Comparative efficacy and safety of pharmacological interventions for smoking cessation in healthy adults: A network meta-analysis

Abstract

Smoking is the leading cause of morbidity and mortality in different non-communicable diseases, and cessation leads to immense health benefits. The present network meta-analysis has been conducted to evaluate and compare the effects of available pharmacological interventions for smoking cessation in adults. A standard meta-analysis protocol was developed and after performing a comprehensive literature search on MEDLINE/PubMed, Cochrane databases, and International Clinical Trials Registry Platform, reviewers extracted data from 97 randomized controlled trials. PRISMA guidelines were followed in data extraction, analysis and reporting of findings. Random effects Bayesian network meta-analysis was done to pool the effects across the interventions. Network graph was built, and for closed triangles in the network graph, node splitting analysis was performed. The primary outcome measure was self-reported biochemically verified smoking abstinence at six months. The number of participants achieving continuous abstinence was reported. Data for the number of participants reporting at least one adverse event was also extracted, if available. Combination of nicotine receptor agonist and nicotine replacement therapy had a significant odd of 4.4 (95%CrI:2.2–8.7), bupropion and nicotine receptor agonist 4.0 (95%CrI:2.1–7.7), bupropion and nicotine replacement therapy 3.8 (95%CrI:2.3–6.2), combination nicotine replacement therapy has an odd of 2.6 (95%CrI:1.8–3.8), and nicotine receptor agonist had a significant odd of 2.7 (95%CrI:2.3–3.2) when compared to placebo (moderate quality of evidence) for continuous abstinence at 6 months. When compared with behavioural therapy, the odds ratio of interventions was not statistically significant. Combination of nicotine receptor agonist and nicotine replacement therapy has the highest probability of being the best treatment for abstinence from smoking.