Cannabis appeared in 84% of teens’ substance-related emergency visits or hospitalizations, study finds

A national analysis of emergency department visits and hospitalizations linked to substance use found that cannabis was involved in most such acute-care encounters among adolescents and young adults, while alcohol was more common among older adults, according to a study published Wednesday, November 4, in The American Journal of Psychiatry.

The study was led by researchers at Washington University School of Medicine in St. Louis. They examined de-identified insurance data for more than 1.3 million people ages 16 to 64 who had at least one substance-related emergency care encounter between January 2016 and December 2023. The dataset included more than 5 million emergency visits and hospital admissions related to alcohol, cannabis, opioids, stimulants and other addictive substances.

Among patients ages 16 to 18 who had a substance-related emergency visit or hospitalization, 84% had at least one encounter involving cannabis, the highest share of any age group in the study. By contrast, 19% of adults ages 55 to 64 had at least one such encounter involving cannabis. Young adults ages 19 to 24 showed the fastest growth in the proportion of encounters involving cannabis, rising by about 1.8 percentage points per year over the study period. Among teens, that proportion increased by about 0.8 percentage points per year.

Alcohol followed a different pattern. Among adolescents in the study, 33% had an alcohol-related encounter. That share increased steadily with age, reaching 70% among adults ages 55 to 64. The sharpest increase in the entire analysis was for alcohol as the primary reason for hospitalization among adults ages 25 to 34, which rose by 3.6 percentage points per year.

Kevin Y. Xu, an assistant professor of psychiatry at WashU Medicine and the study’s senior author, said the findings point to clear age differences in the substances linked to acute harm. He said those differences could help doctors give patients more individualized advice about alcohol and cannabis use.

The researchers said the findings also add detail to a broader shift that has been visible in surveys, with cannabis use rising and alcohol use declining overall in some populations. Xu said the new analysis goes further by focusing on cases serious enough to lead to emergency treatment or hospitalization.

The study also pointed to signs of growing harm tied to cannabis among younger people, including increases in hospital admissions involving psychosis and severe vomiting. Xu noted that cannabis products available in recent years often contain higher levels of THC, the main psychoactive ingredient in cannabis, than products sold in the past. He said that raises concerns for adolescents and young adults, whose brains are still developing. The paper also cited prior research suggesting that heavier cannabis use can increase the risk of cannabis use disorder and, in some adolescents, may be associated with psychotic symptoms, though most people who use cannabis do not develop those problems.

At the same time, the authors said the cannabis findings need to be interpreted carefully. The study counted any emergency visit or hospitalization in which cannabis appeared anywhere in the diagnosis list. That means cannabis may not have been the main reason for the visit. A patient could have been treated after a car crash, during a psychiatric crisis or for another medical problem and still have cannabis listed among the diagnoses. Nearly half of cannabis-related encounters also involved at least one other substance.

That distinction mattered in one of the study’s more cautious findings. When researchers looked only at hospitalizations where cannabis was listed as the primary reason for admission, the trend among adolescents went down rather than up. The authors said that suggests cannabis is increasingly present in acute-care encounters involving younger people, but it may not be the sole cause of hospitalization in many of those cases.

The data came from large commercial insurance and Medicaid databases and were analyzed across six age groups: 16 to 18, 19 to 24, 25 to 34, 35 to 44, 45 to 54 and 55 to 64. Because the analysis relied on insurance claims, it reflects diagnosed encounters recorded in billing data rather than direct clinical interviews with patients. The authors did not analyze which diagnosis was the main reason for admission in every cannabis-related case, beyond the separate review of primary cannabis admissions.

Jacob T. Steinle, the study’s first author, began the research while he was a psychiatry resident at WashU Medicine and is now a clinical fellow at Mass General Brigham. He said more work is needed to explain why patterns of substance-related harm differ so much across the lifespan.

The findings may also carry implications beyond clinical care. For alcohol producers, retailers and public health groups tied to the beverage sector, the age split in emergency care could add pressure to evaluate alcohol-related harm more closely by demographic group and to adapt prevention efforts and responsible-drinking messages for different stages of adulthood, especially as alcohol remains a major driver of substance-related hospital admissions among adults even while cannabis draws growing attention among younger patients.

The research team said the results do not suggest that cannabis causes most substance-related emergencies on its own, nor do they show that alcohol harms are falling to low levels. Instead, the study shows that the substances appearing in acute-care settings differ sharply by age, with cannabis showing a larger presence in teen and young adult encounters and alcohol remaining a dominant factor in serious hospital care for older adults.