Effect of witnessing an overdose on the use of drug checking services among people who use illicit drugs in Vancouver, Canada

Abstract

Background

Since 2013, fentanyl-contaminated drugs have been driving North America’s opioid-overdose epidemic. Drug checking, which enables people who use illicit drugs (PWUD) to test and receive feedback regarding the contents of their drugs, is being considered as a potential tool to address the toxic drug supply. While some PWUD witness overdoses, little is known about the impact of these experiences on subsequent risk reduction practices.

Objective

The purpose of this study was to examine the effect of witnessing an overdose on drug checking service use.

Methods

Data were derived from prospective cohorts of PWUD in Vancouver, Canada, a setting with a community-wide fentanyl overdose crisis, between June 1, 2018 and December 1, 2018. Multivariable logistic regression was used to estimate the effect of witnessing an overdose on drug checking service use.

Results

1,426 participants were eligible for the study, including 530 females; 767 (53.8%) participants reported witnessing an overdose and 196 (13.7%) reported using drug checking services in the last six months. In multivariable analyses, after adjusting for a range of confounders including use of fentanyl, witnessing an overdose was positively associated with drug checking service use (adjusted odds ratio = 2.32; 95% confidence interval: 1.57 – 3.49).

Conclusion

Our findings suggest that witnessing an overdose may motivate PWUD to use drug checking services. Given that only a small proportion of PWUD in the study reported using drug checking services, our findings highlight the need to continue to scale-up a range of overdose prevention interventions.

INTRODUCTION

Opioid overdose mortality continues to increase at an alarming rate in the United States and Canada; this upsurge is largely attributable to the rising prevalence of fentanyl-contaminated drugs (1–3). In the United States, the number of overdose deaths involving fentanyl or fentanyl analogues (e.g., carfentanil) nearly doubled from 2016 to 2017 (4,5). The US Centers for Disease Control and Prevention estimates that more than 70,000 drug overdose deaths occurred in the US in 2017 alone (5,6). In Canada, fentanyl or fentanyl analogue detection rates among opioid-overdose deaths increased 21% between January 2016 and September 2018 (2,7), and more than 10,300 opioid-overdose deaths occurred during this period (7). Despite growing investment in legislative and policy change, dissemination of guidelines on the safe prescribing of opioids, and the implementation and scale-up of a number of evidence-based treatments and interventions, the overdose crisis continues (2,3,6,8).

Innovative in this context, drug checking services are being considered as an intervention to mitigate drug-related risks during the fentanyl overdose crisis (1,9). Available in Europe for over two decades (predominately in nightclub and festival settings) (10–13), drug checking enables people who use drugs (PWUD) to test and receive feedback regarding the contents of their drug samples. Evidence in specific settings across Europe and more recently in North America indicates a high willingness to use drug checking services; and that use of drug checking services has a positive effect on behavioural intentions (9,10,14–18). In 2017, a study involving 334 PWUD in three areas in the United States (Baltimore, MD; Boston, MA; and Rhode Island) found that 85% of participants were interested in checking their drugs (16). Of those interested, 82% reported that they would change their drug-using behaviour (e.g., dose reduction, or cease of use) if their drugs tested positive for fentanyl (16). Yet gaps remain in the literature. For instance, there is a sparsity of literature on motivational factors behind the use of drug checking services, and actual uptake of drug checking services has been low in some settings (19). However, it may be that bearing witness to the negative consequences of ingesting adulterated drugs may motivate some to avail themselves of drug checking services.

The objective of this study was to determine the effect of witnessing a fatal or non-fatal overdose on use of drug checking services among PWUD in Vancouver, Canada, a setting with a community-wide fentanyl overdose crisis with a range of harm reduction services, including drug checking - which are available at various supervised injection and overdose prevention sites in the region (20–23). The drug checking technologies used in this setting (i.e., pairing Fourier transform infrared [FTIR] testing with fentanyl strips) are unique in their ability to test quantitatively (via the FTIR) a wide range of contents in a drug sample while qualitatively detecting a low concentration of fentanyl (via the fentanyl strips). A recent qualitative study found that clients prefer this combination technology, which is described by clients on the use of the FTIR, “That’s the best one then because it shows the quantity. It shows how much. That’s the most accurate one”, and “I think people want to know the percentage so they know, like say it’s got caffeine and fentanyl in it…a person who’s got a lower tolerance to fentanyl will want to find out the percentage so they’re not gonna overdose on it.”(24) This was also evidenced by findings indicating that a pilot program which offered FTIR testing paired with fentanyl strips resulted in significantly higher utilization rates as compared with fentanyl strips alone (1). We hypothesized that PWUD who witnessed an overdose would be more likely to use drug checking services, as observing such an event may lead PWUD to protect themselves from overdose and other negative unintended consequences by adopting harm reduction practices.

METHODS

Study design and population

The Vancouver Injection Drug Users Study (VIDUS), AIDS Care Cohort to evaluate Exposure to Survival Services (ACCESS), and the At-Risk Youth Study (ARYS) are prospective cohort studies involving PWUD in Vancouver, Canada. The methods for these cohorts have been previously described (25,26). In brief, VIDUS enrolls HIV-negative individuals (aged 18 or older) who injected drugs in the last month. ACCESS enrolls HIV-positive individuals (aged 18 or older) who used an illicit drug in the last month. ARYS enrolls street-involved youth (aged 14 to 26) who used an illicit drug in the last month. Participation involves the provision of informed consent and completion of an interviewer-administered questionnaire at baseline and every six-months thereafter. The questionnaire elicits information on behaviours, drug use patterns, the occurrence of traumatic events (e.g., violence, sexual assault, witness of an overdose), and access to health and social services (e.g., access to detox, and drug checking services). Following each interview, a 40$ CAD honorarium is provided. These cohorts have received ethics approval from Providence Health Care/University of British Columbia Research Ethics Board.

For the present analysis, the sample was restricted to those who completed a baseline or follow-up visit between June 1, 2018 and December 1, 2018.

Variable selection

The primary outcome of interest was having used drug checking services in the last six months (yes vs. no). The primary explanatory variable was having witnessed an overdose in the last six months (yes vs. no). Specifically, participants were asked “Have you witnessed an overdose in the last 6 months?”. We did not distinguish between whether fatal or non-fatal overdoses were witnessed.

We considered various confounders that may have impacted the relationship between having witnessed an overdose and having used drug checking services, including: age (per one-year increase); sex at birth (male vs. female); white ethnicity (yes vs. no); homelessness (yes vs. no); at least daily injection drug use (yes vs. no); at least daily heroin use (yes vs. no); at least daily cocaine use (yes vs. no); at least daily crack cocaine use (yes vs. no); at least daily crystal methamphetamine use (yes vs. no); at least daily illicit prescription opioid use (yes vs. no); suspected exposure to fentanyl (yes vs. no); public injection (yes vs. no); selling drugs (yes vs. no); sex for money (yes vs. no); and non-fatal overdose (yes vs. no). With the exception of time-stable variables, all variables were considered as activities reported in the last six months.

Statistical analyses

First, we provided study sample characteristics, stratified by use of drug checking services in the last six months using Pearson’s chi-square test for categorical variables and the Mann-Whitney test for continuous variables. Then, bivariable logistic regression was used to estimate the relationship between use of witnessing an overdose and use of drug checking services. Next, we constructed a multivariable logistic regression model, including all potential confounding variables with a cut-off of p < 0.05 in bivariable analyses. All p-values were two-sided. Analysis was conducted using R version 3.5.0 (Foundation for Statistical Computing, Vienna, Austria).

RESULTS

Of the 1,426 PWUD eligible for inclusion, 530 (37.2%) were female, 652 (45.7%) self-reported white ethnicity, and the median age at baseline was 45 years (quartile [Q] 1 - Q3: 32 – 55). In total, 767 (53.8%) reported witnessing an overdose in the last six months. Of which, the most commonly reported relationship between the witness and the overdose victim included no relationship (i.e., stranger) (60%); casual friend (36%); and close friend (13%). One hundred ninety-six (13.7%) reported use of drug checking services in the last six months. Baseline characteristics stratified by the outcome are presented in Table 1.

Table 1.

Study sample characteristics, stratified by use of drug checking services in the last six months

		Use of drug checking services in the last six months
Characteristic	Total n=1426, N (%)	Yes, n=196, N (%)	No, n=1228, N (%)
Witnessed an overdose*
Yes	767 (53.8)	155 (79.1)	612 (49.8)
No	653 (45.8)	41 (20.9)	612 (49.8)
Age†
Median	45.1	40.4	45.7
IQR	(31.8–55.1)	(30.2–52.2)	(32.3–55.6)
Sex at birth
Male	896 (62.8)	120 (61.2)	775 (63.1)
Female	530 (37.2)	76 (38.8)	453 (36.9)
White ethnicity
Yes	652 (45.7)	103 (52.6)	547 (44.5)
No	761 (53.4)	89 (45.4)	672 (54.7)
Homelessness*
Yes	284 (19.9)	64 (32.7)	220 (17.9)
No	1139 (79.9)	131 (66.8)	1006 (81.9)
≥ Daily injection drug use*
Yes	443 (31.1)	118 (60.2)	324 (26.4)
No	983 (68.9)	78 (39.8)	904 (73.6)
≥ Daily heroin use*
Yes	361 (25.3)	90 (45.9)	270 (22.0)
No	1065 (74.7)	106 (54.1)	958 (78.0)
≥ Daily cocaine use*
Yes	38 (2.7)	2 (1.0)	36 (2.9)
No	1388 (97.3)	194 (99.0)	1192 (97.1)
≥ Daily crack use*
Yes	115 (8.1)	10 (5.1)	105 (8.6)
No	1311 (91.9)	186 (94.9)	1123 (91.4)
≥ Daily methamphetamine use*
Yes	259 (18.2)	56 (28.6)	202 (16.4)
No	1167 (81.8)	140 (71.4)	1026 (83.6)
≥ Daily illicit prescription opioid use*
Yes	30 (2.1)	10 (5.1)	20 (1.6)
No	1396 (97.9)	186 (94.9)	1208 (98.4)
Suspected exposure to fentanyl*
Yes	736 (51.6)	165 (84.2)	571 (46.5)
No	681 (47.8)	29 (14.8)	652 (53.1)
Public injection*
Yes	320 (22.4)	83 (42.3)	236 (19.2)
No	1105 (77.5)	113 (57.7)	991 (80.7)
Drug dealing*
Yes	216 (15.1)	57 (29.1)	158 (12.9)
No	1210 (84.9)	139 (70.9)	1070 (87.1)
Sex work*
Yes	119 (8.3)	28 (14.3)	91 (7.4)
No	1307 (91.7)	168 (85.7)	1137 (92.6)
Non-fatal overdose*
Yes	162 (11.4)	40 (20.4)	122 (9.9)
No	1261 (88.4)	156 (79.6)	1105 (90)

IQR: interquartile range

^†Per one-year increase

^*Activities reported in the last six months

As presented in Table 2, in bivariable logistic regression, we found that witnessing an overdose was positively and significantly associated with use of drug checking services, as compared with those that had not witnessed an overdose (Odds Ratio [OR] = 3.78; 95% CI: 2.66 – 5.49). This association remained significant in a multivariable analysis after adjusting for various confounders (Adjusted Odds Ratio [AOR] = 2.32; 95% CI: 1.57 – 3.49).

Table 2.

Bivariable and multivariable logistic regression models on factors associated with use of drug checking services

	Unadjusted		Adjusted
	OR (95% CI)	P	OR (95% CI)	P
Witnessed an overdose*	3.78 (2.66 – 5.49)	<0.001	2.32 (1.57 – 3.49)	<0.001
Age†	0.98 (0.97 – 1.00)	0.005	1.00 (0.99 – 1.02)	0.784
Sex at birth	0.92 (0.68 – 1.26)	0.612
White ethnicity	1.34 (0.98 – 1.85)	0.067	1.40 (1.01 – 1.96)	0.045
Homelessness*	2.23 (1.60 – 3.11)	<0.001	1.25 (0.83 – 1.89)	0.284
≥ Daily injection drug use*	4.22 (3.09 – 5.79)	<0.001	2.00 (1.31 – 3.06)	0.001
≥ Daily heroin use*	3.01 (2.20 – 4.11)	<0.001	0.97 (0.65 – 1.44)	0.862
≥ Daily cocaine use*	0.34 (0.06 – 1.13)	0.141
≥ Daily crack use*	0.57 (0.28 – 1.07)	0.104
≥ Daily methamphetamine use*	2.03 (1.43 – 2.86)	<0.001	1.07 (0.71 – 1.59)	0.749
≥ Daily illicit prescription opioid use*	3.25 (1.44 – 6.89)	0.003	1.45 (0.55 – 3.53)	0.426
Use of drugs that contained fentanyl*	6.50 (4.38 – 9.97)	<0.001	3.35 (2.09 – 5.48)	<0.001
Public injection*	3.08 (2.24 – 4.23)	<0.001	1.03 (0.68 – 1.55)	0.879
Drug dealing*	2.78 (1.95 – 3.93)	<0.001	1.46 (0.98 – 2.15)	0.062
Sex work*	2.08 (1.30 – 3.24)	0.002	1.05 (0.62 – 1.74)	0.851
Non-fatal overdose*	2.32 (1.55 – 3.42)	<0.001	1.34 (0.86 – 2.08)	0.190

OR: odds ratio

CI: confidence interval

^†Per one-year increase

^*Activities reported in the last six months

DISCUSSION

In the present study, we observed only a small proportion of PWUD using drug checking services, and more than half of PWUD had recently witnessed an overdose. Consistent with our primary hypothesis, our findings demonstrated a positive and independent relationship between witnessing a fatal or non-fatal overdose and use of drug checking services, compared to those who had not witnessed an overdose, after adjusting for a range of possible confounders.

Consistent with other studies (27–29), we observed a high prevalence of having witnessed an overdose. Our finding of a positive association between witnessing an overdose and use of drug checking services is also consistent with other studies (16). A study conducted in three settings in the eastern United States found that participants who had witnessed a fatal overdose were 1.57 times likely to express interest in using drug checking, as compared with those who had not witnessed a fatal overdose (16). Our findings revealed that participants who had witnessed an overdose were 2.32 times likely to use drug checking services, as compared with those who had not witnessed an overdose. Differences between the two studies could be a function of intention versus action, or differences in the study sample. Nevertheless, these findings merit further exploration as little remains known about how overdose exposure might drive uptake of drug checking services.

That only a small proportion of PWUD in the study reported using drug checking services was surprising given that drug checking services are available at various supervised injection and overdose prevention sites in the region (20–23), and given that studies that have shown a high willingness to use drug checking services among PWUD populations (16,17,30). There appears to be a discordance between availability and willingness to use drug checking services and uptake of drug checking services. This discordance may be may be partially attributable to a sense of invincibility, or an ambivalent view of death (31,32). In Vancouver, a qualitative study involving 18 people who inject drugs (PWID) found that perceived invincibility was a common narrative which diminished the potential impact of an overdose warning campaign (32). In Australia, a qualitative study involving 60 PWID found that 28 (47%) were apathetic toward death. It could be that PWUD have a high level of trust in their dealer (33), or they are aware of the limitations of current drug checking interventions (e.g., inability to detect some analogues present at low concentrations, or inability to provide quantitative information) (1,34). Some may not utilize drug checking services because they may not have an alternative source for acquiring opioids (19,32). Other reasons for the discordance between availability and uptake may include time, or competing risk priorities (e.g., mitigating risk associated with income-generation activities, securing accommodations, avoiding arrest) (35,36).

Our study was not without limitations. First, our study was reliant on data from observational research, which may lack measures of some other potential confounders. Second, this was a cross-sectional study, which limits the ability temporality between having witnessed an overdose in the last six months and having used drug checking services in the last six months. Third, there may be potential for misclassification of overdoses; however, we believe this to be minimal given that overdoses are common in this setting, and it is likely that participants would be able to identify one if it occurred. Fourth, all data was self-report, which introduces the possibility of social desirability and recall bias. Finally, these results may not be generalizable to local PWUD or beyond PWUD in this setting.

In sum, the present study found that PWUD who witnessed an overdose were more likely to use drug checking services, compared to those who had not witnessed an overdose. Our findings suggest that witnessing an overdose may motivate PWUD to use drug checking services. Given that only a small proportion of PWUD in the study reported using drug checking services, our findings highlight an urgent need to continue to scale-up a range of overdose prevention interventions.