Researchers compared roughly 150,000 adults with a diagnosis of cannabis use disorder against a matched group of the same size without one. The first group developed lung or bronchus cancer at 3.87 times the rate of the second.
The study appeared in the journal Lung Cancer on April 22, 2026. It is a retrospective cohort analysis built on 20 years of electronic medical records from 67 U.S. health care organizations in the TriNetX US Collaborative Network, a database covering more than 118 million patients. The authors used propensity score matching to make the two groups comparable on demographic and clinical factors. The confidence interval for the headline figure runs from 3.43 to 4.38.
What the numbers show
The risk did not fade over time. Relative risk was 3.48 at one year after the diagnosis and 3.92 at five years. It also held across the main lung cancer types: 2.70 for small cell carcinoma, 2.54 for adenocarcinoma, 2.90 for squamous cell carcinoma and 2.75 for non-small cell lung cancer.
Cannabis use disorder is a clinical diagnosis for heavy, compulsive use that causes real impairment or distress. The study did not look at occasional users, and its authors do not claim the 3.87 figure applies to them.
Limits the authors acknowledge
The study cannot say how much marijuana drives the risk. It also could not fully separate out tobacco, vaping or other inhaled exposures that may overlap in people with the disorder. A causal mechanism remains unproven.
Earlier work was thinner. Smaller samples, shorter follow-up and confusion with tobacco use limited what could be concluded, and pooled analyses often found no clear link. A 2019 JAMA Network Open review pointed to low cumulative exposure, young participants and poor measurement of use. A 40-year Swedish cohort study found that heavy use starting in young adulthood carried about twice the lung cancer risk (HR 2.12) after adjusting for tobacco.
What researchers and clinicians said
Dr. Brooks Udelsman, a thoracic surgeon at Keck Medicine of USC and a co-author, told ScienceDaily that the link between tobacco and lung cancer is well established while the picture for cannabis is much less clear. Growing evidence, he said, suggests heavy use may carry significant risks. Scientists are still working out the biology. Marijuana smoke contains many of the same combustion carcinogens as tobacco smoke, including benzene, naphthalene and vinyl chlorides.
Healio ran a commentary on the study from two physicians, Robert Glatter and Peter Papadakos. They wrote that the evidence is strengthening but not yet definitive, and that clinicians should raise cannabis use with patients, especially adolescents and young adults.
The study authors called for research on dose and response and on the biological mechanisms, and said clinicians should weigh these risks when advising patients. Anyone worried about their own use or lung health should talk to a doctor.
Sources
- Heavy Marijuana Smoking Was Linked to a Nearly 4-Times Higher Lung Cancer Risk in the Largest U.S. Study of Its Kind
- Shifting evidence indicates greater risks for lung cancer with cannabis use disorder
