People Hospitalized for Cannabis Use Disorder Had Nearly 3x the Death Risk Over 5 Years | Research

The takeaway

In a population-based Ontario study, individuals with hospital-based care for cannabis use disorder were 2.79 times more likely to die within five years than matched general population members, with especially elevated risks for suicide and trauma.

Emergency clinicians treating CUD patients, addiction researchers, public health policymakers evaluating cannabis harm.

3.5% of CUD patients died within 5 years vs 0.6% of matched controls (aHR 2.79)

What the researchers found

Within 5 years of incident hospital-based CUD care, 3.5% of individuals died compared to 0.6% of matched general population members. After adjusting for comorbid conditions, CUD patients had 2.79 times higher death risk overall, with particularly elevated risks for suicide (9.7x), trauma (4.55x), opioid poisoning (5.03x), other drug poisonings (4.56x), and lung cancer (3.81x).

Why it matters

This is one of the largest studies to quantify mortality risk associated with cannabis use disorder severe enough to require emergency or hospital care. The findings suggest CUD is a marker for serious health vulnerability, even after accounting for other mental health and substance use conditions.

The numbers in context

11,622,571 individuals studied; 106,994 with incident CUD; 3,770 CUD patients (3.5%) died within 5 years vs 3,770 (0.6%) matched controls; adjusted HR 2.79; suicide aHR 9.70; trauma aHR 4.55; opioid poisoning aHR 5.03; lung cancer aHR 3.81; median follow-up 5 years.

How the study worked

Population-based retrospective cohort study of all Ontario residents aged 15-105 (n=11.6 million) from 2006-2021, comparing mortality between 106,994 individuals with incident hospital-based CUD care and age/sex-matched general population members using cause-specific hazard models adjusted for comorbidities.

What this study cannot tell us

Only captures individuals severe enough to present to emergency departments or hospitals, not all people with CUD. Cannot establish whether cannabis use itself causes the elevated mortality or whether CUD is a marker for other risk factors. Comorbid substance use may confound results despite statistical adjustment.

How to read the evidence

Strong: Very large population-based cohort (11.6 million) published in JAMA Network Open with comprehensive adjustment for comorbidities and long follow-up.

When this study was published

Published in 2025 with data from 2006-2021.

The bigger picture

While CUD carried lower mortality risk than alcohol (1.30x), stimulant (1.69x), or opioid (2.19x) use disorders, the nearly threefold elevation over the general population challenges the perception of cannabis as relatively harmless. This study adds to growing evidence that severe CUD carries significant health consequences.

Questions still open

How much of the elevated mortality is directly attributable to cannabis versus co-occurring conditions? Would these mortality patterns differ in populations with legal cannabis access? What interventions could reduce mortality risk in people presenting with CUD?

Common questions

Does this mean cannabis use causes death?

The study found an association between hospital-level CUD and elevated mortality risk, but it cannot prove direct causation. People with CUD severe enough for hospital care may have other risk factors that contribute to mortality.

How does CUD mortality compare to other substance use disorders?

CUD carried lower mortality risk than alcohol (1.30x higher than CUD), stimulant (1.69x), and opioid (2.19x) use disorders, but was still nearly three times higher than the general population.

Read the original research

Cannabis Use Disorder Emergency Department Visits and Hospitalizations and 5-Year Mortality.

JAMA network open, 8(2), e2457852

View on PubMed ↗Publisher’s paper ↗

Citation

Myran, Daniel T; Pugliese, Michael; McDonald, André J; Xiao, Jennifer; Fischer, Benedikt; Finkelstein, Yaron; Tanuseputro, Peter; Firth, Joseph; Pakpour, Amir; Hsu, Chih-Wei; Chang, Wing-Chung; Solmi, Marco. (2025). Cannabis Use Disorder Emergency Department Visits and Hospitalizations and 5-Year Mortality.. JAMA network open, 8(2), e2457852. https://doi.org/10.1001/jamanetworkopen.2024.57852