Marijuana testing must be accurate

Nobody wants mold in their cannabis. So states that have legalized the drug, including Massachusetts, generally require companies to test products to make sure that they’re free from contaminants.

But for the system to work, labs hired to conduct the tests have to use proven, accurate methods to look for things like yeast and mold. State regulators may need to step in though because emerging evidence suggests some cannabis labs may not be as rigorous as they should be.

Data obtained from regulators in Maryland show a suspicious pattern: a steep drop-off in how much yeast and mold labs identified in cannabis products — right under the limit at which a product would be unable to be sold, according to data compiled by Yasha Kahn, vice president of marketing and data science at Massachusetts-based MCR Labs, who shared with the Globe data collected from multiple states via public records requests. In Michigan, data show another suspicious pattern: new labs opened up that had low failure rates and the market shifted to those labs, driving down failure rates overall.

In both states, when regulators updated rules to ensure the lab tests used had been scientifically validated according to standards approved by an independent third party, the patterns disappeared. The steep drop-off in Maryland disappeared, while failure rates increased again in Michigan.

Testing problems have been found in Massachusetts, too. A study of Massachusetts products bought off the shelves, published in the Journal of Testing and Evaluation by ASTM International, a standards organization formerly known as American Society for Testing and Materials, whose authors include Kahn and Jeff Rawson of the Cambridge-based Institute of Cannabis Science, found “that levels of cannabinoid compounds are systematically overlabeled, that flower with forbidden levels of microbial contaminants reaches consumers, and that papers of prerolled joints may contain heavy metals.” The study concludes: “The errors in labeling and safety that we highlighted here are part of an array of misinformation that puts consumers at risk.”

There are built-in incentives for labs to tell growers what they want to hear. The marijuana growers who pay for testing want lab results showing high potency because it commands a higher price and an absence of contaminants because contaminated flower can’t be sold. Growers can “lab shop” to find labs that deliver these results. There have been cases of outright fraud — the license of Praxis Laboratory in Washington was suspended for falsifying testing data to provide high THC potency results, while Cannex Nevada had its license suspended for intentionally passing contaminated cannabis samples and inflating THC levels. But generally, data cannot show whether misconduct or inaccurate methods are involved.

When the industry started, there were not clear standards for cannabis testing. Luckily, that’s changing. David Vaillencourt, a Colorado-based cannabis consultant and vice chair of an ASTM committee developing cannabis standards, said the organization is “getting really close” to developing standards that regulators can choose to adopt to properly regulate cannabis. There are already standards regarding testing methods and how to choose a representative sample of a crop, among others.

There are models Massachusetts can look to to ensure lab testing is based on good science. Michigan requires that any method used to conduct cannabis safety tests is peer-reviewed and validated by an independent third party in accordance with accepted standards. Maryland regularly updates a technical bulletin with rules for cannabis testing labs, including how methods must be scientifically validated.

Experts say accurate methods must be coupled with enforcement to discourage dishonest practices. This may include holding cultivators, not just labs, accountable if products are mislabeled. Colorado issues a public health and safety advisory whenever contaminants or pesticides are found in a marijuana product on the shelves. Some states have secret shopper programs, where regulators buy products from stores, then retest them.

Bob Miller, chief scientific officer at ACT Laboratories, which operates in six states (not Massachusetts), said Michigan seems to have done a better job regulating laboratories than other states because regulators have standards and enforce them. Michigan’s computer system automatically flags unusual test results, and regulators will then contact the testing lab and request raw data, confirmatory testing, and sometimes a piece of the sample to retest elsewhere. The state submits its own samples to multiple labs to compare results. Michigan is also creating its own standards lab. “Will there always be cheating? Yes,” Miller said. “Can cheating be minimized? Yes.”

In Massachusetts, the Cannabis Advisory Board made recommendations, which this board supports, that include: making testing data public; providing guidance on calculating THC; conducting quarterly audits of labs; considering building a standards lab; starting a secret shopper program; and establishing a testing-related working group.

In an interview with the editorial board, Cannabis Control Commissioner Kimberly Roy voiced a commitment to improving method standardization and eventually doing regular confirmatory testing. She said internal conversations are ongoing about policy proposals to improve testing accuracy. Roy said regulators provide guidance around THC calculations. She said she understands the importance of a secret shopper program but could not comment further to keep any program secret. Labs are subject to unannounced inspections, and the commission recently appointed a new chief of investigations and enforcement, Nomxolisi Khumalo.

Commissioners have a lot on their plate — including their own leadership struggles. But better testing regulation needs to be a priority. The promise of the ballot question legalizing marijuana was that consumers could access a product that was tested and safe, and that’s a promise the commission needs to deliver on.

Editorials represent the views of the Boston Globe Editorial Board. Follow us @GlobeOpinion.

No Byline Policy

Editorial Guidelines

Corrections Policy

Source