strategy for at least one database, including any limits used, such that 25; Appendix B          184 it could be repeated.  Study selection  9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).  25 Data collection process  10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.  25 Data items  11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.  25 Risk of bias in individual studies  12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.  26 Summary measures  13 State the principal summary measures (e.g., risk ratio, difference in means).  Table 2.1 Synthesis of results  14 Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I2) for each meta-analysis.  N/A Risk of bias across studies  15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies).  26; 32-36; 39 Additional analyses  16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified.  N/A RESULTS  Study selection  17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.  26; Figure 2.1 Study characteristics  18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations.  27-28; Table 2.2          185 Risk of bias within studies  19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12).  26-27; Appendix C Results of individual studies  20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.  27-32; Table 2.2 Synthesis of results  21 Present results of each meta-analysis done, including confidence intervals and measures of consistency.  N/A Risk of bias across studies  22 Present results of any assessment of risk of bias across studies (see Item 15).  26; Appendix C Additional analysis  23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]).  N/A DISCUSSION  Summary of evidence  24 Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).  32-36 Limitations  25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias).  39-40 Conclusions  26 Provide a general interpretation of the results in the context of other evidence, and implications for future research.  40 FUNDING  Funding  27 Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review.  xvi *From 182:  Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(7): e1000097. doi:10.1371/journal.pmed1000097                186 Appendix B  Sample search strategy to search the Medline database via the OVID platform; May 01, 2017 No. Search Terms 1 (Supervised ADJ2 inject*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  2 (Supervised ADJ2 smok*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  3 (Supervised ADJ2 consum*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  4 (Supervised ADJ2 shoot*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp 5 (Supervised ADJ2 inhal*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  6 (Safe* ADJ2 inject*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  7 (Safe* ADJ2 smok*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  8 (Safe* ADJ2 consum*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp 9 (Safe* ADJ2 shoot*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  10 (Safe* ADJ2 inhal*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp 11 (Drug* ADJ2 inject*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  12 (Drug* ADJ2 smok*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  13 (Drug* ADJ2 consum*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp 14 (Drug* ADJ2 shoot*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp  15 (Drug* ADJ2 inhal*) ADJ3 (Facilit* OR Room* OR Galler* OR Cent* OR site* OR service*).mp 16 OR/1-15 17 Limit 16 to Humans 18 Remove duplicates from 17          187 Appendix C  Quality assessment of included studies C.1 Quality assessment of observational cohort and cross-sectional studies (NHBLI) Study Quality assessment criteria*  1 2 3 4 5 6 7 8 9 10 11 12 13 14 Zurhold et al. 2003 Yes Yes Yes Yes No No No NA No No Yes CD NA No Van Beek et al. 2004  Yes Yes Yes Yes No Yes Yes NA No No Yes CD NA Yes Kerr et al. 2005 Yes Yes Yes Yes No No No NA No No Yes CD NA Yes Thein et al. 2005 Yes Yes Yes Yes No No No NA No No Yes CD NA No Wood et al. 2005 Yes Yes Yes Yes No No No NA No No Yes CD NA Yes Wood et al. 2006b Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Kerr et al. 2007 Yes Yes Yes Yes No No No NA No No Yes CD NA Yes McKnight et al. 2007 Yes Yes Yes Yes No No No NA No No Yes CD NA Yes Stoltz et al. 2007 Yes Yes Yes Yes No No No NA No No Yes CD NA Yes Wood et al. 2007 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD NA Yes Kimber et al. 2008 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Lloyd-Smith et al. 2008 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Milloy et al. 2008a Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Richardson et al. 2008 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Wood et al. 2008 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD NA Yes Bravo et al. 2009 Yes Yes Yes Yes Yes No No NA No No Yes CD NA Yes Marshall et al. 2009 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Milloy et al. 2009 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Baars et al. 2010 Yes Yes Yes Yes No No No NA No No Yes CD NA Yes Milloy et al. 2010 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Lloyd-Smith et al. 2010 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Scherbaum et al. 2010 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes No Debeck et al. 2011 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Lloyd-Smith et al. 2012 Yes Yes Yes Yes No Yes Yes NA No Yes Yes CD Yes Yes Kinnard et al. 2014 Yes Yes Yes Yes No No No NA No No Yes CD NA No Toth et al. 2016 Yes Yes Yes Yes No No No NA No No Yes CD NA No CD = cannot determine; NA = not applicable; NR = not reported; NHBLI = National Heart, Blood and Lung Institute. *(1) Was the research question or objective in this paper clearly stated? (2) Was the study population clearly specified and defined? (3) Was the participation rate of eligible persons at least 50%? (4) Were all the subjects selected or recruited from the same or similar populations?          188 Were inclusion and exclusion criteria for being in the study pre-specified and applied uniformly to all participants? (5) Was a sample size justification, power description, or variance and effect estimates provided? (6) For the analyses in this paper, were the exposure(s) of interest measured prior to the outcome(s) being measured? (7) Was the timeframe sufficient so that one could reasonably expect to see an association between exposure and outcome if it existed? (8) For exposures that can vary in amount or level, did the study examine different levels of the exposure as related to the outcome (e.g., categories of exposure, or exposure measured as continuous variable)? (9) Were the exposure measures (independent variables) clearly defined, valid, reliable, and implemented consistently across all study participants? (10) Was the exposure(s) assessed more than once over time? (11) Were the outcome measures (dependent variables) clearly defined, valid, reliable, and implemented consistently across all study participants? (12) Were the outcome assessors blinded to the exposure status of participants? (13) Was loss to follow-up after baseline 20% or less? (14) Were key potential confounding variables measured and adjusted statistically for their impact on the relationship between exposure(s) and outcome(s)?                       189 C.2 Quality assessment of before-after (pre-post) studies (NHBLI) Study Quality assessment criteria*  1 2 3 4 5 6 7 8 9 10 11 12 Poschadel et al. 2003 Yes Yes CD CD Yes Yes Yes CD CD Yes Yes No Wood et al. 2004 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No No Freeman et al. 2005 Yes Yes CD Yes Yes Yes Yes CD Yes Yes Yes No Thein et al. 2005 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No Yes Kerr et al. 2006 Yes Yes CD Yes Yes Yes Yes CD Yes Yes Yes No Wood et al. 2006a Yes Yes CD Yes Yes Yes Yes CD Yes Yes No No NCHECR 2007 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No Yes Salmon et al. 2007 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No Yes Fitzgerald et al. 2010 Yes Yes CD Yes Yes Yes Yes CD Yes Yes Yes No Salmon et al. 2010 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No Yes Marshall et al. 2011 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No Yes Donnelly et al. 2013 Yes No CD Yes Yes Yes Yes CD Yes Yes Yes No Vecino et al. 2013 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No No Skelton et al. 2016 Yes No CD CD CD Yes CD CD CD Yes No No Lysyshyn et al. 2017 Yes Yes CD Yes Yes Yes Yes CD Yes Yes No No CD = cannot determine; NHBLI = National Heart, Blood and Lung Institute. *(1) Was the study question or objective clearly stated? (2) Were eligibility/selection criteria for the study population pre-specified and clearly described? (3) Were the participants in the study representative of those who would be eligible for the test/service/intervention in the general or clinical population of interest? (4) Were all eligible participants that met the pre-specified entry criteria enrolled? (5) Was the sample size sufficiently large to provide confidence in the findings? (6) Was the test/service/intervention clearly described and delivered consistently across the study population? (7) Were the outcome measures pre-specified, clearly defined, valid, reliable, and assessed consistently across all study participants? (8) Were the people assessing the outcomes blinded to the participants' exposures/interventions? (9) Was the loss to follow-up after baseline 20% or less? Were those lost to follow-up accounted for in the analysis? (10) Did the statistical methods examine changes in outcome measures from before to after the intervention? Were statistical tests done that provided p values for the pre-to-post changes? (11) Were outcome measures of interest taken multiple times before the          190 intervention and multiple times after the intervention (i.e., did they use an interrupted time-series design)? (12) If the intervention was conducted at a group level (e.g., a whole hospital, a community, etc.) did the statistical analysis take into account the use of individual-level data to determine effects at the group level?                             191 C.3 Quality assessment of cost-effectiveness studies using the Joanna Briggs Institute’s Critical Appraisal Checklist for Economic Evaluations Author (year) Critical appraisal criteria* 1 2 3 4 5 6 7 8 9 10 11 Hedrich 2004 Yes Unclear Unclear Unclear Yes Yes Unclear Unclear Unclear Yes Yes Bayoumi et al. 2008 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Milloy et al. 2008b Yes Yes Unclear Yes Yes Yes Unclear Unclear Yes Yes Yes Andresen et al. 2010 Yes Yes Yes Yes Yes Yes Yes Unclear Yes Yes Yes Pinkerton et al. 2010 Yes Yes Yes Yes Yes Yes Yes Unclear Yes Yes Yes Pinkerton et al. 2011 Yes Yes Yes Yes Yes Yes Yes Unclear Yes Yes Yes Andresen et al. 2012 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Jozaghi et al. 2014 Yes Yes Yes Yes Yes Yes Yes Unclear Yes Yes Yes *1. Is there a well-defined question? 2. Is there comprehensive description of alternatives? 3. Are all important and relevant costs and outcomes for each alternative identified? 4. Has clinical effectiveness been established? 5. Are costs and outcomes measured accurately? 6. Are costs and outcomes valued credibly? 7. Are costs and outcomes adjusted for differential timing? 8. Is there an incremental analysis of costs and consequences? 9. Were sensitivity analyses conducted investigate uncertainty in estimates of cost or consequences? 10. Do study results include all issues of concern to users? 11. Are the results generalizable to the setting of interest in the review?   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0376545/manifest

Comment

Related Items