UBC Faculty Research and Publications

An age-based analysis of nonmedical prescription opioid use among people who use illegal drugs in Vancouver,… Cheng, Tessa; Small, Will; Dong, Huiru; Nosova, Ekaterina; Hayashi, Kanna; DeBeck, Kora Nov 27, 2018

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

Item Metadata


52383-13011_2018_Article_180.pdf [ 620.14kB ]
JSON: 52383-1.0374306.json
JSON-LD: 52383-1.0374306-ld.json
RDF/XML (Pretty): 52383-1.0374306-rdf.xml
RDF/JSON: 52383-1.0374306-rdf.json
Turtle: 52383-1.0374306-turtle.txt
N-Triples: 52383-1.0374306-rdf-ntriples.txt
Original Record: 52383-1.0374306-source.json
Full Text

Full Text

RESEARCH Open AccessAn age-based analysis of nonmedicalprescription opioid use among people whouse illegal drugs in Vancouver, CanadaTessa Cheng1,2, Will Small1,2,4, Huiru Dong2,3, Ekaterina Nosova2, Kanna Hayashi1,2 and Kora DeBeck2,5*AbstractBackground: Nonmedical prescription opioid use (NMPOU) is a serious public health problem in North America. Ata population-level, previous research has identified differences in the prevalence and correlates of NMPOU amongyounger versus older age groups; however, less is known about age-related differences in NMPOU among peoplewho use illegal drugs.Methods: Data were collected between 2013 and 2015 from two linked prospective cohort studies in Vancouver,Canada: the At-Risk Youth Study (ARYS) and the Vancouver Injection Drug Users Study (VIDUS). Factors independentlyassociated with NMPOU among younger (ARYS) and older (VIDUS) participants were examined separately usingbivariate and multivariate generalized estimating equations.Results: A total of 1162 participants were included. Among 405 eligible younger participants (Median age = 25;Inter-Quartile Range [IQR]: 22–28), 40% (n = 160) reported engaging in NMPOU at baseline; among 757 olderparticipants (Median age = 48, IQR: 40–55), 35% (n = 262) reported engaging in NMPOU at baseline. In separatemultivariate analyses of younger and older participants, NMPOU was positively and independently associated withheroin use (younger: Adjusted Odds Ratio [AOR] = 3.12, 95% Confidence Interval [CI]: 2.08–4.68; older: AOR = 2.79, 95%CI: 2.08–3.74), drug dealing (younger: AOR = 2.22, 95% CI: 1.58–3.13; older: AOR = 1.87, 95% CI: 1.40–2.49), and difficultyaccessing services (younger: AOR = 1.47, 95% CI: 1.04–2.09; older: AOR = 1.74, 95% CI: 1.32–2.29). Among the youthcohort only, NMPOU was associated with younger age (AOR = 1.12, 95% CI: 1.05–1.19), crack use (AOR = 1.56, 95% CI: 1.06–2.30), and binge drug use (AOR = 1.41, 95% CI: 1.00–1.97); older participants who engaged in NMPOU were morelikely to report crystal methamphetamine use (AOR = 1.97, 95% CI: 1.46–2.66), non-fatal overdose (AOR = 1.76, 95% CI: 1.20–2.60) and sex work (AOR = 1.49, 95% CI: 1.00–2.22).Discussion: The prevalence of NMPOU is similar among younger and older people who use drugs, and independentlyassociated with markers of vulnerability among both age groups. Adults who engage in NMPOU are at risk for non-fataloverdose, which highlights the need for youth and adult-specific strategies to address NMPOU that include better accessto health and social services, as well as a range of addiction treatment options for opioid use. Findings also underscorethe importance of improving pain treatment strategies tailored for PWUD.Keywords: Prescription opioids, Risk behavior, Addictions, Youth, Overdose* Correspondence: bccsu-kd@bccsu.ubc.ca2British Columbia Centre on Substance Use, Providence Health Care,400-1045 Howe St, Vancouver, BC V6Z 2A9, Canada5School of Public Policy, Simon Fraser University, 515 West Hastings Street,Suite 3271, Vancouver, BC V6B 5K3, CanadaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Cheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 https://doi.org/10.1186/s13011-018-0180-3BackgroundRates of medical and nonmedical prescription opioid use(NMPOU) are rising across Canada and the United States[1], along with shifts towards increases in the global con-sumption of opioids [2]. Research consistently indicatesthat adolescents and young adults are more likely to en-gage in NMPOU than older youth and adults [3–11], andevidence suggests that these differences in prescriptionopioid (PO) use are associated with statistically significantbirth cohort effects; more recent birth cohorts have higherlifetime and past-year prevalence of prescription opioid-use disorder due to NMPOU [12]. Adolescents andyoung adults in the United States are more likely to initi-ate NMPOU than older age groups [3, 13–15], and a studyof the American general population found that the mostfrequently reported age of NMPOU initiation was 16–18years [16]. This effect has been attributed to the increasedavailability of POs [4], and other research amongstreet-involved youth has found that the easy availabilityof POs facilitates NMPOU [17]. In addition, those whoengage in NMPOU are more likely to be younger thanthose engaging in other illegal drug use [18], and adoles-cents and young adults are also more likely to share andreceive any class of prescription medications, includingPOs, than older age groups [19]. For the purposes of thisstudy, “youth”, “young adult”, and “younger age groups”are used to describe individuals in their late teens and upto the late 20’s [20]. “Adult” and “older age groups”include those over the age of majority but focuses onthose in their mid-life and older.Youth-specific strategies to address substance use areoften prioritized due to the developmental harms associ-ated with licit and illegal substances [21]; however, theprevalence of NMPOU among adults up to age 64 is sig-nificantly higher than among adults over the age of 65[22], and the ubiquity of PO use has resulted in a signifi-cant risk of engaging in NMPOU that increases with age[12]. The increase in PO use among older individuals isespecially problematic given that age-related physiologicalchanges (e.g., drug absorption) increase the harms of POuse among older adults, such as hyperalgesia [23].Research has also identified a key difference related toNMPOU among younger and older age groups, wherepain is a more frequent motivator for engaging inNMPOU among adults [24] and youth are more likely toengage in NMPOU for its euphoric effects [6].Although younger and older age groups are both at riskfor engaging in NMPOU, the majority of research findingsrelated to NMPOU rely on population-level sampling andnational surveys that often exclude marginalized groupswho are unstably housed or have low incomes. Despiteprevious research investigating NMPOU among peoplewho use illegal drugs (PWUD) in Canadian and Americansettings [17, 25–34], there have been few if any studiescharacterizing age-based differences associated withNMPOU among PWUD. Given that this populationalready experiences numerous risks and harms related tosubstance use [35], this study investigates age-based differ-ences associated with NMPOU among younger and olderPWUD in Vancouver.MethodsData from this study were drawn from two prospectiveopen cohort studies: the At-Risk Youth Study (ARYS) andthe Vancouver Injection Drug Users Study (VIDUS).ARYS and VIDUS use a harmonized study questionnaireand study participants can attend a study visit at eitherstudy office regardless of their cohort enrollment. ARYSand VIDUS have both been described in detail previously[36, 37]. VIDUS is a cohort of HIV-negative adult PWUDwho injected illegal drugs at least once in the month priorto enrolment. Participants in the VIDUS cohort arerecruited through self-referral and street outreach, whichhas been ongoing since 1996. In brief, ARYS has beenoperating since 2005 as a cohort study of street-involvedyouth. To be eligible, participants must be aged 14–26years at recruitment and also have used illegal drugs otherthan cannabis in the past 30 days, provide writteninformed consent, and be “street-involved”. In this cohort,“street-involved” is defined as being absolutely, periodic-ally, or temporarily homeless (e.g., having no fixedaddress, sleeping on the street, couch surfing, or staying ina shelter or hostel), and includes those who are not home-less but have used services designated for street-youth(e.g., youth-specific drop-in centres, social services, andharm reduction services) in the last year. Youths’ streetinvolvement and eligibility to participate are assessed dur-ing a semi-structured in-person interview with an ARYSstaff member.At enrolment, and on a semi-annual basis, participantsin ARYS and VIDUS complete a questionnaire that isadministered by trained study staff. The questionnaireincludes questions related to demographic informationand drug use patterns. At each study visit, participantsare provided with a stipend ($30 CDN) for their time.The ARYS and VIDUS studies have been approved bythe University of British Columbia/Providence HealthCare Research Ethics Board.All ARYS and VIDUS participants who completed astudy visit between December 2013 and May 2015 wereeligible for the present analyses, as PO-related questionswere added to the study instrument during the summer of2013. The dependent variable for these analyses wasengaging in NMPOU through any route of administration,based on responses to the question: “In the last 6 months,when you were using, which of the following non-injection prescription opiates did you use when they werenot prescribed for you or that you took only for theCheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 2 of 11experience or feeling they caused, and how often did youuse them?” and “In the last 6 months, have you injectedany of the following prescription opiates?” (yes vs. no). Toidentify factors associated with engaging in NMPOU, weconsidered a number of potential explanatory variables ofinterest. The following socio-demographic variables ofinterest were included: age (per year younger); sex; Cauca-sian ancestry; and homelessness, defined as having nofixed address, sleeping on the street, couch surfing, orstaying in a shelter or hostel in the last 6 months. The fol-lowing variables related to drug use were also included:any injection or non-injection heroin use; any injection ornon-injection crack cocaine use; any injection ornon-injection cocaine use; any injection or non-injectioncrystal methamphetamine use; binge drug use, defined asa period of using injection or non-injection drugs moreoften than usual; and experiencing a non-fatal drug over-dose due to injection or non-injection drug use, based onresponses to the question “In the last 6 months, have youoverdosed by accident (i.e., where you had a negative reac-tion from using too much drugs)?”. Behavioural andsocio-structural risk factors hypothesized to be associatedwith NMPOU included: regular employment, defined ashaving a regular job, temporary work, or being self-employed; drug dealing, defined as selling drugs as asource of income; engaging in sex work, defined as ex-changing sex for money, drugs, gifts, food, clothes, shelteror favours; incarceration, defined as being in detention,jail, or prison; and difficulty accessing services, based onresponses to the question “In the last 6 months, was therea time you were in need of a service (e.g., housing, coun-selling) but could not obtain it?”. All variables were binaryand referred to activities, behaviours, and experiences inthe previous 6 months unless otherwise indicated.All analyses were conducted separately for ARYS andVIDUS participants, using cohort enrollment as a proxymarker for younger and older age groups, respectively. Atthe time of enrollment, ARYS participants must be between14 and 26 years old; however, as a prospective cohort studywith substantial resources devoted to maintaining follow-up,the ARYS participant pool necessarily includes participantswho are older than 26. These participants and their data arenot excluded from the analyses given that this data providesa rich source for tracking and understanding long-term risktrajectories associated with street-involvement during a keydevelopmental phase. While this practice may limit the ap-plicability of our results to street-involved adolescents, wecontrolled for possible biases associated with these olderparticipants in the youth cohort by including the continuous“per year younger” variable in the analyses to ensure thatage differences within the cohort were accounted for. Forconsistency and to similarly control for a wide age rangewithin the cohort, we also included the “per year younger”variable in the VIDUS analysis.First, a descriptive analysis of the study sample was con-ducted using Pearson’s chi-square test. Characteristics forparticipants who reported nonmedical prescription opioiduse (NMPOU) were measured at their first visit (duringthe study period: 2013–2015), which involved a report ofNMPOU; characteristics for all other participants weremeasured from the first study visit during the studyperiod. Second, to model factors associated with engagingin NMPOU over time and to analyse longitudinal corre-lated within-subject data [38, 39], generalized estimatingequation (GEE) analyses were performed. These methodsprovide standard errors adjusted by multiple observationsper person using an exchangeable working correlationstructure. The GEE estimating mechanism uses all avail-able pairs method to encompass any missing data fromdropouts or other intermittent missing. All non-missingpairs of data are used in the estimators of the working cor-relation parameters [40].As a first step, GEE bivariate analyses were conductedto determine factors associated with engaging inNMPOU. Variables significant in the bivariate analysesat p < 0.10 were considered for a full multivariate model.A backwards selection procedure was used to identifythe model with the best overall fit as indicated by thelowest quasi-likelihood under the independence modelcriterion (QIC) value [41]. The QIC value was selectedinstead of the more recently developed CorrelationInformation Criterion (CIC), as the CIC is used to selectthe appropriate intracluster correlation structure, and isnot used for covariate selection; we required a mechan-ism that could address both these needs simultaneously[42]. All statistical analyses were performed using SASsoftware version 9.4 (SAS, Cary, NC). All p-values weretwo sided.ResultsA total of 405 ARYS and 757 VIDUS participants wereeligible for this study. For ARYS, 313 (77.3%) partici-pants were enrolled before December 2013 and 92(22.7%) participants were newly enrolled during thestudy period. For the 313 participants, the median age atthe first study visit within the study period for thecurrent analysis was 26 (IQR: 23–28); and for the 92participants, the median age at study enrollment was 21(IQR: 20–23). For VIDUS, 697 (92.1%) participants wereenrolled before December 2013 and 60 (7.9%) partici-pants were newly enrolled during the study period.Among these 697 participants, the median age at thefirst study visit within study period was 49 (IQR: 42–55);and for the 60 newly recruited participants, the medianage at study enrollment was 31 (IQR: 28–34).The number of ARYS participants with at least onestudy follow-up visit was 294 (72.6%) and ARYS partici-pants attended a median of 2 study visits (IQR: 1–3). ACheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 3 of 11total of 646 (85.3%) VIDUS participants had at least onestudy follow-up visit and attended a median of 3 studyvisits (IQR: 2–3). The first ARYS observation used inthis study was recorded on December 2, 2013 and thelast observation was recorded on May 28, 2015. ARYSparticipants contributed a total of 889 observations, ofwhich 239 (26.9%) included a report of NMPOU. Thefirst VIDUS observation used in this study was recordedon December 2, 2013 and the last was recorded on May29, 2015. VIDUS participants contributed 1877 observa-tions, of which 411 (21.9%) included a report ofNMPOU. Among 405 ARYS participants included inthis analysis, 39.5% (n = 160) reported engaging inNMPOU throughout the study period. Among a total of757 VIDUS participants, 34.6% (n = 262) reportedengaging in NMPOU throughout the study period.Among ARYS participants in this sample, 135 (33.3%)were female, and 250 (61.7%) were of Caucasian ethni-city; VIDUS participants were 33.9% (n = 257) femaleand 59.7% (n = 452) Caucasian. Descriptive statistics foryounger (ARYS) and older (VIDUS) participants are dis-played in Tables 1 and 2. The bivariate and multivariateanalyses for younger participants are displayed in Table 3,and Table 4 displays the bivariate and multivariate ana-lyses for older participants. In the multivariate analyses,engaging in NMPOU was independently and positively as-sociated with the following factors among both youngerand older participants: injection or non-injection heroinuse (ARYS: Adjusted Odds Ratio [AOR] = 3.12, 95% Con-fidence Interval [CI]: 2.08–4.68; VIDUS: AOR = 2.79, 95%CI: 2.08–3.74); drug dealing (ARYS: AOR = 2.22, 95% CI:1.58–3.13; VIDUS: AOR = 1.87, 95% CI: 1.40–2.49); anddifficulty accessing services (ARYS: AOR = 1.47, 95% CI:1.04–2.09; VIDUS: AOR = 1.74, 95% CI: 1.32–2.29).Factors positively and independently associated withNMPOU among younger (ARYS) participants only in-cluded: younger age (AOR = 1.12, 95% CI: 1.05–1.19);crack cocaine use (AOR = 1.56, 95% CI: 1.06–2.30);and binge drug use (AOR = 1.41, 95% CI: 1.00–1.97).Among older (VIDUS) participants only, engaging inNMPOU was independently and positively associatedwith injection or non-injection crystal methampheta-mine use (AOR = 1.97, 95% CI: 1.46–2.66), non-fataloverdose (AOR = 1.76, 95% CI: 1.20–2.60), and sexwork (AOR = 1.49, 95% CI: 1.00–2.22).DiscussionSimilar proportions of younger and older PWUD re-ported engaging in NMPOU in this study (40% of ARYS;35% of VIDUS), and in separate multivariate analyses,NMPOU was positively associated with heroin use, drugdealing, and difficulty accessing services among both agegroups. In addition, younger participants who engagedin NMPOU were more likely to be younger, use crackTable 1 Characteristics of younger participants stratified by engaging in nonmedical prescription opioid use over the study period,2013–2015 (n = 405)Characteristica Total (%)(n = 405)Nonmedical Prescription Opioid Use p -valueYes (%)(n = 160)No (%)(n = 245)Age (M [IQR]) 25 (22–28) 23 (21–26) 26 (23–28) < 0.001Any cocaine useb, c 123 (30.4) 60 (37.5) 63 (25.7) 0.012Any crack cocaine useb, c 138 (34.1) 70 (43.8) 68 (27.8) < 0.001Any crystal meth useb, c 268 (66.2) 121 (75.6) 147 (60.0) 0.001Any heroin useb, c 193 (47.7) 114 (71.3) 79 (32.2) < 0.001Any non-fatal overdoseb, c 93 (23.0) 54 (33.8) 39 (15.9) < 0.001Binge drug useb, c 208 (51.4) 108 (67.5) 100 (40.8) < 0.001Caucasian ancestry 250 (61.7) 100 (62.5) 150 (61.2) 0.796Difficulty accessing servicesb, d 151 (37.3) 73 (45.6) 78 (31.8) 0.005Drug dealingb 131 (32.3) 78 (48.8) 53 (21.6) < 0.001Female 135 (33.3) 49 (30.6) 86 (35.1) 0.350Homelessb 202 (49.9) 95 (59.4) 107 (43.7) 0.002Incarcerationb 57 (14.1) 26 (16.3) 31 (12.7) 0.297Regular employmentb 191 (47.2) 81 (50.6) 110 (44.9) 0.259Sex workb 48 (11.9) 26 (16.3) 22 (9.0) 0.027a. Comparison is yes versus no unless otherwise specifiedb. Refers to activities, behaviours, and experiences in the last six monthsc. Includes injection and non-injection drug used. Includes health and social servicesCheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 4 of 11Table 2 Characteristics of older participants stratified by engaging in nonmedical prescription opioid use over the study period,2013–2015 (n = 757)Characteristica Total (%)(n = 757)Nonmedical Prescription Opioid Use p -valueYes (%)(n = 262)No (%)(n = 495)Age (M [IQR]) 48 (40–55) 47 (38–53) 49 (41–56) < 0.001Any cocaine useb, c 205 (27.1) 98 (37.4) 107 (21.6) < 0.001Any crack cocaine useb, c 332 (43.9) 135 (51.5) 197 (39.8) 0.002Any crystal meth useb, c 223 (29.5) 124 (47.3) 99 (20.0) < 0.001Any heroin useb, c 362 (47.8) 200 (76.3) 162 (32.7) < 0.001Any non-fatal overdoseb, c 60 (7.9) 43 (16.4) 17 (3.4) < 0.001Binge drug useb, c 235 (31.0) 118 (45.0) 117 (23.6) < 0.001Caucasian ancestry 452 (59.7) 164 (62.6) 288 (58.2) 0.239Difficulty accessing servicesb, d 149 (19.7) 78 (29.8) 71 (14.3) < 0.001Drug dealingb 171 (22.6) 95 (36.3) 76 (15.4) < 0.001Female 257 (33.9) 92 (35.1) 165 (33.3) 0.622Homelessb 149 (19.7) 76 (29.0) 73 (14.7) < 0.001Incarcerationb 51 (6.7) 28 (10.7) 23 (4.6) 0.002Regular employmentb 202 (26.7) 59 (22.5) 143 (28.9) 0.059Sex workb 72 (9.5) 36 (13.7) 36 (7.3) 0.004a. Comparison is yes versus no unless otherwise specifiedb. Refers to activities, behaviours, and experiences in the last six monthsc. Includes injection and non-injection drug used. Includes health and social servicesTable 3 Bivariate and multivariate GEE analyses of factors associated with engaging in nonmedical prescription opioid use amongyounger participants (n = 405)Characteristica Unadjusted AdjustedOdds Ratio(95% CI)p - value Odds Ratio(95% CI)p - valueAge (per year younger) 1.18 (1.12–1.25) < 0.001 1.12 (1.05–1.19) < 0.001Any cocaine useb, c 1.68 (1.18–2.39) 0.004 1.31 (0.88–1.95) 0.181Any crack useb, c 2.16 (1.53–3.05) < 0.001 1.56 (1.06–2.30) 0.023Any crystal meth useb, c 2.06 (1.42–2.98) < 0.001 –Any heroin useb, c 4.82 (3.34–6.96) < 0.001 3.12 (2.08–4.68) < 0.001Any non-fatal overdoseb, c 2.24 (1.61–3.12) < 0.001 1.43 (0.97–2.10) 0.070Binge drug useb, c 2.36 (1.76–3.15) < 0.001 1.41 (1.00–1.97) 0.049Caucasian ancestry 1.24 (0.85–1.80) 0.269 –Difficulty accessing servicesb, d 1.70 (1.27–2.27) < 0.001 1.47 (1.04–2.09) 0.030Drug dealingb 2.76 (2.03–3.75) < 0.001 2.22 (1.58–3.13) < 0.001Female 0.91 (0.61–1.36) 0.662 –Homelessb 1.59 (1.16–2.17) 0.004 –Incarcerationb 1.32 (0.88–1.99) 0.177 –Regular employmentb 1.23 (0.92–1.64) 0.168 –Sex workb 2.11 (1.35–3.29) < 0.001 –a. Comparison is yes vs. no unless otherwise specifiedb. Refers to behaviours, activities, and experiences in the last six monthsc. Includes injection and non-injection used. Includes health and social servicesCheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 5 of 11cocaine, and engage in binge drug use; older participantswho engaged in NMPOU were more likely to use crystalmethamphetamine, report a recent non-fatal overdose,and engage in sex work. While younger and olderPWUD in these analyses shared risk factors for engagingin NMPOU, this study also found important differencesbetween these two age groups that highlight opportun-ities to develop targeted efforts that address NMPOUand unique risks for each age group.The association between NMPOU and age was notfound to be statistically significant in the analysis of olderparticipants, which suggests that birth cohorts were not ameaningful indicator of NMPOU among this sample.Conversely, the association between NMPOU and youn-ger age was statistically significant among younger partici-pants, and this finding may reflect recent trends wherePO use is increasingly prevalent among young age groups[4, 12, 43]. The increasing use of POs has been attributedto the availability of POs [4, 17, 44], although theincreased risk for younger birth cohorts using POs may bemore strongly related to the linkages between youthobserving their parents modelling substance use (i.e., POuse) and then concluding that PO use is safe [4]. Prevent-ing initiation into injection drug use for youth who engagein NMPOU is key to reducing the sequelae of harms asso-ciated with intensifying substance use, and future researchinvestigating prevention mechanisms is needed.Younger and older participants who engaged inNMPOU were significantly more likely to use heroin,which previous research has found to be used as a replace-ment for POs when PO availability is low [45]. Althoughheroin and POs are both opioids and central nervous sys-tem depressants, POs are originally obtained from regu-lated healthcare sources, and heroin is only availablethrough unregulated illegal drug markets. Acquiring sub-stances from the street is especially problematic in ourstudy setting, where the toxic synthetic opioid fentanylhas adulterated a substantial proportion of the illegal drugsupply [46, 47], and fentanyl-related overdose mortalityhas increased alarmingly in various settings across Canadaand the United States [48–51]. To reduce reliance onillegal heroin, oral medications such as buprenorphine/na-loxone (Suboxone), methadone, naltrexone, and slowrelease morphine have been recommended in recentopioid treatment guidelines as effective treatments [52],and injectable opioids such as diacetylmorphine andhydromorphone are emerging as options for treatment-refractory opioid dependence [53, 54]. The internationaland North American evidence base for heroin assistedtreatment (diacetylmorphine) is strong [53–55], and al-though less research has been conducted on hydromor-phone as an injectable opioid treatment, scaling up thesetreatments may be an important tool to reduce NMPOUand exposure to contaminated illegal drugs in Vancouver.Table 4 Bivariate and multivariate GEE analyses of factors associated with engaging in nonmedical prescription opioid use amongolder participants (n = 757)Characteristica Unadjusted AdjustedOdds Ratio(95% CI)p - value Odds Ratio(95% CI)p - valueAge (per year younger) 1.03 (1.10–1.04) < 0.001 0.99 (0.97–1.00) 0.120Any cocaine useb, c 1.53 (1.18–1.99) 0.001 –Any crack useb, c 1.42 (1.12–1.81) 0.004 –Any crystal meth useb, c 2.95 (2.25–3.88) < 0.001 1.97 (1.46–2.66) < 0.001Any heroin useb, c 3.81 (2.90–4.99) < 0.001 2.79 (2.08–3.74) < 0.001Any non-fatal overdoseb, c 3.09 (2.12–4.48) < 0.001 1.76 (1.20–2.60) 0.004Binge drug useb, c 1.71 (1.37–2.14) < 0.001 –Caucasian ancestry 1.20 (0.90–1.60) 0.225 –Difficulty accessing servicesb, d 2.11 (1.65–2.69) < 0.001 1.74 (1.32–2.29) < 0.001Drug dealingb 2.75 (2.10–3.58) < 0.001 1.87 (1.40–2.49) < 0.001Female 0.97 (0.73–1.31) 0.864 –Homelessb 2.05 (1.50–2.79) < 0.001 –Incarcerationb 2.28 (1.47–3.54) < 0.001 1.55 (0.98–2.44) 0.061Regular employmentb 0.78 (0.60–1.00) 0.055 –Sex workb 2.05 (1.41–2.98) < 0.001 1.49 (1.00–2.22) 0.049a. Comparison is yes vs. no unless otherwise specifiedb. Refers to behaviours, activities, and experiences in the last six monthsc. Includes injection and non-injection used. Includes health and social servicesCheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 6 of 11It is important to note that PWUD have long advocatedfor access to a wider spectrum of opioid agonist treat-ments through collaborations with researchers andcoordinated advocacy efforts [56–61]. This study did notcontrol for intentional or unintentional exposure tofentanyl or other illicitly manufactured synthetic opioids,and future research using the ARYS and VIDUS cohortswould benefit from including exposure to fentanyl inanalyses.More differences than similarities were found with il-legal substance use patterns between younger and olderparticipants. Despite both younger and older PWUD whoengaged in NMPOU being significantly more likely to en-gage in heroin use, younger participants who engaged inNMPOU were more likely to also engage in crack cocaineuse and binge drug use, while older participants weremore likely to use crystal methamphetamine in additionto engaging in NMPOU. Crystal methamphetamine usemay be an important marker of risk for NMPOU amongolder individuals in Vancouver, and this finding is con-cerning given local reports of increasing crystal metham-phetamine use among adults in Vancouver [62–64].Historically, crystal methamphetamine use has been moreprevalent among street-involved youth in Vancouver [36],as 66% of all participants in the younger age groupreported using crystal methamphetamine at baseline; inaddition, crystal methamphetamine has been associatedwith initiating injection drug use among this cohort ofyouth [65].The findings from the multivariate model indicate thatNMPOU among younger participants was not associatedwith a significantly increased risk for non-fatal overdose,whereas older participants who engaged in NMPOU weresignificantly more likely to report a recent non-fatal over-dose; however, it should be noted that the confidence inter-vals for these adjusted odds ratios overlap considerably andthese findings may be attributable to differences in selectioncriteria for the ARYS and VIDUS cohorts. This null resultfor youth was unexpected given the increased rate of over-dose associated with PO use [66], and that the comparisongroup in this analysis included non-opioid users. Thesefindings align with previous research findings that older ageis associated with mortality due to unintentional PO-related overdose [67]. Further research investigating pro-tective factors associated with a lower risk of overdoseamong youth who engage in NMPOU is needed, whichmay include specific routes of administration or harm redu-cing practices related to using a substance with a knowndosage and purity (e.g., prescription opioids versus unregu-lated heroin of unknown purity and composition). Regard-less, harm reduction services are critical to ensure thatolder PWUD, as well as younger PWUD, who engage inNMPOU have access to overdose prevention and reversalservices such as supervised drug consumption spaces/services and the widespread distribution of Naloxone/Nar-can in the community and among PWUD. The reach ofharm reduction services may also be increased by strategic-ally mobilizing key peers within PWUDs’ social networks,which are an untapped resource and could becomeimportant facilitators of harm reduction supplies andinformation [68, 69].Improved access to health and social services is alsoespecially important in Vancouver, as this study found thatolder and younger PWUD who engage in NMPOU weremore likely to report difficulty accessing services. Despitea saturation of services in the neighbourhoods whereARYS and VIDUS participants primarily live and congre-gate (the Downtown South and Downtown Eastside, re-spectively), there remain important gaps in service designand access. Previous research among youth in the ARYScohort found that local youth-focused shelters and hous-ing services had strict rules governing entry into- and con-tinued use of- the service that deterred participants;conversely, adult or non-youth-specific services were per-ceived as unsafe or inappropriate for youth [70]. Qualita-tive research among the VIDUS cohort has found that alocal supervised injection site often has long wait timesthat result in people giving up and using drugs elsewhere[71] and “area restrictions” used by police to prohibit entryinto “drug scenes” impedes access to services and supportstailored for PWUD and are often specifically located inareas with high drug use [72]. This finding among youngerPWUD is consistent with previous research from aroundthe world; youth who use illegal drugs often experiencedifficulty accessing services due to stigma and discrimin-ation from service providers, as well as a lack ofyouth-centric health and social services that are preferredbut not widely available [73, 74]. Although the results in-dicate high rates of illegal polysubstance use in the ARYSand VIDUS cohorts, participants who only engage inNMPOU may experience additional difficulties accessingharm reduction services that are tailored to people whouse illegal drugs, since they and their social networks maybe outside the scope of outreach activities conducted bythese services [75, 76]. More research is required to betterunderstand specific barriers to accessing health and socialservices among youth and adults who engage in NMPOU,as well as inform effective solutions to fill an importantservice gap for these individuals.Participants in both age groups who engage in NMPOUwere more likely to report recent drug dealing, and olderparticipants were more likely to report sex work. Drugdealing has been associated with more intense substanceuse among PWUD of Caucasian or white ethnicity [77],and dealing illegal drugs has been associated with a higherlikelihood of NMPOU among American youth [78]. Ourfindings reflect the well-established relationship betweensocio-economic marginalization and drug dealing [40, 79],Cheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 7 of 11and align with consistent research findings that incomefrom drug dealing [80] and sex work [81] is often used tosustain ongoing substance use. Despite aligning with pre-vious research, it is not clear why drug dealing and sexwork remained significantly associated with engaging inNMPOU after controlling for other illegal drug use; moreresearch is needed investigating NMPOU and incomegeneration among individuals not typically recruited inpopulation-level surveys. Given that difficulty accessingservices was also significantly associated with NMPOUamong younger and older participants, there is a clearneed to improve employment and other services forPWUD who engage in NMPOU as an intervention to in-crease socio-economic independence, particularly amongpeople who engage in sex work. Similar efforts to facilitateentry into evidence-based addiction treatments for opioiduse such as opioid agonist treatment, heroin assisted treat-ment, and injectable opioid therapy may reduce the preva-lence of NMPOU and risky income generating activities.As there is a lack of evidence to support the effective-ness of POs for treating chronic pain [82], safer prescrib-ing to limit the supply and availability of POs isimportant for reducing the incidence and prevalence ofNMPOU in the study setting. However, as the supply ofPOs becomes more restricted, close surveillance and asuite of interventions are needed to ensure that thosewho engage in NMPOU are not at greater risk for sub-stituting PO use with contaminated illegal drugs thathave increased overdose risks. Low barrier harm reduc-tion services to connect with those who engage inNMPOU and have had difficulty accessing services mayalso be an important link to supportive healthcare ser-vices, social services, and a range of treatment optionsfor opioid use among younger and older age groups. Inaddition, future research investigating the correlates ofincident and recurrent NMPOU over time is needed tobetter understand how access to services and changes inhealthcare practices impact NMPOU among high-riskindividuals who use illegal drugs.There are limitations to this research. First, ARYS andVIDUS participants may not be representative of allPWUD in Vancouver and the results therefore notgeneralizable to other settings in the city. However, exten-sive street-based and outreach efforts were undertaken inorder to recruit a representative sample, and thesocio-demographic characteristics of participants in theARYS and VIDUS studies are similar to other studies inVancouver [83, 84]. Second, this study compares datafrom two cohort studies with different inclusion criteria,which may result in cohort or selection effects. Thesecohort effects may affect the results related to homeless-ness and injection drug use, as these risk factors are alsoinclusion criteria for ARYS and VIDUS, respectively.Given the longitudinal nature of these cohort studies andextensive efforts to track participants over multiple studyvisits, the ARYS and VIDUS studies have observedchanges in behaviours and risk factors over time. For ex-ample, we have previously reported on transitions out ofhomelessness among ARYS participants [85] and injectioncessation among VIDUS participants [86, 87]. In addition,ARYS and VIDUS have previously been combined inquantitative and qualitative analyses [72, 88–90]. Third,social desirability and recall bias may have resulted in er-roneous reporting of our outcome and independent vari-ables. Previous research on substance use has founddiscrepancies between self-reported substance use andbioassay test results among American adult male arrestees[91] and noted concerns that youth may not be truthfulabout substance use when speaking with authority figureswho are able to assign punishment [92]. Despite thesefindings, self-reported substance use, criminality, andHIV-related risk-taking among PWUD has also beendeemed sufficiently reliable and valid [93]. Training andengaging PWUD (“peers”) in survey administration anddata collection methods when conducting substance useresearch may be an important mechanism to reduce socialdesirability response biases and increase capacity withincommunities of PWUD [94, 95]. To reduce socially desir-able reporting from participants, the ARYS and VIDUS in-terviewers are trained in building trust and rapport, andstudy instruments situate sensitive questions towards theend of the questionnaire to allow interviewers to buildrapport with participants. Although some socially desir-able reporting is inevitable, any such reporting fromparticipants would be expected to under-estimate theprevalence of sensitive risk factors and therefore our find-ings likely represent conservative estimates. Less is knownabout the accuracy of self-reporting NMPOU amongPWUD; however, efforts to improve the accuracy ofreporting NMPOU among ARYS and VIDUS participantsincluded using both the generic and brand name of POs,and showing pictures of a wide variety of POs duringstudy visits. Fourth, no other information about the cir-cumstances surrounding non-fatal overdose were includedin this analysis (e.g., what substance was used at the timeof overdose, whether substances were used alone or withanother person, or whether fentanyl contributed to theoverdose); however, the focus of this study was whetherrecent NMPOU was an independent marker for overdoseamong other risk factors. A more detailed investigation ofthe circumstances of overdose among people who engagein NMPOU is outside the scope of this analysis but is apromising direction for future research.ConclusionsThe shared risk factors among younger and older partici-pants who engage in NMPOU underscore the importanceof addressing barriers to accessing health and socialCheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 8 of 11services, as people who engage in NMPOU appear to beparticularly under-supported and under-served by existingservices for PWUD. Adult PWUD who engage inNMPOU are also at greater risk of overdose, which high-lights the need for youth and adult-specific strategies thatfocus on reducing high intensity substance use amongyouth and providing low barrier overdose prevention andreversal services for adults. There is an urgent need todesign and implement initiatives to improve healthcareproviders’ skills with managing and treating substance de-pendence [96], as well as developing pain treatment strat-egies tailored for PWUD [26]. In addition to developingservices that address youths’ and adults’ unique needs,policy-makers and healthcare providers are urged toreduce systematic barriers to a range of addiction treat-ment options for opioid use that may contribute to reduc-tions in the prevalence of NMPOU and provide anadditional benefit of preventing other PWUD from initiat-ing injection drug use [97].AbbreviationsAOR: Adjusted Odds Ratio; ARYS: At-Risk Youth Study; CI: Confidence Interval;GEE: Generalized Estimating Equation; IQR: Inter-Quartile Range; NMPOU: Non-Medical Prescription Opioid Use; PO: Prescription Opioid; PWUD: People whouse Illegal Drugs; QIC: Quasi-Likelihood under the Independence modelCriterion; VIDUS: Vancouver Injection Drug Users StudyAcknowledgmentsThe authors thank the ARYS and VIDUS study participants for their contributionto the research, as well as current and past researchers and staff. Special thanksare also extended to Dr. Robert Hogg who provided valuable advice on thismanuscript. Dr. Kora DeBeck is supported by a MSFHR / St. Paul’s HospitalFoundation–Providence Health Care Career Scholar Award and a CanadianInstitutes of Health Research (CIHR) New Investigator Award. Dr. Will Small issupported by a Michael Smith Foundation for Health Research CareerInvestigator Scholar Award. Dr. Kanna Hayashi is supported by a CIHR NewInvestigator Award (MSH-141971) and a Michael Smith Foundation for HealthResearch (MSFHR) Scholar Award.FundingThis work was supported by the US National Institutes of Health under GrantU01DA038886; and Canadian Institutes of Health Research under GrantMOP–286532.Availability of data and materialsThe datasets generated and analysed during the current study are not publiclyavailable due to assurances of strict confidentiality given to participants duringthe informed consent process.Authors’ contributionsTC contributed to the study design, statistical analyses, and took primaryresponsibility for preparing the manuscript. HD and EN were responsible forconducting the statistical analyses. WS, HD, EN, KH, and KD contributedsubstantially to the study design and main content of the manuscript. Allauthors read and approved the final manuscript.Authors’ informationNot applicable.Ethics approval and consent to participateAll participants gave informed consent before enrolling in ARYS and VIDUS.The ARYS and VIDUS studies have been approved by the University of BritishColumbia/Providence Health Care Research Ethics Board.Consent for publicationNot applicable.Competing interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Faculty of Health Sciences, Simon Fraser University, Blusson Hall, Room11300, 8888 University Drive, Burnaby, BC V5A 1S6, Canada. 2British ColumbiaCentre on Substance Use, Providence Health Care, 400-1045 Howe St,Vancouver, BC V6Z 2A9, Canada. 3School of Population and Public Health,University of British Columbia, 2206 East Mall, Vancouver, BC V6T 1Z3,Canada. 4Centre for Applied Research in Mental Health and Addiction, SFUFaculty of Health Sciences, 515 W. Hastings Street, Vancouver, BC V6B 5K3,Canada. 5School of Public Policy, Simon Fraser University, 515 West HastingsStreet, Suite 3271, Vancouver, BC V6B 5K3, Canada.Received: 22 May 2018 Accepted: 13 November 2018References1. Rudd RA, Seth P, David F, Scholl L. Increases in drug and opioid-involvedoverdose deaths — United States, 2010–2015. MMWR Morb Mortal WklyRep. 2016;65:1445–52.2. Nolan S, Socias ME, Wood E. The Threat of an International Opioid Crisis.Current Addiction Reports; 2018.3. Meier EA, Troost JP, Anthony JC. Extramedical use of prescription painrelievers by youth aged 12 to 21 years in the United States: nationalestimates by age and by year. Arch Pediatr Adolesc Med. 2012;166(9):803–7.4. Miech R, Bohnert A, Heard K, Boardman J. Increasing use of nonmedicalanalgesics among younger cohorts in the United States: a birth cohorteffect. J Adolesc Health. 2013;52(1):35–41.5. Fischer B, Ialomiteanu A, Kurdyak P, Mann RE, Rehm J. Reductions in non-medical prescription opioid use among adults in Ontario, Canada: are recentpolicy interventions working? Subst/ Abus Treatment Prev Pol. 2013;8:4.6. Currie CL, Wild TC. Adolescent use of prescription drugs to get high inCanada. Can J Psychiatry. 2012;57(12):745–51.7. Schepis TS, Krishnan-Sarin S. Characterizing adolescent prescriptionmisusers: a population-based study. J Am Acad Child Adolesc Psychiatr.2008;47(7):745–54.8. Veliz PT, Boyd C, McCabe SE. Playing through pain: sports participation andnonmedical use of opioid medications among adolescents. Am J PublicHealth. 2013;103(5):e28–30.9. Deandrea DC, Troost JP, Anthony JC. Toward primary prevention of extra-medical OxyContin(R) use among young people. Prev Med. 2013;57(3):244–6.10. Back SE, Payne RL, Simpson AN, Brady KT. Gender and prescription opioids:findings from the National Survey on drug use and health. Addict Behav.2010;35(11):1001–7.11. Shield KD, Ialomiteanu A, Fischer B, Mann RE, Rehm J. Non-medical useof prescription opioids among Ontario adults: data from the 2008/2009CAMH Monitor. Can J Public Health = Revue canadienne de santepublique. 2011;102(5):330–5.12. Martins SS, Keyes KM, Storr CL, Zhu H, Grucza RA. Birth-cohort trends inlifetime and past-year prescription opioid-use disorder resulting fromnonmedical use: results from two national surveys. J Stud AlcoholDrugs. 2010;71(4):480–7.13. Suryaprasad AG, White JZ, Xu FJ, Eichler BA, Hamilton J, Patel A, et al.Emerging epidemic of hepatitis C virus infections among young nonurbanpersons who inject drugs in the United States, 2006-2012. Clin Infect Dis.2014;59(10):1411–9.14. Back SE, Lawson KM, Singleton LM, Brady KT. Characteristics and correlatesof men and women with prescription opioid dependence. Addict Behav.2011;36(8):829–34.15. Osgood ED, Eaton TA, Trudeau JJ, Katz NP. A brief survey to characterizeoxycodone abuse patterns in adolescents enrolled in two substance abuserecovery high schools. Am J Drug Alcohol Abuse. 2012;38(2):166–70.Cheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 9 of 1116. Cerdá M, Santaella J, Marshall BDL, Kim JH, Martins SS. NonmedicalPrescription Opioid Use in Childhood and Early Adolescence PredictsTransitions to Heroin Use in Young Adulthood: A National Study. J Pediatr.2015;167(3):605–12.e1 2.17. Lankenau SE, Teti M, Silva K, Bloom JJ, Harocopos A, Treese M. Initiation intoprescription opioid misuse amongst young injection drug users. Int J DrugPolicy. 2012;23(1):37–44.18. Lawson KM, Back SE, Hartwell KJ, Maria MMS, Brady KT. A comparison oftrauma profiles among individuals with prescription opioid, nicotine, orcocaine dependence. Am J Addict. 2013;22(2):127–31.19. Caviness CM, Anderson BJ, de Dios MA, Kurth M, Stein M. Prescriptionmedication exchange patterns among methadone maintenance patients.Drug Alcohol Depend. 2013;127(1–3):232–8.20. Arnett JJ. Emerging adulthood: what is it, and what is it good for? ChildDev Perspect. 2007;1(2):68–73.21. Degenhardt L, Stockings E, Patton G, Hall WD, Lynskey M. The increasingglobal health priority of substance use in young people. The LancetPsychiatry. 2016;3(3):251–64.22. Saha TD, Kerridge BT, Goldstein RB, Chou SP, Zhang H, Jung J, et al.Nonmedical prescription opioid use and DSM-5 nonmedical prescriptionopioid use disorder in the United States. J Clin Psychiatry. 2016;77(6):772–80.23. Chau DL, Walker V, Pai L, Cho LM. Opiates and elderly: use and side effects.Clin Interv Aging. 2008;3(2):273–8.24. Han B, Compton WM, Blanco C, Crane E, Lee J, Jones CM. Prescriptionopioid use, misuse, and use disorders in u.s. adults: 2015 national survey ondrug use and health. Ann Intern Med. 2017;167(5):293–301.25. Lake S, Milloy MJ, Dong H, Hayashi K, Wood E, Kerr T, et al. Initiation intoprescription opioid injection and associated trends in heroin use amongpeople who use illicit drugs. Drug Alcohol Depend. 2016;169:73–9.26. Voon P, Callon C, Nguyen P, Dobrer S, Montaner JS, Wood E, et al. Denial ofprescription analgesia among people who inject drugs in a Canadiansetting. Drug Alcohol Rev. 2015;34(2):221–8.27. Voon P, Callon C, Nguyen P, Dobrer S, Montaner J, Wood E, et al. Self-management of pain among people who inject drugs in Vancouver. PainManag. 2014;4(1):27–35.28. DeBeck K, Wood E, Dong H, Dobrer S, Hayashi K, Montaner J, et al. Non-medical prescription opioid use predicts injection initiation among street-involved youth. Int J Drug Policy. 2016;34:96–100.29. Silva K, Schrager SM, Kecojevic A, Lankenau SE. Factors associated withhistory of non-fatal overdose among young nonmedical users ofprescription drugs. Drug Alcohol Depend. 2013;128(1–2):104–10.30. Fischer B, Patra J, Cruz MF, Gittins J, Rehm J. Comparing heroin users andprescription opioid users in a Canadian multi-site population of illicit opioidusers. Drug And Alcohol Rev. 2008;27(6):625–32.31. Havens JR, Lofwall MR, Frost SDW, Oser CB, Leukefeld CG, Crosby RA.Individual and network factors associated with prevalent hepatitis Cinfection among rural Appalachian injection drug users. Am J PublicHealth. 2013;103(1):e44–52.32. Gwin Mitchell S, Kelly SM, Brown BS, Schacht Reisinger H, Peterson JA, RuhfA, et al. Uses of diverted methadone and buprenorphine by opioid-addicted individuals in Baltimore, Maryland. Am J Addict. 2009;18(5):346–55.33. Young AM, Havens JR, Leukefeld CG. A comparison of rural and urbannonmedical prescription opioid users’ lifetime and recent drug use. Am JDrug and Alcohol Abuse. 2012;38(3):220–7.34. Pollini RA, Banta-Green CJ, Cuevas-Mota J, Metzner M, Teshale E, Garfein RS.Problematic use of prescription-type opioids prior to heroin use amongyoung heroin injectors. Subst Abus Rehabil. 2011;2:173–80.35. Rhodes T, Lilly R, Fernandez C, Giorgino E, Kemmesis UE, Ossebaard HC,et al. Risk factors associated with drug use: the importance of ‘riskenvironment’. Drug-Educ Prev Polic. 2003;10(4):303–29.36. Wood E, Stoltz J-A, Montaner J, Kerr T. Evaluating methamphetamine useand risks of injection initiation among street youth: the ARYS study. HarmReduct J. 2006;3(1):18.37. Strathdee SA, Patrick DM, Currie SL, Cornelisse PG, Rekart ML, Montaner JS,et al. Needle exchange is not enough: lessons from the Vancouver injectingdrug use study. AIDS. 1997;11(8):F59–65.38. Ballinger GA. Using generalized estimating equations for longitudinal dataanalysis. Organ Res Methods. 2004;7(2):127–50.39. Hanley JA, Negassa A, Edwardes MD, Forrester JE. Statistical analysis ofcorrelated data using generalized estimating equations: an orientation. AmJ Epidemiol. 2003;157(4):364–75.40. Ferguson KM, Bender K, Thompson SJ, Maccio EM, Pollio D. Employmentstatus and income generation among homeless young adults. Youth Soc.2011;44(3):385–407.41. Pan W. Akaike's information criterion in generalized estimating equations.Biometrics. 2001;57(1):120–5.42. Hin L-Y, Wang Y-G. Working-correlation-structure identification ingeneralized estimating equations. Stat Med. 2008;28(4):642–58.43. Fischer B, Ialomiteanu A, Boak A, Adlaf E, Rehm J, Mann RE. Prevalence andkey covariates of non-medical prescription opioid use among the generalsecondary student and adult populations in Ontario, Canada. Drug AndAlcohol Rev. 2013;32(3):276–87.44. Substance Abuse and Mental Health Services Administration. Results fromthe 2009 National Survey on Drug Use and Health: Volume I. Summary ofNational Findings. Rockville, MD: (Office of Applied Studies - NSDUH SeriesH-38A, HHS Publication No. SMA 10–4586 Findings); 2010.45. Cicero TJ, Ellis MS, Surratt HL, Kurtz SP. The changing face of heroin use inthe United States: a retrospective analysis of the past 50 years. JAMAPsychiatry. 2014;71(7):821–6.46. Fairbairn N, Coffin PO, Walley AY. Naloxone for heroin, prescription opioid, andillicitly made fentanyl overdoses: Challenges and innovations responding to adynamic epidemic. Int J Drug Policy. 2017;46(Supplement C):172–9.47. Lysyshyn M, Dohoo C, Forsting S, Kerr T, McNiel R, editors. Evaluation of afentanyl drug checking program for clients of a supervised injection site,Vancouver, Canada. Montreal: Harm reduction international; 2017.48. British Columbia Coroners Service. Fentanyl-detected illicit drug overdosedeaths January 1, 2012 to august 31, 2017. Burnaby, BC: Ministry of PublicSafety and Solicitor General; 2017.49. Ciccarone D. Fentanyl in the US heroin supply: a rapidly changing riskenvironment. Int J Drug Policy. 2017;46:107–11.50. Centers for Disease Control and Prevention. Increases in Fentanyl DrugConfiscations and Fentanyl-related Overdose Fatalities [press release].Atlanta: Office of Public Health Preparedness and Response; 2015.51. Fischer B, Vojtila L, Rehm J. The ‘fentanyl epidemic’ in Canada – some cautionaryobservations focusing on opioid-related mortality. Prev Med. 2018;107:109–13.52. British Columbia Centre on Substance Use and B.C. Ministry of Health. Vancouver:A Guideline for the Clinical Management of Opioid Use Disorder; 2017.53. Oviedo-Joekes E, Guh D, Brissette S, Marchand K, MacDonald S, Lock K, et al.Hydromorphone compared with diacetylmorphine for long-term opioiddependence: a randomized clinical trial. JAMA psychiatry. 2016;73(5):447–55.54. Oviedo-Joekes E, Brissette S, Marsh DC, Lauzon P, Guh D, Anis A, et al.Diacetylmorphine versus methadone for the treatment of opioid addiction.New Engl J Med. 2009;361(8):777–86.55. Ferri M, Davoli M, Perucci CA. Heroin maintenance for chronic heroin-dependent individuals. Cochrane Database Syst Rev. 2010;(8):Cd003410.https://doi.org/10.1002/14651858.CD003410.pub3.56. Thomson E, Lampkin H, Maynard R, Karamouzian M, Jozaghi E. The lessonslearned from the fentanyl overdose crises in British Columbia, Canada.Addiction. 2017;112(11):2068–70.57. 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.Subst Abuse Treat Prev Policy. 2018;13(1):18.58. Jozaghi E, Marsh S. Missing the trends in the fentanyl overdose crisis: Theneed for immediate intervention in small and rural communities. Can JPublic Health = Revue canadienne de sante publique. 2017;108(4):e457.59. Boyd S, Association NP. Yet they failed to do so: recommendations basedon the experiences of NAOMI research survivors and a call for action. HarmReduct J. 2013;10(1):6.60. Boyd S, Murray D, Snap, MacPherson D. Telling our stories: heroin-assistedtreatment and SNAP activism in the Downtown Eastside of Vancouver.Harm Reduct J. 2017;14:27.61. McNeil R, Kerr T, Anderson S, Maher L, Keewatin C, Milloy MJ, et al.Negotiating structural vulnerability following regulatory changes to aprovincial methadone program in Vancouver, Canada: a qualitative study.Soc Sci Med. 2015;133:168–76.62. Woo A. The quiet rise of crystal meth. The Globe and Mail; 2016.63. Amlani A, McKee G, Khamis N, Raghukumar G, Tsang E, Buxton JA. Why theFUSS (fentanyl urine screen study)? A cross-sectional survey to characterizean emerging threat to people who use drugs in British Columbia, Canada.Harm Reduct J. 2015;12(1):54.64. Merali F. ‘The elephant in the room’: crystal meth use in Vancouver rises600% over last decade. CBC News; 2017.Cheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 10 of 1165. Werb D, Kerr T, Buxton J, Shoveller J, Richardson C, Montaner J, et al. Crystalmethamphetamine and initiation of injection drug use among street-involved youth in a Canadian setting. CMAJ : Canadian Medical AssociationJournal. 2013;185(18):1569–75.66. Rudd RA, Aleshire N, Zibbell JE, Matthew GR. Increases in drug and opioidoverdose deaths—United States, 2000–2014. Am J Transplant. 2016;16(4):1323–7.67. Paulozzi LJ, Kilbourne EM, Shah NG, Nolte KB, Desai HA, Landen MG, et al. Ahistory of being prescribed controlled substances and risk of drug overdosedeath. Pain medicine (Malden, Mass). 2012;13(1):87–95.68. Brothers S. Merchants, samaritans, and public health workers: secondarysyringe exchanger discursive practices. Int J Drug Policy. 2016;37:1–8.69. Bouchard M, Hashimi S, Tsai K, Lampkin H, Jozaghi E. Back to the core: anetwork approach to bolster harm reduction among persons who injectdrugs. Int J Drug Policy. 2018;51:95–104.70. Krüsi A, Fast D, Small W, Wood E, Kerr T. Social and structural barriers tohousing among street-involved youth who use illicit drugs. Health Soc Carecommunity. 2010;18(3):282–8.71. Small W, Shoveller J, Moore D, Tyndall M, Wood E, Kerr T. Injection drugusers’ access to a supervised injection facility in Vancouver, Canada: theinfluence of operating policies and local drug culture. Qual Health Res.2011;21(6):743–56.72. McNeil R, Cooper H, Small W, Kerr T. Area restrictions, risk, harm, and healthcare access among people who use drugs in Vancouver, Canada: a spatiallyoriented qualitative study. Health Place. 2015;35:70–8.73. Krug A, Hildebrand M, Sun N. “We don’t need services. We have no problems”:exploring the experiences of young people who inject drugs in accessingharm reduction services. J Int AIDS Soc. 2015;18:19442 (1758–2652 (Electronic)).74. Merkinaite S, Grund JP, Frimpong A. Young people and drugs: nextgeneration of harm reduction. Int J Drug Policy. 2010;21(2):112–4.75. Frank D, Mateu-Gelabert P, Guarino H, Bennett A, Wendel T, Jessell L,et al. High risk and little knowledge: overdose experiences andknowledge among young adult nonmedical prescription opioid users.2015(1873–4758 (Electronic)).76. Bruneau J, Roy E, Arruda N, Zang G, Jutras-Aswad D. The rising prevalenceof prescription opioid injection and its association with hepatitis Cincidence among street-drug users. Addiction. 2012;107(7):1318–27.77. Floyd LJ, Alexandre PK, Hedden SL, Lawson AL, Latimer WW, Giles N.Adolescent drug dealing and race/ethnicity: a population-based study ofthe differential impact of substance use on involvement in drug trade. Am JDrug and Alcohol Abuse. 2010;36(2):87–91.78. Sung HE, Richter L, Vaughan R, Johnson PB, Thom B. Nonmedical use ofprescription opioids among teenagers in the United States: trends andcorrelates. J Adolesc Health. 2005;37(1):44–51.79. DeBeck K, Shannon K, Wood E, Li K, Montaner J, Kerr T. Income generatingactivities of people who inject drugs. Drug Alcohol Depend. 2007;91(1):50–6.80. Werb D, Kerr T, Li K, Montaner J, Wood E. Risks surrounding drug tradeinvolvement among street-involved youth. Am J Drug Alcohol Ab. 2008;34(6):810–20.81. Deering KN, Shoveller J, Tyndall MW, Montaner JS, Shannon K. The streetcost of drugs and drug use patterns: relationships with sex work income inan urban Canadian setting. Drug Alcohol Depend. 2011;118(2–3):430–6.82. Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids forchronic pain—United States, 2016. JAMA. 2016;315(15):1624–45.83. Ochnio JJ, Patrick D, Ho M, Talling DN, Dobson SR. Past infection withhepatitis a virus among Vancouver street youth, injection drug users andmen who have sex with men: implications for vaccination programs. CanMed Assoc J. 2001;165(3):293–7.84. Miller CL, Strathdee SA, Kerr T, Li K, Wood E. Factors associated with earlyadolescent initiation into injection drug use: implications for interventionprograms. J Adolesc Health. 2006;38(4):462–4.85. Cheng T, Wood E, Feng C, Mathias S, Montaner J, Kerr T, et al. Transitionsinto and out of homelessness among street-involved youth in a Canadiansetting. Health Place. 2013;23:122–7.86. Werb D, Kerr T, Buxton J, Shoveller J, Richardson C, Montaner J, et al. Patternsof injection drug use cessation during an expansion of syringe exchangeservices in a Canadian setting. Drug Alcohol Depend. 2013;132(3):535–40.87. DeBeck K, Kerr T, Li K, Milloy MJ, Montaner J, Wood E. Incarcerationand drug use patterns among a cohort of injection drug users.Addiction. 2008;104(1):69–76.88. Marshall BDL, Wood E, Shoveller JA, Patterson TL, Montaner JSG, Kerr T.Pathways to HIV risk and vulnerability among lesbian, gay, bisexual, andtransgendered methamphetamine users: a multi-cohort gender-basedanalysis. BMC Public Health. 2011;11(1):20.89. Ti L, Socías ME, Wood E, Milloy MJ, Nosova E, DeBeck K, et al. Theimpact of methadone maintenance therapy on access to regularphysician care regarding hepatitis C among people who inject drugs.PLoS One. 2018;13(3):e0194162.90. Nolan S, Dias Lima V, Fairbairn N, Kerr T, Montaner J, Grebely J, et al. Theimpact of methadone maintenance therapy on hepatitis C incidenceamong illicit drug users. Addiction. 2014;109(12):2053–9.91. Hunt DE, Kling R, Almozlino Y, Jalbert S, Chapman MT, Rhodes W.Telling the truth about drug use: how much does it matter? J DrugIssues. 2015;45(3):314–29.92. Harris B, Shaw B, Lawson H, Sherman B. Barriers to addressing adolescentsubstance use: perceptions of New York school-based health centerproviders. J Sch Health. 2016;86(2):96–104.93. Darke S. Self-report among injecting drug users: a review. Drug AlcoholDepend. 1998;51(3):253–63.94. Jozaghi E, Buxton JA, Thomson E, Marsh S, Gregg D, Bouchard M. Buildingnew approaches to risk reduction with social networks and people whosmoke illegal drugs from participatory community-based research. Int JQual Methods. 2018;17(1):1609406918771247.95. Damon W, Callon C, Wiebe L, Small W, Kerr T, McNeil R. Community-basedparticipatory research in a heavily researched inner city neighbourhood:perspectives of people who use drugs on their experiences as peerresearchers. Soc Sci Med. 2017;176:85–92.96. Jozaghi E, Dadakhah-Chimeh Z. A call for action to combat the growingsynthetic opioid epidemic: the need for the creation and expansion ofaddiction and psychiatric nursing programs. J Sch Nurs. 2017;34(1):11–3.97. Werb D, Buxton J, Shoveller J, Richardson C, Rowell G, Wood E.Interventions to prevent the initiation of injection drug use: a systematicreview. Drug Alcohol Depend. 2013;133(2):669–76.Cheng et al. Substance Abuse Treatment, Prevention, and Policy           (2018) 13:41 Page 11 of 11


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items