The use of public spaces for injecting drugs presents a range of public health and public order challenges for cities coping with open drug use.1 A small number of cross-sectional studies suggest that individuals who inject drugs in public settings may employ riskier injecting practices than those injecting in private venues, thereby increasing their chances of contracting blood-borne and other injection-related infections.2^–4 Further, in areas where pubic injecting occurs, local businesses and community members are often disturbed by the sight of people openly injecting drugs,5 6 as well as the presence of drug-related debris, including discarded needles, syringes and other litter.7

The area of Vancouver, Canada, known as the Downtown Eastside has historically been the local centre of open drug market activity and public injecting.8 9 Efforts taken on the part of policy-makers to address the health and social harms associated with public injecting in this area have had varying impacts. Enforcement-based interventions targeting public drug use have had transient success in reducing the visible signs of street-based injecting in the short term.10 However, these operations have been found to displace IDUs to neighbouring areas, impede access to health and harm reduction programmes among street-based injectors, and thereby have potential to negatively impact the health of people who inject drugs in public.11^–13 The identification of unintended negative consequences stemming from intensified policing activities sparked public debate regarding the merit of addressing public drug use with enforcement-based approaches.10 14 15 These concerns have been echoed in other cities where enforcement activities have been employed to manage high rates of public drug use.1 16^–18

An alternative approach to addressing public drug use that also precipitates controversy is supervised injection facilities (SIFs).19 SIFs offer injecting drug users (IDUs) a place to inject pre-obtained drugs in a hygienic environment under the supervision of nurses without fear of arrest.20 A number of cities across Europe as well as Sydney, Australia, and most recently Vancouver, Canada, currently operate SIFs.21 22 Vancouver’s SIF, called “InSite”, was opened in September 2003 and scientific evaluations have found the site to have beneficial public health and public order impacts, including reduced syringe discards and reduced public drug use.23 24 However, InSite is currently a pilot project that has been operating at capacity for over 2.5 years at the time of writing,25 and coverage estimates indicate that the existing facility can accommodate at most 5–10% of all injections that occur in the Downtown Eastside (Vancouver’s drug use epicentre).26 Despite emerging scientific evidence of the effectiveness of supervised injection facilities in addressing both public health and public order concerns related to open drug scenes,24 the current Canadian Government does not presently support the expansion of supervised injection facilities beyond the single tightly restricted pilot facility.27

Despite these efforts, public injecting behaviour continues to be an ongoing concern in Vancouver’s Downtown Eastside. Most recently the Vancouver Police Department identified the reduction of public psychoactive substance use as a strategic priority that they plan to address through increased targeted law enforcement.28 Additionally, beginning in the year 2000 representatives from the City of Vancouver, the Province of British Columbia and the Government of Canada initiated an urban development strategy for the Downtown Eastside, called the “Vancouver Agreement”, which presently involves addressing the open drug scene through enforcement and housing initiatives.29

Given the health-related harms associated with public drug consumption and the political debates surrounding potential interventions,27 scientific evidence regarding public injecting behaviour is required to develop appropriate policy responses. Although public injecting behaviour has been a concern locally for some time, the profile of public injectors in Vancouver has yet to be fully described. This study sought to examine the proportion of IDUs who frequently inject in public, and identify factors related to recent frequent public injecting among a cohort of IDUs in Vancouver during the period of 1 December 2003 to 30 November 2005. It is hoped that our analysis will serve to better inform public policy-makers on how to best deliver targeted interventions that protect the health of those who inject drugs and minimise the prevalence of public injecting.

METHODS

Beginning in May 1996, IDUs were recruited through self-referral and street outreach into the Vancouver Injection Drug User Study (VIDUS), a prospective cohort study described in detail previously.30 31 Briefly, people were eligible if they had injected drugs at least once in the previous month, resided in the greater Vancouver region and provided written informed consent. At baseline and semi-annually, subjects provide blood samples and complete a questionnaire administered by any one of the trained study interviewers. The questionnaire elicits demographic data as well as information about drug use, HIV risk behaviour and addiction treatment. All participants are given a stipend ($C20 (£16)) at each study visit. The study has received the appropriate ethical approval from St. Paul’s Hospital and the University of British Columbia’s Research Ethics Board.

The present analysis was restricted to active IDUs, defined as participants who reported injecting drugs in the 6 months prior to their study visit, and those IDUs seen for follow-up during the period of 1 December 2003 to 30 November 2005. All follow-up visits during the study period that included a report of active injecting drug use in the previous 6 months were included in the analysis; similarly, all follow-up visits that did not include a report of injecting drugs in the previous 6 months were excluded from the analysis. The primary variable of interest in this analysis was self-reported recent frequent public injecting defined as “usually” or “always” injecting in public over the 6 months prior to the study visit. Public spaces were defined as public lavatories, streets, alleys, parks, abandoned buildings and other public settings. Explanatory variables of interest included sociodemographic information: gender (female vs male), age (per year older), Aboriginal ethnicity (yes vs no) and homelessness, defined as having no fixed address for the last 6 months, (yes vs no). Drug use variables considered refer to behaviours in the past 6 months and included frequent heroin injection (⩾ daily vs <daily), frequent cocaine injection (⩾daily vs <daily), frequent crack cocaine smoking (⩾ daily vs < daily), non-fatal overdose, self-identified by participants (yes vs no), requiring help injecting (yes vs no), binge drug use, defined as a period of using drugs more often than usual (yes vs no), borrowing and lending syringes already used to inject drugs (yes vs no), regularly cooking and filtering drugs prior to injection, defined as heating and filtering drug to sterilise and remove impurities (always, usually vs sometimes, never), and participation in any addiction treatment programme (yes vs no). Other risk characteristics considered included sex trade involvement, defined as selling sex for money or in kind exchange (yes vs no), participation in drug dealing (yes vs no), incarceration in the past 6 months (yes vs no) and residing in the Downtown Eastside, that is, Vancouver’s illicit drug use and HIV epicentre (yes vs no).

Since the analyses of factors potentially associated with recent frequent public injecting included serial measures for each subject, we used generalised estimating equations (GEEs) for binary outcomes with logit link for the analysis of correlated data to determine factors associated with recent public injecting throughout the 24-month follow-up period. These methods provided standard errors adjusted by multiple observations per person using an exchangeable correlation structure. Therefore, data from every participant follow-up visit were considered in this analysis. Missing data were addressed through the GEE estimating mechanism, which uses the all available pairs method to encompass the missing data from dropouts or intermittent missing data. All non-missing pairs of data are used in the estimators of the working correlation parameters. As a first step, we used GEE univariate analysis to determine factors associated with recent frequent public injecting. In order to adjust for potential confounding, all variables that were p<0.05 in GEE univariate analyses were entered in a multivariate logistic GEE model. All statistical analyses were performed using SAS software version 8.0 (SAS, Cary, North Carolina). All p values are two sided.

RESULTS

A total of 620 active injecting participants were eligible for this analysis, including 251 (40.5%) women and 203 (32.7%) persons of Aboriginal decent. The median age of all participants was 31.9 years (interquartile range (IQR) 25–39). This sample contributed to 1530 observations, the median number of follow-up visits was three (IQR 2–4) and 552 (89.0%) participants completed at lease two study visits. Among our sample of 620 active IDUs a total of 142 (22.9%) reported recent frequent public injecting at some point during follow-up. The characteristics of the study sample stratified by recent frequent public injecting are presented in table 1. Characteristics for recent frequent public injectors were measured at their first visit (during the study period), which involved a report of frequent public injecting. Characteristics for all other participants were measured from the first study visit during the study period.

The univariate GEE analyses of behavioural and sociodemographic variables are presented in table 2. Factors found to be positively associated with recent frequent public injecting in univariate analyses included homelessness (OR 8.81, 95% CI 6.13 to 12.65); frequent crack cocaine smoking (OR 2.06, 95% CI 1.54 to 2.76); frequent heroin injection (OR 2.05, 95% CI 1.53 to 2.75); recent incarceration (OR 1.92, 95% CI 1.30 to 2.83); and requiring help injecting (OR 1.40, 95% CI 1.01 to 1.93). Factors negatively associated with recent frequent public injecting included older age (OR 0.92, 95% CI 0.89 to 0.94); participation in addiction treatment (OR 0.57, 95% CI 0.42 to 0.76); and cooking and filtering drugs prior to injecting (OR 0.44, 95% CI 0.32 to 0.60).

In the multivariate GEE analysis, also shown in table 2, factors that remained independently associated with recent frequent public injecting in our logistic model included older age (AOR 0.95, 95% CI 0.93 to 0.98); homelessness (AOR 6.70, 95% CI 4.55 to 9.87); frequent crack cocaine smoking (AOR 1.48, 95% CI 1.04 to 2.11); frequent heroin injection (AOR 1.56, 95% CI 1.10 to 2.20); and cooking and filtering drugs prior to injecting (AOR 0.50, 95% CI 0.35 to 0.71).

DISCUSSION

In the present study, at some point during the 2-year follow-up period, 23% of participants reported having regularly injected in public in the 6 months prior to their study visit. A number of characteristics were found to be independently and positively associated with recent frequent public injecting in longitudinal analyses, including homelessness, frequent crack use and frequent heroin injection. Older age was found to be negatively associated with recent frequent public injecting, and those who regularly inject in public were less likely to cook and filter their drugs prior to injecting.

Consistent with previous studies, our findings indicate that public injectors are a vulnerable population characterised by high-intensity addiction and risky behaviour.30 Specifically, recent frequent public injectors in our sample were more likely to be younger in age (under 30), a factor found to be associated with syringe sharing in this setting.24 This underscores the need for interventions targeting high-risk drug use among younger injectors. Recent frequent public injectors were also disproportionately homeless; a status among local injectors associated with an elevated likelihood of HIV infection.32 In addition, recent frequent public injectors in our sample were less likely to cook and filter when preparing their drugs, thereby increasing their chances of developing infections and injection-related complications.33^–35

Our previous qualitative research involving drug users in Vancouver has identified contextual factors that may explain the associations between public injecting and the failure to cook and filter drugs prior to injection. Specifically, fear of arrest and street violence has been found to encourage rushed injecting and impede attempts to follow safer injection practices.12 Given the presence of urine and faeces in alleyways, as well as the absence of public washing facilities in Vancouver’s Downtown Eastside where injectors can wash their hands and follow recommended hygiene routines,36 it is concerning that frequent pubic injectors in this study were less likely to cook and filter drugs.

Given the identified health and social harms associated with public injecting, it is critical that policy interventions designed to address public drug consumption are appropriately targeted. The findings of the current analysis have important implications for policy-makers as they identify homelessness as the factor most strongly associated with recent frequent public injecting. Currently IDUs face a range of challenges when attempting to secure affordable housing in Vancouver,37 and a number of social and structural factors have been found to interplay with and perpetuate homelessness among IDUs. Identified barriers to becoming housed include a lack of supportive housing options, difficulties locating vacant units, obtaining required damage deposits, overcoming prejudice and biases of landlords and accessing social assistance.38 Although the City of Vancouver has developed a housing strategy for the Downtown Eastside,39 the strategy has been unable to provide secure housing for many high-risk IDUs. In fact, from 2002 to 2005 there was a reported doubling in the number of homeless adults in Vancouver (from 628 in 2002 to 1291 in 2005)40 and a local community organisation has documented a net loss of 415 housing units for low-income singles in the Downtown Eastside from 2003 to 2005.41 The current analysis indicates that reversing these trends and providing housing for IDUs is a necessary step to improve public health and public order in Vancouver’s Downtown Eastside. These data suggest that if housing initiatives are able to provide affordable and supportive housing in the Downtown Eastside for individuals with serious addictions, they may have considerable potential to reduce public drug consumption activities and related risk behaviour. Indeed, the importance of addressing contextual factors in the production of HIV risk for IDUs has been previously emphasised.42 43

Public injecting among homeless IDUs may be a result of homeless individuals not having access to alternative injecting environments. Supervised injection sites do offer an alternative injection setting for homeless populations. Evaluations of supervised injection facilities (SIFs) internationally find that these facilities are heavily utilised by homeless individuals22 and a large proportion of Vancouver’s SIF clientele report being homeless.25 However, given the limited capacity of the 12-seat pilot facility, the current Vancouver SIF is not able to accommodate all IDUs engaging in public injecting. This is problematic since SIFs represent one of the few opportunities to couple law enforcement and public health approaches to address public drug use, and prior studies have noted how police commonly refer IDUs to the SIF.24 This suggests that expanding access to the local SIF may provide an important opportunity to further reduce the prevalence of public injecting in Vancouver’s drug use epicentre (the Downtown Eastside).

Furthermore, a high proportion of crack cocaine smokers in Vancouver’s Downtown Eastside also report engaging in injection drug use,44 45 and it is noteworthy that in the current study frequent crack use remained independently associated with recent frequent public injecting. It may be that injectors who smoke crack are deterred from using the supervised injection facility, as smoking is not permitted in the facility. This finding supports previous works46 47 suggesting that operational regulations, such as the ban on smoking drugs in the facility, may pose barriers for a range of IDUs and hinder the potential impact of this intervention in reducing public drug use. However, despite the ability of supervised injection facilities to accommodate the needs of more IDUs by addressing operation issues identified in this and other studies, our findings indicate that the lack of supportive housing and subsequent prevalence of homelessness among IDUs remain the key factor perpetuating public drug use.

There are several limitations in this study to be noted. First, as with most other cohort studies of IDUs, VIDUS is not a random sample and therefore these findings may not generalise to other IDU populations. Second, this study relied on self-reported information concerning stigmatised behaviours, such as injecting drugs in public spaces and engaging in other forms of risk behaviour, and is hence susceptible to socially desirable reporting. In the present study this may have led to an under-reporting of public injecting and related risk behaviours,48 resulting in the prevalence of and risks associated with public injecting being underestimated.

In summary, the present analysis supports previous findings that people who inject drugs in public spaces are a vulnerable population facing significant health hazards. This analysis indicates that interventions to address and mediate the risks associated with public injecting are required and secure housing for IDUs may have the greatest potential to protect the health of IDUs and significantly decrease the prevalence of public injecting in Vancouver. In addition, findings support previous works suggesting that removing barriers to the use of Vancouver’s existing supervised injection facility may serve to further reduce public drug use.