Commentary on Imtiaz <i>et al</i>. (2016): The burden of disease attributable to cannabis—implications for policy

The contribution that current patterns of cannabis use make to the burden of disease is very useful information, but it cannot decide policy. Sameer Imtiaz and colleagues have used the methods developed as part of the 2010 Global Burden of Disease Comparative Risk Assessment to estimate the contribution that cannabis use made to the burden of disease (BoD) in 2012 in Canada, a developed country where the prevalence of cannabis use is among the highest in the world 1. They included among the cannabis-related causes of disease burden: cannabis use disorders (CUD), road traffic injuries and deaths, schizophrenia and lung cancer. CUD are a consequence of cannabis use by definition; recent meta-analyses suggest that cannabis use modestly increases the risk of road traffic injuries and deaths, and there is reasonable evidence that regular cannabis is a contributory cause of the onset of psychoses 2, although some argue that we cannot exclude the possibility of residual confounding 3. The inclusion of lung cancer is more contentious, given that a recent meta-analysis concluded that there was not sufficient evidence to assess whether cannabis smoking increased the risk of lung cancer 4. The authors included uncertainty bounds around their estimates to reflect the fact that many of these estimated risks are based on a small number of epidemiological studies which have weakly quantified cannabis use. Their comparison of the BoD attributable to alcohol, tobacco and cannabis shows that cannabis contributes a very small fraction of the contribution of alcohol and tobacco. This reflects the combined effects of the very large difference in the prevalence of the regular use of these three drugs and the fact that there is a much larger research literature linking alcohol and tobacco to many more causes of death and disability than cannabis. A comparison of BoD attributed to cannabis and opioids shows that cannabis use is much more prevalent than opioid use, but the opioid drugs make a much larger contribution to BoD because they cause many more deaths in young adults by overdose, suicide, accidents and blood-borne virus (BBV) infections. Only 20 years ago, statements about the comparative harms of cannabis, alcohol, tobacco and opioids were highly controversial. Some authors (e.g. 5) insisted that such comparisons could not and should not be made. A paper that Robin Room, Susan Bondy and I prepared for a 1997 World Health Organization (WHO) report on the health effects of cannabis 6 caused enormous public controversy and was disowned by WHO (see 7). The successful development of the comparative risk assessment approach used in the GBD estimates has thankfully put paid to these objections to comparative assessments of contributions that different types of drugs make to BoD. Imtiaz et al. acknowledge an important qualification that applies to all such comparisons; namely, that they reflect the BoD attributable to current patterns of cannabis use. Those who defend current cannabis policies can argue reasonably that these policies can claim some credit for this fact. Many adults in developed countries have used cannabis at some time in their lives, but only a very small proportion of these (approximately 10%) have used regularly, and a still smaller proportion have used daily over decades. The latter is a miniscule proportion by comparison with the proportions of tobacco smokers who smoke daily and alcohol consumers who drink in a risky manner. It is clear that CUD can affect a substantial number of young people in developed societies. This fact demands the policy response suggested by the authors; namely, screening and early intervention in medical settings where problem cannabis users are likely to be seen. Is the contribution that cannabis use makes to BoD enough to justify a change from current prohibitionist policies to more liberal policies that may include legalization for recreational use, as has occurred in four US states? This is properly a matter for democratic deliberation that considers the social and economic costs and benefits of adopting different policies towards cannabis. Data on the effects of cannabis on BoD, as used currently, will play a useful role in such deliberations, but it cannot be decisive 8. None. I thank Sarah Yeates for her assistance in preparing this commentary.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.