E-cigarettes are less harmful than smoking

By David J Nutt, Lawrence D Phillips, David Balfour, H Valerie Curran, Martin Dockrell, Jonathan Foulds, Karl Fagerstrom, Kgosi Letlape, Riccardo Polosa, John Ramsey, David Sweanor

(The Lancet)

The Lancet Editorial1 criticising Public Health England’s review of electronic cigarettes (e-cigarettes) focused on three supposed short-comings of our paper:2 a lack of hard evidence, no formal criteria used, and so relied on the opinions of participants, and potential bias arising from the selection of participants and the declared conflicts of interest of some authors.2 As authors of the original paper,2 we believe that these three criticisms have over-generalised the evidence issue, did not respect the knowledge and experience of the experts selected, and did not take into account the many measures used to minimise potential bias.

First, regarding the lack of evidence, an abundance of evidence is available about the harm of cigarettes. The paucity of evidence for serious harm to users of e-cigarettes over the years since they were first marketed in 2006, with millions purchased, in itself is evidence. Additionally, biomarkers of potential harm of e-cigarettes are broadly reassuring.3

Second, we used the approach of decision conferencing,4 sought from participants their expert judgments and not opinions. The criteria and their definitions were taken from three drug harm studies, the Advisory Council on the Misuse of Drug’s original formulation,5 the 2010 study of UK drug harms published in The Lancet,6 and the 2013 replication for EU drug harms.7 Judgments about scores were based on data along with our own knowledge and experience of the extent of harm and plausible causal mechanisms for harm. If data were available, these were discussed openly about their validity and reliability by the group, but if data were sparse or absent we relied on logical inferences (eg, the dearth of evidence of dying directly from an overdose of smoking led us to infer that cigarettes are not very harmful on that criterion and gave it a low score, but assigned e-cigarettes a higher harm score for that effect because the nicotine solution in the cartridges could potentially be directly accessed). A strength of the multicriteria decision analysis (MCDA) model8 is that it incorporates data and the judgments about the relevance of the data, thus capturing meaningful differences in the importance of their effects.

Third, we selected experts on the basis of their publications, experience, and generally acknowledged professional standing to have diverse perspectives and expertise that could be relevant to assess harms from nicotine products. We included experts on behavioural pharmacology, legal aspects of tobacco control, smoking policy, toxicology, neuropsychopharmacology, psychopharmacology, public health sciences, and internal medicine, who collectively have published more than 300 scientific reports relevant to understand nicotine and tobacco harms. We feel that it was misleading of The Lancet1 to characterise the authors2 as having “no prespecified expertise in tobacco control” because the project was about relative harms of nicotine products not tobacco policy.

Regarding the concern about author conflicts of interest,1 the decision conference process is designed to ensure that participants challenge each other. Additionally, the facilitator ensured that peer review operated on-the-spot throughout the creation and exploration of the MCDA model.8 Consistency checks and sensitivity of overall results to the input scores and weights were thoroughly explored; the model results were very robust to imprecision in data and the few disagreements among the experts. As a result, a single participant with a potential bias could not have had any meaningful influence on the process outcome.

Potential sources of conflicts of interests were disclosed at the 2013 MCDA meeting (July, 2013, London, UK). Any conflicts from the previous 3 years before the meeting were disclosed in the published paper.2 We were informed by EuroSwiss Health (Trélex, Switzerland) that they do not receive funding from any tobacco or e-cigarette manufacturers; a requirement we had before accepting their funding. We received no funds from tobacco or e-cigarette manufacturers and, as stated in our paper,2 EuroSwiss Health and Lega Italiana Anti Fumo (LIAF) had no influence on the MCDA process.

We are confident that the nicotine products we studied were assessed by an appropriately structured process with a requisite diversity of research experts who engaged in constructive discourse in building a model that represented the most scientifically sound assessment of the relative harms of nicotine products. Our model’s results for harms to users of e-cigarettes provided Public Health England with the basis for their correct calculation to estimate that e-cigarettes are 95% less harmful to users than smoking. Or, as we prefer, smoking is estimated to be twenty times more harmful to users than vaping e-cigarettes.

DJN, DB, HVC, MD, KL, JR, and DS declare no competing interests. LDP reports personal fees from DrugScience. JF reports personal fees, grants, and non-financial support from Pfizer and reports personal fees and non-financial support from GlaxoSmithKline (GSK), outside the submitted work. KF reports consulting for Pfizer, Chrono Therapeutics, Novartis, Niconovum, and Nicoventures, and non-financial support from Swedish Match. RP reports grants from Pfizer and Boehringer Ingelheim, personal fees from Novartis and GSK (outside the submitted work), is a consultant for Cancer Research UK, Italian Ministry of Health’s Technical Committee on electronic cigarettes, the UK All Party Parliamentary Group, and previously was a consultant for Global Health Alliance for treatment of tobacco dependence, Arbi Group Srl (an Italian e-cigarette distributor), and ECITA (Electronic Cigarette Industry Trade Association, UK). RP is a scientific advisor for Italian Antismoking League (LIAF).


  1. The Lancet. E-cigarettes: Public Health England’s evidence-based confusion. Lancet. 2015; 386: 829
  2. Nutt, DJ, Phillips, LD, Balfour, D et al. Estimating the harms of nicotine-containing products using the MCDA approach. Eur Addict Res. 2014; 20: 218–225
  3. McRobbie, H, Phillips, A, Goniewicz, ML et al. Effects of switching to electronic cigarettes with and without concurrent smoking on exposure to nicotine, carbon monoxide, and acrolein. Cancer Prev Res (Phila). 2015; 8: 873–878
  4. Phillips, LD. Decision conferencing. in: W Edwards, RF Miles Jr, D von Winterfeldt (Eds.) Advances in decision analysis: from foundations to applications. Cambridge University Press, Cambridge; 2007
  5. UK Advisory Council on the Misuse of Drugs. Consideration of the use of multi-criteria decision analysis in drug harm decision making. Home Office, London; 2010
  6. Nutt, DJ, King, LA, Phillips, LD, and on behalf of the Independent Scientific Committee on Drugs. Drug harms in the UK: a multicriteria decision analysis. Lancet. 2010; 376: 1558–1565
  7. van Amsterdam, J, Nutt, D, Phillips, L, and van den Brink, W. European rating of drug harms. J Pharmacol. 2015; 29: 655–660
  8. Dodgson, J, Spackman, M, Pearman, A, and Phillips, L. Multi-criteria analysis: a manual. UK Department of the Environment, Transport and the Regions, London; 2009

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s