A plain-language synthesis of peer-reviewed research comparing the health effects of alcohol and marijuana consumption.
By Ashley Miami with help from AI. You're welcome.
Research is unambiguous that alcohol carries a substantially higher overall mortality burden than cannabis. But cannabis is not without risks — and those risks are evolving as THC potency has risen dramatically in the legal market. Here's how the two compare across major health categories.
This is perhaps the starkest difference between the two substances. The gap in direct lethality is enormous.
The WHO's 2024 report confirmed 2.6 million annual deaths attributable to alcohol — accounting for roughly 4.7% of all deaths worldwide. In the United States alone, CDC data puts alcohol-related deaths at approximately 88,000 per year, with binge drinking accounting for roughly half.
Lethal dose comparison: Alcohol is one of the most toxic common drugs — consuming just 10 times the dose needed for intoxication can be fatal. Cannabis, by contrast, would theoretically require a dose thousands of times greater than an intoxicating amount to cause death. No such case has ever been documented.
It is worth noting that while cannabis carries no overdose lethality, heavy cannabis use is not without long-term health consequences, and the category of "cannabis-related deaths" is likely to grow as cardiovascular research matures (see the Heart section).
Alcohol's carcinogenicity is one of the most well-established findings in modern epidemiology — and remains deeply underappreciated by the public. Cannabis's cancer profile is more uncertain.
Public awareness gap: A 2025 Annenberg survey found that very few Americans are aware that alcohol causes cancer. The Surgeon General's 2025 advisory called this a major public health failure. The science has been clear for decades; the messaging has not.
Both substances affect the brain — but in different ways, with different risk profiles. Neither is benign, and the picture for cannabis has become more complicated as THC concentrations have risen dramatically.
Alcohol is a central nervous system depressant with well-established neurotoxic effects at high doses. Heavy drinkers show measurable loss of gray matter, impaired executive function, and disrupted memory formation. Recent research found moderate-to-heavy drinkers are at a 57% increased risk of developing dementia. Chronic alcohol use also depletes thiamine (vitamin B1), which can cause Wernicke-Korsakoff syndrome — a severe form of irreversible brain damage.
Cannabis's effects on the brain are more complex and heavily dependent on three factors: age of first use, frequency of use, and increasingly, THC potency.
The THC potency problem (2025): A sweeping review of nearly 100 studies published in September 2025 found strong links between high-potency cannabis and psychosis, schizophrenia, and cannabis use disorder. THC content has increased approximately 5-fold since the 2000s — from ~4% to ~20% in most legal dried cannabis. "Cannabis from the 2000s is not the same as in 2025," noted co-author Dr. Nicholas Fabiano of the University of Ottawa. This is considered a major driver of the increasing psychosis link.
The bottom line on brain effects: alcohol's neurotoxicity is better established at the population level, while cannabis's mental health risks — particularly psychosis — are real but concentrated in specific high-risk groups (young, heavy, high-THC users). Neither is safe for developing brains.
This category is more nuanced for both substances, and features some of the most recent and surprising findings in the literature.
For decades, moderate alcohol consumption was associated with a reduced risk of cardiovascular disease — the so-called "J-curve." This finding is now heavily contested. Mendelian randomization studies suggest much of the apparent benefit reflects confounding (moderate drinkers tend to be healthier in general). Heavy drinking is clearly harmful: it causes alcoholic cardiomyopathy, atrial fibrillation, hypertension, and stroke. The 2024 National Academies review found some evidence of reduced all-cause mortality at moderate levels, but noted significant uncertainty.
A significant February 2024 study found that daily marijuana use raises stroke risk by 42% and heart attack risk by 25% — even with no prior heart disease history and no tobacco use. Cannabis has also been linked to cardiac arrhythmias, myocarditis, arterial spasms, and increased heart failure risk. Importantly, many of these studies involve smoked or vaped cannabis. Whether the cardiovascular risk extends equally to edibles — which avoid combustion and the associated acute heart rate spike — is not yet fully established.
Method of consumption matters: Much of cannabis's cardiovascular research involves smoking or vaping, which causes an acute increase in heart rate and blood pressure. The risk profile of edibles is likely substantially different, though long-term edible-specific data is limited.
Alcohol has minimal direct effect on lung health, except through aspiration during overdose. Cannabis, when smoked, is a different story.
Both substances can produce physical and psychological dependence, but alcohol's withdrawal syndrome is one of the few in medicine that can be directly fatal.
Alcohol dependence produces one of the most dangerous withdrawal syndromes of any substance — delirium tremens can include seizures and is potentially fatal without medical supervision. Cannabis withdrawal is real (irritability, sleep disruption, anxiety) but not medically dangerous. Alcohol's higher addiction rate (~15%) and the severity of its dependence syndrome make it considerably more problematic from a public health standpoint — though CUD, affecting roughly 1-in-10 users, should not be dismissed.
Addiction potential increases substantially with early onset of use, daily use patterns, high-potency products (cannabis), and co-occurring mental health conditions — for both substances.
As cannabis legalization expands and "California sober" culture grows, an important research question has emerged: can cannabis substitute for alcohol, and does it reduce drinking?
A landmark November 2025 randomized controlled trial from Brown University — the first of its kind — found that smoking cannabis led participants who were heavy drinkers to consume less alcohol in the short term. The effect was causal, not merely correlational. The study included 157 adults ages 21–44 who drink heavily and use cannabis at least twice a week.
Key nuance: The substitution effect may benefit heavy drinkers in the short term, but it is not a blanket endorsement of cannabis as a "healthier" alternative. The long-term cardiovascular, respiratory, and mental health implications of cannabis use are still being studied — and simultaneous use of both substances appears to increase risk beyond either alone.
Medical cannabis users in particular have shown reductions in alcohol consumption across several studies — suggesting that structured, intentional cannabis use may serve a harm-reduction role for some populations. This is distinct from recreational heavy use, where co-use tends to be associated with greater overall harm.
By nearly every major epidemiological measure — mortality, cancer causation, addiction severity, and organ damage — alcohol is more harmful to human health than cannabis at the population level. This is the scientific consensus, and it has been consistent across decades of research.
However, cannabis is not harmless. Its risks — particularly psychosis in young heavy users of high-potency products, cardiovascular effects from smoking, and respiratory effects — are real and growing in importance as legal high-THC products proliferate.
The most evidence-based summary: if you're going to use a substance recreationally, the research suggests cannabis carries a lower overall health burden than alcohol — with significant caveats around age, frequency, potency, and method of consumption. The safest option, of course, is neither.