Maryland Cannabis Administration to Launch Five-Lab Testing Accuracy Pilot

The Maryland Cannabis Administration announced on September 28, 2026, that it will launch a proficiency testing pilot in October 2026 comparing cannabinoid testing of Maryland-licensed cannabis flower across five laboratories. In the release, issued from Linthicum, Maryland, the agency described lab shopping, potency manipulation, and testing fraud as a widespread issue in the cannabis industry and said the pilot is intended to measure statewide testing accuracy and address lab shopping.

The Maryland CSQ CLEAR Pilot Program will compare cannabinoid testing of Maryland flower products across five laboratories: four MCA registered independent testing laboratories and the MCA Office of Laboratory Services reference laboratory. The pilot will assess inter-laboratory agreement, intra-laboratory repeatability, and differences between current measurements and product label and original Certificate of Analysis values. All assessment findings will be anonymized; the agency said the pilot will be used for informational purposes only, with no punitive consequences. The first round of testing will commence in early October 2026, with results expected to be compiled and released in December, according to the administration.

The CSQ CLEAR Program

The pilot was developed by Cannabis Safety & Quality, which the release describes as the first accredited certification program of its kind. The release states that the CSQ CLEAR Program uses a modified P-BAT approach and forensic statistical analysis to identify instances of testing manipulation and fraud, calling the program a significant step forward in repairing the integrity of cannabis testing as a result of lab shopping.

“We created the CSQ CLEAR Program as a more cost-effective, robust, transparent, and iterative approach to solving the problem of ‘lab shopping’ than what is currently available,” said Darwin Millard, technical director at CSQ. Millard said the pilot being conducted in Maryland with the MCA will provide a proof of concept that the CSQ CLEAR Program is a reliable model that can be replicated in other states.

The release states that previous studies have shown THC levels are routinely inflated by as much as 25 percent in several legal markets, and it says the CSQ CLEAR Program aims to analyze that discrepancy, addressing systemic issues impacting data integrity and the trustworthiness of laboratories in the U.S. cannabis industry.

“MCA has always had a strong focus on ensuring cannabis product quality in Maryland. The CSQ CLEAR Pilot program provides us an opportunity to measure our efforts industry-wide and adjust if necessary,” said Tabatha Robinson, director of the Maryland Cannabis Administration.

Prior Research on THC Potency Inflation

In a study published April 12, 2023, in PLOS ONE, researchers affiliated with the University of Northern Colorado School of Biological Sciences and Mile High Labs examined the accuracy of THC potency values printed on retail cannabis packaging. The paper noted that legal cannabis products in the United States are required to report THC potency, expressed as total THC percentage by dry weight, on packaging, and it cited earlier research showing that THC potency is a primary factor in determining pricing for cannabis flower.

Using high-performance liquid chromatography, the researchers analyzed THC potency in 23 cannabis flower samples purchased from 10 dispensaries across the Colorado Front Range and compared the results with the potency reported on the packaging. The samples, each one to two grams, represented 12 strains with reported THC values ranging from 12.8 percent to 33 percent. Testing was performed at Mile High Labs, a third-party laboratory in Loveland, Colorado; the paper discloses that two of its authors were Mile High Labs employees while the data was collected and that the laboratory provided support for the study in the form of their salaries.

The researchers reported an average observed THC potency of 14.98 percent, plus or minus 2.23 percent. Observed potency averaged 23.1 percent lower than the lowest label-reported values and 35.6 percent lower than the highest label-reported values, with a maximum percentage change of -56.5 percent. Eighteen of the 23 samples, or 78.26 percent, tested below the lowest value reported on the label; 16 of the 23 were more than 15 percent lower relative to the lowest reported value, and 13 of those samples were more than 30 percent lower. Three samples contained only half of the reported maximum THC potency. A Wilcoxon Signed Rank Test found a statistically significant difference between reported and observed potency (Z = 3.833; p < 0.001).

The authors wrote that sampling was largely opportunistic, conducted in Denver, Garden City, and Fort Collins, and that the study was designed from a consumer perspective to provide an overview of label accuracy in Colorado rather than to compare dispensaries, cities, or strains. They stated the study was not designed to pinpoint the source of the discrepancies, but identified a lack of standardized testing protocols, limited regulatory oversight, and financial incentives to market high THC potency as likely playing a significant role. The paper describes lab shopping as a phenomenon in which cultivators and dispensaries seek out labs that generate the most desirable lab results, and it says the practice is believed to be problematic in many states.

The authors described their work as the first peer-reviewed study to empirically examine commercial THC potency. They recommended that rigorous standard operating procedures be developed to minimize testing variation among labs and that independent testing be implemented to compare THC potency values at the time of harvest and at the point of sale.

Citation: Schwabe AL, Johnson V, Harrelson J, McGlaughlin ME (2023). Uncomfortably high: Testing reveals inflated THC potency on retail Cannabis labels. PLOS ONE 18(4): e0282396. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0282396