It’s easy to ingest cannabis irresponsibly. According to a recent study in the Journal of the American Medical Association, the labels on marijuana edibles are seldom trustworthy: Fewer than 1 in 5 edibles (17 percent) had accurate labels of tetrahydrocannabinol (THC) content, the main psychoactive ingredient in marijuana.
Edibles are a popular option in states that have allowed marijuana to be legally sold, coming in the form of candy bars, cookies and even gummy bears. By December of 2014, the first year of legal recreational marijuana sales in Colorado, edibles made up 45 percent of that state’s marijuana market, which also includes marijuana sold for medical use. Nearly 5 million units of edible products were sold in Colorado last year.
Colorado has offered a model for other states considering legalizing and regulating edibles but is still working out the kinks of its new system. Colorado’s regulations focus on capping the total amount of THC allowed in any edible at 100 milligrams per unit. In 2014, 98 percent of the edibles that Colorado tested met this standard. While Colorado’s regulations set some helpful limits for edibles, they are not designed to catch the labeling problems brought to light by the study, which examined edibles sold in California and Washington state. Colorado’s rules don’t address the risk of getting a higher dose than expected if a THC label is inaccurate.
The study showed that 23 percent of the edibles analyzed delivered a dose of THC that was at least 10 percent higher than the labeled amount. On average, these edibles delivered a THC dose 28 percent higher than a customer would expect. For some drugs, that level of variation wouldn’t be a big deal, but, according to Ryan Vandrey, an associate professor of psychiatry and behavioral sciences at John Hopkins University and the lead author of the study, “cannabis has a narrow window to deliver a therapeutic dose without side effects.” At higher doses, THC can cause anxiety, drowsiness and vomiting.
For recreational users, some of the side effects — particularly the high — may be the point. But some medical users prefer to use cannabis to relieve their existing symptoms (commonly seizures, lost appetite, etc.), not cause new ones. And even someone consuming marijuana edibles for their psychoactive effects probably prefers to avoid anything resembling Maureen Dowd’s overdose.
Mislabeling is also a consumer-service issue. Sixty percent of the products that were tested had at least 10 percent less THC than their wrappers promised. The average amount of “missing” THC for these underdosed edibles was 47 percent. For users of medical marijuana, underdosing on marijuana isn’t any less serious than coming up short on any other prescribed medication. And, for recreational users, underdosing can have more serious consequences than simply feeling cheated out of the high that that they’ve purchased.
That’s because the more serious danger of inaccurate labeling might not be the risk of a single bad experience, but a systematic uncertainty that makes it difficult to be a responsible consumer. Someone who has relied on one mislabeled brand, for example, could wind up misdosing herself when she switches to another one.
For customers who use edibles that are sold with multiple doses in a single package (candy bars and other formulation), Vandrey’s study exposed another difficulty in accurate dosing: An edible might, on the whole, have the amount of THC that the label promised, but if the drug was inconsistently mixed into the edible, the individual servings could have very different doses. Inconsistent mixing was enough of a problem that Vandrey couldn’t stick with his planned experimental protocol. He had intended to take a few samples from each edible and test them for THC content, but in almost half of all the edibles he wanted to test (37 of 75 products), the samples he extracted from a single product differed by more than 10 percent from one another. In those cases, Vandrey had to homogenize the products first by crushing or mixing them.
Vandrey’s protocol considered an edible accurately labeled if its THC content fell within 10 percent of the amount on the label, but an “accurate” product that fails the consistency test can still cause problems for its users. People consuming edibles whose recommended serving is only a portion of the product can still wind up being under- or overdosed by any individual serving. Also, a medical marijuana consumer whose edibles vary significantly within a single package will not be able to be confident in the dose he takes when his symptoms flare up.
Only half of all the products tested in Vandrey’s study had a consistently distributed dose, leading Vandrey to “caution against packaging multiple doses in one product.” A coin-flip chance of having the dose in your edible vary by 10 percent or more between bites is a poor gamble to take.
This study suggests that manufacturers of cannabis edibles have some work to do to ensure that customers can consume their products responsibly, but there are no industrywide standardized protocols for testing or thresholds for accuracy in labeling for businesses or customers to rely on at present. Paul Armentano, deputy director of the National Organization for the Reform Marijuana Laws (NORML), said states trying to require testing and accurate labeling of cannabis products have run into legal problems. “There exists a lack of standardized rules, regulations and methods for those engaged in product testing since such activities are in conflict with federal law,” he said. Marijuana is classified as a Schedule I drug by the Food and Drug Administration, the most restrictive category, which requires researchers to get special permission to do any studies involving cannabis.
Until cannabis edibles are reliably tested and labeled, recreational and medical users have little guidance besides caveat stoner.