This analysis of drug harms in Canada found that alcohol causes the most harm overall. This finding is consistent with the results of drug harm studies conducted in the UK, EU, Australia, and New Zealand, which all ranked alcohol first (Bonomo et al., 2019; Crossin et al., 2023; Nutt et al., 2010; Van Amsterdam et al., 2015). Beyond alcohol, results differ. In Canada, the drugs causing the most harm following alcohol are tobacco, nonprescription opioids, cocaine, and methamphetamine. The UK and EU ranked heroin second, while Australia and New Zealand ranked methamphetamine second. The third-ranked substances were methamphetamine in the UK and EU, heroin in Australia, and synthetic cannabinoids in New Zealand. The lower ranking of tobacco in the other studies is likely because they did not take prevalence into account when scoring harm to users. In the Australian study’s supplementary analysis incorporating prevalence, the adjusted scores moved cigarettes from fifth to second place (Bonomo et al., 2019).

Our findings are also consistent with the conclusions of the most recent Canadian Substance Use Costs and Harms study (2023), which estimated that alcohol accounts for 40% of total drug-attributable costs (healthcare, lost productivity, criminal justice, etc.), followed by tobacco, opioids, cocaine, amphetamine/methamphetamine, and cannabis. The top six rankings in that study correspond with the top six rankings in this MCDA, lending face validity to our findings.

In a commentary on the original UK drug harms study, Rolles and Measham (2011) described drug-related harm as “pharmacology translated through a complex prism of social, behavioral, and environmental variables, in turn influenced by a range of differing policy regimes, from prohibition through to legal regulation.” Indeed, the harm score for each drug reflects not only its pharmacological risk profile but also the current policy context in Canada. For example, alcohol is the most commonly used drug and is associated with a variety of acute and chronic health harms, but independent of its pharmacological properties, the way it is (or is not) regulated can cause additional harm. Alcohol’s availability, price, and promotion drive its consumption—and as population-level consumption increases, so does harm (Babor et al., 2023). Over the past two decades, governments across the country have loosened regulations to make alcohol more affordable and more available, and in many cases, these changes have been associated with increases in deaths, hospitalizations, and economic costs (Giesbrecht and Myran, 2024). In the case of illegal drugs, prohibition impacts how people use them, encouraging risky practices like using alone (which increases the risk of mortality) or sharing equipment (which increases the risk of infectious disease) (Papamihali et al., 2020). It also impacts the production chain of the drugs themselves, creating an unregulated supply in which the content, quality, and potency of drugs are variable and not readily known (Csete et al., 2016). The consumption of opioids, cocaine, and methamphetamine comes with inherent pharmacological risks, and while those risks are reflected in this study’s harm scores, so are the external factors influencing the composition of the drug supply and the consumption patterns of people who use them.

Several limitations of MCDA have been outlined by the authors of the first drug harm ranking studies (Nutt et al., 2010; Van Amsterdam et al., 2015) and their critics (Caulkins et al., 2011; Rolles and Measham, 2011). This study only considers harms and does not account for any benefits a drug may have at the individual or societal level. It does not account for individual differences: drug harms differ depending on the person, the setting, the route of administration, and many other factors, and this study does not shed light on those elements. Since this analysis focuses on the general population aged 15+, it does not provide insight into the harms to youth or specific sub-populations. It does not consider medical/therapeutic use or polydrug consumption. Importantly, it rests on the current state of knowledge and experience. Many questions remain unanswered about the harms associated with some drugs, and even the prevalence of some drugs remains unclear; the expert panel had to generate its own prevalence estimates for some illegal drugs due to the lack of objective prevalence data. There are also important regional differences in drug markets and drug use patterns. For example, drug poisoning deaths are decreasing in western and central Canada but increasing in the east of the country (Fischer et al., 2025). Expert panel members also noted that crack is common in Ontario, Québec, and the Maritimes but much less so in western Canada; conversely, methamphetamine appears to be much more common in the Prairies and British Columbia than in most of eastern Canada (O’Keefe-Markman et al., 2022). Some members of the panel felt that our prevalence estimate for methamphetamine was too low. Increasing it would have moved methamphetamine into fourth place in the rankings, above cocaine. And finally, the decision to account for prevalence limits this drug harm study’s comparability to the ones that came before it. However, this change addresses a key methodological criticism of the previous studies; further, we believe it makes the present rankings more policy-relevant, because public policy—including health policy—is designed based on desired aggregate outcomes and evaluated against aggregate indicators (Kingdon, 1984; Rose et al., 2008). Total harm is thus more relevant to policymakers than individual-level harmfulness.