ESTIMATING THE POPULATION IMPACTOF E-CIGARETTES ON SMOKINGCESSATION IN ENGLAND

An important consideration when assessing the publichealth impact of e-cigarettes is how far they contributeto, or detract from, smoking cessation in the population.There has been speculation about this [1], but withoutadequate data to inform it. England has data that can helpto address this question, at least so far. Addiction’s readersmay be interested in the following analysis. It focuses on2014, the most recent year for which full data areavailable. It leads to an estimate of 16 000–22 000 asthe number of additional long-term quitters generated bye-cigarettes in that year.

ESTIMATION

1. At the start of 2014 there were approximately 8.46million adult smokers in England (19.3% of 43.83million people aged 16+) [2].

2. The percentage of smokers in 2014 who reported thatthey had tried to stop at least once is estimated at37.3% (3.16 million people) [3].

3. The percentage of those who tried to quit who used ane-cigarette (and not a prescription medicine or behav-ioural support) in 2014was 28.2% (891 000 people) [3].

4. The expected long-term (1 year) success rates of a quitattempt made without assistance or using a licensednicotine product (LNP) bought from a shop is approxi-mately 5% [4,5]. Note that in England no benefit hasbeen found for LNPs bought from a shop, whereas theyhave been associated with increased success rateswhen accompanied by at least some professional sup-port [5,6].

5. Evidence from RCTs and from surveys in England indi-cate that using an e-cigarette in a quit attempt increasesthe probability of success on average by approximately50% compared with using no aid or LNP bought froma shop—similar to use of a licensed medicine withlimited behavioural support but less than medicationplus specialist behavioural support [6,7].

6. Therefore, it is estimated that 2.5% of the smokers whoused an e-cigarette in their quit attempt in England(22 000 people) succeeded who would have failed ifthey had used nothing or LNP bought from a shop.

7. As e-cigarette usage has increased, use of prescriptionstop-smoking medications and specialist behaviouralsupport has decreased [3]. The decline in these methodsof stopping since e-cigarettes started to become popularis approximately 10% of quit attempts, which represents3.7% of smokers in 2014 (313 000 smokers). Thetrajectories of the declines have not mirrored theincrease in e-cigarette use, so there may be no

connection. However, we consider that an upper esti-mate for the contribution of e-cigarettes to that declineis 80%, which represents 250 000 smokers(313 000 × 0.8).

8. Therefore, if e-cigarettes have detracted from the useof methods of stopping that are equally or more effec-tive, the net increase in smokers using a method ofstopping yielding an approximately 50% increase inlong-term success is approximately 630 000 people(891 000–250 000). The net number estimated tohave quit in England during 2014 who would nothave quit if e-cigarettes had not been available wouldtherefore be 16 000 (630 000 × 0.025).

COMMENTS AND CAVEATS

1. Estimated prevalence of e-cigarette use in a quit at-tempt as a proportion of all smokers is subject to 95%confidence intervals of ±1%.

2. There have been highly publicized studies purporting tohave found that e-cigarettes promote uptake of ciga-rettes, or are taken up in substantial numbers, by peoplewho would not have smoked and that this outweighsany impact of e-cigarettes on quitting. These claims areundermined by highly plausible alternative explanations[8]. In England and the United States, the evidence thusfar contradicts the hypothesis: regular use of e-cigarettesby never smokers is extremely rare and the decline insmoking prevalence in young people has been as greator greater than in previous years [8].

3. Our estimate does not take account of any effect ofe-cigarettes on the incidence of quit attempts. Sincee-cigarettes became popular the incidence of quitattempts has risen and then fallen again [3], so acausal connection is unlikely.

4. It has been proposed that using an e-cigarette while con-tinuing to smoke may reduce subsequent quitting sothat, even if using an e-cigarette in a quit attemptincreased the chances of success of that attempt, thenet effect of having e-cigarettes on the market hasreduced quitting [9]. However, smokers who use e-ciga-rettes may have a lower pre-existing ability to stopsmoking [10]. If the proposal were correct, one wouldexpect a reduction in population quitting rates as dualuse of e-cigarettes among smokers increased, whereasin England the overall rate of smoking cessation in2014was higher than in anyof the previous 7 years [3].

5. It is possible that smokers who quit with the aid of ane-cigarette may be at greater risk of longer-term relapseto smoking. However, it is also possible that they are atlower or similar risk. This is an issue that requiresfurther study.

6. The figures relate to the population as a whole, notindividual smokers.

1118 Letters to the Editor

© 2016 Society for the Study of Addiction Addiction, 111, 1114–1119

7. No differentiation can be made between different typesof e-cigarette because of the lack of trial and populationlevel data on relative effectiveness and usage.

8. For smokers who used more than one method of quit-ting in their quit attempt, we counted the method thatthe evidence indicated would be most effective [6].

9. The figure of approximately 16 000–22 000 is muchlower than the population estimates of e-cigarette userswho have stopped smoking (approximately 560 000 inEngland at the last count, according to the SmokingToolkit Study). However, the reason for this can beunderstood from the following:a. Only some e-cigarette users who have stopped

smoking will have done so in the past year; 3.6%(252 000) of the 7.01 million long-term ex-smokers(16% of adult population in 2014) used e-cigarettes,according to the Smoking Toolkit Study (leaving308 000).

b. Some 9% (28 000) of the 3.11 million recent ex-smokers (according to the Smoking Toolkit Study)started their e-cigarette use after they stoppedsmoking, possibly to avoid relapse to smoking (leav-ing 280 000).

c. It has to be assumed on the basis of the evidence[6,7] that only a third of e-cigarette users whostopped smoking would not have succeeded hadthey used no cessation aid (leaving 93 000).

d. It is assumed that, as with other smoking cessationaids, 70% of those recent ex-smokers who use e-cig-arettes will relapse to smoking in the long term [11](leaving 28 000).

e. Some people (estimated at 6000, based on the cal-culations in parts 7 and 8 of our estimate) whostopped smoking with the aid of an e-cigarettemay otherwise have used a prescription medicineand/or behavioural support (leaving 22 000).

f. Therefore, by this alternative method, the range is22 000–28 000, which is only slightly higher thanthe 16 000–22 000 estimated earlier.

g. It is, of course, important to appreciate that esti-mates of the numbers of e-cigarette users are subjectto quite wide margins of error. Nevertheless, thepopulation figure for numbers of additional ex-smokers generated by e-cigarettes in 2014 inEngland appears to be in the tens of thousands.

Declaration of interests

R.W. undertakes research and consultancy for companiesthat market smoking cessation medicines but not e-ciga-rettes. He is honorary adviser to the UK’s national Centrefor Smoking Cessation and Training. His salary is funded

by Cancer Research UK. J.B. received an unrestricted re-search grant from Pfizer in 2012 to study smoking ces-sation trends in England. L.S. has received a researchgrant, an honorarium for a talk and travel expenses toattend meetings and workshops from a pharmaceuticalcompany that makes smoking cessation products (note-cigarettes).

Acknowledgements

R.W. is funded by Cancer Research UK. J.B. is funded by theSociety for the Study of Addiction.

ROBERT WEST, LION SHAHAB & JAMIE BROWN

Department of Epidemiology and Public Health

University College London, London, UK

E-mail: robertwest100@gmail.com

References

1. Kalkhoran S., Glantz S. A. Modeling the Health Effects ofExpanding e-Cigarette Sales in the United States and UnitedKingdom: A Monte Carlo Analysis. JAMA Inter Med 2015Oct; 175: 1671–80.

2. West R. Smoking Pipe Model 2014. www.smokinginenglandinfo/sts-documents/. 2015. [Accessed on 20 February 2016]

3. West R, Brown J. Smoking in England 2007–2014. www.smokinginenglandinfo/sts-documents/. 2015. [Accessed on20 February 2016]

4. Hughes J. R., Keely J., Naud S. Shape of the relapse curve andlong-term abstinence among untreated smokers. Addiction2004; 99: 29–38.

5. Kotz D., Brown J., West R. ‘Real-world’ effectiveness ofsmoking cessation treatments: a population study. Addiction2014; 109: 491–9.

6. Brown J., Beard E., Kotz D., Michie S., West R. Real-worldeffectiveness of e-cigarettes when used to aid smoking cessa-tion: a cross-sectional population study. Addiction 2014;109: 1531–40.

7. McRobbie H., Bullen C., Hartmann-Boyce J., Hajek P. Elec-tronic cigarettes for smoking cessation and reduction.Cochrane Database Syst Rev 2014; 12: CD010216.

8. McNeill A., Brose L., Calder R., Hitchman S., Hajek P.,McRobbie H.. Electronic Cigarettes: An Evidence Update.London: Public Health England; 2105.

9. Kalkhoran S., Glantz S. A. E-cigarettes and smoking cessationin real-world and clinical settings: a systematic review andmeta-analysis. Lancet Respir Med 2016 Feb; 4: 116–28.

10. Hajek P., Etter J. F., Benowitz N., Eissenberg T., McRobbie H.Electronic cigarettes: review of use, content, safety, effects onsmokers and potential for harm and benefit. Addiction2014; 109: 1801–10.

11. Stapleton J. A., West R. A direct method and ICER tables forthe estimation of the cost-effectiveness of smoking cessationinterventions in general populations: application to a newcytisine trial and other examples. Nicotine Tob Res 2012; 14:463–71.

Letters to the Editor 1119

© 2016 Society for the Study of Addiction Addiction, 111, 1114–1119