Cannabis Use Was Linked to Less Severe COVID-19 in One Hospital-Data Study
A 2024 study of U.S. hospital records found that patients with a documented history of cannabis use had less severe COVID-19 outcomes than comparable patients without such a record. The findings are noteworthy, but they show an association—not that cannabis prevents or treats COVID-19.
Published in Cannabis and Cannabinoid Research, the Northwell Health study analyzed 322,214 adult COVID-19 hospitalizations from the National Inpatient Sample. Of those encounters, 2,603 involved patients identified through diagnostic coding as cannabis users. Researchers matched cannabis users with non-users on age, race, sex and 17 comorbidities to reduce differences between the groups.
Before and after matching, the cannabis-use group had lower rates of a composite measure of severe COVID-19, which included acute respiratory failure, intubation, acute respiratory distress syndrome and severe sepsis with multiorgan failure. The group also had lower in-hospital mortality and shorter hospital stays.
Those results are consistent with an earlier retrospective study of 1,831 hospitalized patients, which linked current cannabis use with lower rates of intensive-care admission and mechanical ventilation. However, that study did not find a statistically significant improvement in overall survival after adjustment.
The newer analysis has important limitations. It used hospital billing and diagnostic codes rather than direct measurements of cannabis exposure. The database could not establish how much cannabis patients used, when they used it, whether they inhaled or ingested it, or whether use continued during hospitalization. It also did not identify cannabis as a treatment administered for COVID-19. Because the study was observational, unmeasured differences between users and non-users could explain some or all of the apparent benefit.
Other research has produced different results. A 2024 cohort study published in JAMA Network Open found that cannabis use was associated with higher risks of COVID-19 hospitalization and intensive-care admission, although it was not associated with higher all-cause mortality after adjustment for smoking, vaccination, comorbidities and other factors. Taken together, the evidence remains mixed rather than conclusive.
Laboratory studies provide one possible biological explanation for a protective association, but they do not establish a clinical benefit. For example, researchers at Oregon State University and Oregon Health & Science University reported that the cannabinoid acids CBGA and CBDA could bind to the SARS-CoV-2 spike protein and inhibit viral entry in cell-based experiments. Their university research summary makes clear that the work involved laboratory models—not people receiving cannabis products as COVID-19 therapy.
These findings should not be interpreted as evidence that smoking marijuana, taking CBD, or using another cannabis product can prevent infection, reduce the risk of severe disease, or treat long COVID. The U.S. Food and Drug Administration has warned companies against marketing unapproved cannabis-derived products with claims that they prevent or treat COVID-19. Clinical evidence for using cannabis or cannabinoids to manage COVID-19 or its post-acute symptoms remains insufficient.
The Northwell analysis does justify further research, particularly prospective studies that measure cannabis dose, product composition, route of administration, timing and clinical outcomes. Until such studies are conducted, the most accurate conclusion is that cannabis use was associated with better outcomes in one large hospital-record analysis—not that cannabis itself caused those outcomes or should be added to COVID-19 treatment.
