Potent cannabis products linked to psychosis, mental health risks

Cannabis products containing high concentrations of delta-9-tetrahydrocannabinol (THC) — the main psychoactive compound in marijuana — were linked to psychosis, schizophrenia, and addiction, a systematic review found.

Across 99 studies examining the effects of high-concentration THC products on mental health outcomes, 70% of the nontherapeutic studies (i.e., those not attempting to treat a medical condition or symptom) showed an unfavorable association with psychosis or schizophrenia and 75% found a relationship to cannabis use disorder (CUD).

Anxiety and depression appeared to be worsened in 53% and 41% of such studies, respectively, with even higher rates when studying healthy individuals specifically, reported Jonathan M. Samet, MD, of the Colorado School of Public Health in Aurora, and co-authors, writing in the Annals of Internal Medicine.

The findings complement those of prior systematic reviews in “reinforcing the conclusion that higher THC concentrations increase the risk for adverse mental health outcomes, particularly for psychosis or schizophrenia (in both healthy persons and those with preexisting psychosis) and CUD (among healthy persons),” Samet and team wrote.

Particularly concerning are effects at high doses and among healthy users, agreed Patricia Cavazos-Rehg, PhD, of the Washington University School of Medicine in St. Louis, in an editorial accompanying the study.

“Unfortunately, although some studies signaled therapeutic benefits for anxiety and depression, occurring particularly among persons with cancer and neurologic conditions, many studies detected that use of high concentration cannabis products was associated with adverse outcomes, especially among healthy persons,” said Cavazos-Rehg.

In studies looking at therapeutic use of potent THC products such as for treatment of cancer symptoms, the findings were mixed. Close to half of the studies identified benefits for anxiety and depression (47% and 48%, respectively) compared with a lower-concentration THC product or no THC, while another 24% and 30% found unfavorable associations.

While more research is needed, “physicians and other healthcare professionals should educate their patients who use marijuana about the potential harms related to the use of high-concentration cannabis products and that they may not be as safe as traditional plant-based marijuana products,” Cavazos-Rehg concluded.

Cannabis flower THC concentrations have increased over time, from the typical 2-4% in the 1960s to 1980s to now an average 20% and even higher in states where use is legal. Vaping devices can deliver THC concentrations as high as 70-90%, Samet’s group noted.

In 1996, California became the first state to legalize medical cannabis. Thirty-nine states and the District of Columbia have legalized medical cannabis, as of Feb. 1, 2025, and 24 states and the District of Columbia have legalized recreational cannabis.

As Colorado was one of the first to legalize both medical and recreational use, its state legislature called for a review of the literature into the health effects of “high-potency marijuana and THC concentrates.”

For that purpose, Samet and team identified 99 studies, involving 221,097 participants, focused on high-concentration THC products exceeding 5 mg or 10% THC per serving or products described as “high-potency concentrate,” “shatter,” or “dab,” even when specific concentration levels were not provided. Among these studies, 42% were randomized controlled trials, 47% were observational, and 11% leveraged other interventional study designs.

Studies were conducted from 1977 to 2023 in 15 countries. About one-third focused on healthy people, around one-fifth included people whose primary health condition was a mental health disorder, and another third recruited participants with other conditions.

Most of the studies involved participants ages 18 to 59, while just 6% included adults 60 or older and 4% included adolescents. Participants’ race and ethnicity was reported in about a third of the studies. Tracked participants were largely white.