Abstract

Introduction: Web-based treatments can deliver broad reaching, relatively inexpensive, and clinically tested methods for smoking cessation. We performed a systematic review of randomized controlled trials (RCTs) of smoking cessation to evaluate the efficacy of Web-based interventions in adults, college students, and adolescents.

Methods: MEDLINE, EMBASE, The Cochrane Library, CINAHL, and PsycINFO were searched from January 1, 1990 through February 12, 2010 for RCTs examining the efficacy of Web-based smoking cessation programs. Paired reviewers abstracted data on study design, patient characteristics, and outcomes sequentially and did quality assessments independently.

Results: Twenty-one RCTs met eligibility criteria, with 15 conducted among adults. Among adults, 2 RCTs found that a multicomponent intervention with Web and non-Web–based elements was more efficacious than a self-help manual, and one of 2 RCTs found that Web-based interventions may be more effective than no treatment. Three trials provided insufficient evidence to demonstrate whether Web-based interventions were more efficacious than counseling. By contrast, tailored Web sites in 2 RCTs and greater Web site exposure in 6 of 7 RCTs were associated with higher rates of abstinence. Among college students, evidence supporting use of Web-based interventions was insufficient because the one RCT conducted was also a multicomponent intervention. Five RCTs among adolescents demonstrated mixed results, with insufficient evidence supporting their efficacy.

Conclusions: Evidence supporting the use of Web-based interventions for smoking cessation is insufficient to moderate in adults and insufficient in college students and adolescents. These RCTs have, however, elucidated clinical, methodological, and statistical practices that are likely to improve future trial design and treatment delivery.

Introduction

Tobacco smoking is a leading global public health problem. More than 1 billion people in the world smoke, with more 80% living in low- and middle-income countries (Jha, Ranson, Nguyen, & Yach, 2002). In 2005, over 5 million deaths were attributed to tobacco (Mathers & Loncar, 2006), with about half occurring in low- and middle-income countries (Ezzati & Lopez, 2004). Effective tobacco control efforts are needed to reduce tobacco-related morbidity and mortality. Web-based treatments can deliver broad reaching, relatively inexpensive, and clinically tested methods for smoking cessation. A systematic review of 22 computer and Web-based treatments concluded that there was sufficient clinical evidence to support the use of both these modalities for adult smokers (Myung, McDonnell, Kazinets, Seo, & Moskowitz, 2009). Byron and Cobb (2009), however, commented that the studies were too heterogeneous to conclude which components (other than tailoring) contributed to success, how they should be delivered, or to what populations. A second systematic review of 11 interactive Web-based randomized controlled trials (RCTs) from 1990 to 2008 concluded that smoking cessation interventions that are tailored and completely automated can be effective among smokers motivated to quit. Trial heterogeneity and unresolved methodological issues, such as specifying intervention reach, representativeness of participants, theoretical underpinnings of the interventions, and recruiting/retaining participants, limited conclusions about the efficacy of these Web-based interventions (Shahab & McEwen, 2009).

We performed a systematic review of RCTs to specifically evaluate the efficacy of Web-based interventions in adults, college students, and adolescents. Our systematic review expands on the review of Shahab & McEwen (2009) by examining type of Web site use (log-ons and dose–response characteristics) and type of Web sites that contribute to cessation efficacy. Our review through 2009 also identifies an additional 10 RCTs (5 in adolescents) published in the same time period, a variability in article detection that Rosen and Ben (2010) observed in their review of tobacco control systematic reviews.

Methods

This systematic review was part of a larger review evaluating the effectiveness of tobacco control measures among the general population (Wilson LM, Chander G, Hutton HE, Avila Tang E, Odelola O, Elf JE, et al. The Impact of Tobacco Control Interventions on Smoking Patterns: A Systematic Review. Prepared for the International Union Against Tuberculosis and Lung Disease, November 2009, Paris, France). PRISMA guidelines were followed in the reporting of this systematic review (Moher, Liberati, Tetzlaff, Altman, & the PRISMA Group, 2010).

Eligibility Criteria

We included RCTs if they tested the efficacy of a Web-delivered smoking cessation program and had a minimum of 1-month follow-up after intervention. Adolescents, college students, and adults of either gender from any setting or country were included. The primary outcome of interest was self-reported smoking cessation at the longest point of follow-up.

Information Sources and Search Strategy

We used electronic and hand-searching strategies based on medical subject headings and text words from key articles identified (see Table 1). We searched: MEDLINE, EMBASE, The Cochrane Library, CINAHL, and PsycINFO from January 1990 through December 2009. We also reviewed the list of references in eligible articles and the table of contents of 10 journals that published the highest number of included abstracts and articles.

Study Selection

Two team members independently evaluated each title, abstract, and article to identify eligible studies. We included articles in any language if both reviewers agreed that they were RCTs, evaluated the use of Web-based interventions, reported smoking cessation outcomes, had a minimum of 1-month follow-up after intervention, or were published since 1990. Eligible articles were then reviewed for data abstraction. We resolved differences of opinion through consensus or adjudication by a third reviewer if agreement could not be reached.

Data Collection Process

We created a standardized form for data extraction. A primary reviewer completed all data abstraction forms, and a second reviewer with clinical or methodological expertise checked the forms for completeness and accuracy. Reviewers were not masked to the articles’ authors, institutions, or journal.

Data Items, Data Entry, and Quality Assessment

For each study, we extracted data on specific aims; study characteristics; population characteristics, including age, gender, and race/ethnicity; the interventions and controls; definition of smoking cessation; and results. In addition to self-reported outcomes, biochemical validation and use of pharmacological aids were recorded if available.

For assessment of study quality, each reviewer independently rated quality using the Jadad criteria: (a) presence and quality of the randomization scheme, including concealment of allocation, (b) appropriateness of blinding, and (c) description of withdrawals and dropouts (Jadad et al., 1996). All information was entered into SRS 4.0, a Web-based database (TrialStat! Corporation, Ottawa, Ontario, Canada).

Data Synthesis and Analysis

Odds ratios (ORs) and 95% CIs at each follow-up period were either abstracted or calculated from study data. We contacted study authors (Munoz et al., 2006; Pike, Rabius, McAlister, & Geiger, 2007; Prochaska et al., 2008) to provide additional data to calculate ORs. We analyzed data on an intention-to-treat (ITT) basis and synthesized data qualitatively in narrative and tabular format. Data were not quantitatively synthesized because of significant qualitative heterogeneity. Specifically, interventions varied widely in type and duration, type and number of supplementary methods of intervention delivery (E-mail and bulletin boards [BB]), type of study control groups used, nicotine replacement therapy (NRT) use, and length of study follow-up.

For comparability, we depict 7-day point prevalence abstinence on the forest plot, though some studies reported additional outcomes. Quantity, quality, and consistency of the evidence were assessed by adapting an evidence-grading scheme recommended by the Agency for Healthcare Research and Quality. We also assessed limitations affecting study quality, precision, and strength of findings. We thereby classified the strength of the evidence on each subset of studies as: high, moderate, low, or insufficient (see Appendix 2).

Results

We identified 17,629 unique titles, of which 21 studies met final eligibility (Figure 1, flow diagram). Most trials were conducted in the United States (n = 13). Fifteen studies were conducted among adults (Brendryen, Drozd, & Kraft, 2008; Brendryen & Kraft, 2008; Etter, 2005; Japuntich et al., 2006; McKay, Danaher, Seeley, Lichtenstein, & Gau, 2008; Munoz et al., 2006, 2009; Oenema, Brug, Dijkstra, de Weerdt, & de Vries, 2008; Pike et al., 2007; Pisinger, Jorgensen, Moller, Dossing, & Jorgensen, 2010; Prochaska et al., 2008; Stoddard, Augustson, & Moser, 2008; Strecher, Shiffman, & West, 2005; Strecher et al., 2008b; Swartz, Noell, Schroeder, & Ary, 2006), one among college students aged 18–24 years (An et al., 2008) and five among adolescents aged 11–17 years (Buller et al., 2008; Mermelstein & Turner, 2006; Norman, Maley, Li, & Skinner, 2008; Patten et al., 2006; Woodruff, Conway, Edwards, Elliott, & Crittenden, 2007).

Figure 1.

Summary of literature search (number of articles). *Total may exceed number in corresponding box as articles could be excluded for more than one reason at this level.

Figure 1.

Summary of literature search (number of articles). *Total may exceed number in corresponding box as articles could be excluded for more than one reason at this level.

Adults

The overall study quality of 15 adult trials was moderate. All were described as randomized with seven describing adequate randomization schemes and concealed allocation (Supplementary Table 1). All trials enrolled women and men (percent male ranged from 29.5% to 84%), and all but one had a majority of Caucasian participants (Supplementary Table 2). Rates of study retention were highly variable ranging from 27% to 86%. About half of the trials were based on identifiable cognitive behavioral theories, including the transtheoretical model (TTM), problem solving, motivational interviewing (MI), and self-efficacy. There was considerable heterogeneity in length of intervention (8–54 weeks) and length of follow-up (1–12 months). Different methods were used to survey smoking status (mail, E-mail, telephone, and Web), with some trials using multiple methods (Supplementary Table 3). Seven studies compared the efficacy of a Web site–based intervention to an alternative or delayed treatment and three studies compared the efficacy of different types of Web sites. Other studies examined the effect of: tailoring (wherein a program assesses individual characteristics relevant to smoking cessation, uses algorithms to generate interventions addressing the specific needs of the user, and generates a feedback protocol that delivers clear messages to the smoker; three studies), adding specialized components (three), and exposure frequency (seven).

Unless otherwise indicated, studies were conducted in the United States and in a community setting, defined adult as 18 years or older and smoking status as ≥5 cigarettes daily, measured abstinence as a 7-day point prevalence of “not even a puff”; were self-report; and used ITT analyses.

Web Site–Delivered Intervention Compared With Self-Help Brochures, Delayed Intervention, or Counseling

Two trials known as “Happy Endings” compared the efficacy of a self-help manual with a fully automated Web-based intervention supplemented by treatment using over 400 automated contacts by E-mail, Web pages, and cellular telephone messages as well as a craving helpline and a relapse prevention system (Figure 2). The intervention uses chronological modules that channel the smoker through the different psychological stages of quitting. In the first trial, Brendryen and Kraft (2008) recruited 396 NRT users and at 12 months found that both repeated point abstinence rates (22.3% vs. 13.1%; p < .05) and point abstinence rates (37.6% vs. 24.1%; p < .05) were higher in the Internet compared with the control group. A second trial (Brendryen et al., 2008) using the same procedure recruited 290 smokers willing to quit without NRT. Again, the intervention group had higher repeated point abstinence rates than the control group (20% vs. 7%; p < .05); however, the difference in point abstinence rates was smaller (33% vs. 23 %; p < .07) largely because control subjects made a second quit attempt. The intervention was equally effective across age, gender, and NRT users. Although abstinence rates were higher in both intervention groups, it is difficult to isolate the effect of the Web from the many supplemental treatments that were also delivered (Supplementary Table 3).

Figure 2.

Effects of Web-based interventions on the odds of smoking cessation. Cont = Control; Int = Intervention; MI = motivational interviewing; mos = months; NR = not reported; NRT = nicotine replacement therapy; SoCA = stage of change assessments; BB = bulletin board. *Among Spanish speakers. †Among English speakers. ‡For one, two, and three high-depth intervention components, respectively. §For each additional tailoring item added.

Figure 2.

Effects of Web-based interventions on the odds of smoking cessation. Cont = Control; Int = Intervention; MI = motivational interviewing; mos = months; NR = not reported; NRT = nicotine replacement therapy; SoCA = stage of change assessments; BB = bulletin board. *Among Spanish speakers. †Among English speakers. ‡For one, two, and three high-depth intervention components, respectively. §For each additional tailoring item added.

Two studies compared a Web site intervention to a delayed intervention control group (Oenema et al., 2008; Swartz et al., 2006). Swartz et al. recruited 692 smokers from work-sites and assigned them to personalized video treatments or a waitlist. The 1-2-3 Smokefree intervention was based on benefits of, and overcoming barriers to quitting, a tailored quit plan and modeled on a live counseling session. At 3-month follow-up, the cessation rate was significantly higher in the treatment than in the delayed group (12.3% vs. 5.0%; p < .05). No specific component of the program predicted abstinence. Oenema et al. recruited 2,159 Dutch-speaking adults for a computer-tailored lifestyle intervention addressing fat intake, physical activity, and smoking. The intervention addressed motivational enhancement and self-efficacy. Smokers (undefined, n = 692) were randomized to a 1-month exposure to a Web site or received the intervention 1 month later. At 1-month follow-up, quit rates did not differ by group (4% vs. 2.9%; p > .05), which was attributed to lower than expected rates of smokers (60%) electing to use the smoking modules.

Two trials compared a Web- with a counselor-delivered smoking cessation program (Pisinger et al., 2010; Prochaska et al., 2008) and a third (Japuntich et al., 2006) compared Web plus counseling to counseling alone. In Pisinger et al. general practitioners (GPs) were randomized to briefly counsel and refer 1,518 Danish smokers to a Web site or a counseling program or to deliver usual care. The Web site program was 13 sessions, and the counseling program was five 2-hr sessions. At 12 months, quit rates in the Web site intervention (5.9%) and the group behavioral/pharmacological intervention (6.7%) did not differ from usual care from their GP (5.7%; p > .05). Biochemical rates were lower and also did not differ by condition (2.5% Web site, 3.5% group/behavioral and 2.7% usual care). As a subset of a larger behavioral risk reduction study, Prochaska et al. randomized 136 smokers (undefined) to three conditions: static Web site providing a health risk intervention (HRI), HRI plus ProChange LifeStyle online TTM assessments and tailored feedback sessions, or HRI plus person/telephone-delivered MI counseling. At 6-month follow-up, participants showed no difference in respective smoking outcomes (16.7% HRI, 21.1% HRI + TTM, and 34.6% HRI + MI; p > .05). In both studies, insufficient sample size likely affected power to detect differences between groups.

Japuntich et al. (2006) developed the Comprehensive Health Enhancement Support System for Smoking Cessation and Relapse Prevention, a clinically validated program based on Public Health Service Clinical Practice Guidelines. Participants (n = 284) were randomized to a tailored, intensive Web-based intervention plus face-to-face counseling + bupropion or to face-to-face counseling + bupropion treatment alone. Support tools such as “ask-an-expert” were accessed more than informational material. There was no difference at 6 months in quit rates between Web site and control groups (15.0% vs. 11.8%; p > .05), suggesting that adding a Web-based intervention to standard cessation treatment did not significantly improve abstinence rates (Figure 2).

In summary, two trials show that intensive, multicomponent Web-based interventions produce higher quit rates than self-help manuals. It is difficult, however, to isolate the effect of a Web-based program when its efficacy is not tested as a single component. Web-based interventions appear more effective than no (delayed) intervention in one study but equivocal in a second study, which had too a low percentage of Web site “log-ons” to accurately assess its efficacy. Delayed interventions also can have the effect of suppressing quit attempts, potentially obscuring real treatment effect. Three studies compared Web-based interventions to counseling for smoking cessation and found no difference in quit rates. In the absence of a comparison with a no treatment control group, however, it cannot be determined whether Web-based interventions were as effective as counseling or if neither intervention was effective. Thus, the strength of the evidence comparing efficacy of Web-based interventions to self-help and counseling is insufficient, and evidence is moderate comparing Web-based interventions to no treatment (Table 1).

Table 1.

Rating the Strength of the Evidence of Randomized Controlled Trials on the Use of Web-Based Interventions for Smoking Cessation

Population Comparison Strength of the evidence Summary of conclusions 
Adults Web-based intervention + other automated interventions (text messages/E-mail) vs. self-help brochures Insufficient Multicomponent interventions of long treatment duration superior to self-help, however, cannot isolate individual effect of Web-based intervention 
 Web-based intervention vs. delayed intervention Moderate One study shows an effect, one is equivocal. Also, delayed treatment may suppress quit attempts 
 Web-based intervention vs. counseling Insufficient Web site vs. counseling interventions show mixed results with low statistical power and high attrition in two studies; no control groups to provide adequate comparison 
 Different types of smoking cessation Web sites Moderate No clear efficacy for a type of Web site; supportive may be more beneficial than informational Web site 
 Static/minimal-interaction vs. high-interaction Web sites Insufficient Highly interactive Web sites appear as effective as static Web sites but methodology too diverse to make a conclusion 
 Tailored Web site vs. untailored Web sites Moderate Tailoring associated with higher quit rates, although tailored conditions had more program interaction 
 Web site with additional components: bulletin boards and mood management modules Insufficient BB and mood management did not provide additional benefit, however, low use of BB, unclear usage of mood management 
 Amount of Web site exposure Moderate Greater Web site exposure or repetition of intervention material associated with higher quit rates; however, this has not been examined as an independent variable 
College students Web-based intervention vs. self-help or link to online academic resources Insufficient Insufficient evidence to support Web-based interventions over control in college students. Conclusions limited by small number of studies 
Adolescents Web-based intervention vs. measurement control, attention control, brief office intervention, group counseling + Web vs. group counseling only Insufficient Insufficient evidence to draw conclusions on efficacy of Web-based interventions among adolescents. Studies with widely varying intervention and control conditions 
Population Comparison Strength of the evidence Summary of conclusions 
Adults Web-based intervention + other automated interventions (text messages/E-mail) vs. self-help brochures Insufficient Multicomponent interventions of long treatment duration superior to self-help, however, cannot isolate individual effect of Web-based intervention 
 Web-based intervention vs. delayed intervention Moderate One study shows an effect, one is equivocal. Also, delayed treatment may suppress quit attempts 
 Web-based intervention vs. counseling Insufficient Web site vs. counseling interventions show mixed results with low statistical power and high attrition in two studies; no control groups to provide adequate comparison 
 Different types of smoking cessation Web sites Moderate No clear efficacy for a type of Web site; supportive may be more beneficial than informational Web site 
 Static/minimal-interaction vs. high-interaction Web sites Insufficient Highly interactive Web sites appear as effective as static Web sites but methodology too diverse to make a conclusion 
 Tailored Web site vs. untailored Web sites Moderate Tailoring associated with higher quit rates, although tailored conditions had more program interaction 
 Web site with additional components: bulletin boards and mood management modules Insufficient BB and mood management did not provide additional benefit, however, low use of BB, unclear usage of mood management 
 Amount of Web site exposure Moderate Greater Web site exposure or repetition of intervention material associated with higher quit rates; however, this has not been examined as an independent variable 
College students Web-based intervention vs. self-help or link to online academic resources Insufficient Insufficient evidence to support Web-based interventions over control in college students. Conclusions limited by small number of studies 
Adolescents Web-based intervention vs. measurement control, attention control, brief office intervention, group counseling + Web vs. group counseling only Insufficient Insufficient evidence to draw conclusions on efficacy of Web-based interventions among adolescents. Studies with widely varying intervention and control conditions 

Note. (1) “high” grade, indicating confidence that further research is very unlikely to change our confidence in the estimated effect in the abstracted literature; (2) “moderate” grade, indicating that further research is likely to have an important impact on our confidence in the estimates of effects and may change the estimates in the abstracted literature; (3) “low” grade, indicating that the further research is very likely to have an important impact on confidence in the estimates of effects and is likely to change the estimates in the abstracted literature; and (4) insufficient evidence to make a conclusion

Web Site–Delivered Interventions Compared With Other Web Site–Delivered Interventions

Type of Web Site

McKay et al. (2008) developed the Quit Smoking Network (QSN) as an informational and cognitive behavioral smoking cessation Web site with multimedia components and compared it with Active Lives (AL), an attention placebo control Web site encouraging physical activity as a cessation tool (Figure 2). At 6 months, abstinence rates did not differ between QSN and AL (n = 2,318; 9.7% vs. 10.4%; p > .05); however, higher education was correlated with increased abstinence. Rabius et al. (Pike et al., 2007; Rabius, Pike, Wiatrek, & McAlister, 2008) evaluated the effectiveness of five Web-based smoking cessation service providers (Oregon Center for Applied Sciences, ProChange, QuitNet, SmokeClinic, and Centre for Addiction and Mental Health). Participants (n = 6,451) were randomized to one of these five interactive Web sites or to a static Web site containing a downloadable self-help booklet. At 13 months postrandomization, quit rates using 30-day point prevalence abstinence did not differ across Web sites or from the static Web site (9.6%–12.9% vs. 10.1%; p > .05). Even with the large number of participants across six Web sites, it is not definitive that interactive Web sites are as effective as a static Web site because of the low rate of follow-up (Supplementary Table 1) and potentially low Web site use, specifically “most participants did not make more than two visits.” Furthermore, the study had insufficient power to detect differences of less than 3% among the treatment arms. Significant differences were found in the use of the interactive Web sites and correspondingly in quit rates; however, the characteristics of the highly used Web sites were not reported. Etter (2005) compared two French Web sites called Stop.tabac.ch among 8,682 current smokers. Participants were assigned either to the original counseling program containing information on health risks and coping strategies or to a modified program with less counseling and more information on NRT and nicotine dependence. At 2.5 months postassignment, quit rates were higher in the original program than in the modified program (10.9% vs. 8.9%; p < .05). There were no age or gender differences. A sensitivity analysis, conducted because of a lower response rate in the modified program, showed that the original program was still more effective.

The evidence to conclude whether a particular type of Web site is more effective is insufficient (Table 1). One study found that a more supportive Web site was associated with higher quit rates than a more informational one, whereas two other studies comparing a variety of Web sites did not detect differences in quit rates. Even if findings had not been equivocal, all three studies had very low rates of study retention (Supplementary Table 1), which limits ability to identify potentially effective Web site content.

Tailoring

Strecher et al. (2005) tested the efficacy of a tailored versus a generic smoking cessation Web site among 3,971 English and Irish NRT patch users. The Committed Quitters Stop Smoking Plan (CQ) was a tailored intervention consisting of messages and E-mails, a cessation guide, and three tailored newspapers. At 3 months, 10-week continuous abstinence showed that CQ had significantly higher abstinence rates than the generic Web site (20.1% vs. 15.9%; p < .05). This effect remained significant after controlling for NRT use. The authors noted that the tailored participants had more interactions with the program, which may have contributed to the effect on smoking cessation. In a second RCT, Strecher et al. (2008b) investigated the importance of “depth” of tailoring in 1,866 health maintenance organization patients who smoked ≥10 cigarettes daily. Program content tested the use of success stories, personalization (pictures/messages from individuals on the treatment team), outcome expectations, efficacy expectations, and Web site exposure. Within these, the effect of the depth of tailored messages was also tested, such as using largest quit barrier and type of social support. In complete respondent analysis at 6 months, quit rates were higher among participants receiving high-depth success (34.3%) or high-personalization stories (33.6%) compared with those receiving low-depth (26.8%) or low-personalization stories (27.4%; p’s < .05). Tailoring depth was significantly related to smoking cessation in per-protocol analysis for each component added but marginally related in ITT analyses (zero components quit rate: 28%; one, two, and three components’ respective quit rates: 25.8%, 33.7%, 38.6%). Younger, male, or less formally educated participants were more likely to disengage from the program.

The evidence for the effectiveness of tailoring is moderate with two studies finding that tailoring is associated with higher quit rates. One study systematically examined types of tailoring that may be used by a Web site and found that depth of tailoring specifically influenced quit rates.

Additional Components

Three studies examined the effect of adding BB and mood management modules to Web sites. Stoddard et al. (2008) recruited 1,375 federal employees to enroll in Smokefree.gov, an online quit guide with targeted self-help materials, links for reaching a cessation counselor and changes in risk of death by quit dates. Smokers were randomized to Web site only or to Web site plus asynchronous BB condition. At 3 months, quit rates for the Web site/BB condition and Web site only condition did not differ (6.6% vs. 6.9%; p > .05). Few participants, however (11.8%), used BBs, therefore its effectiveness cannot be determined.

Munoz et al. (2006, 2009) added mood management, an eight-lesson social learning course, to an 8-week Web-based smoking intervention called “Guia para Dejar de Fumar.” “Guia” is the National Cancer Institute's evidence-based guide adapted for Web site use and provides reason to stop smoking, relapse prevention, and refusal skills. It was linked to educational E-mails delivered before and after quit dates. From 74 countries, 288 Spanish speakers and 280 English-speaking participants were enrolled. At 12 months, English speakers using Guia + E-mails plus mood management had lower quit rates than those using Guia + E-mails alone (8.6% vs.17.0%; p < .05), and among Spanish speakers, there was no difference between Guia + ITEMs plus mood management and GUIA + ITEMs (20.4% vs.22.6%; p > .05). In a more recent trial, Munoz et al. (2009) randomized smokers from 68 countries to a static Web site or to static Web site augmented with other media. At 12 months, no difference in quit rates was detected in the four groups: Web site only (19.8%), Web site plus E-mail (19.1%), Web site/E-mail plus mood management module (20.7%), or Web site/E-mail/mood management plus BB (22.7%; p > .05). Utilization rates of BBs and mood management were not reported, so it is difficult to assess their efficacy.

The strength of the evidence is insufficient that adding BBs or mood management modules to Web sites can improve quit rates (Table 1).

Web Site Exposure: Log-ons and Dose–Response

Studies varied widely in reporting participant log-ons to the Web site. Four studies reported a range only and three reported a percentage, which varied widely from 15.8% to 88% (Brendryen and Kraft, 2008; Brendryen et al., 2008; Pike et al., 2007; Strecher et al., 2008b). The two studies reporting low log-on rates (15.58% and 60%) also found no difference between Web-based interventions and the control group (Oenema et al., 2008; Pisinger et al., 2010). Similarly, nine RCTs did not report log-on rates and seven of those found no difference between Web-based interventions and control groups.

In six of seven RCTs, higher quit rates were found among smokers who had greater Web site exposure, variously defined as number of Web site sections opened, number of Web site contacts, or time spent on the Web site. Strecher et al. (2008a) found that number of Web sections opened was related to subsequent smoking cessation. Smokers who opened three to five sections had a significantly higher quit rate than those who opened zero to two sections (37.4% vs. 27.3%; p < .05). Women opened significantly more treatment sections than men; however, gender differences in abstinence rates were not reported.

A dose–response effect was also found by Pike et al. (2007) and Rabius et al. (2008). High utilization Web sites (>20% of participants making >5 visits) had quit rates of 12.2%, whereas low utilization sites (<10% of participants making >5 visits) had quit rates of 10.2% (χ2 not provided, p < .05). Furthermore, high Web site users had higher quit rates than low Web site users (21.7% vs. 9.1%; p < .05). Japuntich et al. (2006) found that smokers who “logged in” more than three times weekly had higher quit rates at 6 months than those logging in less frequently (<3; OR = 2.13, 95% CI, 1.25–3.61). Importantly, this dose–response relationship remained after adjusting for age, smoking dependence, and smoking history. Munoz et al. (2006) found higher abstinence rates among English (14.6%) and Spanish (31%) speakers who opened ≥4 lessons compared with those completing ≤3 (4.8% and 9.9%; p < .05). In another study (Munoz et al., 2009), site utilization (use of an online journal and cigarette counter) was significantly related to abstinence across static and interactive Web site conditions. Using time as a measure of Web site exposure, Stoddard et al. (2008) found that exposure times were longer in those who achieved abstinence compared with those still smoking (23.4 min vs. 13.8 min; p < .01) and in those making a serious quit attempt by abstaining for at least 24 hrs (22.4 vs. 10.4 min; p < .05). The longest exposure times were among those whose quit attempt began in the 5 days prior to the study (29.4 min), suggesting that exposure may be influenced by motivation to quit and/or quit attempts.

By contrast, QSN (McKay et al., 2008) found that number of Web site visits and total visit duration were not associated with abstinence either within QSN or in combined QSN and AL Web sites (complete case analyses). However, most participants stopped using both programs soon after initiating them (rate not reported), and exposure rates were low in both groups.

In summary, few studies reported the rate of log-ons, and studies that did had rates that varied widely making it difficult to discriminate failure of participants to log-on versus failure of a Web-based intervention. Once logged on, the strength of the evidence is high, supporting the association between frequency of Web site use and higher quit rates, although the direction of this effect has not been clarified.

College Students

We identified one RCT among college students 18- to 24-year old (total participants, n = 517; An et al., 2008; Figure 2). The study quality was good (Supplementary Table 1) using concealed allocation, biochemical validation of smoking status, and ITT analyses. In addition, loss to follow-up was less than 10%. The trial was conducted among occasional and light smokers (Supplementary Table 2) and was based on an identifiable cognitive behavioral theory.

An et al. (2008) randomized half of 517 students to a multicomponent intervention including weekly E-mail invitations to visit an online Web site, interactive quizzes with tailored feedback, and weekly E-mails from peer coaches. The other half received a single E-mail with links to online academic and health resources. Thirty-day abstinence at 30-week follow-up was 40.5% in the multicomponent intervention group and 23% in the comparison group (p < .05). Biochemically validated abstinence rates were lower (33% Internet group and 17% control group; p < .05). In summary, this single study suggests that Web-based interventions may be effective in promoting smoking cessation in college students, with the intervention effects favoring the treatment groups compared with the control condition. We graded the evidence in college students as insufficient because the one study was a multicomponent intervention. With Web and non-Web–based elements, the effect of the Web-based element cannot be isolated.

Adolescents

Our search identified five trials among adolescents (total participants, N = 4,542; Buller et al., 2008; Mermelstein and Turner, 2006; Norman et al., 2008; Patten et al., 2006; Woodruff et al., 2007. Study follow-up ranged from 3 to 12 months. The overall study quality was fair, but none described concealed allocation. Losses to follow-up ranged from 13% to 47% (Supplementary Table 1). These five trials were heterogeneous. Overall, cigarette consumption varied widely among smokers (Supplementary Table 2). The interventions ranged from Web-based computer sessions to multicomponent interventions combining Web sites and E-mails. Trials were based on identifiable cognitive behavioral theories, including social cognitive theory, MI, and social learning theory. All but one study were school based.

Buller et al. (2008) designed “Consider This,” a program of 73 online activities organized into six tailored modules for Australian and U.S. adolescents in Grades 6 through 9. Among Australian smokers (n = 184), the intervention was associated with a lower 30-day prevalence of smoking a whole cigarette compared with control (intervention/control difference = −0.045, p = .02). Five percent of smokers in the intervention condition stopped smoking compared with 3% in the control (p > .05). In contrast, there was no significant change in 30-day smoking prevalence among U.S. smokers (n = 45).

Norman et al. (2008) tested a multicomponent intervention consisting of a tailored Web-assisted tobacco cessation intervention combined with a small group MI session and tailored E-mails. It was compared with an attention control condition on climate change. E-mails were sent monthly after the classroom session. Overall, 1,402 students in Grades 9 through 11 were randomized, with 211 smoking at baseline. At 24 weeks, there was no change in smoking rates among smokers in either group.

Woodruff et al. (2007) randomized high-school students to either a Web-based virtual reality world with MI or a measurement control. The intervention group was asked to spend 45 mins per week in a virtual reality world with other teenagers and a counselor to explore smoking. Immediately postintervention, the intervention group (N = 77) had higher rates of 7-day abstinence than the control condition (N = 59; 35% vs. 22%; p < .01); however, at 12 months, there was no difference between the two groups (39% vs. 38%; p > .05).

Mermelstein et al. (2006) developed the American Lung Association's NOT program consisting of 10 group therapy sessions compared with the NOT Plus program, 10 group sessions plus a Web-based adjunct and proactive phone calls. At 3-month follow-up, the intervention condition was associated with increased cessation compared with the control (20.4% vs. 10.6%). Lighter smokers, younger age, female, and non-White participants were more likely to be abstinent.

Patten et al. (2006) compared a Web-based intervention to a brief office intervention for 11- to 18-year-old smokers. Adolescents were randomized to either four weekly brief office visits with a counselor (BOV; N = 69) or access to “Stomp out Smokes” (SOS; N = 70), an interactive Web-based cessation program. At 36 weeks, the abstinence rate was 13% in the BOV group and 6% in the SOS group (p > .05).

In summary, the literature on Web-based smoking cessation programs for adolescents is heterogeneous, with widely varying interventions and mixed results. While the Internet-based virtual reality world appeared promising, immediately postintervention, these effects were not sustained. When group therapy was combined with telephone counseling and a Web-based adjunct, there was an effect at 3 months, but there were no results reported at 6 or 12 months, making it unclear if there was a sustained effect. Furthermore, it is difficult to determine whether proactive telephone calls, the Web, or a combination of the two accounted for increased cessation rates. In two school-based studies that combined smoking prevention and cessation, only a small proportion of the sample smoked, making it unclear if the lack of effect was secondary to lack of statistical power or if the intervention itself had little effect. In the final study, face-to-face counseling was superior to a computerized intervention, though only a third of individuals randomized to the Web-based intervention logged on to the Web site. Based on these results, the evidence on the efficacy of Web-based interventions for adolescents is insufficient.

Table 1 provides a summary of the findings along with the strength of evidence for each of the sections.

Discussion

The purpose of this review was to examine the effectiveness of Web-based interventions for smoking cessation among adults, college students, and adolescents. Among adults, the efficacy of Web-based interventions for smoking cessation appears modest so far. Multicomponent interventions using the Web and other components, such as E-mail and text messages, were more effective than self-help booklets; however, the specific effect of the Web could not be isolated. Web-based interventions appeared more effective than no (delayed) intervention, although delayed treatment can suppress quit attempts in the control group. Comparisons of Web-based interventions with in-person counseling produced similar quit rates; however, these studies had sample sizes that were too small to draw firm conclusions. One study found that combining a Web-based intervention with counseling did not provide additional benefit to counseling alone. In the absence of a comparison with a no treatment control group, however, it cannot be determined if Web-based interventions were equally as efficacious as counseling or if neither intervention was efficacious. Global comparisons of different interactive Web sites produced similar quit rates, although it appeared that a Web site offering support may have been more effective than one that offered information. Static Web sites also produced similar quit rates as interactive Web sites, although more systematic evaluation of such Web site presentations is warranted. Adding components to Web sites such as BB or mood management modules did not appear to improve quit rates, possibly because it required additional time and/or navigation.

By contrast, certain elements of Web-based interventions were associated with smoking cessation. Tailoring, particularly depth of tailoring, and amount of Web site exposure were associated with higher rates of quitting. More personalized and intensive Web-based interventions may provide a greater effect on smoking cessation.

Among college students, a multicomponent Web-based intervention compared with no intervention appears to be effective; however, given the number of components included, some of which were not Web based, it is difficult to disentangle which elements were associated with cessation. College students on college campuses may be particularly responsive to Web-based interventions, given the widespread availability of the Internet in these settings and the increased flexibility of college student's schedules. To date, however, there is only one RCT targeting college students specifically, and while Web-based smoking cessation interventions may be promising among this group, the evidence is currently insufficient.

Similarly, evidence of the effectiveness Web-based interventions in adolescents is insufficient. With few RCTs among this population consisting of heterogeneous content and formats, it is difficult to determine which intervention components, if any, are effective. Furthermore, participation in the Web-based programs varied. In Mermelstein et al., face-to-face counseling was superior to a Web-based program; however, only 1/3 of participants randomized to the Web-based program logged on. Assessing barriers to participation in Web-based programs may improve uptake and efficacy of Internet-based interventions. Despite the mixed results in adolescents, Web-based treatments can deliver clinically tested methods for smoking cessation that are broad reaching and relatively inexpensive. Further investigation of theory, methods, and mode of delivery/assessment will advance understanding of effective strategies.

Several methodological issues diminish the ability to estimate effects of Web-based treatment. It is not clear what specific Web-based treatments work. Only about half of the adult trials cited a particular theory used, and across all trials, behavior change methods varied widely. Furthermore, many adult studies combined Web-based treatments with other treatments or other modes of delivering the intervention, thus confounding evaluation of the Web site's efficacy. A few studies (Munoz et al., 2006, 2009) did attempt to test the effect of systematically adding other treatments to a static Web site but found no additive benefit of other interventions. Information on participants was also limited. Few studies examined differences by gender or other sociodemographic variables, such as race/ethnicity or education. Some evidence indicated that women accessed more online treatment and that younger males tended to have higher rates of study withdrawal, but in post-hoc analysis, this did not appear to be associated with smoking outcome. Education showed some association with retention and abstinence. Most trials were conducted in developed countries, in Western European languages, and among Caucasian smokers; therefore, generalizability to developing non-White populations is unknown. In addition, smokers often used multiple nontrial cessation methods (Munoz et al., 2009), though most Web-based RCTs did not report concurrent therapies. Only a minority of studies incorporated or controlled for NRT use; therefore, its impact on smoking cessation in Web-based treatments is unknown. A major limitation was the high loss to follow-up, which is especially problematic in smaller sample trials where power is attenuated by withdrawals and dropouts. It was also unclear who these missing participants were as most RCTs provided little or no information. Some participants dropped out before logging on to a Web site and hence were never exposed to treatment. The few studies reporting the percentage of “log-on” participants showed that it was highly variable (16%–82%). Thus, it is unclear whether they failed to benefit from treatment or failed to log-on to treatment. In follow-up, trials using multiple methods identified additional quitters who would otherwise have been classified as “smokers” in single method paradigms (Munoz et al., 2009). Also, just three trials used incentives to encourage follow-up adherence (An et al., 2008; Mermelstein et al., 2006, Munoz et al., 2006). Finally, ITT analysis was typically used to fill in missing data on the conservative assumption that missing participants were smokers. This may be problematic because differential losses by group could bias study results.

Clinical, methodological, and statistical heterogeneity is expected in an emerging area of research, but such heterogeneity may be limiting evaluations of the efficacy of a potentially promising method of treatment. Apart from the advantages of lower cost and broad reach, Web-based interventions also uniquely permit a standardized administration of an intervention that in-person counseling by definition can never achieve. In addition, these interventions can more readily test the efficacy of different therapeutic components (such as decisional balance vs. normative feedback) and supplemental delivery modes (E-mails vs. BB) and their efficacy in different populations (adolescents, heavy smokers, etc.). Addressing such issues can assist future studies in answering the core question of treatment effectiveness: Which treatment works best, for whom, and under what conditions? As for which treatment works and under what conditions, the field will benefit from RCTs, which compare a single Web-based treatment with an attention or delayed control group, especially when comparing Web-based treatments with other established interventions. Such a methodological design will establish the magnitude of the effect of Web-based interventions. It will also be important for Web-based interventions to define the theory that informs intervention design and systematically test which techniques within Web sites promote the greatest change in smoking cessation rates. Across the broader array of Internet interventions (such as Web sites, E-mail, and BB), theory-driven treatments have been more effective than interventions with no theoretical basis, and certain behavior change techniques have had differential efficacy in changing behavior (Webb, Joseph, Yardley, Michie, 2010). Central to validating which treatment works, however, is verifying actual Web site exposure. It would be useful to first examine what percent and what characteristics of participants enroll in versus actually log on to the Web site. Interventions may appear ineffective when in fact a high percentage of participants never viewed the Web site. In a similar vein, it would be useful to examine the actual dose of Web site use (number and length of sessions used) as dose appears to be associated with higher quit rates. It would also be useful to systematically assess the additive effect of supplemental modes of delivery and the use of nontrial cessation methods, especially pharmacotherapy. NRT is effective in combination with other smoking cessation methods (Stead, Perrara, Bullen, Mant, & Lancaster, 2008) and may boost effectiveness of Web-based treatments. Finally, future studies would advance evaluation of efficacy if length of treatment, type of treatment, and length of follow-up were more homogeneous across trials.

As for whom treatment works and under what conditions, we suggest that future studies obtain more sociodemographic information about study participants, including factors that influence outcome such as depression (Centers for Disease Control and Prevention, 2010) or intention to change. Also future trials may adjust for baseline differences in participants, which could otherwise moderate treatment effects. Reducing attrition will continue to be a challenge. Using multiple follow-up methods and incentives has been successful in boosting follow-up rates (Munoz, et al., 2009). Problems of missing information may also be addressed in advance of data collection by more sophisticated statistical methods using multiple imputation techniques or other likelihood methods, such as random regression, mixed models or generalized estimating equations, survival analysis (Liu, Wei, & Zhang, 2006).

Web-based interventions can potentially provide a low-cost broad reaching treatment for smoking cessation, which can be standardized and tailored as needed. To date, evidence supporting the use of Web-based interventions for smoking cessation is insufficient to moderate in adults, and insufficient in college students and adolescents. These RCTs have, however, elucidated clinical, methodological, and statistical practices that are likely to improve future trial design and treatment delivery.

Supplementary Material

Supplementary Table 1, 2, and 3 can be found online at http://www.ntr.oxfordjournals.org

Funding

This work was supported by the International Union Against Tuberculosis and Lung Disease, November 2009, Paris, France. The Grant # USA-TW-601 funding agency (grant # USA-TW-601) reviewed the final work plan prior to the project's initiation, reviewed the manuscript for the purpose of approval to assert copyright, and reviewed the acknowledgment of funding statement and the disclaimer. Aside from these exceptions, the funding source had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.

Declaration of Interests

None declared.

References

An
LC
Klatt
C
Perry
CL
Lein
EB
Hennrikus
DJ
Pallonen
UE
, et al.  . 
The RealU online cessation intervention for college smokers: A randomized controlled trial
Preventive Medicine
 , 
2008
, vol. 
47
 (pg. 
194
-
199
doi:10.1016/j.ypmed.2008.04.011
Brendryen
H
Drozd
F
Kraft
P
A digital smoking cessation program delivered through internet and cell phone without nicotine replacement (happy ending): Randomized controlled trial
Journal of Medical Internet Research
 , 
2008
, vol. 
10
 pg. 
e51
  
doi:10.2196/jmir.1005
Brendryen
H
Kraft
P
Happy ending: A randomized controlled trial of a digital multi-media smoking cessation intervention
Addiction
 , 
2008
, vol. 
103
 (pg. 
478
-
484
doi:10.1111/j.1360-0443.2007.02119.x
Buller
DB
Borland
R
Woodall
WG
Hall
JR
Hines
JM
Burris-Woodall
P
, et al.  . 
Randomized trials on consider this, a tailored, internet-delivered smoking prevention program for adolescents
Health Education & Behavior: The Official Publication of the Society for Public Health Education
 , 
2008
, vol. 
35
 (pg. 
260
-
281
doi:10.1177/1090198106288982
Byron
MJ
Cobb
NK
Concerns about a meta-analysis of computer smoking cessation programs
Archives of Internal Medicine
 , 
2009
, vol. 
169
 pg. 
1814
  
doi:10.1001/archinternmed.2009.375
Centers for Disease Control and Prevention
Depression and smoking in the U.S. household population aged 20 and over, 2005–2008
 , 
2010
 
Etter
JF
Comparing the efficacy of two internet-based, computer-tailored smoking cessation programs: A randomized trial
Journal of Medical Internet Research
 , 
2005
, vol. 
7
 pg. 
e2
  
doi:10.2196/jmir.7.1.e2
Ezzati
M
Lopez
AD
Regional, disease specific patterns of smoking-attributable mortality in 2000
Tobacco Control
 , 
2004
, vol. 
13
 (pg. 
388
-
395
doi:10.1136/tc.2003.005215
Jadad
AR
Moore
RA
Carroll
D
Jenkinson
C
Reynolds
DJ
Gavaghan
DJ
, et al.  . 
Assessing the quality of reports of randomized clinical trials: Is blinding necessary?
Controlled Clinical Trials
 , 
1996
, vol. 
17
 (pg. 
1
-
12
doi:10.1016/0197-2456(95)00134-4
Japuntich
SJ
Zehner
ME
Smith
SS
Jorenby
DE
Valdez
JA
Fiore
MC
, et al.  . 
Smoking cessation via the internet: A randomized clinical trial of an internet intervention as adjuvant treatment in a smoking cessation intervention
Nicotine & Tobacco Research: Official Journal of the Society for Research on Nicotine and Tobacco
 , 
2006
, vol. 
8
 
Suppl. 1
(pg. 
S59
-
S67
doi:10.1080/14622200601047900
Jha
P
Ranson
MK
Nguyen
SN
Yach
D
Estimates of global and regional smoking prevalence in 1995, by age and sex
American Journal of Public Health
 , 
2002
, vol. 
92
 (pg. 
1002
-
1006
Liu
M
Wei
L
Zhang
J
Review of guidelines and literature for handling missing data in longitudinal clinical trials with a case study
Pharmaceutical Statistics
 , 
2006
, vol. 
5
 (pg. 
7
-
18
doi:10.1002/pst.189
Mathers
CD
Loncar
D
Projections of global mortality and burden of disease from 2002 to 2030
PLoS Medicine
 , 
2006
, vol. 
3
 pg. 
e442
  
doi:10.1371/journal.pmed.0030442
McKay
HG
Danaher
BG
Seeley
JR
Lichtenstein
E
Gau
JM
Comparing two web-based smoking cessation programs: Randomized controlled trial
Journal of Medical Internet Research
 , 
2008
, vol. 
10
 pg. 
e40
  
doi:10.2196/jmir.993
Mermelstein
R
Turner
L
Web-based support as an adjunct to group-based smoking cessation for adolescents
Nicotine & Tobacco Research: Official Journal of the Society for Research on Nicotine and Tobacco
 , 
2006
, vol. 
8
 
Suppl. 1
(pg. 
S69
-
S76
doi:10.1080/14622200601039949
Moher
D
Liberati
A
Tetzlaff
J
Altman
DG
the PRISMA Group
Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement
International Journal of Surgery
 , 
2010
(pg. 
336
-
341
doi:10.1016/j.ijsu.2010.02.007
Munoz
RF
Barrera
AZ
Delucchi
K
Penilla
C
Torres
LD
Perez-Stable
EJ
International Spanish/English internet smoking cessation trial yields 20% abstinence rates at 1 year
Nicotine & Tobacco Research: Official Journal of the Society for Research on Nicotine and Tobacco
 , 
2009
, vol. 
11
 (pg. 
1025
-
1034
doi:10.1093/ntr/ntp090
Munoz
RF
Lenert
LL
Delucchi
K
Stoddard
J
Perez
JE
Penilla
C
, et al.  . 
Toward evidence-based internet interventions: A Spanish/English web site for international smoking cessation trials
Nicotine & Tobacco Research: Official Journal of the Society for Research on Nicotine and Tobacco
 , 
2006
, vol. 
8
 (pg. 
77
-
87
doi:10.1080/14622200500431940
Myung
SK
McDonnell
DD
Kazinets
G
Seo
HG
Moskowitz
JM
Effects of web- and computer-based smoking cessation programs: Meta-analysis of randomized controlled trials
Archives of Internal Medicine
 , 
2009
, vol. 
169
 (pg. 
929
-
937
doi:10.1001/archinternmed.2009.109
Norman
CD
Maley
O
Li
X
Skinner
HA
Using the internet to assist smoking prevention and cessation in schools: A randomized, controlled trial
Health Psychology: Official Journal of the Division of Health Psychology, American Psychological Association
 , 
2008
, vol. 
27
 (pg. 
799
-
810
doi:10.1037/a0013105
Oenema
A
Brug
J
Dijkstra
A
de Weerdt
I
de Vries
H
Efficacy and use of an internet-delivered computer-tailored lifestyle intervention, targeting saturated fat intake, physical activity and smoking cessation: A randomized controlled trial
Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine
 , 
2008
, vol. 
35
 (pg. 
125
-
135
doi:10.1007/s12160-008-9023-1
Patten
CA
Croghan
IT
Meis
TM
Decker
PA
Pingree
S
Colligan
RC
, et al.  . 
Randomized clinical trial of an internet-based versus brief office intervention for adolescent smoking cessation
Patient Education and Counseling
 , 
2006
, vol. 
64
 (pg. 
249
-
258
doi:10.1016/j.pec.2006.03.001
Pike
KJ
Rabius
V
McAlister
A
Geiger
A
American cancer society's QuitLink: Randomized trial of internet assistance
Nicotine & Tobacco Research: Official Journal of the Society for Research on Nicotine and Tobacco
 , 
2007
, vol. 
9
 (pg. 
415
-
420
doi:10.1080/14622200701188877
Pisinger
C
Jorgensen
MM
Moller
NE
Dossing
M
Jorgensen
T
A cluster randomized trial in general practice with referral to a group-based or an internet-based smoking cessation programme
Journal of Public Health
 , 
2010
, vol. 
32
 (pg. 
62
-
70
doi:10.1093/pubmed/fdp072
Prochaska
JO
Butterworth
S
Redding
CA
Burden
V
Perrin
N
Leo
M
, et al.  . 
Initial efficacy of MI, TTM tailoring and HRI's with multiple behaviors for employee health promotion
Preventive Medicine
 , 
2008
, vol. 
46
 (pg. 
226
-
231
doi:10.1016/j.ypmed.2007.11.007
Rabius
V
Pike
KJ
Wiatrek
D
McAlister
AL
Comparing internet assistance for smoking cessation: 13-month follow-up of a six-arm randomized controlled trial
Journal of Medical Internet Research
 , 
2008
, vol. 
10
 pg. 
e45
  
doi:10.2196/jmir.1008
Rosen
LJ
Ben Noach
M
Systematic reviews on tobacco control from cochrane and the community guide: Different methods, similar findings
Journal of Clinical Epidemiology
 , 
2010
, vol. 
63
 (pg. 
596
-
606
doi:10.1016/j.jclinepi.2009.09.010
Shahab
L
McEwen
A
Online support for smoking cessation: A systematic review of the literature
Addiction
 , 
2009
, vol. 
104
 (pg. 
1792
-
1804
doi:10.1111/j.1360-0443.2009.02710.x
Stead
LF
Perrara
R
Bullen
C
Mant
D
Lancaster
T
Nicotine replacement therapy for smoking cessation
Cochrane Database of Systematic Reviews
 , 
2008
 
Issue 1. Art. No.: CD000146. doi:10.1002/14651858. CD000146.pub3
Stoddard
JL
Augustson
EM
Moser
RP
Effect of adding a virtual community (bulletin board) to smokefree.gov: Randomized controlled trial
Journal of Medical Internet Research
 , 
2008
, vol. 
10
 pg. 
e53
  
doi:10.2196/jmir.1124
Strecher
VJ
McClure
J
Alexander
G
Chakraborty
B
Nair
V
Konkel
J
, et al.  . 
The role of engagement in a tailored web-based smoking cessation program: Randomized controlled trial
Journal of Medical Internet Research
 , 
2008
, vol. 
10
 pg. 
e36
  
doi:10.2196/jmir.1002
Strecher
VJ
McClure
JB
Alexander
GL
Chakraborty
B
Nair
VN
Konkel
JM
, et al.  . 
Web-based smoking-cessation programs: Results of a randomized trial
American Journal of Preventive Medicine
 , 
2008
, vol. 
34
 (pg. 
373
-
381
doi:10.1016/j.amepre.2007.12.024
Strecher
VJ
Shiffman
S
West
R
Randomized controlled trial of a web-based computer-tailored smoking cessation program as a supplement to nicotine patch therapy
Addiction
 , 
2005
, vol. 
100
 (pg. 
682
-
688
doi:10.1111/j.1360-0443.2005.01093.x
Swartz
LH
Noell
JW
Schroeder
SW
Ary
DV
A randomised control study of a fully automated internet based smoking cessation programme
Tobacco Control
 , 
2006
, vol. 
15
 (pg. 
7
-
12
doi:10.1136/tc.2003.006189
Webb
TL
Joseph
J
Yardley
L
Michie
S
Using the internet to promote health behavior change: a systematic review and meta-analysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy
Journal of Medical Internet Research
 , 
2010
, vol. 
12
 pg. 
e4
  
doi:10.2196/jmir.1376
Woodruff
SI
Conway
TL
Edwards
CC
Elliott
SP
Crittenden
J
Evaluation of an internet virtual world chat room for adolescent smoking cessation
Addictive Behaviors
 , 
2007
, vol. 
32
 (pg. 
1769
-
1786
doi:10.1016/j.addbeh.2006.12.008