Use of On-site Detoxification Services co-located with a Supervised Injection Facility

Abstract

Objectives

Supervised injection facilities (SIFs) are increasingly being implemented worldwide in response to the harms associated with injection drug use. Although SIFs have been shown to promote engagement of people who use injection drugs (PWID) with external health services, little is known about the potential of co-locating on-site detoxification services with SIFs. The aim of this study was to characterize use of detoxification services co-located at Insite, North America’s first SIF, among PWID in Vancouver, Canada.

Methods

Data were derived from two prospective cohorts of PWID in Vancouver, Canada between November 2010 and December 2012. Using multivariable generalized estimating equation logistic regression, we identified factors independently associated with reporting use of detoxification services at the SIF.

Results

Among 1316 PWID, 147 (11.2%) reported enrolling in detoxification services co-located with the SIF at least once during the two year study period. In multivariable analyses, after adjustment for other potential cofounders, factors independently and positively associated with use of this service included residence <5 blocks from the SIF (Adjusted Odds Ratio [AOR] = 1.70), enrollment in methadone maintenance therapy (AOR = 1.90), public injection (AOR = 1.53), binge injection (AOR = 1.93), recent overdose (AOR = 1.90) and frequent SIF use (AOR = 8.15) (all p < 0.05).

Discussion

Use of on-site detoxification services offered at the SIF was common among PWID and associated with frequent SIF use and various markers of vulnerability and drug-related risk. These findings highlight the potential role of SIFs as a point of access to detoxification services for high-risk PWID. Future studies should examine if co-location leads to higher uptake of addiction services in comparison to services that create geographic or other obstacles.

1.1 INTRODUCTION

People who inject drugs (PWID) contend with an array of health-related harms, including overdose (Mitra, Wood, Nguyen, Kerr, & DeBeck, 2015), HIV/AIDS (Montain et al., 2016), Hepatitis C (HCV) (Kim et al., 2009) and other infectious diseases (Lloyd-Smith et al., 2010). This population also experiences significant barriers in accessing addiction treatment and other health care and supportive services (McCoy, Metsch, Chitwood, & Miles, 2001). In response to these challenges, supervised injection facilities (SIFs) have increasingly been implemented in cities worldwide. These facilities provide PWID with sterile injecting equipment and a safe and hygienic space in which to inject pre-obtained illicit drugs under the supervision of nurses or other trained staff. The goals of SIFs are to engage high-risk PWID, reduce injection-related harms and infections, facilitate access to health and social services, reduce morbidity and mortality associated with overdose, and improve public order (Wood, Kerr, Montaner, et al., 2004; Wood et al., 2005). There are currently over 90 SIFs operating in over 60 cities across the world (EMCDDA, 2016).

Insite, North America’s first legally sanctioned SIF, was opened in 2003 in Vancouver, Canada. To date, numerous peer-reviewed studies have demonstrated the various health and community benefits of this facility. For example, overdose mortality decreased by 35% in the neighbourhood surrounding Insite in the two years after the facility opened (Marshall, Milloy, Wood, Montaner, & Kerr, 2011). In addition, use of the SIF has been associated with reductions in syringe sharing and injection-related injuries (Kerr, Small, Moore, & Wood, 2007; Marshall et al., 2011; Milloy, Kerr, Mathias, et al., 2008; Milloy, Kerr, Tyndall, Montaner, & Wood, 2008; Wood, Tyndall, Montaner, & Kerr, 2006) without increasing either the number of local PWID (Kerr, Tyndall, et al., 2007) or rates of relapse (Kerr et al., 2006). At the community level, the establishment of the facility has contributed to improvements in public order through reductions in public injection and publicly-discarded injection-related litter (Wood, Kerr, Small, et al., 2004), and has not been associated with increases in drug-related crime (Wood, Tyndall, Lai, Montaner, & Kerr, 2006).

SIFs such as Insite have also been shown to serve as important entry points to external drug treatment and other health and social services for PWID (Kimber, Dolan, van Beek, Hedrich, & Zurhold, 2003; Wood, Tyndall, Qui, et al., 2006; Wood, Tyndall, Zhang, Montaner, & Kerr, 2007; Wood, Tyndall, Zhang, et al., 2006; Wood, Zettel, & Stewart, 2003). Indeed, an earlier study of PWID in Vancouver found that frequent Insite use and contact with addictions counselors within the facility were independently associated with more rapid entry into external residential detoxification services (Wood, Tyndall, Zhang, et al., 2006). Subsequent analyses demonstrated a 30% increase in the uptake of external detoxification services in the year after Insite opened compared to the year prior to the establishment of the facility (Wood et al., 2007). This study also found that such entry into detoxification services was associated with an increased likelihood of enrolment in other addiction treatment programs, such as methadone maintenance therapy (MMT) and residential treatment, as well as subsequent declines in use of the SIF (Wood et al., 2007). Further, regular use of the facility has also been directly associated with increased uptake of addiction treatment and, in turn, an increased likelihood of injection drug use cessation (DeBeck et al., 2011). Similarly, a study of a SIF in Sydney, Australia found that frequent users of the facility were more likely to receive referrals to addiction treatment (Kimber et al., 2008).

Despite this evidence demonstrating the role of SIFs in connecting high-risk PWID with external detoxification and addiction treatment services, little is known about the uptake of on-site detoxification services co-located with SIFs. An on-site detoxification facility, known as “Onsite”, has been co-located with Insite since 2007, offering 12 beds of detoxification services. There are no restrictions on the types of substances used by clients prior to enrollment in this program. Average length of stay ranges from one to two weeks, and residents can also access 18 beds of transitional housing post-detoxification (PHS, 2017). Since opening, this detoxification service has seen over 2,800 intakes (PHS, 2017). However, use of this service among PWID has not yet been characterized. Identifying relevant factors affecting uptake of this service could provide important information to guide the development of SIFs in other regions, particularly given that a number of municipalities across Canada and elsewhere are presently conducting planning and feasibility work to establish SIFs, many of which are considering integrating these facilities within existing health services for PWID (Jozaghi, Reid, & Andresen, 2013). The present study was therefore undertaken to characterize use of an on-site detoxification facility at a SIF among a community-recruited prospective cohort of PWID in Vancouver, Canada.

2.1 MATERIAL AND METHODS

Data for this study were derived from two ongoing prospective cohort studies of people who use illicit drugs: The Vancouver Injection Drug Users Study (VIDUS) and the AIDS Care Cohort to evaluate Exposure to Survival Services (ACCESS). These cohorts have been described in detail previously (Strathdee et al., 1998; Tyndall et al., 2003). Briefly, since May 1996, participants have been recruited through self-referral, word-of-mouth and street outreach. Eligibility criteria included residing in the Greater Vancouver region and being at least 18 years of age at the time of recruitment, with the primary distinction being that HIV-positive individuals who use illicit drugs other than cannabis in the month prior to enrollment are followed in ACCESS whereas HIV-negative individuals who injected drugs in the month prior to enrollment are followed in VIDUS. The recruitment and follow-up procedures for the two cohorts are largely identical, allowing for combined analyses. At baseline and semi-annually thereafter, participants complete a harmonized interviewer-administered questionnaire and provide blood samples for serological testing. All participants provide written informed consent and are provided with a $30 CAD stipend at each study visit.

All participants who completed a study visit between November 1, 2010 and December 31, 2012, and who reported injecting drugs in the previous six months at baseline were included in the present analyses. The primary outcome of interest was reporting enrollment in on-site detoxification services at the SIF in the previous six months. Based on existing literature, we selected explanatory variables that we hypothesized might be associated with access to addiction treatment services, including detoxification services (DeBeck et al., 2011; Milloy et al., 2010; Rapp et al., 2006). These included sociodemographic and behavioural variables, including: age (per year older); gender (male vs. female); ancestry (White vs. Non-White); unstable housing (yes vs. no); sex work involvement (yes vs. no); and residence within five blocks of the SIF (<5 blocks vs. ≥5 blocks). Drug use variables considered included: ≥ daily injection cocaine use (yes vs. no); ≥ daily injection heroin use (yes vs. no); ≥ daily injection crystal methamphetamine use (yes vs. no); ≥ daily injection prescription opioid use (yes vs. no); ≥ daily crack smoking (yes vs. no); binge injection drug use (yes vs. no); non-fatal overdose (yes vs. no); participation in MMT (yes vs. no); public injecting (yes vs. no); syringe sharing (yes vs. no); requiring help injecting (yes vs. no); difficulty accessing needles (yes vs. no) and frequent SIF use (≥weekly vs. <weekly). Other variables assessed included being HCV antibody positive (yes vs. no) and HIV seropositive status (yes vs. no). All variables were treated as time-updated and refer to the six-month period prior to the interview unless otherwise indicated.

2.2 Statistical Analyses

We first examined the baseline sample characteristics stratified by reporting accessing on-site detoxification services, using the Pearson’s Chi-squared test (for binary variables) and Mann-Whitney test (for continuous variables). Fisher’s exact test was used for binary variables when one or more of the cells contained expected values less than or equal to five. Since analyses of factors potentially associated with accessing on-site detoxification included serial measures for each participant, we used generalized estimating equations (GEE) with logit link. This method provides standard errors adjusted by multiple observations per person using an exchangeable correlation structure and therefore considers data from every participant follow-up visit. As a first step, we used bivariable GEE analyses to determine factors associated with use of on-site detoxification services at the SIF. Next, we constructed an explanatory multivariable model using an a priori-defined backward model selection procedure based on examination of quasilikelihood under the independence model criterion statistic (QIC) and Type-III p-values. In brief, we first fit a full model that included all explanatory variables that were significantly associated with the outcome at the level of p< 0.20 in bivariable analyses. After examining the QIC of the model, the variable with the largest p-value was removed sequentially. We continued this iterative process to build a final multivariable model that included the set of explanatory variables associated with the lowest QIC (Pan, 2001).

We also recognized that the factors associated with use of detoxification services at the SIF use might differ for SIF users specifically compared to the larger sample of PWID. Therefore, as a subanalysis, we restricted the sample to participants who reported use of the SIF in the previous six months at each interview and examined bivariable and multivariable associations for this subsample using the same approach outlined above. All p-values were two-sided. All statistical analyses were performed using RStudio, version 0.99.892.

3.1 RESULTS

In total, 1316 participants were included in the present study. Of these, 883 (67.1%) were male, 777 (59.0%) self-reported Caucasian ancestry and the median age at baseline was 46.2 years (interquartile range [IQR] = 40.2 – 52.1). Overall, the 1316 individuals contributed 4053 observations to the analysis with a median follow-up time of 17.2 (Interquartile range [IQR]: 11.7 – 18.4) months. At baseline, 75 participants (5.7%) reported use of the on-site detoxification service in the previous six months. In total, 147 (11.2%) reported enrolling in on-site detoxification services at the SIF at least once over the two-year study period. The baseline characteristics of all participants, stratified by reported enrollment in on-site detoxification services, are presented in Table 1.

Table 1.

Factors associated with use of on-site detoxification services at a supervised injection facility (SIF) at baseline among 1316 people who inject drugs in Vancouver, Canada (2010–2012).

Characteristic	Yes n (%)n = 75	No n (%)n = 1241	Odds Ratio (95% CI)	p - value
Age
Median (IQR)	43 (39 – 48)	47 (40 – 52)		0.006
Gender
Male	50 (66.7)	833 (67.1)	0.98 (0.60 – 1.61)	0.935
Female	25 (33.3)	408 (32.9)
Ancestry
White	48 (64.0)	729 (58.7)	1.25 (0.77 – 2.03)	0.369
Non-White	27 (36.0)	512 (41.3)
HIV serostatus†
Positive	23 (30.7)	539 (43.4)	0.58 (0.35 – 0.95)	0.030
Negative	52 (69.3)	702 (56.6)
HCV serostatus†
Positive	68 (90.7)	1116 (89.9)	1.09 (0.49 – 2.42)	0.836
Negative	7 (9.3)	125 (10.1)
Unstable housing†
Yes	60 (80.0)	744 (60.0)	2.63 (1.48 – 4.69)	0.001
No	15 (20.0)	490 (40.0)
Home residence†
<5 blocks from SIF	35 (46.7)	439 (35.4)	1.58 (0.99 – 2.53)	0.053
≥5 blocks from SIF	40 (53.3)	794 (64.0)
Sex work involvement†
Yes	10 (13.3)	92 (7.4)	1.95 (0.97 – 3.92)	0.057
No	64 (85.3)	1147 (92.4)
Methadone maintenance therapy†
Yes	43 (57.3)	664 (53.5)	1.20 (0.75 – 1.94)	0.445
No	31 (41.3)	576 (46.4)
Daily heroin injection†
Yes	30 (40.0)	135 (10.9)	5.46 (3.33 – 8.96)	<0.001
No	45 (60.0)	1105 (89.0)
Daily methamphetamine injection†
Yes	1 (1.3)	38 (3.1)	0.43 (0.06 – 3.16)	0.391
No	74 (98.7)	1202 (96.9)
Daily cocaine injection†
Yes	14 (18.7)	60 (4.8)	4.51 (2.39 – 8.51)	<0.001
No	61 (81.3)	1178 (94.9)
Public injection†
Yes	36 (48.0)	178 (14.3)	5.51 (3.41 – 8.90)	<0.001
No	39 (52.0)	1062 (85.6)
Binge injection†
Yes	33 (44.0)	222 (17.9)	3.60 (2.23 – 5.80)	<0.001
No	42 (56.0)	1016 (81.9)
Non-fatal overdose†
Yes	9 (12.0)	40 (3.2)	4.09 (1.91 – 8.79)	<0.001
No	66 (88.0)	1201 (96.8)
Syringe sharing†
Yes	4 (5.4)	31 (2.5)	2.23 (0.77 – 6.49)	0.132
No	70 (94.6)	1209 (97.4)
Require help injecting†
Yes	15 (20.0)	112 (9.0)	2.52 (1.38 – 4.58)	0.002
No	60 (80.0)	1128 (90.9)
Difficulty accessing needles†
Yes	5 (6.7)	47 (3.8)	1.83 (0.71 – 4.76)	0.206
No	69 (93.2)	1189 (95.8)
Frequent SIF use†
Yes	45 (60.8)	154 (12.4)	10.88 (6.63 – 17.88)	<0.001
No	29 (39.2)	1080 (87.5)

^†Denotes activities in the previous six months

IQR= Interquartile range; SIF= supervised injection facility; HIV= human immunodeficiency virus; HCV= Hepatitis C.

The results of the bivariable and multivariable GEE analyses of factors associated with enrollment in on-site detoxification services at the SIF are presented in Table 2. As shown, in the final multivariable model, factors that remained independently associated with use of on-site detoxification services included: residence <5 blocks from the SIF (Adjusted Odds Ratio [AOR] = 1.70, 95% Confidence Interval [CI] = 1.19 – 2.45), participation in MMT (AOR = 1.90, 95% CI = 1.34 – 2.68), public injection (AOR = 1.53, 95% CI = 1.03 – 2.26), binge injection (AOR = 1.93, 95% CI = 1.31 – 2.85) recent overdose (AOR = 1.90, 95 % CI = 1.07 – 3.41) and frequent SIF use (AOR = 8.15, 95 % CI = 5.38 – 12.34). Older age (AOR = 0.98, 95% CI = 0.96 – 1.00) was independently and negatively associated with the outcome (all p <0.05).

Table 2.

Bivariable and multivariable generalized estimating equations (GEE) analyses of factors associated with use of on-site detoxification services at a supervised injection facility (SIF) among 1316 people who inject drugs in Vancouver, Canada (2010–2012).

Characteristic	Unadjusted		Adjusted
	Odds Ratio (95% CI)	p - value	Odds Ratio (95% CI)	p - value
Age
(per year older)	0.97 (0.95 – 0.98)	<0.001	0.98 (0.96 – 1.00)	0.025
Gender
(men vs. women)	0.98 (0.68 – 1.42)	0.928
Ancestry
(White vs. non-White)	1.30 (0.90 – 1.87)	0.166	1.28 (0.89 – 1.86)	0.189
HIV positive
(yes vs. no)	0.66 (0.46 – 0.95)	0.024	0.76 (0.52 – 1.11)	0.150
HCV positive
(yes vs. no)	1.39 (0.71 – 2.73)	0.331
Unstable housing†
(yes vs. no)	2.28 (1.56 – 3.34)	<0.001	1.43 (0.92 – 2.22)	0.116
Home residence†
(<5 blocks vs. ≥5 blocks from SIF)	1.64 (1.19 – 2.27)	0.003	1.70 (1.19 – 2.45)	0.004
Sex work involvement†
(yes vs. no)	1.99 (1.23 – 3.22)	0.005
Methadone maintenance therapy†
(yes vs. no)	1.65 (1.17 – 2.31)	0.004	1.90 (1.34 – 2.68)	<0.001
Daily heroin injection†
(yes vs. no)	2.85 (1.96 – 4.13)	<0.001
Daily methamphetamine injection†
(yes vs. no)	0.83 (0.31 – 2.19)	0.705
Daily cocaine injection†
(yes vs. no)	3.15 (1.84 – 5.38)	<0.001
Public injection†
(yes vs. no)	4.14 (2.89 – 5.91)	<0.001	1.53 (1.03 – 2.26)	0.036
Binge injection†
(yes vs. no)	3.51 (2.47 – 4.99)	<0.001	1.93 (1.31 – 2.85)	0.001
Non-fatal overdose†
(yes vs. no)	2.97 (1.81 – 4.88)	<0.001	1.90 (1.07 – 3.41)	0.030
Syringe sharing†
(yes vs. no)	2.30 (0.94 – 5.59)	0.067
Require help injecting†
(yes vs. no)	2.08 (1.35 – 3.21)	0.001
Difficulty accessing needles†
(yes vs. no)	1.26 (0.62 – 2.58)	0.523
Frequent SIF use†
(≥weekly vs. < weekly)	12.24 (8.38 – 17.87)		8.15 (5.38 – 12.34)	<0.001

^†Denotes activities in the previous six months

SIF= supervised injection facility; HIV= human immunodeficiency virus; HCV= Hepatitis C.

Among the subsample of 554 participants who reported recently using the SIF, a total of 131 (23.7%) reported enrolling in on-site detoxification services at the SIF over the two-year study period. In multivariable analyses, residence <5 blocks from the SIF (AOR = 1.53, 95% CI = 1.06 – 2.20), binge drug use (AOR = 1.51, 95% CI = 1.03 – 2.21), participation in MMT (AOR = 1.59, 95% CI = 1.10 – 2.31), and frequent SIF use (AOR = 2.86, 95% CI = 1.92 – 4.28) were independently and positively associated with use of on-site detoxification services among this group.

4.1 DISCUSSION

In this community-recruited prospective cohort study of over 1300 PWID in Vancouver, Canada, we observed that more than 10% of the sample reported accessing on-site detoxification services at a SIF over a median of 17 months of follow up. However, when we restricted our sample to PWID who reported recent use of the SIF, 23.7% reported accessing on-site detoxification services during follow-up. In multivariable analyses, residence in the neighbourhood immediately surrounding the SIF, public injection drug use, binge injection drug use, recent non-fatal overdose, participation in MMT, and frequent SIF use were positively associated with enrollment in on-site detoxification services at the SIF among PWID. In the subanalysis restricted to recent SIF users, residence in the neighbourhood of the SIF, binge injection drug use, participation in MMT, and frequent SIF use were independently and positively associated with use of detoxification services co-located with the SIF.

Our results show that enrollment in detoxification services co-located with a SIF was associated with a number of common markers of vulnerability and drug-related risk, including public injection, binge injection and recent overdose. These findings are largely consistent with previous studies demonstrating that the low-threshold programming offered at SIFs effectively attracts higher-risk PWID (Toth, Tegner, Lauridsen, & Kappel, 2016; Wood, Tyndall, Qui, et al., 2006) and helps to connect these individuals to critical healthcare and addiction treatment services (Kimber et al., 2008; Lloyd-Smith et al., 2012; Wood et al., 2007). These findings are also consistent with existing literature demonstrating that high-risk drug use practices and use of other harm reduction programs, such as needle exchange programs, are associated with increased use of detoxification services among PWID (Strathdee et al., 1999). However, this study is the first, to our knowledge, to demonstrate that detoxification programs co-located with SIFs are also effective in engaging PWID that are highly marginalized and at elevated risk of health-related harms.

Given that PWID with the characteristics and exposures identified herein often face considerable barriers to conventional addiction treatment utilization and referral uptake (Kimber et al., 2008; Longshore, 1999), these findings are encouraging in demonstrating that SIFs may serve as a direct point of access to detoxification services for higher-risk PWID. These results have important policy and programmatic implications, providing evidence for increasing the availability of detoxification services within other SIFs in effort to reach populations that share these markers of vulnerability and risk.

We should note that certain variables of interest, such as public injection and overdose, were significantly associated with on-site detoxification use among the full sample of PWID but not once restricted to the sub-sample of recent SIF users. We suspect that these discrepancies in findings are likely due to the fact that SIF users are more likely to be structurally disadvantaged and to experience drug-related risks and harms (Wood et al., 2005; Wood, Tyndall, Qui, et al., 2006). These differences in characteristics between those in the cohort who do and do not use the SIF further suggests the need to co-locate treatment programming with SIFs that is responsive to the unique needs of the vulnerable populations who use these facilities.

We also found that frequent SIF use was associated with increased uptake of on-site detoxification services, in both the full sample of PWID and the sub-sample who had recently used the SIF. This association may be due to frequent SIF users having more opportunities to engage with SIF healthcare staff, including addiction counselors, to discuss health needs in a non-judgmental and supportive environment that minimizes spatial barriers to treatment (McNeil & Small, 2014). Our finding of an association between frequent SIF use and use of on-site detoxification provides additional evidence of the importance of co-locating detoxification services within existing and emerging SIFs. These results also build on previous work demonstrating greater uptake of external addiction treatment among frequent SIF users (DeBeck et al., 2011; Wood, Tyndall, Zhang, et al., 2006) and further highlight the role of SIFs in facilitating entry into treatment services, despite the fact that these facilities are harm reduction programs designed for active drug users.

Our results also show that access to on-site detoxification services was associated with increased likelihood of participation in MMT. Although our analysis does not allow us to establish the temporal direction of this association, previous local studies have shown that that PWID who have engaged with external detoxification services are more likely to subsequently engage with sustained treatment options such as MMT (Wood et al., 2007) and, in turn, to cease injecting drugs (DeBeck et al., 2011). Thus, it appears that on-site detoxification services at SIFs may offer similar benefits to external detoxification services in acting as a referral mechanism to evidence-based addiction treatment. Although we suspect that this is the most plausible explanation for this finding, it is also possible that individuals in MMT could demonstrate greater concern for their health, and might therefore be more likely to enroll in on-site detoxification services and MMT during the same time period. Thus, future research is needed to better understand potential explanations for the observed association between on-site detoxification use and MMT participation.

This study has several other limitations that warrant discussion. As noted above, the analyses undertaken for the present study limited our ability to determine the temporal nature of the observed associations. Further, the individuals who comprise the cohorts used in this study are not recruited at random, and we therefore urge caution in using our results to make generalized statements about people who inject drugs. Our data was also largely questionnaire-based, and therefore relied upon self-report. Thus, study participants may have underreported less socially desirable behaviours such as illicit drug use and other stigmatized behaviors. Despite these concerns, self-reported information among people who used drugs has been shown to be largely accurate and reliable (Darke, 1998).

4.2 Conclusions

In summary, we found that a substantial number of PWID enrolled in detoxification services at a local SIF. Frequent SIF use and a number of known markers of vulnerability and high-risk drug use were associated with uptake of on-site detoxification services co-located with the SIF. As PWID that utilize these services represent a marginalized and exceedingly difficult to reach population (Wood, Tyndall, Qui, et al., 2006), these findings are encouraging considering that this and previous studies have found enrollment in detoxification services to be associated with further engagement in other forms of treatment (Wood et al., 2007) and cessation of injecting (DeBeck et al., 2011; Wood et al., 2007). Given that a number of settings in Canada and internationally are currently considering establishing SIFs, this information is useful in highlighting the potential benefits of providing detoxification services co-located with these facilities.

HIGHLIGHTS.

• Supervised Injection Facilities (SIFs) engage high-risk people who use injection drugs (PWID)

• Little is known about co-locating on-site detoxification services with SIFs

• Characteristics from a cohort of PWID using on-site detoxification at a SIF are reported.

• Use of detoxification services is associated with markers of vulnerability and drug-related risk

• Findings highlight the role of SIFs in access to detoxification services for high-risk PWID