Cahill · The Cochrane database of systematic reviews 2013 · Systematic review and network meta-analysis · n=267 studies (101,804 participants)

Pharmacological interventions for smoking cessation: an overview and network meta-analysis.

Cited 1249 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and network meta-analysis of randomized controlled trials

PubMed 23728690 · doi:10.1002/14651858.CD009329.pub2 · record verified 2026-08-30

What was done

This Cochrane overview and network meta-analysis evaluated pharmacological interventions for smoking cessation in adult smokers, drawing from 12 Cochrane systematic reviews published through November 2012. Reviews focusing on pregnant women, specific clinical disease populations, or specific settings were excluded. The primary efficacy outcome was continuous or prolonged abstinence at least six months from treatment initiation. The primary harm outcome was the incidence of serious adverse events (SAEs). Network meta-analyses compared NRT, bupropion, and varenicline with placebo and head-to-head for efficacy, as well as bupropion and varenicline for SAE risks.

What was found

Across 267 randomized trials involving 101,804 participants: - Versus placebo, abstinence increased with NRT (OR 1.84; 95% CredI 1.71 to 1.99), bupropion (OR 1.82; 95% CredI 1.60 to 2.06), varenicline (OR 2.88; 95% CredI 2.40 to 3.47), cytisine (RR 3.98; 95% CI 2.01 to 7.87), nortriptyline (RR 2.03; 95% CI 1.48 to 2.78), and clonidine (RR 1.63; 95% CI 1.22 to 2.18). - Bupropion and NRT showed equivalent efficacy head-to-head (OR 0.99; 95% CredI 0.86 to 1.13). - Varenicline outperformed single-form NRT (OR 1.57; 95% CredI 1.29 to 1.91) and bupropion (OR 1.59; 95% CredI 1.29 to 1.96), but did not differ from combination NRT (OR 1.06; 95% CredI 0.75 to 1.48). - Combination NRT was superior to single NRT formulations. - Neither bupropion nor varenicline showed statistically significant increases in total SAEs, neuropsychiatric events (bupropion RR 0.88, varenicline RR 0.53), or cardiovascular events (bupropion RR 0.77, varenicline RR 1.26) compared with placebo.

Why it matters

This overview consolidates extensive randomized evidence to establish that varenicline and combination NRT provide the highest cessation rates among available pharmacotherapies, while trial data do not demonstrate excess serious neuropsychiatric or cardiovascular risks for varenicline or bupropion.

Limits

The analysis was limited to trials captured in Cochrane reviews through November 2012 and excluded pregnant smokers and specific disease cohorts. Trial-reported SAEs may be underpowered to detect very rare complications (such as bupropion-induced seizures, observed at approximately 1 in 1,500). Several promising or legacy agents (e.g., cytisine, mecamylamine) had smaller evidence bases or limited global licensing.

Cited by