Antidepressants for smoking cessation

John R Hughes; Lindsay F Stead; Jamie Hartmann‐Boyce; Kate Cahill; Tim Lancaster

doi:10.1002/14651858.CD000031.pub4

. 2014 Jan 8;2014(1):CD000031. doi: 10.1002/14651858.CD000031.pub4

Antidepressants for smoking cessation

John R Hughes ^1,^✉, Lindsay F Stead ², Jamie Hartmann‐Boyce ², Kate Cahill ², Tim Lancaster ²

Editor: Cochrane Tobacco Addiction Group

PMCID: PMC7027688 PMID: 24402784

Abstract

Background

There are at least three reasons to believe antidepressants might help in smoking cessation. Firstly, nicotine withdrawal may produce depressive symptoms or precipitate a major depressive episode and antidepressants may relieve these. Secondly, nicotine may have antidepressant effects that maintain smoking, and antidepressants may substitute for this effect. Finally, some antidepressants may have a specific effect on neural pathways (e.g. inhibiting monoamine oxidase) or receptors (e.g. blockade of nicotinic‐cholinergic receptors) underlying nicotine addiction.

Objectives

The aim of this review is to assess the effect and safety of antidepressant medications to aid long‐term smoking cessation. The medications include bupropion; doxepin; fluoxetine; imipramine; lazabemide; moclobemide; nortriptyline; paroxetine; S‐Adenosyl‐L‐Methionine (SAMe); selegiline; sertraline; St. John's wort; tryptophan; venlafaxine; and zimeledine.

Search methods

We searched the Cochrane Tobacco Addiction Group Specialised Register which includes reports of trials indexed in the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, and PsycINFO, and other reviews and meeting abstracts, in July 2013.

Selection criteria

We considered randomized trials comparing antidepressant medications to placebo or an alternative pharmacotherapy for smoking cessation. We also included trials comparing different doses, using pharmacotherapy to prevent relapse or re‐initiate smoking cessation or to help smokers reduce cigarette consumption. We excluded trials with less than six months follow‐up.

Data collection and analysis

We extracted data and assessed risk of bias using standard methodological procedures expected by the Cochrane Collaboration.

The main outcome measure was abstinence from smoking after at least six months follow‐up in patients smoking at baseline, expressed as a risk ratio (RR). We used the most rigorous definition of abstinence available in each trial, and biochemically validated rates if available. Where appropriate, we performed meta‐analysis using a fixed‐effect model.

Main results

Twenty‐four new trials were identified since the 2009 update, bringing the total number of included trials to 90. There were 65 trials of bupropion and ten trials of nortriptyline, with the majority at low or unclear risk of bias. There was high quality evidence that, when used as the sole pharmacotherapy, bupropion significantly increased long‐term cessation (44 trials, N = 13,728, risk ratio [RR] 1.62, 95% confidence interval [CI] 1.49 to 1.76). There was moderate quality evidence, limited by a relatively small number of trials and participants, that nortriptyline also significantly increased long‐term cessation when used as the sole pharmacotherapy (six trials, N = 975, RR 2.03, 95% CI 1.48 to 2.78). There is insufficient evidence that adding bupropion (12 trials, N = 3487, RR 1.19, 95% CI 0.94 to 1.51) or nortriptyline (4 trials, N = 1644, RR 1.21, 95% CI 0.94 to 1.55) to nicotine replacement therapy (NRT) provides an additional long‐term benefit. Based on a limited amount of data from direct comparisons, bupropion and nortriptyline appear to be equally effective and of similar efficacy to NRT (bupropion versus nortriptyline 3 trials, N = 417, RR 1.30, 95% CI 0.93 to 1.82; bupropion versus NRT 8 trials, N = 4096, RR 0.96, 95% CI 0.85 to 1.09; no direct comparisons between nortriptyline and NRT). Pooled results from four trials comparing bupropion to varenicline showed significantly lower quitting with bupropion than with varenicline (N = 1810, RR 0.68, 95% CI 0.56 to 0.83). Meta‐analyses did not detect a significant increase in the rate of serious adverse events amongst participants taking bupropion, though the confidence interval only narrowly missed statistical significance (33 trials, N = 9631, RR 1.30, 95% CI 1.00 to 1.69). There is a risk of about 1 in 1000 of seizures associated with bupropion use. Bupropion has been associated with suicide risk, but whether this is causal is unclear. Nortriptyline has the potential for serious side‐effects, but none have been seen in the few small trials for smoking cessation.

There was no evidence of a significant effect for selective serotonin reuptake inhibitors on their own (RR 0.93, 95% CI 0.71 to 1.22, N = 1594; 2 trials fluoxetine, 1 paroxetine, 1 sertraline) or as an adjunct to NRT (3 trials of fluoxetine, N = 466, RR 0.70, 95% CI 0.64 to 1.82). Significant effects were also not detected for monoamine oxidase inhibitors (RR 1.29, 95% CI 0.93 to 1.79, N = 827; 1 trial moclobemide, 5 selegiline), the atypical antidepressant venlafaxine (1 trial, N = 147, RR 1.22, 95% CI 0.64 to 2.32), the herbal therapy St John's wort (hypericum) (2 trials, N = 261, RR 0.81, 95% CI 0.26 to 2.53), or the dietary supplement SAMe (1 trial, N = 120, RR 0.70, 95% CI 0.24 to 2.07).

Authors' conclusions

The antidepressants bupropion and nortriptyline aid long‐term smoking cessation. Adverse events with either medication appear to rarely be serious or lead to stopping medication. Evidence suggests that the mode of action of bupropion and nortriptyline is independent of their antidepressant effect and that they are of similar efficacy to nicotine replacement. Evidence also suggests that bupropion is less effective than varenicline, but further research is needed to confirm this finding. Evidence suggests that neither selective serotonin reuptake inhibitors (e.g. fluoxetine) nor monoamine oxidase inhibitors aid cessation.

Plain language summary

Do medications used to treat depression help smokers who are trying to quit

Background and review questions

Some medications and supplements that have been used to treat depression (antidepressants) have been tested to see whether they also help people who are trying to stop smoking. Two antidepressants, bupropion (Zyban) and nortriptyline, are sometimes prescribed to help with quitting smoking. This review set out to determine if using antidepressants increased people's likelihood of successfully quitting smoking at six months or longer and to determine the safety of using these medications to help quit smoking.

Study characteristics

The evidence is current to July 2013. This update includes 24 new studies, and this review includes 90 studies overall. The studies included people who smoked and people who had recently quit smoking. There were 65 trials of bupropion, which is licensed for use as a smoking cessation medication under the trade name 'Zyban'. There were ten trials of nortriptyline which is a tricyclic antidepressant which is not licensed specifically for smoking cessation. We only included studies which measured long term quitting (whether or not people had quit smoking at six months or longer from the start of the study).

Key results and quality of evidence

Trials of bupropion (Zyban) for smoking cessation show high quality evidence that it increases the likelihood of a quit attempt being successful after at least six months (44 trials, over 13,000 participants). The side effects of bupropion include insomnia, dry mouth and nausea and rarely (1:1000) seizures and perhaps psychiatric problems, but the last is unclear. There is also moderate quality evidence, limited by a relatively small number of included studies and participants, that the antidepressant nortriptyline increases quit rates (six trials, 975 participants). The side effects of this medication include dry mouth, constipation, nausea, and sedation, and it can be dangerous in overdose. Selective serotonin reuptake inhibitor antidepressants (for example, fluoxetine), monoamine oxidase inhibitor antidepressants (for example, selegiline), and the antidepressant venlaxafine have not been shown to help smoking cessation, nor has the herbal therapy St John's wort, or S‐Adenosyl‐L‐Methionine (SAMe), a dietary supplement that is thought to have antidepressant properties.

Discussion and considerations

The way in which bupropion and nortriptyline might work is not fully understood. Both appear to help people quit smoking whether or not they have a history of depression, or have depressive symptoms when they stop smoking. The likelihood of quitting using bupropion or nortriptyline appears to be similar to that for nicotine replacement therapy, but the likelihood of quitting using bupropion appears to be lower than the likelihood of quitting using varenicline.

Summary of findings

Background

Whilst nicotine replacement is the most widely used pharmacotherapy for smoking cessation, some people prefer a treatment that does not use nicotine. Others require alternative treatments having failed to quit with nicotine replacement. Observations that a history of depression is found more frequently amongst smokers than nonsmokers, that cessation may precipitate depression, that nicotine may have antidepressant effects, and that antidepressants influence the neurotransmitters and receptors involved in nicotine addiction provided a rationale for the study of antidepressant medications for smoking cessation (Benowitz 2000; Kotlyar 2001).

Description of the intervention

The following medications and substances regarded as having antidepressant properties have been investigated for their effect on smoking behaviour in at least one study:

the tricyclic antidepressants (TCAs) doxepin, imipramine and nortriptyline
the monoamine oxidase inhibitors (MAOI) moclobemide, selegiline, lazabemide, and EVT302
the selective serotonin reuptake inhibitors (SSRIs) fluoxetine, paroxetine, sertraline, citalopram, and zimeledine
the atypical antidepressants bupropion, tryptophan, venlafaxine, imipramine, and doxepin
extracts of St. John's wort (Hypericum perforatum L.)
the dietary supplement S‐Adenosyl‐L‐Methionine (SAMe)

Of the antidepressants tested for smoking cessation, the most commonly used medication is bupropion. This antidepressant has both dopaminergic and adrenergic actions, and appears to be an antagonist at the nicotinic acetylcholinergic receptor (Fryer 1999). It may work by blocking nicotine effects, relieving withdrawal (Cryan 2003; West 2008), or reducing depressed mood (Lerman 2002a). It has been licensed as a prescription aid to smoking cessation in many countries. The usual dose for smoking cessation is 150 mg once a day for three days increasing to 150 mg twice a day continued for 7 to 12 weeks, and the quit attempt is generally initiated a week after starting pharmacotherapy.

Following bupropion, the second most commonly tested medication for smoking cessation is the tricyclic antidepressant nortriptyline. Its presumed mechanism of action is increased noradrenergic activity. It is sometimes prescribed when first‐line treatments have been unsuccessful, and is licensed for smoking cessation in New Zealand. The recommended regimen is 10 to 28 days of titration before the quit attempt, followed by a 12‐week dose of 75 to 100 mg daily (Cahill 2013).

No other antidepressants are currently licensed for use as smoking cessation aids, though others have been tested for possible use. It has been hypothesized that MOAIs may aid smoking cessation because they could substitute for the ability of smoking to act as a monoamine oxidase‐A (MOA) inhibitor. Inhibiting MOA increases dopamine and noradrenaline and hence it has been hypothesized that it should decrease negative affect and make quitting smoking easier. It has been hypothesised that SSRIs might be helpful because they increase serotonin which is also associated with improving negative affect. Other antidepressants including doxepin, tryptophan, and venlafaxine, and alternative therapies for depression such as St. John's wort and SAMe, may help with smoking cessation through similar biological mechanisms. Whether antidepressants work mostly due to reducing negative affect, reducing urges to smoke or withdrawal symptoms, or by acting as nicotine blockers is unclear.

The focus of this review and meta‐analysis is on trials that provide evidence for an effect of antidepressants on long‐term smoking cessation. We describe these in the Effects of interventions section. For pharmacotherapies for which there is still a lack of long‐term data, we briefly describe results from excluded short‐term trials in the Description of studies section.

Objectives

To assess the evidence for the efficacy and safety of medications with antidepressant properties in assisting long‐term smoking cessation, including: bupropion; citalopram; doxepin; fluoxetine; imipramine; lazabemide; moclobemide; nortriptyline; paroxetine; tryptophan; SAMe; selegiline; sertraline; St John's wort; venlafaxine; and zimeledine.

For each medication identified as having been used in a smoking cessation trial we evaluated whether it was more effective than placebo or an alternative treatment in achieving long‐term smoking cessation.

Methods

Criteria for considering studies for this review

Types of studies

For efficacy, we examined randomized trials comparing antidepressant with placebo or with an alternative therapeutic control, or comparing different dosages of an antidepressant, that reported six‐month or longer follow‐ups. For safety, we examined data from randomized controlled trials comparing antidepressant with placebo or no pharmacotherapy controls, and also considered observational data. Studies were included irrespective of their publication status and language of publication.

Types of participants

Current cigarette smokers, or recent quitters (for trials of relapse prevention).

Types of interventions

Treatment with any medication with antidepressant properties to aid a smoking cessation attempt or to prevent relapse, or to reduce the number of cigarettes smoked and aid subsequent cessation. Trials in which all participants received the same pharmacotherapy regimen but different behavioural support were not included.

Types of outcome measures

Efficacy was measured via a) abstinence from smoking or b) incidence of reducing cigarette consumption to 50% or less of baseline, both assessed at follow‐up at least six months from start of treatment. Safety was assessed by incidence of serious and other adverse events, and drop‐outs due to adverse events.

Search methods for identification of studies

We identified studies from the Cochrane Tobacco Addiction Group's Specialised Register. At the time of the updated search in July 2013, the Register included the results of searches of the Cochrane Central Register of Controlled trials (CENTRAL), issue 5, 2013; MEDLINE (via OVID) to update 20130607; EMBASE (via OVID) to week 201324; and PsycINFO (via OVID) to update 20130610. See the Tobacco Addiction Group Module in the Cochrane Library for full search strategies and a list of other resources searched. We searched the Register for reports of studies evaluating bupropion, nortriptyline or any other pharmacotherapy generally classified as having an antidepressant effect. Search terms included relevant individual drug names or antidepressant* or antidepressive*. We checked the citation lists of these studies, recent reviews of non‐nicotine pharmacotherapy, and abstracts from the meetings of the Society for Research on Nicotine and Tobacco. See Appendix 1 for the register search strategy.

Data collection and analysis

One author (LS) conducted the searches, screened titles and abstracts for relevance, and obtained full text of reports of eligible or possibly eligible studies. For conference abstracts reporting potentially eligible studies, attempts were made to contact study investigators to obtain additional data. Papers that reported secondary analyses from eligible studies were listed as additional references under the main study identifier. Two authors (LS and TL or JHB) independently extracted study data and compared the findings. Any discrepancies were resolved by mutual consent. Where available, the following information is recorded in the Characteristics of included studies table:

Type of antidepressant
Country and setting
Recruitment method
Definition of smoker used
Participant demographics (i.e. average age, gender, average cigarettes per day)
Intervention and control description (including dose, schedule, and behavioural support common to all arms)
Outcome(s) used in meta‐analysis, including length of follow‐up, definition of abstinence, and biochemical validation of smoking cessation
Sources of funding

Assessment of risk of bias in included studies

We assessed included studies for risks of selection bias (method of random sequence generation and allocation concealment), performance and detection bias (the presence or absence of blinding), attrition bias (levels and reporting of loss to follow‐up), and any other threats to study validity.

Studies were considered at high risk of performance and detection bias where there was no blinding of participants or personnel or where there was evidence of unblinding, at unclear risk if insufficient information was available with which to judge, and at low risk if the study reported blinding of participants and personnel in detail and there was no evidence of unblinding. Studies were considered to be at low risk of attrition bias where over half of the participants were followed up at the longest follow‐up and where numbers followed up were similar across arms (difference < 20%).

Measures of treatment effect

In each study, we used the strictest available criteria to define cessation, so we used the longest reported follow‐up and extracted figures for sustained abstinence in preference to point prevalence where both were presented. In studies that used biochemical validation of cessation, only those subjects meeting the criteria for biochemically confirmed abstinence were regarded as having stopped smoking. As far as possible, we used an intention‐to‐treat analysis with people who dropped out or were lost to follow‐up treated as continuing smokers. Where subjects appeared to have been randomized but were not included in the data presented by the author we noted this in the study description (see Characteristics of included studies). Assuming that people lost to follow‐up were smokers will ensure that actual quit rates are conservative, but may not necessarily lead to conservative relative treatment effects (e.g. risk ratios) if loss to follow‐up is higher in the control group than in the intervention group (Hall 2001). Some studies now use alternative methods to model effects of missing data (Hall 2001; Niaura 2002). Where differential results using alternative models were reported we considered whether the results of the meta‐analysis were sensitive to the use of different denominators.

Assessment of heterogeneity

To investigate statistical heterogeneity we use the I² statistic, given by the formula [(Q ‐ df)/Q] x 100%, where Q is the Chi² statistic and df is its degrees of freedom (Higgins 2003). This describes the percentage of the variability in effect estimates that is due to heterogeneity rather than sampling error (chance). We used threshold values of 25% and 50% as suggesting moderate and substantial heterogeneity respectively. Although we give a summary statistic, the conclusions that can be drawn from it must be cautious. Where trials are small and few in number the confidence intervals will be wide.

Assessment of reporting biases

The derivation of the summary statistic implicitly assumes that data from all randomized trials are available without any bias due to non‐publication of unpromising results or to exclusion of randomized individuals. There is evidence that publication bias occurs in the field of smoking cessation research (Egger 1997), and this issue is discussed further in the Cochrane review of nicotine replacement therapy (NRT) (Stead 2012). Thus, we included unpublished studies or studies found only as abstracts where sufficient detail was available. We contacted authors for further data if necessary. Where sufficient data were available (ten or more studies in a comparison), we used funnel plots to investigate publication bias.

Data synthesis

We summarized individual study results as a risk ratio (RR), calculated as: (number of quitters in intervention group/ number randomized to intervention group) / (number of quitters in control group/ number randomized to control group). A risk ratio greater than 1.0 indicates a higher rate of quitting in the treatment group than in the control group. For each type of medication where more than one eligible trial was identified, we performed meta‐analysis using a Mantel‐Haenszel fixed‐effect method to estimate a pooled risk ratio with 95% confidence intervals (Mantel 1959). Where studies contributed more than one intervention arm to a pooled analysis, we split the control arm to avoid double counting. We created summary of findings tables for the efficacy of bupropion and nortriptyline, using standard Cochrane methodology.

Throughout, when we discuss effect we are referring to risk ratios, and not to absolute quit rates.

Subgroup analysis and investigation of heterogeneity

We distinguished between trials testing an antidepressant as a single pharmacotherapy and those testing an antidepressant as an adjunct to NRT for initial cessation. We also distinguished between cessation trials and those where the intervention addressed relapse prevention or reduction in number of cigarettes smoked. For trials of bupropion, we have included subgroup analyses by length of follow‐up, recruitment method (clinical/community), and level of behavioural support. For the subgroup analysis based on level of additional support, we used the same criteria applied in the Cochrane NRT review (Stead 2012); low intensity support was regarded as part of the provision of routine care, so the duration of time spent with the smoker (including assessment for the trial) had to be less than 30 minutes at the initial consultation, with no more than two further assessment and reinforcement visits.

Where reported, we also extracted data from analyses evaluating a potential interaction between current depression or past history of depression and quit rates. We relied upon the definition of depression used by study authors, which included both formal diagnoses and scores on validated depression scales.

None of the trials located were specifically designed to directly compare antidepressant pharmacotherapy with non‐pharmacological therapies.

Adverse events

Tables in the results section summarize the adverse events reported in clinical trials for smoking cessation for medications which have shown evidence of efficacy (bupropion and nortriptyline). These tables include all studies of bupropion and nortriptyline in which one arm received the pharmacotherapy and the other arm did not.

In addition, for this update we have created meta‐analyses for serious adverse events (SAEs) for bupropion and nortriptyline. As per the definition provided by the U.S. Food and Drug Adminstration, SAEs were defined as any event that was life‐threatening, resulted in hospitalization, death, disability, or permanent damage, or required intervention to prevent one of the above outcomes (FDA definition). Studies had to meet three criteria to be included in SAE analyses:

the active treatment was compared to a placebo or no pharmacotherapy control;
neither arm received additional pharmacotherapy above the antidepressant being tested; and
SAEs were reported during or within 30 days of active drug treatment.

We differentiate between studies which stated that no SAEs occurred (included in the analyses) and those that did not report on SAEs, in which it is possible that no events occurred, but not explicitly made clear (excluded from analyses and recorded in AE tables). In addition to screening included studies, we also screened excluded studies of nortriptyline and bupropion where the reason for exclusion was short‐term follow‐up. Where these studies met the above three criteria, they were included in the SAE meta‐analysis.

As with smoking cessation meta‐analyses, we used number randomized as the denominator and calculated risk ratios for each comparison, calculated as (number of events in the treatment group/number randomized to the treatment group)/(number of events in the control group/number randomized to the control group). For nortriptyline and bupropion, we used any SAE as the outcome of interest. Due to concerns about specific adverse effects of bupropion, we also analysed psychiatric and cardiovascular SAEs separately for bupropion studies.

Results

Description of studies

We identified twenty‐four additional trials for this update, yielding a total of 90 included trials. The new trials were of bupropion (Cinciripini 2013; Cox 2012; Eisenberg 2013; Gariti 2009; Hall 2011; Hays 2009; Kalman 2011; Levine 2010; Piper 2009; Planer 2011; Rose 2013; Rovina 2009; Schnoll 2010 (previously in 'awaiting classification' as Schnoll 2005) Siddiqi 2013; Smith 2009; Stapleton 2013; Wittchen 2011), fluoxetine (Brown 2013), nortriptyline (Richmond 2013), SAMe (Sood 2012, not previously covered in this review), selegiline ( Kahn 2012; Killen 2010), and St John's wort (Parsons 2009; Sood 2010, not previously covered in this review). One study previously included based on unpublished data is now published (Weinberger 2010). There were sixty‐six trials of bupropion, including seven testing the medication for relapse prevention (Covey 2007; Croghan 2007; Hall 2011; Hays 2001; Hays 2009; Hurt 2003; Killen 2006) and one for reduction (Hatsukami 2004). Eight of the bupropion trials allowed a direct comparison with nicotine replacement therapy (Gariti 2009; Górecka 2003; Jorenby 1999; Piper 2009; Smith 2009; Stapleton 2013; Uyar 2007; Wittchen 2011), and four a direct comparison with the nicotine receptor partial agonist varenicline (Cinciripini 2013; Gonzales 2006; Jorenby 2006; Nides 2006). Ten trials used nortriptyline including three which also used bupropion (Haggsträm 2006; Hall 2002; Wagena 2005). There were five trials of fluoxetine (Blondal 1999; Brown 2013; Niaura 2002; Saules 2004; Spring 2007), five of selegiline (Biberman 2003; George 2003; Kahn 2012; Killen 2010; Weinberger 2010), two of St John's wort (Parsons 2009; Sood 2010), one of paroxetine (Killen 2000), one of sertraline (Covey 2002), one of venlafaxine (Cinciripini 2005), and one of SAMe (Sood 2012). One study included only patients with current depression (Brown 2013). All other studies but two (Ahluwalia 2002; Schnoll 2010) excluded smokers with current depression but almost all included smokers with a past history of depression. Further details of the study designs are given in the table 'Characteristics of included studies'.

We list 80 excluded studies. Most of these were short‐term or laboratory‐based studies. For medications where there is little or no evidence from long‐term studies we briefly describe the results of the excluded short‐term trials. The reasons for exclusion are given in the table 'Characteristics of excluded studies'. Two placebo‐controlled trials studied the use of bupropion for smokeless tobacco cessation (Dale 2002; Glover 2002). These trials are excluded from the present review but are covered in the Cochrane review of interventions for smokeless tobacco cessation (Ebbert 2011). Papers reporting additional outcomes or subgroup analyses from included studies are listed as references under the study identifier. One further study is potentially relevant but did not have sufficient data to assess for inclusion at the time of publication (Rose 2013a).

Bupropion

Sixty‐six studies of bupropion with long‐term follow‐up are included. Outcomes for four studies are based on conference abstracts or pharmaceutical company data (Ferry 1992; Ferry 1994; Selby 2003; SMK20001).

The majority of trials were conducted in North America but studies are also included from Europe (Aubin 2004; Dalsgarð 2004; Fossati 2007; Górecka 2003; Rovina 2009; Stapleton 2013; Wagena 2005; Wittchen 2011; Zellweger 2005 ); Brazil (Haggsträm 2006); Australia (Myles 2004); Israel (Planer 2011); New Zealand (Holt 2005); Pakistan (Siddiqi 2013); Turkey (Uyar 2007); and two multi‐continent studies (Tonnesen 2003; Tonstad 2003). Special populations recruited include smokers with the following conditions: chronic obstructive pulmonary disease (Górecka 2003; Tashkin 2001; Wagena 2005); schizophrenia (Evins 2001; Evins 2005; Evins 2007; George 2002; George 2008); post traumatic stress disorder (Hertzberg 2001); cancer (Schnoll 2010); suspected tuberculosis (Siddiqi 2013); alcoholism (Grant 2007; Hays 2009); and cardiovascular disease (Eisenberg 2013; Planer 2011; Rigotti 2006; Tonstad 2003). Three of the studies in patients with cardiovascular disease, and one other, enrolled hospital inpatients (Eisenberg 2013; Planer 2011; Rigotti 2006;Simon 2009). Other populations included adolescents (Killen 2004; Muramoto 2007); smokers awaiting surgery (Myles 2004); hospital staff (Dalsgarð 2004); healthcare workers (Zellweger 2005); African‐Americans (Ahluwalia 2002; Cox 2012); and Maori (Holt 2005). Two studies recruited smokers who had previously failed to quit smoking using bupropion (Gonzales 2001; Selby 2003), and two included smokers who had just failed to quit using NRT (Hurt 2003; Rose 2013).

More than half the bupropion studies followed participants for at least 12 months from the start of treatment or the target quit day. Twenty‐six studies (40%) had only six months‐follow‐up (Ahluwalia 2002; Aubin 2004; Cinciripini 2013; Collins 2004; Cox 2012; Dalsgarð 2004; Evins 2001; Evins 2005; George 2002; George 2008; Grant 2007; Haggsträm 2006; Hatsukami 2004; Hertzberg 2001; Kalman 2011; Killen 2004; Muramoto 2007; Myles 2004; Rose 2013; Schnoll 2010; Siddiqi 2013; Simon 2009; Smith 2009; Stapleton 2013; Uyar 2007; Wagena 2005). The majority of studies reported an outcome of sustained abstinence. In 18 (28%) only point prevalence rates were given, or the definition of abstinence was unclear (Cox 2012; Evins 2005; Gariti 2009; George 2002; George 2008; Grant 2007; Hall 2011; Killen 2004; Muramoto 2007; Myles 2004; Piper 2007; Piper 2009; Schmitz 2007; Schnoll 2010; Selby 2003; Smith 2009; Swan 2003; Uyar 2007).

Forty‐four trials evaluated bupropion as a single pharmacotherapy to assist initial cessation. Twelve trials that tested bupropion as an adjunct to nicotine replacement therapy are pooled separately (Evins 2007; Kalman 2011; Killen 2004; Jorenby 1999 (part); George 2008; Grant 2007; Piper 2009; Rose 2013; Schnoll 2010; Simon 2004; Smith 2009; Stapleton 2013), as are the eight trials making direct comparisons between bupropion and nicotine replacement therapy (Gariti 2009; Górecka 2003; Jorenby 1999; Piper 2009; Smith 2009; Stapleton 2013; Uyar 2007; Wittchen 2011), three comparing bupropion and nortriptyline (Haggsträm 2006; Hall 2002; Wagena 2005), and four comparing bupropion and varenicline (Cinciripini 2013; Gonzales 2006, Jorenby 2006; Nides 2006). Trials testing the extended use of bupropion for relapse prevention (Covey 2007; Croghan 2007; Hall 2011; Hays 2001; Hays 2009; Hurt 2003; Killen 2006) and its use for assisting in reducing the amount smoked (Hatsukami 2004) are pooled separately.

The seven studies that evaluated bupropion SR for relapse prevention each had slightly different designs. These studies also contribute to a separate Cochrane review on interventions for relapse prevention (Hajek 2013).

One study evaluated bupropion for reducing smoking in people not wanting to make a quit attempt but interested in reducing smoking (Hatsukami 2004). During treatment, if participants decided they wanted to try to quit, they were enrolled in a cessation programme during which they continued to use bupropion and were then followed up for 19 weeks.

Most of the bupropion trials excluded participants with current depression but not those with a history of depression. Two studies did include participants with current depression (Ahluwalia 2002; Schnoll 2010). Two studies explicitly excluded participants with a past history of major depression (Dalsgarð 2004) or any psychiatric disorder (Collins 2004). Amongst the studies recording the prevalence of a past history of depression at baseline, the proportion ranged from 6% (Hatsukami 2004) to 44% (Swan 2003), but was typically 20 to 30%.

Nortriptyline

Ten published studies of the tricyclic antidepressant nortriptyline are included. Richmond 2013 is new since the last review, and was conducted in prisoners. Hall and colleagues conducted three trials, and Prochazka and colleagues two, both in the USA. Two studies were conducted in Brazil (Da Costa 2002; Haggsträm 2006), one in the Netherlands (Wagena 2005), one in the UK (Aveyard 2008), and one in Australia (Richmond 2013). Eight studies excluded participants with current depression but most of these included people with a history of depression. All studies were placebo controlled and used doses of 75 to 100 mg/day or titrated doses to serum levels recommended for depression during the week prior to the quit date. Treatment duration ranged from 12 to 14 weeks. All studies used a definition of cessation based on a sustained period of abstinence. The three studies by Hall and colleagues, Aveyard 2008, and Richmond 2013 reported outcomes after 12 months of follow‐up and the other five studies had six months of follow‐up.

The three studies by Hall and colleagues (Hall 1998; Hall 2002; Hall 2004) used factorial designs to test nortriptyline versus placebo crossed with different intensities of behavioural support, whereas the remaining studies provided a set amount of behavioural support to all participants, ranging from brief behavioural counselling to repeated group and individual sessions. Four studies tested nortriptyline as an adjunct to NRT (Aveyard 2008; Hall 2004; Prochazka 2004; Richmond 2013) and six tested nortriptyline on its own (Da Costa 2002; Haggsträm 2006; Hall 1998; Hall 2002; Prochazka 1998; Wagena 2005).

Selective Serotonin Reuptake Inhibitors (SSRIs)

One new study of an SSRI has been added since the last update of this review (Brown 2013, evaluating fluoxetine).

Fluoxetine

Five trials with long‐term follow‐up are included. Two studies used fluoxetine as the only pharmacotherapy and had six months follow‐up: a multicentre trial compared 30 mg daily, 60 mg daily, or placebo for 10 weeks (Niaura 2002) and a single‐site study used 60 mg or placebo for 12 weeks (Spring 2007). Three trials provided NRT to all participants and evaluated the addition of fluoxetine over 12 months follow‐up: Blondal 1999 used 20 mg/day or placebo for three months as an adjunct to nicotine inhaler; Saules 2004 used 20 or 40 mg/day or placebo for 10 weeks as an adjunct to nicotine patch; and Brown 2013 compared 10 weeks of fluoxetine, 16 weeks of fluoxetine, or no additional treatment in participants using nicotine patch for eight weeks. Brown 2013 was conducted in smokers with elevated depressive symptoms. Participants in all other trials were not currently depressed but may have had a past history of depression. Spring 2007 stratified by history of depression.

We list as excluded other short‐term studies, one with three month abstinence outcomes (Spring 1995) and others which reported outcomes other than abstinence (Cornelius 1997; Cornelius 1997; Dalack 1995; Naranjo 1990; Pomerleau 1991; Stein 1993). Another pharmaceutical company‐sponsored multicentre trial was completed but its results were never presented or published (personal communication).

Paroxetine

One trial with six‐month follow‐up assessed paroxetine (40 mg, 20 mg, or placebo) for nine weeks as an adjunct to nicotine patch (Killen 2000).

Sertraline

One trial with six‐month follow‐up assessed sertraline (200 mg/day) for 11 weeks versus placebo in conjunction with six individual counselling sessions. There were 134 participants, all current smokers with a past history of major depression (Covey 2002). One trial that combined sertraline with buspirone was excluded because the specific effect of sertraline could not be evaluated (Carrão 2007).

Citalopram/zimelidine

There were no long‐term studies of citalopram or zimelidine. One short‐term study used a crossover design to investigate the effect of the SSRIs citalopram or zimelidine on the smoking behaviour of heavy drinkers who were not attempting to stop smoking. Their cigarette use did not change significantly between active medication and placebo periods (Sellers 1987).

Monoamine oxidase inhibitors

Moclobemide

Moclobemide has been tested for smoking cessation in one long‐term placebo‐controlled trial in France (Berlin 1995). Treatment with 400 mg/day began one week before quit day and continued for two months, reducing to 200 mg/day for a further month. No behavioural counselling was provided. Final follow‐up was at 12 months.

Selegiline

Five long‐term trials have been published. Three used 10 mg/day oral treatment (Biberman 2003; George 2003; Weinberger 2010) and two used 6 mg/day patch treatment (Kahn 2012; Killen 2010). Three had treatment durations of nine weeks (George 2003; Kahn 2012; Weinberger 2010), one had a treatment duration of eight weeks (Killen 2010), and one continued therapy for 26 weeks (Biberman 2003). One excluded study was terminated early due to lack of efficacy and reported results to nine weeks only (Le Foll 2009). One other short‐term study (Houtsmuller 2002) and one reporting only preliminary short‐term data (Brauer 2000) are also excluded.

Lazabemide

There were no long‐term studies of this selective MAO‐B inhibitor. It was evaluated in an eight‐week, dose finding, exploratory study (Berlin 2002). The trial was halted early due to liver toxicity observed in trials of the medications for other indications, and lazabemide is not being developed further. Continuous four‐week quit rates at the end of treatment, including all drop‐outs as treatment failures, were 17% (18/108) for 200 mg/day, 11% (12/108) for 100 mg/day, and 9% (10/114) for placebo.

EVT302

There are no long‐term studies of this selective reversible MAO‐B inhibitor. It has been evaluated in an eight‐week placebo controlled study (Berlin 2012), excluded due to short follow‐up. There was no evidence of clinical benefit and no significant difference between active and placebo groups. Continuous four‐week quit rates at the end of treatment, including all drop‐outs as treatment failures were 17.2% (25/145) for EVT302, 15.2 % (22/145) for placebo, 32.8 % (22/61) for EVT302 plus nicotine patch, and 26.8 % (17/61) for placebo plus nicotine patch.

Other antidepressants

Three studies of other antidepressants have been included since the last review: two studies of St John's wort (Parsons 2009; Sood 2010) and one trial of the dietary supplement SAMe (Sood 2012).

Venlafaxine

One long‐term trial with 147 participants compared venlafaxine at a dose of up to 225 mg/day with placebo. All participants also received nicotine patches and nine brief individual counselling sessions; follow‐up was for 12 months (Cinciripini 2005). An unpublished short‐term study (Frederick 1997) reported no difference in abstinence rates at eight weeks, and frequent side effects in the treatment group.

Hypericum (St John's wort)

Two studies with long‐term follow‐up are included (Parsons 2009; Sood 2010). Both reported prolonged abstinence at six months. Parsons 2009 compared 14 weeks of 900 mg/day St John's wort to placebo and Sood 2010 compared 900 mg/day, 1800 mg/day, and placebo for 12 weeks.

Two studies are excluded. One unpublished study of an oral spray and placebo control with 45 participants detected no difference in abstinence at one month follow‐up (Becker 2003). Barnes 2006 compared two doses for smoking cessation in an open randomized study with no placebo control. Quit rates were low and did not differ between dose levels. No participants maintained abstinence for 12 months.

S‐Adenosyl‐L‐Methionine (SAMe)

One long‐term trial with 120 participants compared 1600 mg/day or 800 mg/day SAMe to placebo for 8 weeks with follow‐up at six months.

Doxepin

There are no long‐term studies of this serotonergic tricyclic antidepressant. It has been evaluated in a single small trial with short‐term follow‐up (Edwards 1989). Treatment was with 150 mg doxepin daily for three weeks prior to quit day and four weeks afterwards. Subjects forfeited a US$135 deposit if they failed to stop smoking for seven days. Two months after cessation, 78% (7/9) of the doxepin group and 10% (1/10) of the placebo group reported abstinence, a statistically significant difference (P < 0.02). However one week post‐cessation abstinence rates using stringent validated abstinence criteria failed to show a statistically significant difference. Among withdrawal symptoms, there was a significant group difference only for craving.

Imipramine

There are no long‐term studies of this noradrenergic tricyclic antidepressant. One trial (Jacobs 1971) compared imipramine (25 mg 3 times/day) with lobeline, dextroamphetamine, placebo and a no‐medication control. Some participants attended group support sessions. After three months, success rates, which included a reduction in smoking to less than 10% of baseline, were 56% (10/18) for imipramine, 40% (6/15) for placebo, and 69% (27/39) for the no‐medications control. These differences were not statistically significant.

Tryptophan

There have been no long‐term studies reported. Bowen and colleagues postulated that this serotonin‐enhancing action in conjunction with a high carbohydrate (CHO) diet might relieve the negative affect of cigarette withdrawal. Oral l‐tryptophan (50mg/kg/day) and instructions to follow a high CHO, low‐protein diet were compared with placebo pills and instructions for a low‐carbohydrate diet (Bowen 1991). Participants in both groups also received four two‐hour weekly multi‐component group therapy sessions. Two weeks following the target cessation date 75% (12/16) of the tryptophan and high CHO diet group were abstinent versus 47% (7/15) of the placebo and low CHO diet group. This difference was not statistically significant.

Risk of bias in included studies

The majority of studies were judged to be at unclear risk for selection, performance and detection bias and at low risk for attrition bias. Figure 1 displays risk of bias judgements across each domain.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Risk of bias in each area is summarized below. Additional details about the methodology of individual trials are given in the table Characteristics of included studies. Consistent with Cochrane methods, we included some trials that have only been published as abstracts, which have limited information on methodological issues (Higgins 2008). For some studies we have obtained additional information from authors, or from the pharmaceutical company funding the study. Use of unpublished data in the meta‐analysis is noted in the Characteristics of included studies table.

Allocation

All of the trials were described as randomized, but most failed to report randomization and concealment methods in detail. Twenty‐nine studies (32%) reported a method of sequence generation and allocation concealment judged to place the results at low risk of selection bias. One study was judged to be at high risk of selection bias due to a lack of allocation concealment (Wittchen 2011). All other trials were judged to be at unclear risk of selection bias because the method of sequence generation or allocation was not described in sufficient detail.

Blinding

Twenty‐seven trials (30%) provided sufficient information on blinding to be judged to be at low risk of performance and detection bias. These trials were placebo controlled, with both participants and personnel blinded to treatment allocation. Thirteen studies either did not blind participants or personnel or reported evidence suggesting unblinding, and were hence judged to be at high risk of performance and detection bias. The remaining studies did not provide sufficient information on blinding with which to judge risk of bias. For the most part, these trials reported that they were "double‐blind" but provided no additional detail on the nature of the blinding used.

Incomplete outcome data

The majority of studies (63%, 57 trials) reported low losses to follow‐up and were judged to be at low risk of attrition bias. Eight studies reported high (> 50%) or differential (> 20% difference between arms) loss to follow‐up and were judged to be at high risk of bias for this domain. The remaining 25 studies did not report the number or percentage of participants lost to follow‐up in each arm, and were hence judged to be at unclear risk for attrition bias.

Other potential sources of bias

One cluster‐randomized trial (Siddiqi 2013) was judged to be at high risk of other bias. In this trial, despite adequate reported methods of sequence generation and allocation concealment, the authors found substantial heterogeneity of program effects across clusters and a baseline imbalance in cigarette smoking behaviour (20% of participants in the control arm smoked only waterpipes (no cigarettes) compared to 4% in the intervention arm).

Definition of abstinence

The definition of abstinence was not always explicit and biochemical validation of self‐reported smoking status was not always used. However, all but four of the bupropion studies (Planer 2011; Smith 2009; Swan 2003; Wittchen 2011) and all but one of the nortriptyline studies (Da Costa 2002) used biochemical verification for most self‐reported quitters at some assessment points. In a small number of studies we were able to obtain a sustained outcome that was not given in the published report. Most of the sustained abstinence rates are based on self‐reported slip‐free abstinence from the start of the third week after the target quit date (TQD) onward and validated at intermediate and final follow‐ups.

Effects of interventions

See: Table 1; Table 2

Summary of findings for the main comparison. Bupropion for smoking cessation.

Bupropion for smoking cessation
Patient or population: people who smoke  Intervention: bupropion
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Bupropion
Bupropion versus placebo/control. Abstinence   Follow‐up: 6+ months	115 per 1000¹	187 per 1000   (172 to 203)	RR 1.62   (1.49 to 1.76)	13728  (44 studies)	⊕⊕⊕⊕  high^2,3
Bupropion and NRT versus NRT alone. Abstinence   Follow‐up: 6+ months	186 per 1000¹	221 per 1000   (175 to 281)	RR 1.19   (0.94 to 1.51)	3487  (12 studies)	⊕⊕⊝⊝  low^3,4,5
Bupropion versus NRT. Abstinence   Follow‐up: 6+ months	254 per 1000¹	244 per 1000   (216 to 277)	RR 0.96   (0.85 to 1.09)	4086  (8 studies)	⊕⊕⊕⊝  moderate⁴
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; RR: Risk ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Nortriptyline for smoking cessation
Patient or population: people who smoke  Intervention: nortriptyline
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Nortriptyline
Nortriptyline versus placebo. Abstinence   Follow‐up: 6+ months	99 per 1000¹	201 per 1000   (147 to 275)	RR 2.03   (1.48 to 2.78)	975  (6 studies)	⊕⊕⊕⊝  moderate^2,3
Nortriptyline and NRT versus NRT alone. Abstinence   Follow‐up: 6+ months	116 per 1000¹	141 per 1000   (109 to 180)	RR 1.21   (0.94 to 1.55)	1644  (5 studies)	⊕⊕⊕⊝  moderate^2,3
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; RR: Risk ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Nortriptyline vs placebo. 'No report' = no information, 'None occurred' = explicit statement
Study	Serious events	Other adverse events	Withdrawal due to AE
Aveyard 2008	2 admissions to hospital (1 intervention, 1 control) with collapse or palpitations judged possibly caused by treatment	Dry mouth (80% vs 'more than half, OR 6.67, 5.12 to 8.69), Constipation (OR 2.06, 1.66 to 2.56) and Sweating (OR 1.37, 1.11 to 1.68) significantly more common	No report
Da Costa 2002	None occurred	No significant differences between groups.  Dry mouth 44% vs 24%, constipation 29% vs 12%, irritation 18% vs 24%, anxiety 18% vs 28%.	No report
Haggsträm 2006	None occurred	Dry mouth (67.3% vs 31.4%) and drowsiness (19.2% vs 11.8%) significantly more common with nortriptyline  Constipation (15.4% vs 9.8%) NS	No report
Hall 1998	None occurred	Dry mouth (78% vs 33%), lightheadedness (49% vs 22%), shaky hands (23% vs 11%) and blurry vision (16% vs 6%) were significantly more common.	4 (4%) dropouts due to side effects in drug and 1 (1%) in placebo group. Total dropout rates 30% in drug and 17% in placebo groups
Hall 2002	None occurred	Dry mouth (72% vs 33%) and constipation ( 32% vs 14%) significantly more common with nortriptyline	No report
Hall 2004	None occurred	Dry mouth (85% vs 40%), lightheadedness (44% vs 22%), shaky hands (30% vs 14%) constipation (38% vs 15%), blurry vision (23% vs 7%), difficulty urinating (13% vs 2%), all p<0.01. Sexual difficulties (19% vs 2% p <0.02)	4 (10%) in active brief treatment, 10 (25%) in active extended treatment, 9 (11%) in placebo stopped medication due to adverse effects (active vs placebo p=0.18, Fisher's exact test)
Prochazka 1998	None occurred	Dry mouth (64% vs 23%), dysgeusia (20% vs 8%), GI upset (41% vs 24%), drowsiness (24% vs 8%) significantly more common	10 (9%) treatment withdrawal due to adverse events in nortriptyline group, vs 3 (3%) in placebo group.
Prochazka 2004	None occurred	Dry mouth (38% vs 8%) & sedation (20% vs 3%) significantly more common.	10 (13%) discontinued nortriptyline including 1 subject with a normal baseline ECG who developed asymptomatic prolongation of QT interval, vs 1 placebo
Wagena 2005	One death in placebo group, previously hospitalised with dermatological reactions	Dry mouth (61% vs 20%), diarrhoea or constipation (48% vs 26%) and fatigue (20% vs 9%) significantly more common in nortriptyline group	24% discontinued nortriptyline vs 9% placebo

Date	Event	Description
8 October 2013	New search has been performed	Updated with 24 new included studies. Studies of S‐Adenosyl‐L‐Methionine and St John's wort included for the first time. Meta‐analyses of serious adverse events added.
8 October 2013	New citation required but conclusions have not changed	Conclusions largely unchanged. Efficacy findings unchanged.
22 June 2011	Amended	Additional table converted to appendix to correct pdf format
5 October 2009	Amended	Correction to excluded studies table, detail for Carrão 2007
30 July 2009	New search has been performed	Updated with 13 new included trials including 3 of selegiline, not previously covered. No substantial change to effects, main conclusions not altered
17 June 2008	Amended	Converted to new review format.
11 October 2006	New citation required but conclusions have not changed	Seventeen new trials were added to the review for issue 1, 2007. There were no major changes to the reviewers' conclusions.
16 July 2004	New citation required but conclusions have not changed	New trials of bupropion, nortriptyline and fluoxetine were added for issue 4, 2004, and additional information on adverse effects was included. There were no major changes to the reviewers' conclusions.
8 January 2003	New citation required but conclusions have not changed	New trials of bupropion and nortriptyline were added to the review in Issue 2 2003. There were no major changes to the reviewers' conclusions
19 September 2001	New citation required but conclusions have not changed	Four new studies on bupropion, and one each on nortriptyline and paroxetine were added to the review in Issue 1 2002. In press data from a trial of fluoxetine are included which differ from unpublished data previously used. The reviewers' conclusions about the efficacy of bupropion and nortriptyline were not changed substantively.
28 August 2000	New citation required and conclusions have changed	Updates the earlier Cochrane review 'Anxiolytics and antidepressants for smoking cessation'. Anxiolytics are evaluated in a separate review.

Country	No. of reports	No. of users	Rate of reports	Events reported	Source/ date
UK (safety notice 2002)	7,630  24 July 2002	540,000 patients (31 March 2002)	14/1000	Commonest:  Urticaria/rashes/pruritus (2357, 31% of total reports)  Insomnia (994, 13%)  Headache (779, 10%)  Dizziness (747, 10%)  Seizures: 184 (2.4% of reports, est rate 0.3/1000)  Deaths: 60	Medicines Control Agency  26 July 2002 http://www.mca.gov.uk/ourwork/  monitorsafequalmed/  safetymessages/zyban26702.pdf
UK (MHRA routine data on suspected adverse drug reactions). Includes reports summarised in UK safety notice 2002	8,452 (18,319 reactions) to May 2004	Estimated approximately 1,000,000 community prescriptions dispensed in UK 2000‐April 2004, based on England data of 762,200 for 2000‐2003. Average prescription was a 4 week course.	approx 8/1000 prescriptions	By organ class:  ‐General disorders: 4127  Includes insomnia (1030), dizziness (805), chest pain (384)  ‐Skin & subcutaneous tissue disorders: 3728  Includes pruritus (421), rashes (1094), urticaria (1104), angioedema (481)  ‐Psychiatric disorders: 2417  Includes affective disorders (627), suicidal ideation/ suicide/parasuicide (87)  ‐Neurological disorders: 2338  Includes convulsions & epilepsy (212)  ‐Gastrointestinal disturbances: 2176  Includes nausea / vomiting (775)  ‐Deaths: 70, includes 4 suicides	Medicine and Healthcare products Regulatory Agency (MHRA) Adverse Drug Reactions Information Service. Data provided April 2004.  Prescription data from DoH Prescription Cost Analysis (www.publications.doh.gov.uk/prescriptionstatistics/index.htm)
Canada (safety notice 2001)	1,127  28 May 2001	1,245,000 Zyban, 699,000 Wellbutrin  (April 30 2001)	0.6/1000	Full list not given  Serious: 682  Seizures: 172 (Zyban 120, Wellbutrin 46 bupropion 6) (15% of reports, est rate 0.1/1000 for Zyban  Deaths: 19  Serum sickness: 37	Canadian Adverse Reaction Monitoring Programme (CADRMP)GSK/ Health Canada 'Dear Doctor' letter 3 July 2001  http://www.hc‐sc.gc.ca/hpfb‐dgpsa/tpd‐dpt/zyban_e.html
Australia (safety alert 2001)	1,237  August 2001	Not given	not available	Full list not given  skin reactions (499 reports), psychiatric disturbances (427), nervous system (406, includes convulsions/twitching 74), gastrointestinal tract (258), facial/angioedema (89), serum sickness (63)  Deaths: 18	Therapeutic Goods Administration (TGA) alert 31 August 2001  http://www.health.gov.au/tga/ docs/html/zyban.htm
Australia  (routine data)	1,672 (4,390 reactions)  to March 2004	Approx 534,000 prescriptions 2000‐March 2004, mainly 30 pill/2 week	approx 3/1,000 prescriptions	Deaths: 31  Skin & subcutaneous tissue: 930. Includes urticaria (366), pruritus (90), rash (164), oedema (160)  Nervous system: 792. Includes convulsions (105), Psychiatric disorders: 992. Includes suicide/self‐injurious ideation (32), depression (13)  Gastointestinal disorders: 440. Includes nausea/ vomiting (221)	TGA data supplied 31 March 2004
France  (analysis of pharmacovigilance database & GSK reports)	1682 of which 475 classified as serious, September 2001 to September 2004	698,000 patients treated	approx 2.4/1,0000 people	Deaths: 21 (including 3 suicides)  Neuropsychiatric: 62, includes suicide attempts (21), suicidal ideation (19), seizures: (75, incidence 0.01%).  Skin & subcutaneous tissue: 148, includes angiodema (50), serum sickness like reaction (40), urticaria (27). 11 intentional overdose including 2 deaths	Beyens 2008

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Bupropion versus placebo/control. Subgroups by length of follow‐up	44	13728	Risk Ratio (M‐H, Fixed, 95% CI)	1.62 [1.49, 1.76]
1.1 Twelve month follow‐up	27	9866	Risk Ratio (M‐H, Fixed, 95% CI)	1.59 [1.44, 1.76]
1.2 Six month follow‐up	17	3862	Risk Ratio (M‐H, Fixed, 95% CI)	1.69 [1.45, 1.97]
2 Bupropion versus placebo/control. Subgroups by clinical/recruitment setting	44	13728	Risk Ratio (M‐H, Fixed, 95% CI)	1.62 [1.49, 1.76]
2.1 Community volunteers	21	7524	Risk Ratio (M‐H, Fixed, 95% CI)	1.67 [1.49, 1.87]
2.2 People recruited from health care settings	18	3928	Risk Ratio (M‐H, Fixed, 95% CI)	1.60 [1.38, 1.86]
2.3 Community + health care settings	1	540	Risk Ratio (M‐H, Fixed, 95% CI)	1.33 [0.83, 2.13]
2.4 Health care professionals/ hospital staff	2	1002	Risk Ratio (M‐H, Fixed, 95% CI)	1.32 [0.98, 1.78]
2.5 People with a previously unsuccessful quit attempt using bupropion	2	734	Risk Ratio (M‐H, Fixed, 95% CI)	2.25 [1.29, 3.90]
3 Bupropion versus placebo. Subgroups by level of behavioural support	41	13012	Risk Ratio (M‐H, Fixed, 95% CI)	1.63 [1.49, 1.77]
3.1 Multisession group behavioural support	10	2001	Risk Ratio (M‐H, Fixed, 95% CI)	1.76 [1.44, 2.16]
3.2 Multisession individual counselling	30	10964	Risk Ratio (M‐H, Fixed, 95% CI)	1.60 [1.45, 1.76]
3.3 Low intensity support	1	47	Risk Ratio (M‐H, Fixed, 95% CI)	2.88 [0.32, 25.68]
4 Bupropion dose response. 300 mg/day versus 150 mg/day	3	2042	Risk Ratio (M‐H, Fixed, 95% CI)	1.08 [0.93, 1.26]
5 Bupropion and NRT versus NRT alone	12	3487	Risk Ratio (M‐H, Random, 95% CI)	1.19 [0.94, 1.51]
5.1 Patch alone	9	1774	Risk Ratio (M‐H, Random, 95% CI)	1.24 [0.84, 1.84]
5.2 Lozenge alone	2	1051	Risk Ratio (M‐H, Random, 95% CI)	1.21 [0.81, 1.81]
5.3 Choice of NRT	1	662	Risk Ratio (M‐H, Random, 95% CI)	0.97 [0.73, 1.28]
6 Bupropion for relapse prevention	7	1959	Risk Ratio (M‐H, Fixed, 95% CI)	1.15 [1.00, 1.33]
7 Bupropion versus NRT	8	4086	Risk Ratio (M‐H, Fixed, 95% CI)	0.96 [0.85, 1.09]
7.1 Patch	6	1634	Risk Ratio (M‐H, Fixed, 95% CI)	1.04 [0.84, 1.27]
7.2 Lozenge	2	694	Risk Ratio (M‐H, Fixed, 95% CI)	0.91 [0.67, 1.22]
7.3 Patch + lozenge	2	720	Risk Ratio (M‐H, Fixed, 95% CI)	0.74 [0.55, 0.98]
7.4 Choice of NRT	2	1038	Risk Ratio (M‐H, Fixed, 95% CI)	1.08 [0.87, 1.33]
8 Bupropion versus varenicline	4	1810	Risk Ratio (M‐H, Fixed, 95% CI)	0.68 [0.56, 0.83]
9 Bupropion for harm reduction	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
9.1 Reduction in cotinine >50% from baseline at 1y	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
9.2 Cessation at 6m	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Nortriptyline versus placebo	6	975	Risk Ratio (M‐H, Fixed, 95% CI)	2.03 [1.48, 2.78]
2 Nortriptyline and NRT versus NRT alone	4	1644	Risk Ratio (M‐H, Fixed, 95% CI)	1.21 [0.94, 1.55]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Nortriptyline vs placebo. 'No report' = no information, 'None occurred' = explicit statement			Other data	No numeric data
2 Bupropion versus control. 'No report' = no information, 'None occurred' = explicit statement			Other data	No numeric data

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 SSRI versus placebo/control	4	1594	Risk Ratio (M‐H, Fixed, 95% CI)	0.93 [0.71, 1.22]
1.1 Fluoxetine	2	1236	Risk Ratio (M‐H, Fixed, 95% CI)	0.92 [0.65, 1.30]
1.2 Paroxetine	1	224	Risk Ratio (M‐H, Fixed, 95% CI)	1.08 [0.64, 1.82]
1.3 Sertraline	1	134	Risk Ratio (M‐H, Fixed, 95% CI)	0.71 [0.30, 1.64]
2 SSRI and NRT versus NRT alone	3	466	Risk Ratio (M‐H, Fixed, 95% CI)	0.70 [0.48, 1.03]
2.1 Fluoxetine	3	466	Risk Ratio (M‐H, Fixed, 95% CI)	0.70 [0.48, 1.03]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 MAOIs versus placebo	6	827	Risk Ratio (M‐H, Fixed, 95% CI)	1.29 [0.93, 1.79]
1.1 Moclobemide	1	88	Risk Ratio (M‐H, Fixed, 95% CI)	1.57 [0.67, 3.68]
1.2 Selegiline	5	739	Risk Ratio (M‐H, Fixed, 95% CI)	1.25 [0.88, 1.78]

Methods	BUPROPION  Randomized controlled trial Setting: community‐based health care centre, USA  Recruitment: community volunteers
Participants	600 African American smokers, > 10 CPD; 70% F, av. age 44, av. CPD 17, 27% had possible clinical depression (CES‐D > 16)
Interventions	1. Bupropion 300 mg/day for 7 weeks  2. Placebo  Both arms: 8 sessions of in‐person or telephone counselling & S‐H guide
Outcomes	Abstinence at 26w (prolonged)   Validation: CO <= 10 ppm, discrepancies resolved with cotinine <= 20 mg
Notes	Continuous abstinence rates shown in Figure 3 of paper. Figures obtained from authors. Funding: National Cancer Institute. GlaxoSmithKline provided study medication.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomization codes were generated in blocks of 50 and sent to the pharmaceutical company..."
Allocation concealment (selection bias)	Low risk	Blinded drugs provided to investigator; " ... [the pharmaceutical company]... packaged the treatment and then shipped the blinded drug to the investigator."
Blinding (performance bias and detection bias)   All outcomes	Low risk	Double blind. "Blinding was successful. At the end of treatment, 58% (150/259) of participants correctly guessed that they received bupropion SR, and 41% (104/253) correctly guessed they received placebo."
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Approximately 32% lost to follow‐up in each group; included as smokers.

Methods	BUPROPION Randomized controlled trial  Setting: 74 cessation outpatient clinics, France  Recruitment: volunteers  Randomization: computer‐generated, blind
Participants	504 smokers, >= 10 CPD; 56% F, av age 41, av CPD NS, 16% history of MDD
Interventions	1. Bupropion 300 mg for 7 weeks  2. Placebo  Both arms: motivational support at clinic visits at baseline, w3, w7, w12 & 3 phone calls TQD, 2‐3 days later, w5, w18
Outcomes	Abstinence at 26w (continuous from w4)  Validation: CO < 10 ppm
Notes	First included as Lebargy 2003 based on abstract. Funding: GlaxoSmithKline
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"The computerized randomization schedule, prepared by the sponsor, was inaccessible to the investigator who was provided with a specific set of sequential treatment numbers."
Allocation concealment (selection bias)	Low risk	"The computerized randomization schedule, prepared by the sponsor, was inaccessible to the investigator who was provided with a specific set of sequential treatment numbers."
Blinding (performance bias and detection bias)   All outcomes	Low risk	"Double‐blind" "Blinding was assured by matching the placebo to the bupropion tablets..."
Incomplete outcome data (attrition bias)   All outcomes	Low risk	26% of the placebo and 27% of the bupropion groups lost; included as smokers.

Methods	NORTRIPTYLINE Randomized controlled trial  Setting: National Health Service stop smoking clinics, UK  Recruitment: People attending clinics
Participants	901 smokers, ≥10/day; 46% F, av. age 43, av. CPD 21
Interventions	1. Nortriptyline 75 mg/day, for 8 w including tapering (max dose for 6w)  2. Placebo capsules  All participants received free NRT and had behavioural support, the majority attending group sessions run by cessation specialists
Outcomes	Abstinence at 12 months (prolonged from day 15 post quit)  Validation: CO at 4w, saliva cotinine (collected by post) at 6m & 12m
Notes	Funding: Cancer Research UK and National Institute for Health Research. Medication provided by King Pharmaceuticals.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"An independent statistician generated the randomisation  schedule in Stata. We used block randomisation,  with randomly ordered block sizes of two, four, and six,  stratified by stop smoking adviser."
Allocation concealment (selection bias)	Low risk	Study nurses recruited participants, and the study administrator (who had not met the participants) allocated participants in sequence against the list for each adviser. Only the administrator and the pharmacist knew the allocation.
Blinding (performance bias and detection bias)   All outcomes	Low risk	"Advisers, participants, and study staff...were blind to allocation... tablets were encapsulated, and identical powder filled capsules provided the placebos."
Incomplete outcome data (attrition bias)   All outcomes	Low risk	12% I, 17% C lost at 12m, included as smokers. Authors note that majority of losses were already smoking.

Methods	MOCLOBEMIDE Randomized controlled trial  Setting: clinic, France  Recruitment: By adverts in general practices or from occupational medicine depts
Participants	88 smokers, >20/day and FTQ>=6. No current major depression or anxiety disorders. 57% had history of MDD
Interventions	1. Moclobemide 400 mg/day for 1w pre‐ and 2m post‐TQD, 200 mg for 3rd month  2. Placebo (P)  No behavioural intervention or counselling
Outcomes	Abstinence at 1 year (prolonged)  Abstinence verified at all visits up to 6m by plasma cotinine <= 20 ng/ml. 1 year abstinence based on telephone self report by 6m quitters.
Notes	There were no serious adverse reactions. Insomnia was more common in drug (36%) than P (7%) groups. There were 4 drop‐outs for adverse effects/relapse in drug and 2 in P. Funding: Roche
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization method not described.
Allocation concealment (selection bias)	Unclear risk	Double‐blind, but blinding at allocation not explicit.
Blinding (performance bias and detection bias)   All outcomes	Unclear risk	"Double‐blind" but further detail not provided
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Number lost to follow‐up not reported. "Relapses and subjects lost from follow‐up were considered treatment failures."

Methods	SELEGILINE Randomized controlled trial  Setting: 3 community‐based clinic, Israel  Recruitment: mailing to members of public health service provider
Participants	109 smokers (15+ CPD); 38% F, av. age 42, av. CPD 27‐30
Interventions	1. Selegiline 10 mg/day for 26 weeks, nicotine patch 21 mg for 8 weeks incl tapering  2. Placebo & nicotine patch  Both arms: Behavioural support from trained family physician; weekly then fortnightly visits for 12w
Outcomes	Abstinence at 52 w, continuous with validation at each visit  Validation: negative for urine nicotine, cotinine, 3‐hydroxycotinine (Niccheck)
Notes	No serious AEs, no significant differences in AEs, 2 selegiline discontinuations. Funding not reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Four hundred dice‐throwing generated a randomized sequence code; 199 containers prepacked with selegiline and 201 containers prepacked with placebo were numbered accordingly." Judged adequate.
Allocation concealment (selection bias)	Low risk	"The code was sealed, kept secretly and was revealed for the first time when the last participant finished the 12 months of follow‐up. The first participant who joined the trial after the initial visit run‐in phase received the first bottle from the container set number 001, the second  participant from set number 002 and so on. The trial coordinator arranged participant’s allocation." Judged adequate.
Blinding (performance bias and detection bias)   All outcomes	Low risk	"Double‐blind" (see above) "No discontinuation difference for selegiline or placebo was observed among the groups, which implies masking success."
Incomplete outcome data (attrition bias)   All outcomes	Low risk	19 lost to follow‐up, included as smokers in MA.

Methods	FLUOXETINE Randomized controlled trial  Setting: cessation clinic, Iceland  Recruitment: community volunteers
Participants	100 smokers (excl 5 early withdrawals), > 10 CPD; 62% F, av age 41, av CPD 28, 38% fluoxetine/56% placebo had history of depression
Interventions	1. Nicotine inhaler and fluoxetine for 3m, option of continuing for 3m more. Fluoxetine 10 mg/day initiated 16 days before TQD, increased to 20 mg/day on day 6.  2. Nicotine inhaler and placebo  Both arms: 5 x 1 hr group behaviour therapy. Advised to use 6‐12 inhalers/day for up to 6m.
Outcomes	Abstinence at 1 year (sustained from quit day)  Validation: CO < 10 ppm at all assessments (6w, 3,6, 12 m)
Notes	Funding: Oddur Olafsson Fund, Pharmacia & Upjohn Consumer Health Care. Delta Pharmaceutical Company provided fluoxetine and placebo and fluoxetine analyses.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated randomization; part of the randomization procedure was performed by the manufacturer at another location where the code was also kept until it was broken in May 1997.
Allocation concealment (selection bias)	Low risk	Randomization codes applied to pill boxes which were then allocated sequentially. "This part of the randomization procedure was performed by the manufacturer at another location where the code was also kept."
Blinding (performance bias and detection bias)   All outcomes	Low risk	"Double‐blind." "...pill boxes, with either fluoxetine or an identical appearing placebo containing the same ingredients except fluoxetine, were labelled with numbers ranging from 100 to 210.."
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Low numbers lost to follow up but reported results exclude 5 withdrawals; 3 from fluoxetine group due to adverse effects ‐ nervousness and anxiety, 1 from fluoxetine due to pregnancy, 1 from placebo who had purchased fluoxetine.

Methods	BUPROPION  Randomized controlled trial, 2x2 factorial design Setting: 2 clinical sites (Butler Hospital, Miriam Hospital) Rhode Island, USA  Recruitment: community volunteers
Participants	524 smokers >= 10 CPD; 48% F, av. age 44, av. CPD 25, 17.6% with history of MDD single episode, 3.1% recurrent MDD
Interventions	2 x 2 factorial design. Alternative psychosocial treatments were standard cessation therapy or plus CBT for depression. Both had 12 x 90 min groups twice weekly/ weekly/ monthly for 12w. TQD 5th session. Collapsed in this analysis  1. Bupropion 300 mg/day for 12 weeks  2. Placebo
Outcomes	Abstinence at 12m (sustained at 4 visits)  Validation: CO <= 10 ppm, saliva cotinine <= 15 ng/ml
Notes	First included as Brown 2006, part unpublished data. Some genotyping studies combine these participants with those reported in Collins 2004 Funding: National Institutes of Health
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"Participants were randomly assigned to one of two treatment sites, where they were to receive one of two manualized group treatments ... Participants  were then randomly assigned to receive one of two medication conditions, bupropion or placebo, using the urn randomization technique."
Allocation concealment (selection bias)	Unclear risk	"Whereas we were able to balance the drug and placebo conditions on an individual basis, behavioral treatments were randomized by group and thus were more susceptible to fluctuations in recruitment and to the availability at both sites of pairings of a senior and a junior therapist trained in CBTD". Knowledge of behavioural assignment was probably not concealed but seems unlikely to have lead to individual selection bias.
Blinding (performance bias and detection bias)   All outcomes	Low risk	"Double‐blind." Psychological condition unlikely to be blinded but unlikely to affect comparisons included in this review. "All participants and study staff were blind to medication condition."
Incomplete outcome data (attrition bias)   All outcomes	Low risk	81% provided complete outcome data at all follow ups, not related to treatment condition. All participants included in ITT analyses

Methods	FLUOXETINE Randomized controlled trial Setting: clinic, USA Recruitment: community volunteers
Participants	216 smokers with elevated depressive symptoms (CES‐D score ≥ 6) smoking ≥ 10 cpd. 38.4% F, av age 46, av cpd 21, mean FTND 5.6, mean CES‐D 11.4
Interventions	1. 10 weeks of 20 mg fluoxetine (beginning 2 weeks prior to TQD) 2. 16 weeks of 20 mg fluoxetine (beginning 8 weeks prior to TQD) 3. Control (no placebo) All arms: nicotine patch for 8 weeks starting on TQD (21 mg/day for 4 weeks, 14mg/day for 2 weeks, 7 mg/day for last 2 weeks), 5 sessions of brief behavioural smoking cessation treatment (in person and phone over 4 weeks, 20‐ 30 mins each)
Outcomes	Continuous abstinence at 12m Validation: salivary cotinine < 10 ng/ml
Notes	New for 2013. Significantly higher abstinence in 16 week arm than in 10 week arm, results presented separately in meta‐analysis with control divided. N abstinent not reported, extrapolated from percentages provided. Funding: American Cancer Society
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Urn randomization, no further detail provided
Allocation concealment (selection bias)	Unclear risk	Not specified
Blinding (performance bias and detection bias)   All outcomes	High risk	Open‐label
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Over 90% followed up at 12m. Similar rates across arms.

Methods	VENLAFAXINE Randomized controlled trial  Setting: clinic, USA  Recruitment: community volunteers
Participants	135 smokers, >= 10 CPD; 50% F, av age 46, av CPD 27
Interventions	1. Venlafaxine titrated to max of 225 mg/day from 3w before quit day for 21w, including 2w tapering.  2. Placebo  Both arms: 6w 22 mg nicotine patch, and 9x 15 min behavioural counselling.
Outcomes	Abstinence at 12m (PP)  Validation: CO <= 10 ppm and/or saliva cotinine < 15 ng/ul  Adverse events/withdrawals: not reported
Notes	First included as Cinciripini 1999 based on abstract. Funding: National Institutes for Health and National Institute for Drug Abuse. Medication provided free of charge by Wyeth Ayerst Laboratories.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization method not described. Stratification by depression history.
Allocation concealment (selection bias)	Low risk	Randomization by pharmacy, all study personnel with direct patient contact blind.
Blinding (performance bias and detection bias)   All outcomes	Low risk	"Double‐blind... Blinding of the study staff to the medication  was maintained using prenumbered pill containers, assigned to each participant at randomization by the pharmacy. All study personnel with direct patient contact were blind to group assignment."
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Majority of participants followed‐up (65 intervention; 63 control), participants lost to follow‐up counted as smokers

Methods	BUPROPION Randomized controlled trial Setting: clinic, USA Recruitment: community volunteers
Participants	294 smokers of ≥ 5 cpd 61% M, av age 44, av cpd 20, mean FTND 4.5
Interventions	1. 12 weeks bupropion started 12‐19 days before TQD (150mg/d days 1‐3, 300mg/d thereafter) 2. 12 weeks varenicline on same schedule (0.5mg/day days 1‐3, 1.0mg/day days 4‐7, 2.0mg/day thereafter) 3. Placebo on same schedule All arms: 10 individual counselling sessions (6 in person, 4 via phone, 240 mins total)
Outcomes	Continuous abstinence after 2 week grace period at 6m (Other prolonged abstinence outcomes also reported) Validation: CO < 10 ppm or salivary cotinine < 15 ng/mL
Notes	New for 2013 In less than 1% of the total cases, participants who did not attend a follow‐up were coded as abstinent because they were abstinent at the following data point. All other losses to follow‐up counted as smokers. Author provided further detail on AE measurements via e‐mail. Funding: National Institute on Drug Abuse, National Cancer Institute
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	“Adaptive randomization,” no further detail provided
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding (performance bias and detection bias)   All outcomes	Unclear risk	“Blinded” but no further information provided
Incomplete outcome data (attrition bias)   All outcomes	Low risk	73% followed up at 6m, similar rates across arms, all lost to follow‐up known to be smokers

Methods	SERTRALINE Randomized controlled trial  Setting: clinic, USA  Recruitment: volunteers
Participants	134 smokers with a history of past MDD; 65% F, av age 44.5, 47% had history of recurrent MDD
Interventions	1. Sertraline starting dose 50 mg/day, 200 mg/day by week 4 quit day. 9 day taper. Total duration 10w + 9 day taper, including 1w placebo washout prior to randomization  2. Placebo  Both arms: 9 x 45 min individual counselling sessions at clinic visits
Outcomes	Abstinence 6m after end of treatment (7 day PP)  Validation: serum cotinine < 25 ng/ml
Notes	Funding: Pfizer, Inc and National Institute on Drug Abuse
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization method not described.
Allocation concealment (selection bias)	Unclear risk	Not specified
Blinding (performance bias and detection bias)   All outcomes	Low risk	"double‐blind" "Medications were provided in prepared bottles that were numbered according to the randomization schedule and dispensed at each visit. All study staff at the clinic site were blinded to treatment assignment."
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Total participants lost to follow‐up at 6m not reported. "The subjects lost to follow‐up after random assignment were considered treatment failures."

PERMALINK

Antidepressants for smoking cessation

John R Hughes

Lindsay F Stead

Jamie Hartmann‐Boyce

Kate Cahill

Tim Lancaster

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the intervention

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Search methods for identification of studies

Data collection and analysis

Assessment of risk of bias in included studies

Measures of treatment effect

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Adverse events

Results

Description of studies

Bupropion

Nortriptyline

Selective Serotonin Reuptake Inhibitors (SSRIs)

Fluoxetine

Paroxetine

Sertraline

Citalopram/zimelidine

Monoamine oxidase inhibitors

Moclobemide

Selegiline

Lazabemide

EVT302

Other antidepressants

Venlafaxine

Hypericum (St John's wort)

S‐Adenosyl‐L‐Methionine (SAMe)

Doxepin

Imipramine

Tryptophan

Risk of bias in included studies

1.

Allocation

Blinding

Incomplete outcome data

Other potential sources of bias

Definition of abstinence

Effects of interventions

Summary of findings for the main comparison. Bupropion for smoking cessation.

Summary of findings 2. Nortriptyline for smoking cessation.

Smoking cessation

Bupropion

Compared to placebo or no pharmacotherapy control, no other pharmacotherapy

2.

1.1. Analysis.

Sensitivity to length of follow‐up

Sensitivity to clinical setting

1.2. Analysis.

Effect of level of behavioural support

1.3. Analysis.

Effect of dose

1.4. Analysis.

Combined bupropion and nicotine replacement therapy compared to NRT alone

1.5. Analysis.

Methods	BUPROPION Randomized controlled trial, 2x2 factorial  Setting: Cessation clinic, USA  Recruitment: community volunteers quit after 8w bupropion & nicotine patch
Participants	289 abstainers (excludes 5 withdrawing consent before starting medication); 45% F, av. age 43, av. cpd 21
Interventions	Relapse prevention study. All participants received 8 w open‐label bupropion & nicotine patch (21mg with weaning) for 7w from TQD. Transition procedures preserved blinding for RP phase but allowed weaning from bupropion. Individual counselling including CBT techniques, 15 min x6 during open label, x4 during RP, x2 during follow up.  1. Bupropion (300 mg) & nicotine gum (2 mg, use as needed to manage craving) for 16 w  2. Bupropion & placebo gum  3. Nicotine gum & placebo pill (150 mg bupropion for first week)  4. Double placebo (150 mg bupropion for first week)
Outcomes	Abstinence (no relapse to 7 days of smoking) for 12m (10m after randomization, 6m after EOT) (Primary outcome for study was time to relapse)  Validation: CO ≤8ppm at each visit
Notes	Quit rate after open‐label treatment was 52% so the final quit rate of 30% for combination therapy is equivalent to ˜16% of people starting treatment Funding: National Institute on Drug Abuse. GlaxoSmithKline provided medications.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"A statistician who did not participate in the clinical phases of the study provided computer‐generated randomization lists that were not accessible to the clinical staff", stratified by gender & depression history.
Allocation concealment (selection bias)	Low risk	A research nurse who did not have direct contact with participants prepared individual medication kits based on the randomization schedule.
Blinding (performance bias and detection bias)   All outcomes	Low risk	"Double‐blind" "Participants and clinical researchers with direct participant contact were blinded to the randomization" At end, subjects asked to guess treatment assignment; half guessed correctly and guess was not associated with treatment outcome.
Incomplete outcome data (attrition bias)   All outcomes	High risk	5 randomized participants withdrew before double blind phase. Less than 50% followed up in each group. Greater loss to follow up in double placebo, losses included in ITT analysis.

Methods	BUPROPION & NORTRIPTYLINE Randomized controlled trial  Setting: Smoking cessation clinic, Brazil  Recruitment: community volunteers.
Participants	156 smokers, FTND at least 4; 70% F placebo & nortriptyline, 59% Bup, av. age 44, av. CPD NS
Interventions	1. Bupropion 300 mg/day for 60 days, placebo nortriptyline, TQD during week 2  2. Nortripytyline 75 mg/day for 60 days, placebo bupropion  3. Double placebo  All arms: 6x 15‐min individual CBT, weekly then bi‐weekly.
Outcomes	Abstinence at 6m (continuous from TQD)  Validation: CO <= 10 ppm at 3 & 6m
Notes	Funding not reported
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization method not described.
Allocation concealment (selection bias)	Unclear risk	Allocation concealment not described.
Blinding (performance bias and detection bias)   All outcomes	Low risk	Double blind, double dummy. "Both investigators and patients were blind to the treatment"
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Numbers lost to follow‐up not reported, all included as smokers.

Methods	SELEGILINE Randomized controlled trial Setting: clinics, USA Recruitment: community
Participants	246 smokers of ≥ 15 cpd in 30d prior to enrolment, smoked for past 5 years and expired CO ≥ 9 ppm, motivated to quit. 51% M, av. age 46, av. cpd 22.
Interventions	1. Selegiline patch (6mg/24hr) for 9 weeks, starting 7 days before TQD 2. Placebo patch on same schedule Both arms: 9 weekly individual counselling sessions of approx. 10mins each
Outcomes	Prolonged abstinence at 6m (continuous from week 6 onwards) Validation: CO < 9 ppm
Notes	New for 2013 update. Some additional information on study characteristics provided by author. Mean compliance rates 91.6% and 91.3% for the STS and placebo groups Funding: National Institutes of Health, National Institute on Drug Abuse
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"Adaptive randomization," method not reported
Allocation concealment (selection bias)	Unclear risk	Not specified
Blinding (performance bias and detection bias)   All outcomes	Unclear risk	"Double‐blind," no further details provided
Incomplete outcome data (attrition bias)   All outcomes	Low risk	70% placebo and 74% STS followed up at 12 months

Methods	PAROXETINE Randomized controlled trial  Setting: clinic, USA  Recruitment: Advertisements
Participants	224 smokers, > 10 CPD, no current major depression. 12‐25% had history of MDD; 46% F, av age 46, av CPD 26
Interventions	1. Nicotine patch (24 hr, 21 mg, 8w) + 40 mg paroxetine (9w incl tapering)  2. Patch as 1 + 20 mg paroxetine  3. Patch as 1 + placebo paroxetine  All arms: S‐H manual and 15 min behavioural counselling at weeks 1 & 4.
Outcomes	Abstinence at 6m (7 day PP at 10 & 26w)  Validation: CO < 9 ppm and saliva cotinine < 20 ng/ml at each visit.
Notes	40 mg & 20 mg dose pooled in MA from 2009. 20/75 quit on 40mg, 15/75 on 20mg Funding: University of California Tobacco‐Related Disease Research Program, SmithKline Beecham
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization method not described.
Allocation concealment (selection bias)	Unclear risk	Not described
Blinding (performance bias and detection bias)   All outcomes	Unclear risk	"Double‐blind" but unclear who exactly was blinded
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Number lost to follow‐up not reported. "Those failing to provide confirmation [of smoking status] were reclassified as smokers."

Methods	ST JOHN'S WORT Randomized controlled trial, 2x2 factorial Setting: Smoking cessation clinic, UK Recruitment: direct mail from GP, stop smoking service, newspaper advertisements
Participants	143 adult smokers of at least 10 cpd, motivated to quit 38% M, av age 46, av cpd 21, mean FTND 5.5
Interventions	1. St John's wort 900 mg/day (300mg x 3/day) for 14 w, started 2 w prior to TWD 2. Placebo on same schedule Both arms: 7 weekly individual behavioural support sessions in clinic
Outcomes	Prolonged abstinence at 6m Validation: CO ≤ 10 ppm
Notes	New for 2013. Factorial trial ‐ also tested the use of chromium versus placebo for weight loss. Arms collapsed for analysis; no difference detected. Funding: Cancer Research UK
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Via computer program
Allocation concealment (selection bias)	Low risk	Independent statistician sent randomization codes to medication packing company, medication allocated in sequence. Researchers blind to allocation
Blinding (performance bias and detection bias)   All outcomes	Low risk	“Participants, therapists, and outcome assessors were blind to the treatment allocation.”
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Over 90% followed up at 6m, similar between groups

Methods	BUPROPION Cluster randomized trial Setting: health centres, Pakistan Recruitment: patients from participating health centres with suspected pulmonary tuberculosis
Participants	33 health centres covering 1955 adult smokers with suspected tuberculosis (1299 included in arms relevant to this review), smoking ≥ 1cpd or smoking hookah on a daily basis 95%M, av age 41, av cpd 19 (where one hookah counts as 2 cigarettes)
Interventions	1. 7 weeks bupropion (75mg/d first week, 150mg/d thereafter) 2. No pharmacotherapy Both arms: 2 sessions of brief, in‐person behavioural support (Note, third arm received usual care only, not included in this review)
Outcomes	Continuous abstinence at 6m Validation: CO ≤ 9 ppm
Notes	New for 2013 Reported narratively only due to substantial heterogeneity of program effects across clusters. 275/659 quit intervention vs 254/640 control, adjusted RR 1.1 (0.5–2.3). Funding: International Development Research Centre
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated
Allocation concealment (selection bias)	Low risk	Cluster randomized trial. “A researcher who was blinded to center identity” allocated conditions
Blinding (performance bias and detection bias)   All outcomes	High risk	No blinding
Incomplete outcome data (attrition bias)   All outcomes	Low risk	No clinics dropped out post‐randomization. Over 90% of participants followed up at 6m
Other bias	High risk	Substantial heterogeneity of program effects across clusters. 20% of participants in control arm smoked only hookah (no cigarettes) compared to 4% in intervention arm.

Methods	S‐ADENOSYL‐L‐METHIONINE (SAMe) Randomized controlled trial Setting: clinic, USA Recruitment: community volunteers
Participants	120 smokers of ≥ 10 cpd motivated to quit 53% M, av age 40, av cpd 20, mean FTND 5.2
Interventions	1. SAMe 1600 mg/day (via mouth) for 8w 2. SAMe 800 mg/day same schedule 3. Placebo same schedule All arms: Behavioural counselling using “Smoke Free and Living It” manual at every clinic visit (approx. 7)
Outcomes	7d PP at 6m (prolonged abstinence measured but not reported) Validation: CO ≤ 8 ppm
Notes	New for 2013 update. SAMe is a dietary supplement used to treat depression. No difference between arms 1 and 2, hence combined in meta‐analysis. Funding: National Institutes of Health
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding (performance bias and detection bias)   All outcomes	Unclear risk	"Blinded," no further detail provided
Incomplete outcome data (attrition bias)   All outcomes	Low risk	57% followed up overall, similar rates between groups

Study	Reason for exclusion
Akbarpour 2010	Bupropion ‐ short follow‐up
Banham 2010	Not RCT ‐ review of smoking cessation treatment for people with severe mental illness
Barnes 2006	St John's wort ‐ pilot study comparing two doses of St John's wort, no quitters at 12 months.
Becker 2003	St John's wort ‐ short follow‐up (1 month)
Berlin 2002	Lazabemide (monoamine oxidase‐B inhibitor) ‐ short follow up
Berlin 2005	Befloxatone (reversible monoamine oxidase‐B inhibitor) ‐ data not published, treatment reported to have had no effect on abstinence rates.
Berlin 2012	EVT302 (MAO‐B inhibitor) ‐ short follow‐up. (No evidence of short‐term benefit; 412 participants, RR 1.21, 95% CI 0.83 to 1.76)
Bloch 2010	Bupropion ‐ trial in people with schizophrenia, short follow‐up and cessation not reported
Bowen 1991	Tryptophan ‐ short follow up  Tryptophan 50 mg/kg/day, with high carbohydrate low protein diet (7/1 ratio), vs placebo and low carbohydrate high protein diet (1/1 ratio) for two weeks.
Brauer 2000	Selegiline ‐ only preliminary short‐term results available. Six month follow‐up planned
Breitling 2008	Trial of practitioner education and financial incentives, or cessation drug costs reimbursement
Carrão 2007	Sertraline ‐ combined with buspirone so effect of sertraline could not be isolated
Chan 2005	Bupropion ‐ case control study in pregnant women
Cornelius 1997	Fluoxetine ‐ cessation not an outcome. Fluoxetine reduced the amount smoked by depressed alcoholic smokers.
Cornelius 1999	Fluoxetine ‐ short‐term outcome in a study of depressed alcoholic patients not attempting to quit.
Dalack 1995	Fluoxetine ‐ refers to but does not report on a cessation study.
Dale 2002	Bupropion ‐ used for smokeless tobacco cessation, not smoking cessation.
Dale 2007	Bupropion ‐ for smokeless tobacco cessation, see Ebbert 2011
Daniela 2008	Sertraline and buspirone ‐ effect of antidepressant confounded with that of anxiolytic
Edwards 1989	Doxepin ‐ short follow‐up (2 months)
Elsasser 2002	Bupropion ‐ only 12 week follow‐up reported to date. 17 teenage (14‐19) smokers treated.
Evins 2008	Bupropion ‐ long‐term results not presented due to high loss to follow‐up
Fatemi 2005	Bupropion ‐ short‐term crossover trial
Frederick 1997	Venlafaxine ‐ short follow‐up (8 weeks)
Gawin 1989	Buspirone ‐ open trial
Gifford 2011	Bupropion ‐ test of behavioural therapy, all participants received bupropion
Glover 2002	Bupropion ‐ used for smokeless tobacco cessation, not smoking cessation
Gold 2002	Bupropion ‐ non random assignment, patient preference
Grandi 2011	Bupropion ‐ not RCT, review of bupropion use in patients with CVD
Grassi 2009	Not an RCT, pre‐post study of influence of smoking ban on people's selection of smoking cessation treatment
Gray 2011	Bupropion ‐ short follow‐up
Hall 2009	Bupropion ‐ all participants received bupropion for quitting, test of extended CBT or NRT
Hawk 2008	Bupropion ‐ short follow‐up (12 weeks). Compares 1 week to 4 week prequit use.
Hilberink 2005	Bupropion ‐ test of NRT + counselling, one cluster received bupropion but is not a test of bupropion
Hitsman 1999	Fluoxetine ‐ the majority of patients in this study were also part of the multi‐centre trial reported in Niaura 2002.
Houtsmuller 2002	Selegiline ‐ short‐term laboratory study
Hussain 2010	Bupropion ‐ short follow‐up, trial in unmotivated smokers
Jacobs 1971	Imipramine ‐ short follow‐up. Outcome was reduction in smoking to less than 10% of baseline.
Kalman 2004	Bupropion ‐ short follow‐up (12 weeks)
Karam‐Hage 2011	Bupropion ‐ short follow‐up (to end of medication phase). Pilot study, 11 participants
Kotz 2009	Nortriptyline ‐ pharmacotherapy was confounded with additional counselling from nurse (control group 1), compared to usual care
Kras 2010	St John's wort ‐ short follow‐up
Lawvere 2006	St John's wort ‐ uncontrolled study
Le Foll 2009	Selegiline ‐ study terminated early due to lack of efficacy, results available at 9 weeks only
Li 2009	Bupropion ‐ short follow‐up
Miller 2003	Bupropion ‐ short follow‐up (8 weeks)
Monuteaux 2007	Bupropion ‐ participants were adolescent non‐smokers, not for cessation
Mooney 2008	Bupropion ‐ short follow‐up, bupropion for opioid and tobacco dependence
Naranjo 1990	Fluoxetine ‐ study of short‐term smoking behaviour.
Neumann 2000	Bupropion ‐ smokers randomized to 1 or 2 months of medication (300 mg/day). 91/165 randomized were not included in the analysis, including some 1‐month group participants who requested further medication.
Neumann 2002	Bupropion ‐ short‐term follow‐up. Comparison of 300 mg and 150 mg doses
Niederhofer 2004	Bupropion ‐ short‐term. 22 adolescents followed up during 90 days of treatment
Olmstead 1999	Bupropion ‐ all participants received bupropion. Short‐term follow‐up.
Paluck 2006	Bupropion ‐ uncontrolled prospective observational study
Pomerleau 1991	Fluoxetine ‐ no cessation data reported
Raynor 2005	Bupropion ‐ short (90 day) follow‐up. Sub‐study within a larger trial with long‐term follow up, not yet published
Robinson 1991	Buspirone ‐ case series
Rovina 2003	Bupropion ‐ abstract only, trial report not available. Insufficient information to determine inclusion
Schepis 2006	Bupropion ‐ abstract only, trial report not available. Insufficient information to determine inclusion
Sellers 1987	Zimelidine or citalopram (SSRIs) ‐ placebo‐controlled crossover design study of smoking behaviour and alcohol use in non‐depressed heavy drinkers
Sheng 2013	Bupropion ‐ follow‐up less than 6 months
Sherman 2008	Bupropion ‐ trial of NRT as adjunct to bupropion
Shiffman 2000	Bupropion ‐ placebo‐controlled short‐term study of effects on craving and withdrawal in patients not wanting to quit smoking permanently
Shoptaw 2008	Bupropion ‐ tested for methamphetamine dependence. Reduction in smoking was a secondary outcome. Only 48/73 participants smoked, quitting not reported.
Singh 2010	Bupropion ‐ short‐term follow‐up
Sittipunt 2007	Nortriptyline ‐ only 3‐month follow‐up
Sonntag 2003	Bupropion ‐ abstract only, trial report not available. Insufficient information to determine inclusion
Spring 1995	Fluoxetine ‐ 6‐month cessation not reported. Primarily a study of post‐cessation weight gain.
Stein 1993	Fluoxetine ‐ does not report outcomes from a double‐blind study
Steinberg 2009	Bupropion ‐ confounded with nicotine inhaler and treatment duration in comparison with nicotine patch alone
Strayer 2004	Bupropion ‐ all participants prescribed bupropion. Test of behavioural interventions, not bupropion. Adverse event data from author used.
Swanson 2003	Bupropion +/‐ nicotine patch. Unable to confirm correct denominators.
Tidey 2009	Bupropion ‐ laboratory study, outcomes included urge to smoke, not cessation
Toll 2007	Bupropion ‐ all participants had same pharmacotherapy
Weinberger 2008	Bupropion for people with bipolar disorder. Short follow‐up (8 weeks). Only 5 participants.
Weiner 2001	Bupropion ‐ no control group
Weiner 2012	Bupropion ‐ short follow‐up
White 2005	Bupropion versus gabapentin ‐ Short follow‐up (6 weeks)
Zernig 2008	Bupropion ‐ used as an active control to a psychosocial intervention, cannot estimate pharmacotherapy effect
ZYB30011 2002	Bupropion ‐ only follow‐up to end of treatment (7 weeks)

Trial name or title	Combination varenicline/bupropion treatment for NRT‐nonresponders
Methods	Double‐blind randomized controlled trial
Participants	222 NRT non‐responders
Interventions	1. Varenicline and bupropion 2. Varenicline alone
Outcomes	Abstinence at 8‐11 weeks postquit and at 6 months
Starting date	March 2011
Contact information	Jed Rose, jed.rose@duke.edu
Notes	Results at 8 to 11 weeks found significant benefit of adding bupropion for male participants, but not for female participants