The owner of one analytical testing lab is calling out what he says is a serious problem with the accuracy of THC levels on cannabis flower in Canada. 

Rob O’Brien, CEO and CSO of Supra Research and Development in Kelowna, BC, recently shared online his own independent testing results from 46 different cannabis products he purchased from BC Cannabis Stores.  

The results of his tests show what he says are significant variations in the cannabis flower he tested compared to what was stated on the label. In some cases there was more than a 40 percent difference. In one example results show a product labelled at 34 percent THC to be only 19 percent with his own testing.

Results can be found here, here, and here.

Rather than publicly calling out producers or labs, though, O’Brien says his biggest concern is holding federal and provincial regulators accountable. While he shared the results online, he didn’t share the labs, producers, or product names. But he says he did share the entire, unredacted info with the BC government and with Health Canada. 

“I’m not here to try to shame companies, I’m here to try to solve this problem,” says O’Brien. “The credibility of the entire sector is in question if we don’t get this right.”

Instead, he’s placing that burden on those two levels of government.

“They are the ones that should be responsible for naming names if that does come up.”

Although he doesn’t rule out labs or producers who may “put their thumbs on the scale,” O’Brien says he believes much of the discrepancies he found in his own testing were due to flaws in how producers are required to take samples for testing in the first place. 

One of the problems, he explains, is that the current federal regulations only require testing per harvest, even though there can be significant variation in the THC levels in the flowers from that harvest. Flowers closer to the light tend to have higher THC than those further from it. And while some growers take steps to mitigate this, variation of a biological product like cannabis is nearly inevitable. 

This, combined with a tendency for producers to then send in only the largest, most impressive flowers that have the highest THC, means that what is actually in the consumers’ packaging won’t necessarily match what the THC numbers show on the label. 

“The largest buds are the ones that are closest to the label claim. That’s likely the type of buds that are being sent to the labs for testing. These buds are two to three grams, maybe four at the extreme. And you can’t put a two-gram bud in a one-gram bag. So the smaller package sizes have the smallest buds in them, typically two or three, and those buds are 30-40 percent below the bigger ones. That’s part of the problem.”

Another factor is the tested buds aren’t going through a sometimes disruptive packaging process that can lower the THC level. 

“And when you send the first pristine buds to the lab for testing, then the others go through the process of being sorted and packaged, there’s a lot of tumbling and bouncing around, and that knocks off trichomes that affect the THC level. So if you’re testing the packaged product, you’ll be much closer to what’s there.”

Related Articles

138 cannabis samples tested by Health Canada in 2020, but no labs inspected

March 17, 2021

Can regulators reign in THC inflation without driving consumers to the illicit market?

February 1, 2023

If provincial buyers were held more accountable for the products they are generally the gatekeepers of in their provinces, he contends this would immediately force producers to begin better testing procedures on their own, regardless of the minimal requirement from Health Canada for batch lots covering an entire harvest.

This is especially true since, as he points out, most provincial buyers tend to have a bias against cannabis flower that is under 20 percent THC.

“The provincial buyers need to throw away this thing that if you don’t have 20 percent THC on the COA that they’re not going to buy their product. That is ridiculous and is contributing to the problem.”

“THC content has significantly increased over the last two years. We need to have a better regulatory system, and we need to have better transparency.”

Hubert Marceau, a chemist and the director of development at Laboratoire PhytoChemia Inc., an analytical testing lab in Quebec, says consumers should take this kind of knowledge into account when buying cannabis. 

Even relatively fair and accurate testing results will always have some kind of variance, he says. Instead, he thinks consumers should think of the THC level on the label as a range, give or take a few points in either direction, not a specific number. 

This would account for at least some of the variance in flower size and batch size. 

“You can never test the whole batch. So by testing a sample of that batch, you’re only estimating the average. Consumers need to be aware they are askew. There’s no other product on the market where you have this kind of false sense of security, this level of accuracy down to two decimal points.”

Like O’Brien, he says he is also aware of stories that some labs will simply provide the testing results that a processor demands in order to keep their business.

“Eventually, it just becomes a numbers game where people just want to have the higher number and will do everything in their power to have a higher number. And that’s what’s happening here.”

This also brings the credibility of the industry into question for Marceau. THC levels, from his perspective, seem to often hover just under 20 percent THC, so very high THC levels, sometimes well over 30 percent raise his eyebrows. 

He references a study from 2021 that showed clustering of testing results from cannabis flower in two US states that, once corrected for outliers from labs found to be inaccurate, showed most cannabis flower, when accurately tested, was around the 20% threshold, some a little under, some a little over. 

“Eventually, the plant gets to a point where it can’t produce more THC. There’s no way in hell that THC—that only resides in the trichomes that are on the surface of the flower—is going to be 35 percent of the cannabis.”

Like O’Brien, he says he thinks a lot of it comes down to the 20 percent (or, in the same cases, even higher) THC threshold some provinces prefer. This creates a situation where cultivars, processors, and labs are all encouraged to push up their levels to meet these demands. And this creates a situation where consumers are largely only able to buy these higher THC products, further cementing the idea of higher THC equalling higher quality. 

“You have the consumer who has been told higher THC is better, which is the equivalent of saying only drink everclear alcohol, so the distributor is saying it will only buy lots over 20 percent. This forces the producer to put out more than 20 percent flower.”