Background: Tobacco use is the leading preventable cause of disease, disability, and death in the United States. Interventions to help adults quit smoking might stop or reduce tobacco-related illness.
Purpose: To systematically review evidence for the effectiveness and safety of pharmacotherapy and behavioral tobacco cessation interventions among adults, including pregnant women and those with mental health conditions, and to conduct a de novo search for primary evidence related to electronic nicotine delivery systems for adults.
Methods: We conducted a review of reviews and searched for existing systematic reviews published through August 1, 2014 in the following databases and organizations' websites: PubMed, PsycInfo, the Database of Abstracts of Reviews of Effects, the Cochrane Database of Systematic Reviews, the Centre for Reviews and Dissemination Health Technology Assessment, the Agency of Healthcare Research and Quality, British Medical Journal Clinical Evidence, the Canadian Agency for Drugs and Technologies in Health, Center for Disease Control and Prevention's Guide to Community Preventive Services, the Institute of Medicine, the National Institute for Health and Clinical Excellence, the National Health Service Health Technology Assessment Programme, and the Surgeon General. We included reviews that were published in the English language that systematically reported the effects of tobacco cessation interventions on health, cessation, or adverse outcomes. We excluded nonsystematic meta-analyses and narrative reviews and those that focused on harm reduction or relapse prevention. We conducted an a priori search for primary trial evidence related to the effectiveness and safety of electronic nicotine delivery systems (ENDS) (through March 1, 2015) and a search for pharmacotherapy among pregnant women (through August 15, 2014) to supplement the review of reviews methodology. Two investigators independently reviewed abstracts and full-text articles against a set of a priori inclusion and quality criteria. Discrepancies were resolved by consensus. One reviewer abstracted data into an evidence table and a second reviewer checked these data. We grouped reviews based on population (general adults, pregnant women, individuals with mental health conditions) and intervention (pharmacotherapy, behavioral, or combined interventions). We identified one or more reviews within each population and intervention subgroup that represented the most current and applicable evidence to serve as the basis for the main findings (“primary” reviews) and discussed complementary and discordant findings from other included reviews as necessary. We did not reanalyze any of the individual study evidence; we presented pooled analyses and existing point estimates from included reviews.
Results: We included 54 systematic reviews, 22 of which served as the basis for the primary findings. Among adults, nine reviews addressed the efficacy and/or harms of nicotine replacement therapy (NRT), bupropion hydrochloride sustained release (bupropion SR), and/or varenicline. None of these reviews reported on health outcomes. All three medications were found to be effective in increasing smoking quit rates compared with placebo or nondrug arms at 6 or more months followup. The pooled risk ratio (RR) for abstinence for NRT was 1.60 (95% confidence interval [CI], 1.53 to 1.68); for bupropion SR, RR 1.62 (95% CI, 1.49 to 1.76); and for varenicline, 2.27 (95% CI, 2.02 to 2.55). Combined NRT versus a single form of NRT showed a statistically significantly greater cessation effect in pooled analysis (RR 1.34 [95% CI, 1.18 to 1.51]). None of the drugs were associated with major cardiovascular adverse events, although NRT produced higher rates of all cardiovascular events (driven by minor events). One review on combined pharmacotherapy and behavioral interventions reported a relative increase in quitting by 82 percent versus nonpharmacotherapy usual care (RR 1.82 [95% CI, 1.66 to 2.00]). We included an additional 33 reviews that addressed behavioral tobacco cessation treatments among adults, including those that focused on specific subpopulations such as older adults. Compared with various controls, behavioral interventions such as in-person advice and support from clinicians, self-help materials, and telephone counseling had modest, but significantly increased, relative smoking cessation at 6 or more months (18% to 96%). For example, the pooled RR of physician advice versus no advice was 1.76 (95% CI, 1.58 to 1.96) for smoking cessation at 6 or more months followup. Only two trials addressed the efficacy and harms related to the use of electronic cigarettes and these trials suggested no benefit on smoking cessation among smokers intending to quit. We included eight reviews that focused on pregnant women that found significant benefits for perinatal health, including increased birth weight and reduced preterm birth. These benefits were evident with behavioral interventions, and suggested by data from some of the NRT trials, although that evidence was limited. Cessation during late pregnancy was greater among women receiving any type of behavioral intervention, with evidence most clear for counseling. Rates of validated cessation among women allocated to NRT (5% to 24%) compared with placebo (0% to 15%) were not statistically different, although few studies contributed data. Our reviews among individuals with depression or schizophrenia provided limited trial evidence on the efficacy of pharmacotherapy or behavioral interventions. There was, however, some evidence of a benefit for bupropion among those with schizophrenia and the addition of a mood management component to behavioral interventions for smokers with depression.
Conclusions: This review of reviews suggests that behavioral interventions and pharmacotherapy, alone or in combination, are effective in helping to reduce rates of smoking among the general adult population. Behavioral interventions, in particular, can assist pregnant women to stop smoking. Data on the effectiveness and safety of electronic nicotine delivery systems are limited. Future research should focus on direct comparisons between different combinations and classes of drugs; the incidence of serious adverse events related to medications for cessation; the efficacy and safety of ENDS; and pharmacotherapies for pregnant women and those with mental health conditions including evidence on health outcomes.