Do Ensembling and Meta-Learning Improve Outlier Detection in Randomized Controlled Trials?