Efficacy of an Internet-based,individually tailored smoking cessation program: A randomized-controlled trial

Abstract

Introduction

Self-help computer-based programs are easily accessible and cost-effective interventions with a great recruitment potential. However, each program is different and results of meta-analyses may not apply to each new program; therefore, evaluations of new programs are warranted. The aim of this study was to assess the marginal efficacy of a computer-based, individually tailored program (the Coach) over and above the use of a comprehensive Internet smoking cessation website.

Methods

A two-group randomized controlled trial was conducted. The control group only accessed the website, whereas the intervention group received the Coach in addition. Follow-up was conducted by e-mail after three and six months (self-administrated questionnaires). Of 1120 participants, 579 (51.7%) responded after three months and 436 (38.9%) after six months. The primary outcome was self-reported smoking abstinence over four weeks.

Results

Counting dropouts as smokers, there were no statistically significant differences between intervention and control groups in smoking cessation rates after three months (20.2% vs. 17.5%, p = 0.25, odds ratio (OR) = 1.20) and six months (17% vs. 15.5%, p = 0.52, OR = 1.12). Excluding dropouts from the analysis, there were statistically significant differences after three months (42% vs. 31.6%, p = 0.01, OR = 1.57), but not after six months (46.1% vs. 37.8%, p = 0.081, OR = 1.41). The program also significantly increased motivation to quit after three months and self-efficacy after three and six months.

Discussion

An individually tailored program delivered via the Internet and by e-mail in addition to a smoking cessation website did not significantly increase smoking cessation rates, but it increased motivation to quit and self-efficacy.

Keywords

Tobacco smoking smoking cessation Internet computer program randomized controlled trial

Introduction

Self-help computer- or Internet-based programs are accessible and cost-effective interventions with a great recruitment potential.^1,2 This makes them a valuable method for enhancing the use of support during quit attempts. Meta-analyses have shown the added value of self-help materials compared to no intervention.³ Moreover, individually tailored materials are more effective than standardized self-help materials.^3–5

Internet-based programs are low-threshold interventions and are mostly free of charge, accessible and easily adjustable.⁶ They can be especially effective among smokers with low socioeconomic status.⁷ Furthermore, the medium allows for multiple exposures to preventive messages over time.⁸ Some repetitiveness and omnipresence can be achieved through these interventions.⁹ They allow users to receive messages in a timely manner¹⁰ in different settings and relapse situations.^11,12 Some recent reviews and meta-analyses indicate that self-help interventions based on new technologies are promising large-scale interventions,^11,13 while other reviews suggest that such programs have not yet truly proven effective^14,15 or are effective to a small extent.^4,16

Consequently, the rationales for our study are as follows: first, as the question has not been definitively settled yet, new randomized trials contribute to this discussion. Second, this program is new and has to be evaluated and the results reported. Third, there is a specific need to study more precisely the effects of tailored components over standard smoking cessation websites, as we do here. In fact, Shahab and McEwen state, “More research is needed to evaluate the relative efficacy of interactive web-based interventions compared with static websites.”¹¹ Chen et al. also underlined, “Neither of the main reviews was able to determine, from the available evidence, what form electronic aids should take.”¹⁶

Accordingly, our contribution to the ongoing debate is to assess the added value, or marginal efficacy, of a new Internet-based, individually tailored smoking cessation program, the Coach, delivered in addition to an existing smoking cessation website, Stop-Tabac.ch. The interactive Coach is an original program that was developed by scientists from the University of Geneva based on theories of addiction and behaviour change. It is similar in principle to other existing programs described in the literature, but original in its content. We know of no equivalent program in French.

Stop-Tabac.ch and the Coach

The Stop-Tabac.ch website, created in 1997, provides free information and support for smoking cessation. It covers different aspects of tobacco addiction, such as risks, mechanisms of dependence, smoking substitutes, tips for quitting and relapse prevention. Information is provided through text pages, videos, booklets, discussion forums and testimonials. Most of its content is available in the three Swiss national languages (German, French and Italian). Stop-Tabac has been continuously adapted based on the scientific literature, evaluations and users’ feedback, but it was stable during the trial duration in order to allow the assessment.

In supplement to this comprehensive website, the new Coach program, developed by the same group, provides individualized counselling (information, encouragement, advice and follow-up) through personalized messages in French that are tailored to participants based on their answers to questionnaires. It accompanies participants through the process of smoking cessation and relapse prevention. It is targeted at both those who are motivated and those who are unmotivated to quit, as well as at current and former smokers. The program aims at enhancing participants’ knowledge (risks, benefits, addiction, withdrawal symptoms, relapse situations), skills (coping strategies, use of evidence-based treatments), motivation, self-efficacy and capacity for action (preparation for the quit attempt).

Both Stop-Tabac and the Coach are based on the transtheoretical model of behaviour change^17,18 and theories of relapse prevention¹⁹ and tobacco dependence.²⁰ Both are non-commercial programs. In the present trial, we assessed the incremental effect of the Coach tailored program over and above the Stop-Tabac website.

Methods

Trial design

Through a randomized controlled trial, we compared a control group that had access only to the Stop-Tabac website with an intervention group that additionally benefited from the Coach individually tailored program. The duration of the Coach intervention for the intervention group, as well as the duration of the control, was six months from enrolment. Participants in the control group were informed that they could access the Coach program after the trial.

Recruitment and participants

Participants were recruited between March 2012 and March 2013 from among spontaneous visitors of Stop-Tabac.ch. The experiment was not advertised elsewhere. The eligibility criteria for participating in the trial were: be a current or ex-smoker; be 18 years or older; provide valid e-mail and postal addresses and a phone number; and provide informed consent. We included former smokers because the program also targets them (consolidation of quit attempts). On the website page labelled ‘Coach’ (readable from computers, not from tablets or mobile phones), visitors were invited to join the study and informed about the aims and procedures of the trial, including randomization and the possibility to be assigned to a control group receiving no additional intervention than access to the basic website. Thus, the participants were not blind to treatment conditions. No incentive was offered for participation.

Randomization

A list of random numbers was used. Participants were assigned automatically by a computer either to the intervention group that received the program Coach or to the control condition, which did not.

Questionnaires

All eligible participants completed the same questionnaires: a baseline questionnaire upon entry into the study (T-0) and follow-up questionnaires three and six months after enrolment (T-3 and T-6). The answers were used both to conduct the trial and to tailor the feedback elements of the Coach for the intervention group.

The first part of the questionnaire was used at T-0, T-3 and T-6. It covered sociodemographics, smoking status (smoker/ex-smoker, wants to quit progressively or abruptly, quit date) and level of addiction (cigarettes per day, minutes to first cigarette of the day). It also covered attitudes toward smoking, including motivation to quit, withdrawal symptoms, use of self-change strategies and self-efficacy (i.e. confidence in ability to refrain in typical relapse situations). Motivation to quit was measured by five items (e.g. “Cigarettes will damage my health”), withdrawal symptoms by eight items (e.g. “I am feeling anxious”), self-change strategies by five items (e.g. “In order to refrain from smoking, I avoid places where people smoke”) and self-efficacy by seven items (e.g. “I am able to refrain from smoking after a meal”). The items could be answered on a five-point scale with options ranging from never to always (or from “totally disagree” to “fully agree”). These items were drawn from validated scales. Their test–retest reliability, internal consistency, sensitivity to change and predictive validity were assessed extensively. We used or selected items from the Cigarette Dependence Scale, the Cigarette Withdrawal Scale, the Attitudes Towards Smoking Scale, the Smoking Self-Efficacy Scale, the Self-Change Strategies Scale and the Attitudes Towards Nicotine Replacement Therapy (NRT) Scale, and we also developed several questions specifically for the Coach.^21–26

The second part of the questionnaire concerned the use of smoking cessation aids. We firstly assessed the use of NRT and smoking cessation medications at the three time points. We secondly assessed at T-3 and T-6 the self-reported use of other smoking cessation aids in the two groups (Stop-Tabac website, other websites, physician, smoking cessation clinics, support group, quitline).

In addition to the self-assessment questionnaire, we tracked the frequency of use of the Coach in the intervention group within our system (visits to the personal page) to determine whether the program was actually used by the participants.²⁷

Treatment conditions

Only the intervention group received the Coach intervention. Both groups could access the Stop-Tabac website. All participants were allowed to use smoking cessation aids available elsewhere (websites, quitlines, NRT, etc.).

The Coach program consisted of the three following elements:

A series of automatic, personalized feedback reports that were assembled by the computer based on the participant’s answers to the tailoring questionnaire. Each report sent to the participants consisted of 30 feedback items selected automatically from a stockpile of over 300 items. These items included paragraphs of text, images and graphs showing the respondents’ scores. Different answers to the tailoring questions produced different paragraphs in the feedback reports. A specific set of decision rules was created to link participants’ answers to the questionnaires with the library items. To select the feedback items from the library, we used cutpoints for each variable or multi-item scale. The questionnaires, feedback items, decision rules and algorithms were developed by the research team based on its clinical and research experience and on pretests with smokers. The decision rules and algorithms were empirical and were not formally validated. Participants received a progress report for each of the three answered questionnaires.

A personal web page with progress graphs, for a visual representation of change over time in the levels of tobacco dependence, withdrawal symptoms, motivation and self-efficacy.

A series of automatic, individually tailored, proactive e-mail messages that took into account each participant’s smoking status, quit date (past or future) and level of dependence. These generic statements were also selected from the stockpile of items. The number of messages depended on the content and frequency of participant reactions to the program as reported in the personal page (smoking status and potential relapses) and on their quit date (depending on how close it was).

The hypothesis being tested was whether the Coach program increased smoking cessation rates in addition to the effects of the Stop-Tabac website.

Instruments

At baseline and after three and six months, participants were invited to answer online questionnaires. Up to six reminders were sent by e-mail to non-respondents, who were not contacted through other means.

Outcome measure

The primary outcome was self-reported one-month smoking abstinence (not a puff of tobacco during the past four weeks). No biochemical validation was performed.

Statistical analysis

Regarding the primary outcome, we conducted both intention-to-treat analyses (ITT), in which non-respondents at follow-up were considered smokers, and analyses including only the respondents. We used t-tests to compare means, Wilcoxon tests to compare medians and Fisher tests to compare proportions. We regressed the number of visits to the program on the outcome for determining the effect of the program’s intensity of use on the chance of quitting. When relevant, ex-smokers were analysed separately from baseline smokers.

Results

Study population

The number of participants who registered in the study was 1226. Of these, 1160 participants were randomized because 66 persons did not complete the registration process. Among these 1160 participants, 40 were eliminated (11 because of double registrations and 29 because they did not complete the baseline questionnaire). The analysis was based on the remaining 1120 participants (intervention group: 559; control condition: 561) (Figure 1).

Figure 1.

Consort diagram Coach program – randomized controlled trial.

The response rate was 51.7% (579/1120) after three and 38.9% (436/1120) after six months. Response rates differed significantly between the intervention and control groups at the three-month follow-up (but not at six months): 48.1% in the intervention group and 55.3% in the control group (odds ration (OR) = 1.33; 95% confidence interval (CI) 1.05–1.70). Of the baseline characteristics, age, use of NRT and gender predicted attrition: at the two follow-ups, older participants were more likely to respond (three months: mean age = 38.1 vs. 34.7, P < 0.0001; six months: mean age = 38.8 vs. 35, P < 0.0001), and participants using NRT were more likely to respond (three months: OR = 3.46 (95% CI 2.60–4.63); six months: OR = 1.82 (95% CI 1.40–2.38)). Women were less likely to respond than men at six months (OR = 0.61; 95% CI 0.47–0.81). None of these predictors of dropout differed significantly between groups.

Baseline characteristics

The mean age of participants was 36.5 years, and 65.7% were women (736/1120). Most baseline participants (94.6%, 1060/1120) were current smokers, and 5.4% (60/1120) were former smokers. There were no significant differences at baseline between intervention and control groups regarding age, gender, level of dependence, proportion of baseline smokers, attitudes towards smoking, motivation or use of NRT and medications (Table 1).

Table 1.

Participant characteristics at baseline – no statistically significant differences.

	Control (n = 561)	n missing	Intervention (n = 559)	n missing
Female	66.5% (372)	2	65.2% (364)	1
Mean age (SD)^a	36.8 (11.2)	71	36.1 (10.4)	62
Smoker	93.4% (524)	0	95.9% (536)	0
Mean cigarettes per day (SD)^a	17.5 (8.1)	1	17.2 (8.4)	1
Median time from waking (quartile)^b	15 (5, 45)	8	15 (5, 45)	9
Using bupropion medication	2.6% (14)	12	3.1% (17)	12
Using nicotine replacement therapy	32.2% (180)	2	30.2% (168)	2
Mean Motivation, 1–5 score (SD)^c	4.7 (0.6)	1	4.7 (0.6)	0
Mean Withdrawal, 1–5 score (SD)^c	4.0 (1.1)	1	3.9 (1.1)	4
Mean Strategy, 1–5 score (SD)^c	2.8 (1.1)	1	2.8 (1.0)	7
Mean Self-efficacy, 1–5 score (SD)^c	2.8 (1.1)	0	2.8 (1.1)	0

Standard deviation.

1st quartile, 3rd quartile.

Value was computed by taking the mean of all the rank-ordered scores (1–5) of the items belonging to one variable.

Use of the intervention

In the intervention group, 25.2% (141/559) of participants connected to their personal page only once (i.e. at registration). The median number of connections to the personal page was three, and the median number of e-mail messages received was 47 per person. In the intervention group, the intensity of use of the program was associated with quitting smoking at six months: quitters connected to the program nine times on average, compared with three times for those still smoking (Wilcoxon test [W] = 32808, P < 0.0001).

Main outcome

For the 60 baseline ex-smokers, there was no difference between intervention and control groups regarding the main outcome. Considering all participants (baseline and ex-smokers) in the ITT analysis (i.e. non-respondents were considered as still smoking), 20.2% (113/559) of the intervention group was abstinent at three months, versus 17.5% (98/561) in the control group. This difference was not statistically significant (OR = 1.20; 95% CI 0.88–1.64). The six-month difference was not significant either, with 17% (95/559) abstinence in the intervention group versus 15.5% (87/561) in the control group (OR = 1.12; 95% CI 0.80–1.55). Considering only baseline smokers, there was also no significant difference (Table 2). When analysing the data without dropouts (i.e. only the respondents), the differences between intervention and control groups were statistically significant after three months, whether including all participants (OR = 1.57; 95% CI 1.10–2.23) or only baseline smokers (OR = 1.58; 95% CI 1.09–2.30). After six months, the difference was not significant (Table 3).

Table 2.

Self-reported smoking abstinence (no puff in the previous four weeks) at three- and six-month follow-ups – no statistically significant differences (intention-to-treat analysis).

	Control	Intervention	OR	95% CI	P-value
Including all participants randomized to control or intervention groups
n	561	559
3 months^a	17.5% (98)	20.2% (113)	1.20	0.88–1.64	0.252
6 months^a	15.5% (87)	17.0% (95)	1.12	0.80–1.55	0.518

Including only baseline smokers
n	524	536
3 months^a	15.7% (82)	18.3% (98)	1.21	0.86–1.69	0.288
6 months^a	13.7% (72)	15.1% (81)	1.12	0.78–1.60	0.542

In brackets: number of quitters.

Table 3.

Self-reported smoking abstinence (no puff in the previous four weeks) at three- and six-month follow-ups – statistically significant differences at three-month follow-up (analysis without dropouts).

	Control	Intervention	OR	95% CI	P-value
Including all participants randomized to control or intervention groups
n	561	559
3 months^a	31.6% (98)	42.0% (113)	1.57	1.10–2.23	0.012*
6 months^a	37.8% (87)	46.1% (95)	1.41	0.94–2.10	0.081
Including only baseline smokers
n	524	536
3 months^a	28.6% (82)	38.7% (98)	1.58	1.09–2.30	0.014*
6 months^a	34.1% (72)	42.9% (81)	1.45	0.95–2.22	0.080

In brackets: number of quitters.

p < 0.05.

p < 0.01.

***

p < 0.001.

Use of other cessation aids

At three and six months, there were no statistically significant differences between intervention and control groups regarding the use of other cessation aids. Considering all trial participants, the reported use of NRT at three (OR = 2.59; 95% CI 1.63–4.15) and six months (OR = 2.30; 95% CI 1.32–4.07) was associated with higher quit rates, which either captures the efficacy of NRT or the fact that its use is a sign of high motivation. Similarly, the reported use of any other behavioural cessation aids at three months was associated with higher quit rates (both groups merged: OR = 1.79; 95% CI 1.09–2.96) (Table 4).

Table 4.

Reported use of cessation aids at three- and six-month follow-ups – no statistically significant differences.

	Control	Intervention	OR (95% CI)	P-value
3 months
n	310	269
Nicotine therapy	35.2%	43.1%	0.72 (0.45–1.13)	0.15
n missing	114	116
Behavioural support outside study	67.7%	74.2%	0.73 (0.44–1.19)	0.20
n missing	115	114
6 months
n	230	206
Nicotine therapy	35.6%	42.5%	0.75 (0.44–1.28)	0.30
n missing	84	93
Behavioural support outside study	66.7%	73.5%	0.72 (0.41–1.28)	0.28
n missing	83	93

Attitudes towards smoking

There were differences between the two groups for some of the attitudinal variables at the follow-ups (Table 5). In particular, after three months, self-efficacy and motivation to quit were higher in the intervention than in the control group (P = 0.0002 vs. P = 0.019), as well as self-efficacy after six months among baseline smokers.

Table 5.

Attitudes towards smoking at three- and six-month follow-ups (standard deviation in brackets).

	Control	Intervention	P-value
Including all participants
3 months
Motivation to quit	4.7 (0.8)	4.8 (0.5)	0.019*
Withdrawal symptoms	3.5 (1.0)	3.3 (1.1)	0.0496*
Self-change strategies	2.7 (1.0)	2.7 (1.1)	0.775
Self-efficacy	3.4 (1.3)	4.0 (1.3)	0.0002***
6 months
Motivation to quit	4.8 (0.5)	4.8 (0.4)	0.122
Withdrawal symptoms	3.4 (1.1)	3.3 (1.2)	0.354
Self-change strategies	2.5 (1.1)	2.5 (1.1)	0.826
Self-efficacy	3.7 (1.3)	4.0 (1.2)	0.051
Including only baseline smokers
3 months
Motivation to quit	4.6 (0.8)	4.8 (0.5)	0.012*
Withdrawal symptoms	3.4 (1.0)	3.2 (1.1)	0.0440*
Self-change strategies	2.6 (1.0)	2.7 (1.1)	0.66
Self-efficacy	3.4 (1.3)	4.0 (1.3)	0.0002***
6 months
Motivation to quit	4.7 (0.5)	4.8 (0.4)	0.110
Withdrawal symptoms	3.4 (1.1)	3.2 (1.2)	0.214
Self-change strategies	2.5 (1.1)	2.5 (1.1)	0.841
Self-efficacy	3.6 (1.3)	4.0 (1.2)	0.040*

p < 0.05.

p < 0.01.

***

p < 0.001.

Discussion

Marginal effect of the Coach

We found that an Internet-based, individually tailored smoking cessation program Coach delivered over the Stop-Tabac website did not increase quit rates, but significantly increased self-efficacy and motivation to quit smoking at three months among the whole intervention group, as well as self-efficacy after six months among baseline smokers. The effects on self-efficacy and motivation confirm the capacities of these programs to empower users regarding attitudes towards smoking.

Participants of both groups were spontaneous visitors of the Stop-Tabac website, and were thus already motivated to use Internet-based tools, which might have reduced the likelihood of finding differences. Moreover, Stop-Tabac is already effective in itself: an earlier version of Stop-Tabac (tailored materials by postal mail) produced a 2.6 times greater cessation rate in the intervention group than in the control group which did not receive the intervention.²⁸ In the present trial, the effectiveness of Stop-Tabac (accessible to all participants) might have reduced the differences between both groups. This mitigated result could also be due to the sample size. To detect an OR of 1.35 with a power of 80% at a significance level of 0.05, a sample of 4000 participants would have been needed. Statistically significant results might have been found with a larger sample size.

Strengths and limitations of the study

A first strength of our study was to assess the program via severe criteria: four-week abstinence and six-month follow-up. A second strength was to measure the frequency of use of the program. More frequent use of the Coach was associated with higher quit rates. This might capture the effect of the program or might reflect the fact that more motivated participants were more likely to use the program and succeed in quitting. A third strength was to test the use of other smoking cessation aids. This allowed to know whether there was a disequilibrium between both groups (e.g. the control group could have more actively searched for cessation aids elsewhere), which was not the case.

The sample size (1120 participants) constitutes one limitation of the study. Another limit was the follow-up response rate. High dropout rates often occur in Internet studies and can pose problems of generalizability.²⁹ However, our response rate was in line with similar studies using web-based follow-ups.^30,31 Eysenbach²⁷ states that web-based trials can lose “as many as about half of the patients”. This phenomenon is typical of trials based on low-intensity mediums.³² Finally, the proportion of participants who used the Coach only once was high (25.2%), which decreased its impact. This demonstrates the necessity of systematically assessing participation in low-intensity interventions and of elaborating programs capable of retaining participants and enhancing usage rate.³³

Conclusions

In ITT analysis (i.e. considering dropouts as smokers), our study did not find evidence of the effectiveness of an individually tailored computer intervention (the Coach) delivered over a comprehensive smoking cessation website (Stop-Tabac). It nevertheless showed that the Coach increased self-efficacy and motivation to quit after three months, and increased self-efficacy after six months. Moreover, non-ITT quit rates (i.e. including only the respondents) showed significant difference between intervention and control groups at three months (but not at six months), which suggests that both self-efficacy and motivation are crucial for helpful support in smoking cessation and that programs should not neglect one of these two variables over the medium-term.

Our results provide elements to the ongoing debate on the efficacy of Internet-based programs. While reviews and meta-analyses indicate that self-help interventions based on new technologies are promising,^11,13 authors have recently raised the question of which form electronic aids should take.^11,16 Hence, the question of the combination of different components is crucial. Our study detected a difference when delivering a tailored program over a comprehensive smoking cessation website, but only in non-ITT analysis and as far as the program achieved enhancement of both motivation and self-efficacy among participants. Because of the lack of unanimity about the efficacy of the various forms of Internet-based interventions^14,15 and their combinations, new programs such as the Coach still have to be systematically assessed and the results shared for inclusion in future meta-analyses.

Footnotes

Acknowledgement

The authors want to thank Vincent Baujard of the Health On the Net Foundation, who developed the software for data collection.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Tobacco Control Fund of the Swiss Federal Office of Public Health (grant number 10.003634). The Tobacco Control Fund had no role in the study design, collection, analysis or interpretation of the data, writing of the manuscript or the decision to submit the manuscript for publication. The authors have received the financial support only for the research of this article.

References

Brown

Michie

Geraghty

AWA

. A pilot study of StopAdvisor: a theory-based interactive internet-based smoking cessation intervention aimed across the social spectrum. Addict Behav 2012; 37: 1365–1370.

Graham

Chang

Fang

. Cost-effectiveness of internet and telephone treatment for smoking cessation: an economic evaluation of the iQUITT study. Tob Control 2013; 22: e11–e11.

Lancaster

Stead

Silagy

. Effectiveness of interventions to help people stop smoking: findings from the Cochrane Library. BMJ 2000; 321: 355–358.

Civljak

Sheikh

Stead

. Internet-based interventions for smoking cessation. Cochrane Database Syst Rev 2010; 9: CD007078–CD007078.

Lancaster

Stead

. Self-help interventions for smoking cessation. Cochrane Database Syst Rev 2005; 3: CD001118–CD001118.

Pulverman

Yellowlees

. Smart devices and a future of hybrid tobacco cessation programs. Telemed J E Health 2014; 20: 241–245.

Brown

Michie

Geraghty

AWA

. Internet-based intervention for smoking cessation (StopAdvisor) in people with low and high socioeconomic status: a randomised controlled trial. Lancet Respir Med 2014; 2: 997–1006.

Mason

Gilbert

Sutton

. Effectiveness of Web-based tailored smoking cessation advice reports (iQuit): a randomized trial. Addiction 2012; 107: 2183–2190.

Wells

Forde

. A new frontier in health technology: the role of SMS text-based messaging as a smoking cessation intervention. Univ Toronto Med J 2011; 88: 135–140.

10.

Brendryen

Drozd

Kraft

. A digital smoking cessation program delivered through Internet and cell phone without nicotine replacement (Happy Ending): randomized controlled trial. J Med Internet Res 2008; 10: e51–e51.

11.

Shahab

McEwen

. Online support for smoking cessation: a systematic review of the literature. Addiction 2009; 104: 1792–1804.

12.

Free

Whittaker

Knight

. Txt2stop: a pilot randomised controlled trial of mobile phone-based smoking cessation support. Tob Control 2009; 18: 88–91.

13.

Myung

McDonnell

Kazinets

. Effects of Web- and computer-based smoking cessation programs: meta-analysis of randomized controlled trials. Arch Intern Med 2009; 169: 929–937.

14.

Walters

Wright

Shegog

. A review of computer and Internet-based interventions for smoking behavior. Addict Behav 2006; 31: 264–277.

15.

Hutton

Wilson

Apelberg

. A systematic review of randomized controlled trials: Web-based interventions for smoking cessation among adolescents, college students and adults. Nicotine Tob Res 2011; 13: 227–238.

16.

Chen

Madan

Welton

. Effectiveness and cost-effectiveness of computer and other electronic aids for smoking cessation: a systematic review and network meta-analysis. Health Technol Assess 2012; 16: 1–205.

17.

DiClemente

Prochaska

Fairhurst

. The process of smoking cessation: an analysis of precontemplation, contemplation, and preparation stages of change. J Consult Clin Psychol 1991; 59: 295–304.

18.

Etter

Sutton

. Assessing “stage of change” in current and former smokers. Addiction 2002; 97: 1171–1182.

19.

Marlatt

Gordon

. Relapse prevention: maintenance strategies in the treatment of addictive behaviors, New York: Guilford Press, 1985.

20.

Benowitz

. Cigarette smoking and nicotine addiction. Med Clin North Am 1992; 76: 415–437.

21.

Etter

Le Houezec

Perneger

. A self-administered questionnaire to measure dependence on cigarettes: the Cigarette Dependence Scale. Neuropsychopharmacol 2003; 28: 359–370.

22.

Etter

. A self-administered questionnaire to measure cigarette withdrawal symptoms: the Cigarette Withdrawal Scale. Nicotine Tob Res 2005; 7: 47–57.

23.

Etter

Humair

Bergman

. Development and validation of the attitudes towards smoking scale (ATS-18). Addiction 2000; 95: 613–625.

24.

Etter

Bergman

Humair

. Development and validation of a scale measuring self-efficacy of current and former smokers. Addiction 2000; 95: 901–913.

25.

Etter

Bergman

Perneger

. On quitting smoking: development of two scales measuring the use of self-change strategies in current and former smokers (SCS-CS and SCS-FS). Addict Behav 2000; 25: 523–538.

26.

Etter

Perneger

. Attitudes towards nicotine replacement therapy in smokers and ex-smokers in the general public. Clin Pharmacol Ther 2001; 69: 175–183.

27.

Eysenbach

. Issues in evaluating health websites in an Internet-based randomized controlled trial. J Med Internet Res 2002; 4: e17–e17.

28.

Etter

Perneger

. Effectiveness of a computer-tailored smoking cessation program: a randomized trial. Arch Intern Med 2001; 161: 2596–2601.

29.

Cobb

Graham

Bock

. Initial evaluation of a real-world Internet smoking cessation system. Nicotine Tob Res 2005; 7: 207–216.

30.

Saul

Schillo

Evered

. Impact of a statewide Internet-based tobacco cessation intervention. J Med Internet Res 2007; 9: e28–e28.

31.

Strecher

Shiffman

West

. Randomized controlled trial of a Web-based computer-tailored smoking cessation program as a supplement to nicotine patch therapy. Addiction 2005; 100: 682–688.

32.

Eysenbach

. The law of attrition. J Med Internet Res 2005; 7: e11–e11.

33.

Balmford

Borland

Benda

. Patterns of use of an automated interactive personalized coaching program for smoking cessation. J Med Internet Res 2008; 10: e54–e54.