Cannabis Use Tied to Higher Cardiovascular Odds, With the Strongest Signal for Stroke
Cannabis use was associated with higher odds of cardiovascular disease in a large U.S. survey analysis, with the clearest signal appearing for stroke. The association did not move in a simple dose-response pattern. People reporting relatively infrequent use and those reporting heavy use both had higher odds of cardiovascular disease than people reporting no recent cannabis use, a result that complicates the assumption that lower exposure is necessarily safer from a vascular standpoint.
The analysis drew from nearly half a million adults surveyed through the Behavioral Risk Factor Surveillance System between 2020 and 2024, limited to jurisdictions that included the optional cannabis module. Participants were asked about cannabis use over the prior month, including smoking, vaping, and ingesting gummies or other cannabis-containing foods. They also reported whether a clinician had ever told them they had coronary heart disease, a heart attack, or stroke. After adjustment for a wide range of cardiovascular risk factors, including tobacco smoking, alcohol use, obesity, diabetes, chronic kidney disease, physical inactivity, education, and income, cannabis use remained associated with adverse cardiovascular history.
Compared with no cannabis use, low-frequency use, defined as 1 to 5 days per month, was associated with 23% higher odds of reporting cardiovascular disease. Heavy use, defined as 16 to 30 days per month, was associated with 21% higher odds. The stroke signal was larger: low-frequency use was associated with 40% higher odds of stroke, and heavy use with 25% higher odds. No statistically significant association emerged between cannabis use frequency and heart attack.
Stroke carried the sharpest excess risk.
Why the Stroke Signal May Matter More Than the Overall Cardiovascular Pattern
The biology behind cannabis-related vascular harm remains under active discussion, but several pathways are plausible. THC and other cannabis compounds may affect heart rate, blood pressure, endothelial function, platelet activity, and autonomic balance. In practical terms, those changes can alter vascular tone and may influence clot formation or cerebral perfusion, mechanisms that fit more easily with stroke risk than with a uniform increase across all cardiovascular outcomes. The fact that the association was strongest for stroke, and absent for heart attack in this analysis, suggests that cannabis exposure may interact with cerebrovascular physiology in a way that deserves more careful study.
Aabha Divya, M.B.B.S., M.Ch., an assistant professor in the division of cardiothoracic surgery at Tulane University School of Medicine, said the findings were unexpected because the team anticipated a more gradual rise in cardiovascular risk as use increased. Instead, both low-use and heavy-use groups showed elevated odds of disease history. That pattern raises a familiar epidemiologic problem: frequency alone may not capture the exposure that matters most. Route of administration, THC content, potency, and cumulative dose all remain unknown in this dataset, and those variables could shape vascular risk in ways the survey could not measure.
The study also lands in a setting where cannabis exposure is becoming more common. Divya pointed to prior national survey data showing that 1 in 6 U.S. adults reported cannabis use in the last month in 2024, up from 1 in 10 in 2020. As prevalence rises, even modest excess risk could translate into a larger public health burden, especially if clinicians do not ask about use consistently during stroke and heart disease prevention visits.
What the BRFSS Data Can and Cannot Prove
The cohort included 442,906 U.S. adults with an average age of 49 years and approximately half women. Annual participation in the marijuana module ranged from 16 to 24 states, so the analysis did not represent every U.S. jurisdiction. Exposure and outcome were both self-reported, leaving room for recall error and misclassification. The survey also captured only a single point in time, which means the analysis can show association, not causation.
Reverse causation remains a real possibility. Some participants may have reduced or stopped cannabis after a cardiovascular diagnosis, which would push former users into the no-recent-use group. Without data on potency or total amount consumed, the analysis also cannot distinguish a person who used a low-THC product intermittently from someone who used high-potency products repeatedly. Those gaps matter because a smoked product, an edible, and a vape exposure may not carry the same cardiovascular consequences, even if they were all counted under the umbrella of cannabis use.
Robert L. Page II, Pharm.D., M.S.P.H., FAHA, who chaired the writing group for the American Heart Association’s 2020 cannabis scientific statement, framed the finding in a broader clinical context. Tobacco remains a major confounder, since many people who use cannabis also smoke cigarettes. Even so, Page described this as the second observational study suggesting cannabis exposure, without tobacco smoking, may be a potential risk factor for cardiovascular disease. His view was direct: clinicians should address cannabis use in the same conversations used for tobacco when assessing cardiovascular risk.
Counseling Patients Now Means Asking About Cannabis, Not Only Cigarettes
For bedside practice, the clearest implication is not to wait for perfect mechanistic certainty before asking the question. Cannabis use now belongs in routine cardiovascular history-taking alongside tobacco, alcohol, blood pressure, lipids, diabetes, and kidney disease. Patients may not volunteer the information, and many may assume that a “natural” product carries less vascular risk than other substances. The present analysis does not support that assumption.
The findings are preliminary because they come from an abstract scheduled for presentation at American Heart Association Scientific Sessions 2026, and abstracts are not peer-reviewed in the same way as a full manuscript. Even so, the signal is clinically relevant because it points toward a potentially modifiable exposure in a growing user population. The strongest association centered on stroke history, and the absence of a significant heart attack association should not be read as reassurance; it may simply reflect the limits of self-reported survey data, exposure classification, or the biology of the outcome itself.
For prevention counseling, the practical message is straightforward: cannabis use should be discussed explicitly, documented carefully, and weighed as part of the broader vascular risk profile. In patients with stroke risk factors, that conversation may deserve the same urgency as counseling about tobacco or uncontrolled hypertension, particularly when the product, frequency, and potency are unknown. Future peer-reviewed work will need to define whether route of use, THC exposure, and cumulative dose explain why cerebrovascular risk appears to rise even when heart attack risk does not.