By most scientific measures, alcohol is the more dangerous substance. Large-scale harm-ranking studies conducted in the UK, Australia, and New Zealand have all placed alcohol at the top of the list when combining damage to the user with damage to others, ahead of heroin, methamphetamine, tobacco, and cannabis. But “worse” is not a single dimension. Cannabis carries its own serious risks, particularly for mental health and the developing brain, and the answer shifts depending on which organ system, which pattern of use, and which person you’re talking about.
How Scientists Actually Rank Drug Harms
The most influential attempt to compare drug harms head-to-head was a UK study published in The Lancet. Using panels of experts who scored each drug across sixteen criteria covering both personal and societal damage, alcohol received the highest overall harm score, driven largely by its outsized harm to others through violence, family disruption, and economic costs. Heroin and crack cocaine ranked second and third overall, while cannabis landed much further down the list.1The Lancet. Drug harms in the UK: a multicriteria decision analysis
The same methodology has since been replicated in other countries with consistent results. An Australian study found alcohol was the most harmful substance overall, followed by cigarettes and then crystal methamphetamine, with cannabis coming in fourth.2PubMed. The Australian drug harms ranking study A New Zealand replication scored alcohol at 88 out of 100 points for total harm, with high marks across half of all measured criteria, especially drug-related mortality and community damage.3PubMed Central. The New Zealand drug harms ranking study: A multi-criteria decision analysis These aren’t fringe findings. Across different expert panels in different countries and different years, alcohol consistently scores as the most damaging substance when you weigh the full picture.
Overdose Risk Is Not Even Close
One of the starkest differences between alcohol and cannabis is how easy it is to take a fatal dose. Researchers have quantified this using something called the margin of exposure, which compares a typical recreational dose to the amount that would kill you. Alcohol falls into the “high risk” category by this measure. Cannabis, by contrast, had a margin of exposure greater than 10,000, making a lethal overdose from the drug itself essentially impossible under normal circumstances.4PubMed Central. Comparative risk assessment of alcohol, tobacco, cannabis and other illicit drugs using the margin of exposure approach That doesn’t mean cannabis is harmless. It means the risk of dying acutely from too much of it is vanishingly small compared to alcohol, which kills thousands every year through poisoning alone.
This gap in acute toxicity matters because it shapes how each substance shows up in emergency departments. Alcohol poisoning is a well-known medical emergency that can suppress breathing and heart function. A person who consumes too much cannabis may experience severe anxiety, nausea, or temporary psychosis, but these episodes are rarely life-threatening in an otherwise healthy person.
What Each Substance Does to the Brain
Both alcohol and cannabis affect the brain, but the patterns of damage look quite different. A review focused on adolescent and young adult users found that alcohol is associated with widespread decreases in gray matter volume, thinner cortical tissue, slowed white matter development, and measurable deficits in learning, memory, and impulse control. Heavy cannabis use was linked to changes in subcortical volume and some disruption of executive functioning, but the overall conclusion was that co-use findings pointed to more pronounced effects from alcohol than from cannabis.5PubMed Central. Alcohol and Cannabis Use and the Developing Brain
A neuroimaging study of adolescent substance users reinforced this picture, finding that binge drinking was associated with more differences in brain activation patterns during verbal learning tasks than marijuana use was.6PubMed Central. Neural correlates of verbal learning in adolescent alcohol and marijuana users That said, the researchers also noted that alcohol and marijuana together seemed to interact in ways that altered prefrontal brain function beyond what either did alone, which is a recurring theme across many of these comparisons.
Driving Under the Influence
Alcohol’s effect on driving is one of the most thoroughly documented dangers in all of public health. Cannabis also impairs driving, but the two substances impair it differently, and the magnitude is not comparable. A systematic review found that cannabis impairs lane-keeping and slows reaction times at a level roughly similar to low blood alcohol concentrations. Alcohol, on the other hand, produces much larger effects at typical intoxication levels.7PubMed. The effects of cannabis and alcohol on driving performance and driver behaviour: a systematic review and meta-analysis
The real-world crash data tells a dramatic story. A case-control study of fatal motor vehicle crashes found that drivers who tested positive for alcohol alone had roughly sixteen times the odds of being involved in a fatal crash compared to sober drivers. Drivers positive for marijuana alone had about one and a half times the odds. Drivers positive for both had twenty-five times the odds.8PubMed Central. Interaction of marijuana and alcohol on fatal motor vehicle crash risk: a case-control study An older review noted that while epidemiological studies have been somewhat mixed on whether cannabis alone causes more accidents, there is unanimous agreement that alcohol does.9PubMed Central. The effect of cannabis compared with alcohol on driving
Cannabis-impaired drivers tend to drive more slowly and cautiously, possibly because they’re more aware of being impaired. Alcohol-impaired drivers tend to do the opposite, driving faster and taking more risks. Neither substance belongs behind the wheel, but the crash data makes clear that alcohol is in a different league of danger on the road.
Aggression and Violence
Alcohol’s connection to violence is well established and runs in the opposite direction from cannabis. In a controlled laboratory study, participants given alcohol showed increased aggressive responses compared to placebo, while participants given cannabis showed decreased aggression compared to placebo.10PubMed Central. Subjective aggression during alcohol and cannabis intoxication before and after aggression exposure This tracks with what most people intuitively observe: alcohol disinhibits, cannabis sedates.
In a daily diary study of college women, days of alcohol use were associated with increased odds of both psychological and physical aggression when anger was already high. Marijuana use days were associated with psychological aggression under similar conditions, but not physical aggression.11PubMed Central. The temporal relationship between alcohol, marijuana, angry affect, and dating violence perpetration: A daily diary study with female college students A study on intimate partner violence found that frequency of alcohol intoxication was associated with about a fifty percent increase in the odds of violence perpetration, while marijuana use was linked to bidirectional violence but at a much smaller magnitude.12PubMed Central. Problem Drinking and Marijuana Use as Risks for Unidirectional and Bidirectional Partner Violence
This is one of the dimensions where the harm-ranking studies give alcohol its highest marks for damage to others. Alcohol-fueled violence, domestic abuse, and public disorder represent a category of harm that cannabis simply doesn’t produce at comparable rates.
Where Cannabis Carries Unique Psychiatric Risks
The clearest area where cannabis poses a distinct danger that alcohol does not is mental health, specifically the risk of psychotic disorders. Longitudinal studies have found that regular cannabis use predicts an increased risk of schizophrenia and psychotic symptoms even after controlling for other drug use and personal characteristics.13PubMed Central. Cannabis use and the risk of developing a psychotic disorder
More recent and larger studies have sharpened this finding. A 2025 study of adolescent cannabis users found that cannabis use was associated with roughly double the risk of psychotic disorders and bipolar disorder, along with smaller but meaningful increases in the risk of depression and anxiety.14JAMA Health Forum. Adolescent Cannabis Use and Risk of Psychotic, Bipolar, Depressive, and Anxiety Disorders A Danish registry study found that people with cannabis use disorder had about four times the risk of psychotic bipolar disorder and nearly double the risk of depression compared to people without such a diagnosis.15JAMA Psychiatry. Cannabis Use Disorder and Subsequent Risk of Psychotic and Nonpsychotic Unipolar Depression and Bipolar Disorder
Alcohol, of course, can also worsen depression and anxiety, and chronic heavy drinking devastates mental health. But the specific psychosis connection is more prominent with cannabis, particularly in young people and in those with a family history of psychotic illness. Today’s high-potency cannabis products, which contain much more THC than what was available decades ago, may be amplifying this risk, though that remains an active area of research.
Cannabis withdrawal is also more common than many users expect. About half of regular users experience withdrawal symptoms when they stop abruptly, including anxiety, irritability, disturbed sleep, depressed mood, and appetite loss. Symptoms typically peak within two to six days and can linger for three weeks or more in heavy users.16PubMed Central. Clinical management of cannabis withdrawal Alcohol withdrawal, by comparison, can be medically dangerous and even fatal in severe cases, but the point is that cannabis dependence is a real phenomenon, not the myth it’s sometimes portrayed as.
Lungs, Liver, and Cancer
The two substances attack different organs. Alcohol is famously hard on the liver, progressing through fatty liver disease, hepatitis, cirrhosis, and eventually liver cancer with chronic heavy use. Here, cannabis has an unexpected twist: among alcohol users, those who also use cannabis show significantly lower odds of developing progressive stages of alcoholic liver disease, including cirrhosis and liver cancer.17PubMed. Cannabis use is associated with reduced prevalence of progressive stages of alcoholic liver disease Lab research has suggested a possible mechanism involving cannabinoid receptors that are upregulated in alcohol-related liver fibrosis.18PubMed Central. Cannabinoid receptor type I modulates alcohol-induced liver fibrosis This is intriguing but far from a recommendation to combine the two; it’s more a hint that the endocannabinoid system plays a complex role in liver damage.
Cannabis smoke, however, is not kind to the lungs. A state-of-the-art review found that cannabis smoking irritates the airways, causes chronic bronchitis symptoms, and impairs the lung’s immune defenses, increasing susceptibility to respiratory infections. It’s a risk factor for collapsed lungs and certain forms of lung inflammation, and a growing number of studies point to an independent association with lung cancer. The review found no link to chronic obstructive pulmonary disease, which may surprise some readers, but the overall conclusion was unequivocal: cannabis smoking harms the respiratory system.19PubMed. Effects of cannabis smoking on the respiratory system: A state-of-the-art review
For cancer more broadly, alcohol is classified as a Group 1 carcinogen, meaning there is sufficient evidence that it causes cancer in humans, with established links to cancers of the mouth, throat, esophagus, liver, breast, and colon. When alcohol and tobacco smoking combine, the risk to oral and laryngeal cancer climbs steeply because alcohol acts as a solvent that helps tobacco’s carcinogens penetrate tissue.20Journal of Korean Medical Science. The Combined Effects of Alcohol Consumption and Smoking on Cancer Risk by Exposure Level: A Systematic Review and Meta-Analysis Cannabis’s cancer profile is less settled. The emerging lung cancer signal is concerning, but the evidence base is much thinner than alcohol’s decades of clear epidemiological data.
Pregnancy
Neither substance is safe during pregnancy, but alcohol has a much more thoroughly documented record of harm. Heavy alcohol consumption during pregnancy is associated with reduced birth weight and length, and fetal alcohol spectrum disorders represent one of the most well-characterized patterns of prenatal harm in medicine.21Neurotoxicology and Teratology. A comparison of the effects of prenatal exposure to tobacco, alcohol, cannabis and caffeine on birth size and subsequent growth A review of the broader cognitive and behavioral evidence found that prenatal exposure to alcohol, tobacco, or cannabis was each related to some overlapping outcomes including attention problems, impulsivity, and reduced cognitive functioning in children.22PubMed. Maternal smoking, drinking or cannabis use during pregnancy and neurobehavioral and cognitive functioning in human offspring
Animal research has added detail to the cannabis picture. Prenatal THC exposure in rodents reduced body weight in offspring, delayed early sensorimotor development, and impaired motor coordination into adolescence. When alcohol and THC were combined prenatally, the effects were sometimes additive, particularly for locomotor activity in males.23PubMed Central. Prenatal Alcohol and THC E-Cigarette Exposure Effects on Motor Development The picture for cannabis in human pregnancies is muddied by confounders like poverty, tobacco co-use, and underreporting, but the precautionary consensus is clear: avoid both substances entirely during pregnancy.
What Happens When You Use Both Together
A significant percentage of cannabis users also drink, and the combination is not simply additive. The fatal crash data already showed this: drivers positive for both alcohol and marijuana had twenty-five times the odds of fatal crash involvement, much worse than either substance alone.8PubMed Central. Interaction of marijuana and alcohol on fatal motor vehicle crash risk: a case-control study The systematic review of driving studies confirmed that the combination produced greater impairments in lane-keeping and time spent outside the lane than either drug in isolation.7PubMed. The effects of cannabis and alcohol on driving performance and driver behaviour: a systematic review and meta-analysis
Interestingly, one older controlled study found that the combination did not produce additive effects on mood or behavior in a non-driving context, suggesting that the synergy may be more dangerous for tasks requiring motor coordination than for subjective experience.24PubMed. Separate and combined effects of marijuana and alcohol on mood, equilibrium and simulated driving But from a safety perspective, the combination is clearly more hazardous than either substance used alone.
Cannabis Contaminants Add a Wild Card
One risk unique to cannabis that doesn’t apply to commercially produced alcohol is contamination. In legal and especially in unregulated markets, cannabis products can contain microbes, heavy metals, and pesticides. The direct health effects of these contaminants in humans are poorly quantified, but they include potential for infection, carcinogenicity, and reproductive harm.25PubMed Central. Cannabis contaminants: sources, distribution, human toxicity and pharmacologic effects This is an area where the legal status of cannabis matters enormously. Regulated markets with testing requirements can reduce these exposures. Black-market products carry unknown and potentially serious contaminant risks that have nothing to do with THC itself. For immunocompromised people, fungal contamination of cannabis is a specific and documented concern, with links to invasive pulmonary aspergillosis.
Alcohol, by comparison, is one of the most tightly regulated consumer products in most countries. While contamination incidents do occur (particularly with illicit spirits), a person buying legal alcohol faces essentially no contamination risk. A person buying cannabis from an unregulated source cannot make the same assumption.