Preliminary research presented at the American Heart Association's 2026 meeting found no steady dose-response, and the authors stress what the survey cannot prove

Adults who used cannabis on just one to five days a month had 40 percent higher odds of reporting a past stroke than people who did not use it, according to preliminary research presented at the American Heart Association's Scientific Sessions 2026 in Chicago. Heavy users, who used it 16 to 30 days a month, had 25 percent higher odds.

That pattern surprised the researchers — they expected risk to climb steadily with more frequent use. The findings come from 442,906 adults in a federal health survey, according to the American Heart Association's news release.

The study adds to a push for routine questions about cannabis in heart and stroke checkups. About 1 in 6 U.S. adults reported past-month use in 2024, up from 1 in 10 in 2020, making this a common health question for millions of households.

Inside the 442,906-Person Survey

The team analyzed 2020 to 2024 data from the Behavioral Risk Factor Surveillance System, a telephone survey of U.S. adults. Only states that used the optional marijuana questions were included, between 16 and 24 per year, so not every state is represented. Participants averaged 49 years old, and about half were women.

Compared with nonusers, low-frequency users had 23 percent higher odds of reporting any cardiovascular disease, and heavy users had 21 percent higher odds. Stroke showed the strongest link. Researchers found no statistically significant association between cannabis frequency and heart attack, according to ScienceDaily's summary.

Use was grouped by the number of days in the past month: low frequency meant 1 to 5 days, moderate meant 6 to 15 days, and heavy meant 16 to 30 days. The AHA release did not report results for the moderate group, which makes it harder to judge the overall shape of the relationship.

The analysis adjusted for age, sex, tobacco smoking, alcohol, obesity, diabetes, kidney disease, physical inactivity, education and income. The link appeared regardless of whether people smoked, vaped or ate cannabis.

Lead author Aabha Divya, an assistant professor at Tulane University School of Medicine, said the link with cardiovascular disease "was not straightforward" in the AHA release.

The Puzzle of the Occasional User

Researchers did not explain the higher stroke odds among occasional users, and the survey cannot answer that question. Several possibilities exist — former heavy users who cut back after a health scare could appear in the low-frequency group, and some people who quit after a diagnosis may sit in the no-use group.

The result also differs from an earlier analysis of 2016 to 2020 survey data, published in 2024 in the Journal of the American Heart Association, which found risk rising with more days of use, AJMC reported. That earlier study reported that daily use was tied to 42 percent higher odds of stroke and 25 percent higher odds of heart attack. Conflicting patterns across two analyses of the same national survey are a signal to wait for stronger study designs, such as long-term studies that follow users over time, before drawing firm conclusions.

There is also a reporting gap. HCPLive noted that the conference abstract lists 577,057 adults, while the AHA release cites 442,906, a difference the release does not explain.

Self-Reported Data, Unknown Potency and Tobacco Overlap

The limits are significant and should be stated plainly. This is a cross-sectional survey from a single point in time, so it cannot show whether cannabis use came before or after a stroke, and it cannot prove that cannabis causes heart disease or stroke.

Both cannabis use and medical history were self-reported, which invites memory errors. The survey had no information on THC potency or the amount consumed, a major gap as products have grown stronger. Tobacco smoking was adjusted for, but heavy overlap between cannabis and tobacco users can still blur results.

The findings are an abstract presented at a scientific meeting and have not been published in a peer-reviewed journal. The AHA notes that meeting abstracts are selected by independent review panels but are not peer-reviewed, and it treats such results as preliminary until a full manuscript appears.

Heart Experts Want Cannabis in the Checkup

Robert L. Page II, a University of Colorado pharmacy professor who chaired the AHA's 2020 scientific statement on cannabis and was not involved in the study, said tobacco use among cannabis users clouds the picture. Still, he said clinicians should be "addressing cannabis use with their patients as they would tobacco use," according to Reuters reporting carried by ARY News.

For readers, that means it is reasonable to tell a doctor honestly about cannabis use, including occasional use and edibles, at a routine visit. People with high blood pressure, diabetes, prior heart disease, or a smoking history have the most reason to raise it.

Sudden face drooping, arm weakness, slurred speech, or a severe headache requires an immediate 911 call, whatever the cause. This study does not change emergency guidance, and it does not show that occasional users should switch to heavier use.

A peer-reviewed paper, if published, would allow closer review of the methods and the sample-size discrepancy.

What Readers Want to Know

What did the study find? Cannabis users had higher odds of reporting stroke: 40 percent higher for occasional users and 25 percent higher for heavy users.

Does cannabis cause strokes? The survey cannot prove that. It captured one point in time and relied on self-reported data.

Why did occasional users show higher odds? Researchers did not explain it. Former heavy users who cut back could be one factor, but that is unconfirmed.

Was heart attack risk higher? No statistically significant link with heart attack was found.

Should I tell my doctor about cannabis use? Yes. AHA experts say clinicians should ask about cannabis much as they ask about tobacco.

Is this study final? No. It is a conference abstract that has not yet been published in a peer-reviewed journal.