A systematic review of randomized controlled trials and network meta-analysis of e-cigarettes for smoking cessation.

Abstract Aim E-cigarettes, or nicotine vaping products, are potential smoking cessation aids that provide both nicotine and behavioural substitution for combustible cigarette smoking. This review aims to compare the effectiveness of nicotine e-cigarettes for smoking cessation with licensed nicotine replacement therapies (NRT) and nicotine-free based control conditions by using network meta-analysis (NMA). Methods We searched PubMed, Web of Science and PsycINFO for randomised controlled trials (RCTs) that allocated individuals to use nicotine e-cigarettes, compared to those that used licensed NRT (e.g., nicotine patches, nicotine gums, etc), or a nicotine-free control condition such as receiving placebo (nicotine-free) e-cigarettes or usual care. We only included studies of healthy individuals who smoked. Furthermore, we identified the latest Cochrane review on NRT and searched NRT trials that were published in similar periods as the e-cigarette trials we identified. NMA was conducted to compare the effect of e-cigarettes on cessation relative to NRT and control condition. Cochrane risk-of-bias tool for randomized trials Version 2 was used to access study bias. Results For the e-cigarette trials, our initial search identified 4,717 studies and we included 7 trials for NMA after removal of duplicates, record screening and assessment of eligibility (Total N = 5,674). For NRT trials, our initial search identified 1,014 studies and we included 9 trials that satisfied our inclusion criteria (Total N = 6,080). Results from NMA indicated that participants assigned to use nicotine e-cigarettes were more likely to remain abstinent from smoking than those in the control condition (pooled Risk Ratio (RR) = 2.08, 97.5% CI = [1.39, 3.15]) and those who were assigned to use NRT (pooled RR = 1.49, 97.5% CI = [1.04, 2.14]. There was a moderate heterogeneity between studies (I2 = 42%). Most of the e-cigarette trials has moderate or high risk of bias. Conclusion Smokers assigned to use nicotine e-cigarettes were more likely to remain abstinent from smoking than those assigned to use licensed NRT, and both were more effective than usual care or placebo conditions. More high quality studies are required to ascertain the effect of e-cigarette on smoking cessation due to risk of bias in the included studies.