Tobacco Control Research: A Tale of Two Delphi Groups

Neil McKeganey | 9 September 2015

Tobacco control researchers are nothing if not inconsistent. Take, for example, the recent case of Professor David Nutt who convened a group of experts to assess the relative harm of different nicotine delivery products (Nutt et al 2015). The paper Nutt and colleagues published on the basis of their assessment formed part of Public Health England’s recent conclusion that electronic cigarettes were 95% less harmful than combustible tobacco.

The method that Nutt and colleagues used in this study was a continuation of earlier work in which they had used the approach of Multi-Criterion Decision Analysis (a form of the Delphi technique in which the views of key experts on a given topic are collected and assessed) to rank the relative harm of a range of legal and illegal drugs (Nutt et al 2007). This earlier work on legal and illegal was published in the Lancet medical journal to worldwide attention and some considerable acclaim (Caulkins et al 2011, Owen 2009, Rossow 2011, Nutt 2011).

Nutt and his various colleagues might have felt somewhat taken aback then when the Lancet published an anonymous editorial attacking the nicotine ranking study he had undertaken on the basis that the Multi-Criterion Decision Analysis method he had used was inadequate to assess the relative harm of electronic cigarettes compared to combustible tobacco products (Lancet 2015). The Lancet editorial produced further attacks on the Nutt study in the Guardian and the Telegraph newspapers along with a swathe of critical comments on Twitter from some of the leading tobacco control researchers such as Professor Simon Capewell and Professor Simon Chapman and further correspondence published in the Lancet (McKee and Capewell 2015. Included within these attacks were criticisms of the fact that three of the members of the group that Nutt had convened to assess the relative harm of various nicotine delivery products had declared, past, connections to the electronic cigarette industry. The various experts in Nutt’s group must have begun to wonder at the wisdom of even participating in a study that drew such harsh criticism, and in the case of one of the experts involved (the widely respected Professor Riccardo Polosa) unfair attacks on past, declared, research funding. Rouse the big beast of tobacco control researcher and you’d better be prepared to be bitten.

Scroll back a couple of years and you have another Delphi group that produces a very different reaction from tobacco control researchers. This time the group is convened by Pechey and colleagues (Pechey et al 2013) and is focussed on eliciting the views of various tobacco control experts on the likely impact of tobacco plain packaging. The experts consulted were of one opinion - tobacco plain packaging would significantly reduce smoking prevalence. So convinced were the experts of the impact of a policy that had not yet even been implemented, they were able to give a quantitative assessment of just how much smoking prevalence would reduce two years after the policy had been adopted. Smoking prevalence, they opined, would reduce by one percent in adults and three percent in children. It is likely that the results of this research contributed in part to the decision to implement plain packaging policy within the UK.

In the case of the Pechey study, tobacco control researchers voiced no criticisms of the methods used or the external interests of the experts consulted. Indeed in the latter case that would have been impossible because the names of the experts who provided the assessments, nor their own possible conflicts of interests, were ever declared.

So what does all this tell you about tobacco control research? It tells you that when it comes to weighing the harms of different products, or assessing the impact of different policies, tobacco control researchers embrace the research that reports the findings they like and rail against the studies that report inconvenient or unwelcome findings. That is a hugely regrettable aspect of research in this area. The science of understanding human behaviour is difficult enough without the added burden of political interests around big tobacco and smoking colouring how different studies are assessed and different researchers are attacked.

Centre for Drug Misuse Research


  1. Bowcott, O. (2009) David Nutt’s Dangerous Drugs List Guardian 2 November
  2. Caulkins, J., Reuter, P., Coulson, C. (2011) Basing drug scheduling decisions on scientific ranking of harmfulness: false promise from false premises. Addiction, 2011; DOI: 10.1111/j.1360-0443.2011.03461.x
  3. Lancet (2015) E-cigarettes: Public Health England's evidence-based confusion Lancet Volume 386, No. 9996, p829, 29 August 2015
  4. McKee, M., Capewell, S. (2015) Electronic cigarettes: we need evidence, not opinions Lancet DOI:
  5. Nutt, D (2011) Let Not the Best be the Enemy of the Good Addiction. Volume 106, Issue 11, pages 1892–1893, November 2011
  6. Nutt, D., et al (2014) Estimating the Harms of Nicotine Containing Products Using the MCDA Approach European Addiction Research 20 (5) 218—225
  7. Nutt, D., King, L., Saulsbury, W., Blakemore, C. (2007) Development of a rational scale to assess the harm of drugs of potential misuse. Lancet Volume 369, No. 9566, p1047–1053, 24 March 2007
  8. Pechey, R., Spiegelhalter, D., Marteau, T. (2013) Impact of Plain Packaging of Tobacco Products on Smoking in Adults and Children: An Elicitation of International Experts Estimates BMC Public Health 13, 18
  9. Rossow, I. (2011) Can Harm Ratings be Useful Addiction Volume 106, Issue 11, pages 1893–1894, November 2011