When people argue that cannabis should be legal, they often make a comparison to society’s favored drug, alcohol. Cannabis is much safer than alcohol, people say, so why should we allow alcohol and ban cannabis?
However, you can’t live in the modern world without realizing that some people will say all sorts of things regardless of whether there is evidence supporting them. So the question remains: is cannabis really safer than alcohol?
As part of our series on the best arguments for and against cannabis, we’re taking a deep-dive into this pro-weed talking point.
Summary: Is Cannabis Safer Than Alcohol?
So is marijuana safer than alcohol? Here are both sides of the debate…
Yes…
- Alcohol is responsible for over 140,000 deaths per year in the US. Cannabis is responsible for none.
- Cases taken by poison control centers reveal that cannabis wasn’t solely responsible for any deaths from 2013 to 2021. Alcohol caused deaths each year.
- While both alcohol and cannabis cause health conditions, alcohol causes more dangerous problems like liver cirrhosis.
- Direct comparisons of the risks by experts consistently find that cannabis is safer than alcohol.
- Cannabis is safer than alcohol regardless of how you use it, and edibles reduce the risk even further.
No…
- Comparing the distinct risks of cannabis and alcohol doesn’t lead to any solid conclusions. What’s worse: high blood pressure or shortness of breath? There is no clear answer.
- Although alcohol may be more dangerous than cannabis overall, if used in moderation, neither is particularly risky.
- More than alcohol, cannabis contributes to laziness and a lack of self-fulfillment. In other words, “pot makes you feel fine with being bored.”
The Basics: Deaths From Alcohol vs. Deaths From Cannabis
One of the most popular ways of making this point is the simplest one possible: counting deaths from alcohol and cannabis. In short, alcohol leads to tens of thousands of deaths a year – however you count them. Cannabis probably doesn’t lead to any.
Statistics on Alcohol-Related Deaths
The CDC’s overall figure for excessive alcohol use is 140,557 deaths per year in the U.S.. However, this also includes things like suicides or fall injuries where alcohol had a role to play. If you want to be as strict as reasonably possible about it, using the Alcohol-Related Disease Impact application, there are 33,658 deaths per year that are 100% attributable to alcohol.
Statistics on Cannabis-Related Deaths
For cannabis, it isn’t even possible to go onto something like the CDC’s website and find information about the number of cannabis deaths per year. This is because there are so few that such a page would only consist of a handful of cases where cannabis was probably not the only drug involved.
The National Poison Data System report for 2021 reports show that while cannabis was involved in 1.08% of all fatalities (page 29 of 264, Table 18), there were no single substance exposures that led to death. In contrast, alcohols were involved in 7.73% of all fatalities, and led to 28 single substance deaths. In fact, from 2013 to 2021, there were no cannabis related deaths in any year.
One study from England (Rock et. al., 2022) looked at all reported deaths from 1998 to early 2021 where cannabis was found in the person’s system. Out of 3,455 cases, just 4% (136 cases) involved cannabis without other substances. Of these, there were 14 cases where cannabis was called the underlying cause of death by the coroner. But in most of these, they used cannabis immediately before something happened, like heart failure or an injury.
There was just one case where cannabis was the sole listed cause of death, and this person had between 100 and 150 μg/L of THC in their blood. For reference, after a frequent user smokes a joint, their blood THC will reach a median of 34.4 μg/L (Desrosiers, et. al., 2014). The authors pointed out that it’s unclear how this much THC could kill. They conclude that “The risk of death due to direct cannabis toxicity is negligible.”
In short, it’s unlikely there are really any deaths from cannabis use on an average year. There might even be none at all.
The Health Risks of Alcohol vs. Cannabis
At this stage, someone might protest that death is really quite a serious end-point to focus on when we’re talking about risks. I could break my toe once a year and never die from it, but that doesn’t mean it’s safe or that I should do it on purpose. So in order to really answer the question, we have to take a lightning tour of the risks of both alcohol and cannabis.
The Health Risks of Alcohol
General sources on the risks of alcohol tend to simply list the possible risks, with the assumption that we’re talking about excessive use. These point to risks for the heart (cardiomyopathy and stroke, for instance), the liver (cirrhosis and alcoholic hepatitis), the pancreas (pancreatitis), cancer risk (head/neck, esophageal, liver, breast and colorectal) and other effects on things like mood or general infection risk. However, for comparison it’s useful to be able to attach figures to these claims.
One meta-analysis (Corrao et. al., 2004) combined the results from 156 studies to produce pooled risk estimates for 15 conditions (including injuries and violence) for three different levels of drinking. These levels were 25 g of ethanol a day (1.79 standard US drinks), 50 g a day (3.57 standard drinks) and 100 g a day (7.14 standard drinks). Taking the 50 g results, some key findings are:
- Over 7 times the risk for liver cirrhosis
- Over 3 times the risk for oral cavity cancer
- Approximately doubled risk for larynx and esophageal cancer
- Doubled risk for essential hypertension (high blood pressure)
- 82% increase in risk of hemorrhagic stroke
- 13% decrease in coronary heart disease risk
Note that this doesn’t cover all of the risks of alcohol. The figures are also just pooled estimates, not the definitive risk ratios for the conditions.
The Health Risks of Cannabis
General sources about the risks of cannabis focus more on mental health effects, addiction and the risks of cannabis to youth, with less focus on serious physical disease and death. This is ultimately because there isn’t as much evidence about cannabis as there is for alcohol, and conclusions rest on shakier ground.
Cannabis and alcohol are also very different substances that are used in different ways, so comparing them is hard. For example, cannabis use has no reason to particularly impact the liver, while alcohol’s effects are most noticeable there. Likewise, drinking won’t cause respiratory issues, but smoking marijuana will.
With that in mind, this focuses on the most notable risks of cannabis, but also includes figures for direct comparison with alcohol where possible. However, it is important to keep the limitations of such comparisons in mind.
- Moore et. al. (2007) looked at 35 studies on marijuana use and mental health outcomes. They found that users had a 41% increase in the risk of psychotic conditions, with around double the risk for the most frequent users.
- Martinasek et. al. (2016) looked at 48 studies on the respiratory effects of inhaled marijuana, and found some increased risks. In particular, they found an association between marijuana use and pneumothorax, bullous emphysema or COPD, and respiratory symptoms such as wheezing, coughing and shortness of breath.
- Martinasek et. al. also claim a link to lung cancer. However, looking at the cited studies and other research does not inspire confidence in this conclusion. For example, Callaghan, Allebeck and Sidorchuk (2013) found that heavy cannabis smoking doubles risk, but they only asked about tobacco use at the very start of the 40-year study. They didn’t even know if the weed smokers mixed with tobacco.
- Shah et al. (2021) found smoking cannabis increases your risk of heart attacks by 88% and stroke by 81%. However, this is only for those who smoked cannabis (with no risk from edibles), and who smoked on more than 10 days of the previous 30.
- Carvalho et. al. (2015) looked at head and neck cancer rates (including throat and mouth cancers) in cannabis users compared to non-users. They didn’t find a link with cannabis use. But they did group all users together, so it’s possible that heavy use would be different.
- Adejumo et. al. (2017) looked at non-alcoholic fatty liver disease in cannabis smokers, and found that weed is associated with lower disease rates. Occasional users were 15% less likely to get it, while regular users were 51% less likely to get it.
Direct Comparisons Between Cannabis and Alcohol
Looking at the risks of each substance and comparing them is a simple way to see which is safer, but there is a lot of room for error. The information above certainly gives an overview of the risks of each substance, but it feels a little loose to just eyeball the risks and make a call. Luckily, researchers have also made more direct comparisons between the two substances. These give the most direct and to-the-point answer to the key question.
- Lachenmeier and Rehm (2015) used the “margin of exposure” approach to compare the risks of alcohol, tobacco, cannabis and other illicit drugs. Margin of exposure basically means the ratio between the lethal dose and the usual daily dosage of a substance. A lower number means a usual dose is closer to the lethal dose and the drug is more dangerous. Based on this, alcohol, nicotine, cocaine and heroin were classed as high risk and cannabis was classed as low risk.
- Nutt, King and Phillips (2010) compared the harms of different drugs both to individuals and to others. The scoring on various categories was performed by members of the Independent Scientific Committee on Drugs. This analysis showed that the overall “harm score” for alcohol was 72/100, compared to 20/100 for cannabis. The authors concluded that alcohol both poses more risk to individuals than cannabis and more risk to others.
- Van Amsterdam et. al. (2010) did a similar analysis to Nutt, King and Phillips. A panel of 19 experts scored a selection of illegal and legal drugs on societal and individual-level harms. Overall, the experts determined that alcohol was much more dangerous than cannabis, with scores of 2.16 and 2.36 (out of 3) for individual and population-level harm, respectively. Cannabis, in comparison, scored just 1.19 and 1.26.
Is Cannabis Safer Than Alcohol?
Overall, the evidence does suggest that cannabis is safer than alcohol.
Both cannabis and alcohol can cause strokes, with a similar increase in risk, and moderate alcohol consumption is actually pretty good for heart health. But this is where the good news for alcohol ends. Head and neck cancers are higher in drinkers but not in stoners. Alcohol is awful for your liver but cannabis is beneficial, if anything.
There is a unique risk of cannabis in the form of schizophrenia and other mental health conditions. Smoking cannabis also opens up your lungs to many smoking-related illnesses. However, cannabis also has much more medical potential than alcohol, with actively beneficial effects for many conditions and individuals. You can also use edibles instead of smoking and remove many of the risks identified above.
Finally, while it is admittedly very crude, the raw numbers of deaths from each substance underline the reality of the situation more clearly than comparing odds ratios for disparate illnesses. Alcohol kills; cannabis doesn’t.
These facts are why the expert analysis of this exact question has come to the same conclusion as the cannabis advocates: cannabis is safer than alcohol, however you look at it.
Counterpoint: Non-Physical Risks of Cannabis
In practice, not many people will seriously argue that cannabis is a more dangerous drug than alcohol. Whether they’re pro or anti pot, they likely know perfectly well that pot isn’t going to kill anyone. You’re much more likely to have lung problems from smoking 100% legal cigarettes. The reality is that while “cannabis is safer than alcohol” is a pro-pot argument, “alcohol is safer than cannabis” isn’t a common anti-pot argument.
South Park really captured this in the episode “My Future Self ‘n’ Me,” where the kids are caught taking a puff on a joint they found and the parents go full Reefer Madness to try to get them to avoid drugs. The kids find out, as we have here, that cannabis is not really that dangerous, especially compared to the most widely-used drug in our society. However, at the end of the episode, Randy delivers one of the most poignant anti-pot messages we’ve seen in all media:
“The truth is, marijuana probably isn’t going to make you kill people and most likely isn’t going to fund terrorists, but… Well son, pot makes you feel fine with being bored, and it’s when you’re bored that you should be learning some new skills or discovering some new science or being creative. If you smoke pot, you might grow up to find out that you aren’t good at anything.”
Our Take
It ultimately doesn’t really matter whether cannabis is safer than alcohol. The question of whether people should be allowed to use cannabis doesn’t really come down to comparative risks, it comes down to pragmatic issues (should we regulate the market or leave criminals to handle it?) and philosophical ones (who am I to tell my neighbor what he can do in the privacy of his own home?). Alcohol definitely is harmful but we shouldn’t advocate prohibition, for instance.
If we must compare, though, the evidence shows that cannabis – while not risk-free – is substantially safer than alcohol. This is especially true if you use edibles or vape it instead of smoking.
References
Adejumo, A.C. et al. (2017) “Cannabis use is associated with reduced prevalence of non-alcoholic fatty liver disease: A cross-sectional study,” PLOS ONE, 12(4). https://doi.org/10.1371/journal.pone.0176416
About the source:
- Peer reviewed? Yes, published in PLOS ONE.
- Methodology: Cross-sectional case-control study looking at non-alcoholic fatty liver disease (NAFLD) and cannabis use, using data from the Nationwide Inpatient Survey.
- Sample size: 5,950,391.
- Main results: NAFLD was found to be 15% lower in non-dependent weed users and 51% lower in dependent ones, compared to non-users of cannabis.
- Other notes: Since this was cross-sectional, people may have misremembered details about their cannabis use, and the ratios of THC to CBD (or any details about strains) were not in the dataset.
Callaghan, R.C., Allebeck, P. and Sidorchuk, A. (2013) “Marijuana use and risk of lung cancer: A 40-year cohort study,” Cancer Causes & Control, 24(10), pp. 1811–1820. https://doi.org/10.1007/s10552-013-0259-0
About the source:
- Peer reviewed? Yes, published in Cancer Causes & Control.
- Methodology: Population-based cohort study on Swedish conscripts, recruiting men aged 18 to 20 in 1969 and 1970, and following up with them until 2009. Cannabis use and tobacco use were assessed at the start, with follow-ups mainly focused on lung cancer outcomes.
- Sample size: 49,321
- Main results: There was no association between overall cannabis use and lung cancer, after adjusting for tobacco use (although pre-adjustment, it looked like there was). For those who’d used cannabis more than 50 times, they were found to have 2.12 times the risk of a non-user even in the adjusted models.
- Other notes: The study didn’t track tobacco use during follow-up, and it is unknown whether the cannabis users mixed their weed with tobacco to smoke it. As the authors note, 91% of the heaviest cannabis smokers smoked tobacco at the time of conscription, so the effect could be due to inadequate adjustment for the tobacco-related risk.
Corrao, G. et al. (2004) “A meta-analysis of alcohol consumption and the risk of 15 diseases,” Preventive Medicine, 38(5), pp. 613–619. https://doi.org/10.1016/j.ypmed.2003.11.027
About the source:
- Peer reviewed? Yes, published in Preventative Medicine.
- Methodology: Meta-analysis of epidemiological studies on the link between alcohol and 14 conditions, plus injuries and violence, with literature published from 1966 to 1998.
- Sample size: 156 studies, with a total of 116,702 cases of the 14 conditions plus injuries.
- Main results: While alcohol increased risk of many conditions (especially liver cirrhosis, hypertension and cancers of the oral cavity, esophagus and larynx), and was more weakly associated with many other cancers, moderate consumption was protective against coronary heart disease.
- Other notes: This misses all more modern evidence, and for some conditions, there weren’t too many studies available that met the authors’ criteria. However, the results are broadly confirmed by more recent research.
de Carvalho, M.F.F. et al. (2015) “Head and neck cancer among marijuana users: A meta-analysis of matched Case–Control Studies,” Archives of Oral Biology, 60(12), pp. 1750–1755. https://doi.org/10.1016/j.archoralbio.2015.09.009
About the source:
- Peer reviewed? Yes, published in Archives of Oral Biology.
- Methodology: Found matched case control studies looking into the association between marijuana use and cancers of the head and neck (e.g. oropharyngeal, nasopharyngeal and oral cancers) and performed a meta-analysis of the findings.
- Sample size: 9 case control studies, totaling 13,931 participants (5,732 cases, 8,199 controls)
- Main results: There was no association between marijuana use and head and neck cancer rates.
- Other notes: This analysis only looked at ever having used marijuana vs. never having used it. It’s possible that there would be an association if people used more frequently.
Desrosiers, N.A. et al. (2014) “Phase I and II cannabinoid disposition in blood and plasma of occasional and frequent smokers following controlled smoked cannabis,” Clinical Chemistry, 60(4), pp. 631–643. https://doi.org/10.1373/clinchem.2013.216507
About the source:
- Peer reviewed? Yes, published in Clinical Chemistry.
- Methodology: People who smoked at least occasionally had their blood and plasma cannabinoid levels taken before and after smoking a 6.8% THC joint.
- Sample size: 14 frequent and 11 occasional smokers.
- Main results: Frequent smokers had more THC and its derivatives in their blood most of the way through the experiment. Frequent smokers had at least 5 ng THC/ml of blood for a median of 3.5 hours after smoking, and for occasional smokers it was 1 hour.
- Other notes: The frequent smokers were also generally younger and had smoked for fewer years overall than the occasional smokers. Two participants self-reported as occasional users but were classified as frequent users because their results were more in line with published data on occasional users.
Grønbæk, M. (2009) “The positive and negative health effects of alcohol- and the Public Health Implications,” Journal of Internal Medicine, 265(4), pp. 407–420. https://doi.org/10.1111/j.1365-2796.2009.02082.x
About the source:
- Peer reviewed? Yes, published in the Journal of Internal Medicine.
- Methodology: A (non-systematic) review of studies on the health risks and benefits of alcohol use.
- Main results: For some conditions and all-cause mortality, there is a J-shaped curve, with a reduction in risk at low and moderate levels. However, for many conditions it increases risk if you drink a lot.
- Other notes: There is some uncertainty around the beneficial effects of alcohol, because of biases in the studies (e.g. mainly recruiting middle class, older people) and the comparison groups (e.g. “non-drinkers” may not drink because they have pre-existing risk factors).
Lachenmeier, D.W. and Rehm, J. (2015) “Comparative risk assessment of alcohol, tobacco, cannabis and other illicit drugs using the margin of exposure approach,” Scientific Reports, 5(1). https://doi.org/10.1038/srep08126
About the source:
- Peer reviewed? Yes, published in Nature Scientific Reports
- Methodology: Using already existing data to compare average human intake of substances to their LD50 (the lethal dose for 50% of people).
- Main results: Alcohol, nicotine, cocaine and heroin are “high risk” while cannabis is “low risk.”
- Other notes: LD50 is a fairly extreme benchmark for comparison, since drugs also have negative effects before becoming fatal to users. However, this was all the authors could find
Martinasek, M.P., McGrogan, J.B. and Maysonet, A. (2016) “A systematic review of the respiratory effects of inhalational marijuana,” Respiratory Care, 61(11), pp. 1543–1551. https://doi.org/10.4187/respcare.04846
About the source:
- Peer reviewed? Yes, published in Respiratory Care.
- Methodology: Systematic review of English-language peer-reviewed articles on the effects of cannabis on the respiratory system.
- Sample size: 48 articles with varying sample sizes.
- Main results: Marijuana smokers risk lung cancer, spontaneous pneumothorax, emphysema and COPD, and many report symptoms like wheezing, coughing, shortness of breath, bronchodilation and other symptoms.
- Other notes: Some of the studies included in the review had major oversights. For instance, only some of the lung cancer results controlled for the known effect of tobacco smoking. Not accounting for this would substantially bias the results.
Moore, T.H.M. et al. (2007) “Cannabis use and risk of psychotic or affective mental health outcomes: A systematic review,” The Lancet, 370(9584), pp. 319–328. https://doi.org/10.1016/s0140-6736(07)61162-3
About the source:
- Peer reviewed? Yes, published in The Lancet.
- Methodology: Searched for longitudinal and population-based studies on marijuana use and psychotic or affective mental health conditions and pooled the results.
- Sample size: Included 35 studies.
- Main results: Ever use of cannabis was associated with a 41% increase in risk of psychotic conditions, with the most frequent users being over twice as likely to develop one. There was less clarity about depression, suicidal thoughts and anxiety.
- Other notes: The studies making up this meta-analysis are certainly not perfect, and the authors point out that this doesn’t establish causality and it’s unlikely that these uncertainties will be cleared up in the near future either.
Nutt, D.J., King, L.A. and Phillips, L.D. (2010) “Drug harms in the UK: A multicriteria decision analysis,” The Lancet, 376(9752), pp. 1558–1565. https://doi.org/10.1016/s0140-6736(10)61462-6
About the source:
- Peer reviewed? Yes, published in The Lancet.
- Methodology: Members of the Independent Scientific Committee on Drugs scored 20 drugs on 16 different harm criteria, 9 for harm to users and 7 for harm to others.
- Main results: Heroin, crack cocaine and methamphetamine were judged as the most harmful to users, and alcohol, heroin and crack were the most harmful to others. Overall, alcohol was the most harmful (harm score 72 out of 100) and cannabis was eighth (20 out of 100).
- Other notes: Although the analysis was performed by experts using objective measures, the scoring has an unavoidable element of subjectivity. It is essentially expert opinion, not hard evidence. Also, harms are rated based on their availability and consequences for their use, so the harms resulting from prohibition (or, social acceptability) are included.
Rock, K.L. et al. (2022) “Can cannabis kill? Characteristics of deaths following cannabis use in England (1998–2020),” Journal of Psychopharmacology, 36(12), pp. 1362–1370. https://doi.org/10.1177/02698811221115760
About the source:
- Peer reviewed? Yes, published in the Journal of Psychopharmacology.
- Methodology: Used deaths following cannabis use reported to the English National Program on Substance Abuse Deaths, with records from 1997 to April 2021, to determine whether blood cannabis concentration is useful in post-mortem examinations.
- Sample size: 3,455 people who died with cannabinoids in their system.
- Main results: Most (96%) cases involved other drugs too, with just 136/3,455 involving cannabis alone. Of these, 62% were deaths from traumatic injury. Cannabis toxicity was deemed the sole cause of death in just one case, where the individual had between 100 and 150 μg/L of THC in their blood (with 5 μg/L being a more common value after smoking for regular users).
- Other notes: The authors note that “The risk of death due to direct cannabis toxicity is negligible,” and that it’s unclear how exactly a high blood THC level would cause death. Also, since all deaths weren’t reported to the system, these figures are likely underestimates.
Shah, S. et al. (2021) “Association of marijuana use and cardiovascular disease: A behavioral risk factor surveillance system data analysis of 133,706 US adults,” The American Journal of Medicine, 134(5). https://doi.org/10.1016/j.amjmed.2020.10.019
About the source:
- Peer reviewed? Yes, published in The American Journal of Medicine.
- Methodology: Pooled data from the 2016, 2017 and 2018 Behavioral Risk Factor Surveillance System. They looked for non-cigarette-smokers who either use or don’t use cannabis. The researchers asked about cardiovascular conditions such as coronary artery disease and strokes. They also asked how often participants used marijuana and how they used it.
- Sample size: 133,706
- Main results: Frequent (but not occasional) marijuana users had 88% higher odds of myocardial infarction or coronary artery disease and 81% higher chance of stroke. By mode of use, frequent marijuana smokers (including vapers) had 2.1 times higher odds of myocardial infarction or coronary artery disease and 84% higher chance of stroke. However, users of edibles did not show an increase in risk.
- Other notes: The study was cross-sectional (at a single point in time), so causality is difficult to establish. There is a possibility of some errors in recall or simply not reporting use. The survey also didn’t ask about other drug use, which may bias the results.
van Amsterdam, J. et al. (2010) “Ranking the harm of alcohol, tobacco and illicit drugs for the individual and the population,” European Addiction Research, 16(4), pp. 202–207. https://doi.org/10.1159/000317249
About the source:
- Peer reviewed? Yes, published in European Addiction Research.
- Methodology: 19 drugs were ranked for physical and social harmfulness by a panel of 19 Dutch experts. This includes 7 scientists, 8 clinicians and 4 social harm experts.
- Main results: Crack cocaine was the most dangerous drug overall and magic mushrooms were the least dangerous. Alcohol scored much higher than cannabis, with an individual-level score of 2.16 and a population level score of 2.36. This is compared to 1.19 and 1.26 for cannabis.
- Other notes: This is expert opinion, but still ultimately opinion-based. Whether a single score can capture the “harmfulness” of drugs in this way is debatable.
Zhang, L.R. et al. (2015) “Cannabis smoking and lung cancer risk: Pooled analysis in the International Lung Cancer Consortium,” International Journal of Cancer, 136(4), pp. 894–903. https://doi.org/10.1002/ijc.29036
About the source:
- Peer reviewed? Yes, published in the International Journal of Cancer.
- Methodology: Analyzed lung cancer cases and controls (from six International Lung Cancer Consortium studies) to determine the association (if any) between cannabis smoking and lung cancer.
- Sample size: 2,159 cases and 2,985 controls (5,144 total)
- Main results: The pooled analysis revealed no significant association between cannabis use and lung cancer.
- Other notes: The abstract states, “Weak associations between cannabis smoking and lung cancer were observed in never tobacco smokers.” However, this association is not statistically significant according to the full text of the paper.