A key question in the current debate about the growing use of e-cigarettes or ENDS [electronic nicotine delivery systems] is their safety as compared with cigarettes and other nicotine delivery systems such as snus. Because there can never be a randomised controlled trial of the different nicotine products and because of the huge health damage from cigarettes [6 million premature deaths per year world-wide] it is vital to provide the best available estimates of comparative harm to both users and health policy makers. Building on the success of our earlier estimates of comparative drug harms made using the MCDA approach [1,2] we decided to do the same for nicotine products. Data from the paper reporting out conclusions  were used recently by PHE to argue for the value of ENDS a health improvement measure . This conclusion has been today subjected to a critical editorial in the Lancet that provides no new data but attempts to undermine the value of the PHE report and our nicotine MCDA paper by innuendo of bias and claims of conflicts of interest.
Although our nicotine MCDA study  considered 12 products and 14 harm criteria, (7 harms to users and 7 harms to others), we are concerned here with only two nicotine products, smoked tobacco and e-cigarettes, and four criteria, the only ones that distinguish the two products. Two criteria concern harms to users: product specific damage (morbidity, chronic ill health), and product related mortality (death due to cancer, respiratory illness, cardiovascular disease and fire). There is abundant evidence of those two harms from smoking, but as they are mainly the result of carcinogens and toxins in cigarette smoke, which is absent in e-cigarettes. It does not take a large data base of evidence to judge that e-cigarettes are considerably less harmful on these two criteria. The 95% figure is mainly based on just those two criteria, with a little more widening of the gap between those two products if dependence is taken into account (e-cigarettes judged to be lower).
Interestingly, the PHE figure seems to be based only on harms to users, and takes no account of the only two harms to others that discriminate between the two products: economic cost (health-care, customs, loss of productivity, absenteeism), and injury (injuries to others both directly and indirectly, e.g. traffic accidents, foetal harm, second-hand smoke, accidental poisoning, burns). Again, there is more evidence about those features for cigarettes, less so for e-cigarettes, but again, it is clear that the far greater incidence of smoking generates more harm than e-cigarettes. If these considerations are included, then the 95% figure increases to 97%.
As for the implied objection to the heavy reliance on opinion and the charge of potential bias, note that the starting point for Nutt et al  was the well-thought-through selection and definitions of all the criteria, which was originally initiated by the Advisory Council on the Misuse of Drugs, and has subsequently been used in two studies of the harms of psychoactive drugs, one for the UK , the other for the EU . Both studies relied heavily on the judgements of participants (less than 15% attending both workshops) and the final results correlated 0.993. We believe this exceptional degree of reliability is the result of choosing participants with differing expertise and perspectives, applying a structured process of modelling, and conducting the workshop by an impartial facilitator who applied techniques specifically designed to minimize bias. It is these features, all of which have been found to produce reliable and replicable workshop results , that characterised the nicotine products workshop, and we stand by our findings.
The funding of the MCDA nicotine study was provided by a Swiss-based consulting firm interested in this question as an unrestricted grant to the charity DrugScience. They had no involvement in the subsequent meetings and publications. There are no links or contracts, current or historic, between EuroSwiss Health and any tobacco company. So the funds provided for meetings leading to the development of a rational scale and risk assessment of tobacco and nicotine products were not linked to tobacco companies at all. Furthermore as stated in the paper EuroSwiss Health had no influence the MCDA process. An Italian anti-smoking charity made a small financial contribution at the publication stage to help pay for open access. The expert group was chosen by members of the DrugScience scientific panel with the help of other UK experts to provide international expert group with expertise across the various disciplines of health, psychology, addiction, pharmacology, toxicology, neuroscience etc; the final panel was an impressive group. As is inevitable in such a field several had or had had “interests” in aspects of nicotine and tobacco supply or harm reduction, and these were declared at the time. However it is important to re-emphasise that the MCDA process is such that neither individuals nor even groups with particular biases can significantly influence the outcomes.
In conclusion we believe the DrugScience 2014 nicotine MCDA paper to be by far the best estimate of comparative nicotine product harms we have today. The Lancet experts or others who do not like the results should look to provide funding to repeat the analysis. A repetition would be particularly timely now the evidence base for ENDS has grown.
David Nutt & Larry Phillips
On behalf of DrugScience.org.uk
1. Nutt DJ, King LA, Phillips LD (2010) Drug harms in the UK: a multicriteria decision analysis Lancet 376: 1558-65 http://www.thelancet.com/journals/lancet/article/PIIS0140-6736%2810%2961462-6/abstract
2. van Amsterdam J, Nutt D, Phillips L, van den Brink W (2014) European rating of drug harms. J Psychopharmacol. 2015 Apr 28. pii: 0269881115581980.
3. Nutt DJ, Phillips LD, Balfour D, Curran HV, Dockrell M, Foulds J, Fagerstrom K, Letlape K, Milton A, Polosa R, Ramsey J, Sweanor D (2014) Estimating the harms of nicotine-containing products using the MCDA approach. Europ J Addiction 2014;20:218-225DOI: 10.1159/000360220 http://www.karger.com/Article/FullText/360220
4. McNeill A, Brose L, Calder R, Hitchman S, Hajek P, McRobbie H. E-cigarettes: an evidence update. London: Public Health England; 2015.
5. Phillips LD. Decision Conferencing. In: Edwards W, Miles RF, von Winterfeldt D, editors. Advances in Decision Analysis: From Foundations to Applications. Cambridge: Cambridge University Press; 2007.