Science & Health
Marijuana Product Flavors Are ‘Concerning,’ New American Medical Association Study Of 500,000 Weedmaps Listings Says
A new federally funded study published by the Journal of the American Medial Association (JAMA) analyzed product listings on a popular marijuana advertising site and found that nearly half of all products were “flavored,” based on notes in the product descriptions. Researchers raised concerns that those flavors—along with cannabis-infused edibles, products like concentrates and merely listing retailers and products online—may make marijuana more appealing to young people.
Authors, from the University of California San Diego and California State University Fullerton, first used “automated data collection” and “custom data mining approaches” to scrape product listing information from the cannabis platform Weedmaps, which catalogues products and provides online menus for retailers across the country. All told, the team said it looked at more than 500,000 consumable cannabis product listings.
Researchers then manually annotated the listings “to identify and categorize product characteristics and flavor types,” the paper says, for example looking at product form factor, chemical makeup, flavor description and other information provided on the platform.
The goal, said authors, was to “develop a scheme to classify and describe the breadth of cannabis products commercially available in the US as marketed on a large e-commerce cannabis platform. They noted the study, which received funding from the National Institute on Drug Abuse (NIDA), is part of a “larger project examining new and emerging cannabis products.”
“From over half a million unique listings on a popular e-commerce platform, a large diversity of [cannabis-derived product] types was detected, including myriad product forms (eg, [cannabis combustible products], concentrates, [cannabis electronic delivery systems]) and ways to consume them (eg, epidermal agents, cooking ingredients for cannabis-infused foods, scented candles),” says the new paper, published on Monday in the journal JAMA Network Open. “Additionally, approximately 2 of every 5 products included at least 1 characterizing flavor, with 260 total unique flavors detected including those that appeal to youths such as berry, candy, chocolate, and alcohol.”
“These product features are concerning,” authors added, “as numerous studies have shown that other addictive flavored products (eg, e-cigarettes) can lead to increased uptake by youths, and surveys of adolescent cannabis users have found that the use of flavored cannabis is common.”
One caveat to the team’s conclusions, however, is that “flavored” doesn’t necessarily mean the product has any added flavorings. In other words, it seems a pre-rolled joint that’s described as having citrusy or herbaceous tasting notes would likely have been classified as a flavored product, even if it contained nothing besides cured cannabis flower.
“The three most common flavors,” the study says, “were lemon (22 106 [8.92%]), cake, (19 463 [7.86%]), and strawberry (13 961 [5.63%]).”
Asked by Marijuana Moment whether the analysis might be identifying certain strain names—such as Lemon Haze or Wedding Cake—as flavored products, c0-author Tim Mackey, a UC San Diego professor and director of the Global Health Policy and Data Institute said the team focused on words specifically in product descriptions.
“We included products as flavored if they represented in their product description a flavor characterization,” he wrote in an email, noting that the analysis was limited by what information was posted on Weedmaps. “If a product only had a flavor characterization in the name, we would not have reviewed unless a flavor characterization was also in the description.”
Mackey and other study authors did not respond to separate questions about whether the methodology might overestimate the portion of products that add flavorings to cannabis products. It seems, for example, that a joint rolled with the cannabis strain Dutch Treat would be considered a flavored product because the strain is described as having “citrusy aromas and sugary sweet flavors.”
Mackey did clarify that “if the product simply stated it was a cannabis product, we would not constitute that as having a characterizing flavor.” He did not immediately respond to a follow-up email, however, asking what qualifies as simply “a cannabis product.”
“We think this points to the need for more adequate disclosure of product information to consumers and for purposes of knowing what cannabis product is being sold,” he told Marijuana Moment. “Whether that is a regulatory consideration for state governments that have legalized adult-use/recreational or medical marijuana, is up to the states to decide in the context of their own public policy priorities (e.g., whether flavored products are attractive to certain demographics.)”
Weedmaps, for its part, declined to comment on the study.
Tasting notes are also used to describe other consumable products, such as wine or coffee, despite those products often having no added flavorings. Similarly, many of the terpenes produced by the cannabis plant have their own distinct aromas, such as citrus, pine or floral notes.
Among the other findings in the JAMA paper are that roughly two thirds of marijuana products on Weedmaps have delta-9 THC as the dominant cannabinoid—the component traditionally associated with producing a cannabis high.
In terms of form factors, “multisystem” products—including both cured flower and cannabis concentrates—were the most popular, making up 41.04 percent of listings. After that were respiratory products (36.98 percent), ingestible products like beverages and edibles (19.75 percent), epidermal (1.89 percent) and oral (0.33 percent).
Authors said that when they conducted the study, from September though November of last year, Weedmaps “did not actively use age verification requirements on its main launch page,” although they added that individual vendor pages may have been age-gated.
“Cannabis product and dispensary listings on the platform also enable consumers to connect via social media,” the study says, “further increasing potential market exposure among youths that can lead to increased cannabis use.”
Future studies, it adds, “should proactively monitor the growing physical and virtual cannabis marketplace for aggressive marketing tactics, further product innovation, and new and emerging product that may shift [cannabis-derived product] use patterns toward increased harm.”
As regulators attempt to shape newly legal cannabis markets, some have sought to limit the influence of online advertisers. New York, for example, banned price-advertising platforms similar to Weedmaps, but a state judge struck down that ban after another e-commerce website challenged it.
Meanwhile, a federal report released this summer shows that teen marijuana use in the United States is now lower than it was before individual states began legalizing the drug for adults. The percentage of young people aged 12 to 17 who’ve ever tried marijuana, for example, dropped 18 percent from 2014—when the first legal recreational cannabis sales in the U.S. launched—to 2023. Past-year and past-month rates among young people also declined during that time period.
Multiple other studies have debunked the idea that the reform broadly increases youth use, with most finding that consumption trends are either stable or decrease after the reform is implemented. Use by heavy users may increase, however.
For example, a research letter published by the Journal of the American Medical Association (JAMA) in April said there’s no evidence that states’ adoption of laws to legalize and regulate marijuana for adults have led to an increase in youth use of cannabis.
Another JAMA-published study earlier that month that similarly found that neither legalization nor the opening of retail stores led to increases in youth cannabis use.
Data from a recent Washington State survey of adolescent and teenage students found overall declines in both lifetime and past-30-day marijuana use since legalizations, with striking drops in recent years that held steady through 2023. The results also indicate that perceived ease of access to cannabis among underage students has generally fallen since the state enacted legalization for adults in 2012.
Rates of youth marijuana use in Colorado, meanwhile, declined slightly in 2023—remaining significantly lower than before legalization. That’s according to results of the biannual Healthy Kids Colorado Survey released this month that found that past-30-day use of cannabis among high schoolers was at 12.8 percent in 2023, a dip from the 13.3 percent reported in 2021.
A separate study late last year also found that Canadian high-school students reported it was more difficult to access marijuana since the government legalized the drug nationwide in 2019. The prevalence of current cannabis use also fell during the study period, from 12.7 percent in 2018–19 to 7.5 percent in 2020–21, even as retail sales of marijuana expanded across the country.
In December, meanwhile, a U.S. health official said that teen marijuana use has not increased “even as state legalization has proliferated across the country.”
“There have been no substantial increases at all,” said Marsha Lopez, chief of NIDA’s epidemiological research branch. “In fact, they have not reported an increase in perceived availability either, which is kind of interesting.”
Another earlier analysis from CDC found that rates of current and lifetime cannabis use among high school students have continued to drop amid the legalization movement.
A study of high school students in Massachusetts that was published last November found that youth in that state were no more likely to use marijuana after legalization, though more students perceived their parents as cannabis consumers after the policy change.
A separate NIDA-funded study published in the American Journal of Preventive Medicine in 2022 also found that state-level cannabis legalization was not associated with increased youth use. The study demonstrated that “youth who spent more of their adolescence under legalization were no more or less likely to have used cannabis at age 15 years than adolescents who spent little or no time under legalization.”
Yet another 2022 study from Michigan State University researchers, published in the journal PLOS One, found that “cannabis retail sales might be followed by the increased occurrence of cannabis onsets for older adults” in legal states, “but not for underage persons who cannot buy cannabis products in a retail outlet.”
The trends were observed despite adult use of marijuana and certain psychedelics reaching “historic highs” in 2022, according to separate data released last year.
Photo courtesy of Mike Latimer.
