Background/Objectives. Missing data bias and sampling uncertainty are common concerns in randomized controlled trials (RCT). Widely-recommended methods that address missingness and sampling uncertainty have been studied extensively by themselves and in the context of economic evaluations conducted alongside RCTs. However, little is known on how to appropriately combine these methods, and the potential pitfalls and advantages of different approaches. This study provides a critical assessment of methods used in economic evaluations of substance use disorder interventions conducted alongside RCTs that simultaneously address missingness and sampling uncertainty. Methods. We identified published economic evaluations from 4 prior systematic reviews and a targeted search of the literature that met the following criteria: 1) RCT study design; 2) cost-effectiveness measures based directly on participant-level data; 3) trial reported missing data. All articles meeting the inclusion criteria were read to assess the use of statistical methods to address uncertainty, missing data, and the type of resampling approach employed in combination with the identified method to address missingness. Results. Twenty-nine studies were included in our review. Twelve used complete case analysis for at least one missing outcome variable, 3 applied last-observation-carry-forward, 1 used linear interpolation for non-monotonic missing data, 6 employed inverse probability weighting (IPW), 3 applied baseline-carry-forward, a single study used mean and variance estimates from complete cases to randomly impute missing cases, and all other studies applied some form of single-imputation method, including regression-adjusted, mean-imputation, within-group-matching, or expert opinion. No study in this review incorporated multiple imputation, the most robust approach to addressing missingness. Twenty-three economic evaluations applied the recommended nonparametric bootstrap to address sampling uncertainty; of those, 7 employed compete-case-resampling, 11 followed an imputation before resampling approach, and 5 conducted IPW within resampling. Conclusions. We provide a timely review of statistical methods employed by economic evaluations of substance use disorders. We evaluated how each study addressed missing data bias, whether the recommended nonparametric bootstrap was employed, how these two methods were combined, and provided with recommendations for future applied research.