Certainty of evidence
The certainty of the evidence for each drug is shown in table 3. Varenicline was downgraded by three levels in total due to high heterogeneity (inconsistency) and wide confidence intervals (imprecision), and as such it was rated as having very low evidence. Naltrexone was downgraded by one level due to an unclear risk of bias, and the evidence for its effect was considered moderate. Topiramate was downgraded by two levels due to the small number of studies and high heterogeneity (inconsistency), such that the evidence was rated as low. Bupropion was downgraded by two levels due to a small number of studies and wide confidence intervals (imprecision), resulting in a low evidence rating.