Monthly Reports

September 2023 Monthly Report

Derek Robinson

Oct 12, 2023 • 12 min read

Samples from September 23rd, two fentanyl samples, one side (meth) sample, and one crack sample (top to bottom, left to right)

In this blog post, we discuss our September 2023 report and provide more information on how to interpret the results. The PDF report can be found at the end.

Key findings:

The median fentanyl concentration found across all drug categories was 8.4%
The median fluorofentanyl concentration nearly tripled from August, going from 5.1% to 14.5%
Carfentanil was found in 3 expected opioid - down samples with a median concentration of 0.2% and maximum concentration of 0.4%
Benzodiazepines and/or etizolam were found in 60% (223/374) of expected opioid-down samples
Bromazolam, the most common benzo found within opioid-down, was found in 174 opioid-down samples with a median concentration of 5.8% and maximum concentration of 25.0%
Xylazine was found in 24 samples (all expected opioid-down) with a median concentration of 6.0% and a maximum concentration of 31.6%
Isobutyryl fentanyl was found in 23 samples (all expected opioid-down) with a median concentration of 5.2% and a maximum concentration of >80%. It’s speculated that Isobutyryl fentanyl has roughly ¼ the potency of fentanyl

Insight for the September 2023 Monthly Report

This blog, and the associated pdf report, breakdown our sample counts into six categories:

samples received through direct service provision in Victoria, where service users are bringing samples into the Substance storefront. These samples are labelled as “Substance” samples in the figures/tables of this blog post
samples received through direct service provision in Campbell River, where service users bring samples to the Vancouver Island Mental Health Society (VIHMS). These samples are labelled as “Campbell River”.
samples received through direct service provision in the Comox Valley, where service users are bringing samples to AVI Health & Community Services in Courtenay, BC. These samples are labelled as “Comox Valley”.
samples received through direct service provision in the Cowichan Valley, where service users bring samples to the Duncan Lookout Society OPS in Duncan, BC. These samples are labelled as “Duncan”.
samples received through direct service provision in Port Alberni, where service users bring samples into Port Alberni Shelter Society’s OPS. These samples are labelled as “Port Alberni”.
samples received through indirect service provision, where samples are collected through no-contact drop-off envelopes, are collected by harm reduction workers and other community members at supported housing sites, overdose prevention sites, and supervised consumption locations. These samples are labelled as “Outreach” samples in the figures/tables herein. September’s Outreach data includes samples collected at Rifflandia Electric Avenue and Rifflandia The Park.

Drug types
Fig. 1 shows the prevalence of each expected drug category checked, split by sample collection location/method

Figure 1. Prevalence of drug classes checked during September split by sample collection/method. Bars are stacked by the percentage of samples in each drug class, with the individual sample counts overlaid.

Table 1. Prevalence of expected drug classes checked during September, split by sample collection/method.

September's Superb Summary of Substances

For the majority of samples checked, we confirm the presence of the expected drug with no additional active compounds detected above the limitations of the drug check. The bar charts below highlight a few classes of drugs, differentiating samples where only the expected active component was detected - from situations when other unexpected active components were detected.

Figure 2. Compositional breakdown by drug class and sample collection location/method. Bars are stacked by the percentage of samples in each category, with the individual sample counts (and relative proportions) overlaid. “Dark Blue” groups samples that were as expected with no other notable compounds detected, “Purple” groups samples that contained the expected drug and contained other unexpected active(s), “Magenta” groups samples that did not contain the expected active but did contain unexpected active(s), Salmon groups samples where we were unable to determine the composition (e.g. scenarios where we do not have appropriate reference spectra), and Lime displays samples where no active compounds were detected.

98% (43/44) of expected ketamine samples checked in September were confirmed to be ketamine with no other active compounds detected. The remaining expected ketamine sample, from Campbell River, was found to contain ketamine and phenacetin. Phenacetin (a.k.a. "superbuff") is a tylenol-like drug that is most often found as cut in cocaine samples.

Speaking of cocaine, 88% (117/133) of expected cocaine samples (99 cocaine HCl/soft, 34 cocaine base/hard/crack) were confirmed to be cocaine with no additional active compounds detected.
4 samples were found to only contain inactive components:

Two samples were found to contain only sodium bicarbonate (one from Victoria, one from Duncan). If mixed with vinegar these might turn into the local high school science fair.
One sample was found to contain a mixture of caffeine, erythritol, xylitol.
We were unable to detect any compounds in the remaining sample

11 samples contained an active component in addition to cocaine:

Fentanyl or a fentanyl analogue - Three samples (Victoria, expected cocaine base); detected via strip test only. The presence of fentanyl in samples other than expected opioid - down is rare. In these cases it is likely due to cross contamination.
Phenacetin - Two samples (Victoria, expected cocaine base) were found to contain approximately 3.5% phenacetin.
Levamisole - Six samples (Victoria, expected cocaine HCl); a veterinary dewormer often found as a cutting agent in cocaine.

One expected cocaine base sample did not contain cocaine. This sample was translucent crystals from Victoria and was found to be MDMA with no cuts of adulterants detected.
One expected cocaine base sample did not contain cocaine. This sample was translucent crystals from Victoria and was found to be MDMA with no cuts of adulterants detected.

89% (42/47) of expected methamphetamine samples checked were found to be meth with no other active compounds detected. One sample was found to contain only sodium bicarbonate.

In two of the remaining four samples, meth was found in addition to another active component:

Fentanyl or a fentanyl analogue - Two samples (1x Duncan, 1x Victoria); in both cases fentanyl was only detected via strip test. Thus, it is likely present in only trace amounts and as a result of cross contamination

In the last two samples, meth was not detected but rather cocaine base (1x Port Alberni, 1x Campbell River)

88% (91/103) of expected MDA/MDMA samples checked were confirmed to be MDA (3 samples) or MDMA (88 samples) as expected. The remaining expected MDA/MDMA samples displayed a range of compositions:

Four expected MDA samples were found to contain additional active components.
- One sample contained both MDMA and ketamine
- Two samples contained both MDMA and cocaine base
- One sample contained both MDMA and MDA
Four expected MDA samples were found to be MDMA instead
Three expected MDMA samples were found to be MDA instead
One expected MDMA sample was found to be ketamine instead

Benzodiazepines (n=11)
31% (4/11) of the expected benzodiazepine samples checked in September came to our service sites in the form of pressed pills with the following expected and detected compositions:

Table 2. The composition of benzodiazepine pressed pills checked in September. “Expected To Be” describes the benzo expected/reported by the service user, while “Found To Contain” describes the contents we found through the drug check. “Undifferentiated benzodiazepine” refers to samples where the benzo strip test was positive but no benzos were identified with our other instruments. These “undifferentiated” samples either contain a benzo at very low concentrations and/or novel benzos that are not in our targeted method for the mass spectrometer.

Within the remaining seven samples, two were as expected (alprazolam and bromazolam), one expected bretazenil sample could not be identified, and the remaining four samples, all with an unspecified expected active were as follows:

One sample from Victoria contained bromazolam and fluorofentanyl
Two samples from Victoria contained bromazolam, fentanyl, and fluorofentanyl
One sample from Victoria contained fentanyl and fluorofentanyl

Opioid-positivity in non-opioid-down samples

In September, we checked 415 samples that were not expected to contain fentanyl or other “unexpected” opioids^[1]. Since the opioid-down supply is no longer “just heroin” or “just fentanyl” and is instead a complex, potent, and ever-changing polysubstance market containing other synthetic opioids like fluorofentanyl or nitazenes, here we will examine the prevalence of any unexpected opioid, not just fentanyl, detected in non-opioid-down samples.

Specifically, we are excluding samples that were expected to be “opioid-down” or samples that had an “unknown/missing” expected composition. In the case of “opioid-other” samples, e.g. hydromorphone tablets and oxycodone pills, “unexpected opioids” are defined as any opioid that is not the expected opioid. ↩︎

Table 3. Overview of unexpected opioid detection in non-opioid-down samples in September.

Examining Table 3, we find that thirteen samples tested positive for unexpected opioids in September, representing 3.1% of all non-opioid-down samples checked. These samples were as follows:

Three were expected cocaine base samples. All three tested positive for fentanyl via strip test only with no indication of fentanyl present on our paper spray mass spectrometer
Two were expected methamphetamine samples. Both tested positive for fentanyl via strip test only with no indication of fentanyl present on our paper spray mass spectrometer
Four were expected unknown samples that turned out to be benzo - down samples
One was an expected tucibi sample which turned out to contain bromazolam and fentanyl. Tucibi is a polysubstance mixture also known as “Tusi” or “pink cocaine”; often a mixture of cocaine, MDMA, and ketamine
Two were expected hydromorphone samples that ended up being isotonitazene instead
One was an expected oxycodone sample that ended up being metonitazene instead

In September, no unexpected opioids were detected in samples expected to be MDMA/MDA or dissociatives.

In people’s personal quests for bodily autonomy and informed consumption, there is often evaluation of risk and consequence, but when the consequences can be severe and the risks are unknown or are intentionally exaggerated, these become difficult, if not impossible, conversations to weigh. We believe that drug checking can help provide people with the information needed to evaluate the risks, and provides harm reduction advice to minimize undesired consequences of substance use. These data are not meant to downplay concerns or invalidate past experiences. We recognize the tragic consequences of when fentanyl is found in non-opioid samples and honour the heartbreak that such experiences produce. Instead, we present these data with the intent to combat misinformation and provide an evidence-based context for people to consider when making decisions about substance use. While these numbers reflect what we have seen over the course of the project, these (roughly) 1-in-200 events still occur, so we always encourage folks to get their stuff checked.

Opioid-Down (n=374)

In this section we present results specific to the opioid-down supply, therefore they may differ from the highlighted findings above that are inclusive of all expected drug categories.

80% of expected opioid-down samples contained fentanyl (301/374)
28 samples contained heroin (7% of expected opioid-down samples)
7 samples (Victoria) contained heroin (and related alkaloids like acetylmorphine (MAM), acetylcodeine, and morphine). Caffeine was found as a cut in 1 sample.
The other 21 samples contained heroin with the additional actives such as fentanyl, fluorofentanyl, isobutyryl fentanyl, bromazolam, undifferentiated benzos, and etizolam.
64% of expected opioid-down samples contained fluorofentanyl (241/374)
Three samples contained carfentanil
60% of expected opioid-down samples contained a benzodiazepine and/or etizolam (223/374)
Xylazine was detected in 6% (24/374) of opioid-down samples

In September, 74% (270/333) of all opioid-down samples checked contained an additional active to the expected fentanyl/heroin. These data are shown in Fig. 3 highlighting the prevalence of benzos, fluorofentanyl, and xylazine in the down supply.

Figure 3. The percentage of expected opioid-down samples checked between August 2022 and September 2023 that only contained fentanyl/heroin actives (solid dark blue), opioid-down samples with an additional active detected (dot-dashed purple), opioid-down samples that contained a benzodiazepine-related drug (dotted magenta), opioid-down samples that contained fluorofentanyl (dashed salmon), and opioid-down samples that contained xylazine (dashed yellow).

Fluorofentanyl was the most common additional active found within the opioid-down supply, with 50% (186/374) of opioid-down samples containing fluorofentanyl in addition to fentanyl. Additionally, fluorofentanyl was the only opioid detected in 14% (53/374) of opioid-down samples (i.e. no fentanyl or heroin was detected in these samples).

Benzo-related drugs contribute to a majority of the other additional actives found in expected opioid-down samples, with 60% (223/374) of expected opioid-down samples checked containing a benzo-related drug. Bromazolam continues to be the most common benzo seen in the down supply, with bromazolam being detected in 78% (174/223) of the benzo-positive opioid-down samples. Scattered detections of other drugs are still found and can be reviewed in the pdf report at the end of this blog

Quantification for Expected Opioid-Down^[1]

In September, we quantified fentanyl for 285 of the expected opioid-down samples containing fentanyl and found the median concentration to be 8.5%^[2]. Though the median is a useful indicator, it doesn’t capture the volatility of fentanyl concentrations present in the opioid supply, as half of fentanyl-positive down samples contained between 2.5% and 15.4% fentanyl, and any one sample might be the lowest strength (0.1%) or the strongest (>80%^[3]). Fluorofentanyl was seen at concentrations ranging from 0.2% to greater than 80% as well, with a median concentration of 14.5%. Similarly, the concentration of bromazolam was across the board in expected opioid down samples, with samples ranging from less than 0.1% to 25.0% bromazolam, with a median of 5.8%. For context, a sample containing 4% bromazolam would be roughly equivalent to two full 2mg Xanax bars worth of benzo per point (100mg).

Not all opioid down samples brought to our service can be quantified. This is primarily due to too limited sample collected for our instruments to report a reliable mass percentage. Nevertheless, qualitative detection is still possible. ↩︎
This number is specific to fentanyl quantified in opioid-down samples. The median concentration listed in the Key Findings at the beginning of this blog (8.4%) is inclusive of all samples checked, across all drug classes and unknown samples, that contained fentanyl. ↩︎
For samples that contain more than 80% fentanyl, or more than 25% etizolam by weight, our mass spectrometer is presently unable to reproducibly assign a concentration due to the upper limits of the calibration methods currently adopted ↩︎

Table 4. Quantification of low-concentration active components in expected opioid-down samples, across all service models/location, in September 2023. See Table 5 for a breakdown of this information by location. Weight percentage is reported; “IQR” is the interquartile range: the range that contains half of the quantified samples. *These concentrations reflect the upper limits of the PS-MS calibration.

The fentanyl, fluorofentanyl, and bromazolam concentrations that we quantified in September, across all expected drug categories and service models, are presented in Fig. 4. Small black dots are individual opioid-down samples, the large white dot shows the median concentration, dashed white lines bound half of the quantified samples, and the width of the shaded regions mirrors the number of samples at a given concentration.

Figure 4. Violin plots of fentanyl (top panel) and bromazolam (bottom) positive samples quanitifed during September across all collection locations/methods.

We can also examine the regional variability in the unregulated market. The table below expands on the quantitative data presented above. It focuses only on fentanyl, fluorofentanyl, carfentanil, bromazolam, and xylazine quantified within expected opioid-down samples, separated by collection location/model. Weight percentage is reported; "IQR" is the interquartile range: the range that contains half of the quantified samples.

Table 5. Quantification of low-concentration active components in expected opioid-down samples, split by service models/location, in September 2023. See Table 3 for the overview from all location. Weight percentage is reported; “IQR” is the interquartile range: the range that contains half of the quantified samples. The upper limit of reporting for fentanyl on the PS-MS is 80% – any samples containing more than 80% will be flagged as ">80%".

Vancouver Island Drug Checking Sept 2023 Monthly Report

Vancouver Island Drug Checking Sept 2023 Monthly Report.pdf

2 MB

Want to be notified when we release these reports? Join our mailing list to receive updates about when our reports are out. You can subscribe and unsubscribe yourself from this list at any time.

As always, send us feedback at substance@uvic.ca on how we can continue to offer our drug checking results in a useful way.

Join the mailing list