Quality assessment
This study used the method in the Cochrane Handbook to assess the risk of bias of the included studies because the quality of the study has an important impact on the credibility of the results. A high-quality study design can generate more reliable evidence[43, 44]. Three of the nine included studies had a low risk of bias, while the others were unclear. Although there were no studies with a high risk of bias, four RCTs with an unclear risk of bias were identified due to reporting ambiguous details, especially for blinding and random sequence generation procedures. Alternatively, the three RCTs with a low risk of bias explained their procedures for generating random sequences using computer software or random number tables and reported blinded participants or researchers. These important methodological designs improved the validity and reliability of their results. However, some studies did not mention the details of the methodological design and how the trial was implemented and can, therefore, only be judged as having an unclear risk of bias. There is no doubt that an unclear risk of bias has the potential to be high risk, which reduces the reliability of the evidence. Therefore, randomized controlled design trials should emphasize reporting study design features such as randomization and blinding.
As for the certainty of the evidence, the GRADE results showed that Varenicline was very low, Naltrexone was moderate, and Topiramate and Bupropion were low. Considering that all the included studies were RCTs, only five degradation factors were considered. Among these factors, high heterogeneity (inconsistency), wide confidence intervals (imprecision), and low quality or a small number of included studies reduced the level of evidence. Especially for Varenicline and Topiramate, their high heterogeneity reduced the certainty of the evidence. The extent of the downgrade was based on the degree of heterogeneity, such that if heterogeneity was less than 50%, the study was downgraded one level, otherwise downgrading by two levels was considered. The number of studies was also assessed as it would have an impact on the certainty of the evidence. The drugs in this systematic review only had a few studies each, especially for Topiramate and Bupropion, and as such, the level of evidence was reduced due to the small number of studies.