Article: Laurel P. Gibson, et al., “Effects of cannabidiol in cannabis flower: Implications for harm reduction,” Addiction Biology, (2021).

Reviewed by: Stacey McKenna

As of July 2021, medicinal marijuana is legal and regulated in 37 states and Washington D.C. Eighteen of those states and Washington D.C. also permit the use and sale of recreational cannabis. The legal cannabis market abounds with products of varying potencies as growers and manufacturers select for specific cannabinoids and terpenes, the active components in the plant. Most producers label the relative percentages of the two most common cannabinoids—THC, which gives the drug its characteristic “high,” and cannabidiol (CBD), its non-intoxicating but potentially therapeutic cousin. This percentage based labeling gives consumers information that could potentially impact their purchase and use decisions.

Self-report and observational data support anecdotal accounts that cannabis produces both desired (i.e., euphoria) and undesired (i.e., paranoia) effects, and that CBD and THC content may indeed play an important role in those experiences. Thus, in response to the particular potency and cannabinoid profile of a given strain or product, individuals may adjust their use to meet personal preferences. However, the data on these issues is still emerging and findings are not consistent.

Despite cannabis’ growing popularity as the most commonly used (federally) illegal drug in the United States, research on the effects of cannabinoids such as THC and CBD, especially using state-regulated products, remains extremely limited. In a recent study published in Addiction Biology, Gibson et al. sought to fill gaps in the current evidence by comparing users’ subjective experiences and consumption behaviors across three distinct ratios of THC to CBD purchased in Colorado dispensaries.

The researchers recruited potential participants via social media and mailed flyers. Individuals were screened using a detailed set of inclusion criteria, including: age (21 to 70 years); cannabis use behaviors and preferences; use of non-cannabis substances; mental health; treatment status for mental health or substance use disorders; and pregnancy (neither pregnant nor attempting to become pregnant). The final sample consisted of 159 eligible individuals.

The study process consisted of a baseline appointment and a follow-up visit five days later. During the first visit, participants completed demographics and medical history questionnaires, and surveys about substance use experiences and mood. Upon completion of these tasks, they were randomly assigned to one of three groups based on distinct cannabis chemical profiles, or “chemovars:” THC dominant (24 percent THC and 1 percent CBD); THC+CBD (9 percent THC and 10 percent CBD); or CBD dominant (1 percent THC and 23 percent CBD).

In order to comply with cannabis’ federal Schedule I status, which prohibits researchers from providing state-regulated products in laboratory experiments, Gibson et al. used a “naturalistic at-home administration” approach in which participants purchased the assigned product themselves from a local dispensary. During the five days between the baseline and follow-up appointments, participants were asked to consume the study product at home according to their own preferences in order to become familiar with it.

After the “familiarization” period, two members of the research team brought a mobile pharmacology lab to the participants’ homes for appointment number two. Participants were asked to refrain from using any cannabis on the day prior to their second appointment. During this session, participants consumed their product at a rate and quantity of their choosing, and researchers tested blood concentrations of THC, CBD and other cannabinoids as well as participant-reported effects at three timepoints: pre-use, immediately after use and one hour following use. The team then analyzed the data to identify several relationships between variables, including:

Although there were only a handful of significant correlations between blood concentrations of cannabinoids and immediate subjective effects, Gibson et al. did find associations between product profile and subjective experience. Overall, compared to the CBD dominant group, participants in the THC dominant and THC+CBD conditions reported more “desirable” effects, such as elation and enjoyment. However, these effects did drop off over time.

Contrastingly, those in the THC dominant condition were more likely to report undesired effects such as anxiety and paranoia compared to individuals in either the CBD dominant or THC+CBD, with little time-related change in these undesired effects reported by any group. Participants across groups averaged the same consumption quantity, suggesting that they did not adjust their use to increase THC consumption.

Based on these findings, Gibson et al. conclude that CBD may play a role in reducing potential harms of THC exposure without diminishing desired effects, such as euphoria, perceived high or self-reported enjoyment of the drug. They propose two potential pathways by which CBD has such an effect. The first is that CBD may enhance THC-related happiness, allowing consumers to attain the same desired state of consciousness with less THC, thereby reducing risk for paranoia or anxiety. The other potential way the researchers suggest CBD could mitigate harm is by directly reducing THC-related anxiety.

Methodologically, this study makes an important contribution to the literature, as it applies approaches usually reserved for the laboratory such as chemovar randomization and biometric measures to products that are available to consumers. Unfortunately, as the authors note, some limitations do remain. In particular, although the researchers were blinded to each participant’s condition, the participants themselves, by virtue of purchasing their own cannabis flower, were aware of the chemical makeup. This knowledge could have affected their self-reports of subjective effects, especially since the group consisted of experienced users.

A second weakness is that the researchers did not directly observe the consumption to confirm the quantity of flower consumed or whether it was consumed in a way that might indicate an effort to alter subjective effects (such as taking smaller or bigger hits or inhaling more deeply). Finally, because self-report data were solely gathered via standardized and closed-ended surveys, the researchers may have missed opportunities to fill in these gaps in knowledge by discussing use behaviors with participants.

Despite these limitations the findings have potential implications for both medicinal and recreational cannabis use. If CBD does in fact mitigate the harms associated with THC without undermining the desired effects, it may be warranted to advise individuals who enjoy the “high” or who derive THC-specific therapeutic benefits but are sensitive to paranoia and anxiety to seek out flower and other cannabis products with balanced chemical profiles. Yet, more research is necessary to make any clear recommendations.

Featured Publications