• Alcohol;
  • classification system;
  • drug control policy;
  • drug-related harm;
  • drug scheduling;
  • harm assessment;
  • illicit drugs;
  • tobacco

The publication 4 years ago by Nutt et al. [1] of a scale to assess drug-related harms brought with it a storm of reactions from scientists and the public at large. Of particular interest was the finding that an independent assessment of drug harms was inconsistent with the current system of drug classification, in that harm scores for some legal substances (alcohol and tobacco) were higher than scores for many of the classified substances (e.g. cannabis). The controversy has continued unabated, much of it fuelled by Professor Nutt himself in press interviews and public lectures [2]. In a 2010 report presented as an improvement on the 2007 study, Nutt and colleagues [3] used an expert consensus approach to assess drug harms and arrived at a similar conclusion that the way drugs are classified is not supported by scientific evidence regarding the level of harm associated with the drugs.

As expected, the second study has been as controversial as the first. In their reaction to the paper, Caulkins, Reuter & Coulson [4] argue that attempts to rank-order drugs on the basis of harm is ‘fundamentally misguided’, and that even if a method produced perfect rankings the exercise ‘would not provide sufficient basis’ for scheduling or other policy decisions. Thus, the problems with the study as seen by Caulkins and colleagues are both on the methodology and relevance of its findings. Calling the exercise ‘misguided’ and ‘pseudo-scientific’ might be a little harsh, but it is fair to question the method used to establish harm. Determining harm by consensus can be problematic, as assessments of this nature are likely to be influenced by personal biases and levels of expertise. The method of multi-criteria decision analysis as used by Nutt et al. also did not take into account the benefits of drug use (especially in the case of alcohol), differential availability of illegal substances and polydrug use, all of which are factors that might affect how harm is defined and measured. This should be a concern in the UK context but more so if, as is likely to happen, the findings of this study are generalized to other settings.

Caulkins and colleagues are opposed to rankings, but they seem to recognize the basic human motivation for order that drives efforts to compare different types of harm. They have, therefore, proposed an alternative ‘harm matrices’ (as opposed to ‘harm ratings’) approach that would be more multi-dimensional by taking into consideration the context of drug use, among other factors. What they have suggested is a more complex version of Nutt et al., a version that I suspect will produce similar (but not comparable) results, and one they hope would serve as a ‘flexible’ tool for policy.

The implications for international drug control policy and the suspicion that a political agenda might be at play seem to have contributed to the controversy surrounding the work by Nutt et al., who are obviously critical of the current classification system because it ignores drug-related harms. Of course, a system based on harms alone would be problematic. What, for example, should be done about alcohol, a drug that is well known to be associated with high levels of physical and social harm [5,6]? While the majority in most countries would hardly recommend banning alcohol, the findings by Nutt et al. at least call for greater attention to this drug, including the implementation of interventions that are effective in reducing alcohol-related harm [7].

In spite of the critical response by Caulkins and colleagues to the rank-ordering of drugs according to physical, psychological and social harms, I suspect that rankings of this nature will continue, albeit with refined methods. If the system of harm matrices proposed by the authors can produce more useful information for policy, that would be an important contribution.


  1. Top of page
  2. Declarations of interest
  3. References
  • 1
    Nutt D., King L. A., Saulsbury W., Blakemore C. Development of a rational scale to assess the harm of drugs of potential misuse. Lancet 2007; 369: 104753.
  • 2
    Nutt D. Estimating Drug Harms: A Risky Business ? 2009. Available at: (accessed 29 May 2011; archived by Webcite at
  • 3
    Nutt D. J., King L. A., Phillips L. D. Drug harms in the UK: a multicriteria decision analysis. Lancet 2010; 376: 155865.
  • 4
    Caulkins J. P., Reuter P., Coulson C. Basing drug scheduling decisions on scientific rankings of harmfulness: false promise from false premises. Addiction 2011; Epub ahead of print. doi:10.1111/j.1360-0443.2011.03461.x.
  • 5
    Rehm J., Room R., Monteiro M., Gmel G., Graham K., Rehn N. et al. Alcohol use. In: Ezzati M., Lopez A., Rodgers A., Murray C. J. L., editors. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva: World Health Organization; 2004, p. 9591108.
  • 6
    World Health Organization. Global Status Report on Alcohol and Health. Geneva: World Health Organization; 2011.
  • 7
    Babor T. F., Caetano R., Casswell S., Edwards G., Giesbrecht N., Graham K. et al. Alcohol: No Ordinary Commodity. Oxford: Oxford University Press; 2010.