The evaluation of potential of a new molecular entity (NME) to inhibit P-glycoprotein (P-gp) in vivo is an integral part of drug development and is recommended by regulatory agencies. In this study, we compared the performance of 5 prediction methods and their associated criteria (including those from the European Medicines Agency, the US Food and Drug Administration, and the Pharmaceuticals and Medical Devices Agency of Japan) for assessing the potential of an NME to inhibit P-gp in vivo based on in vitro assessment. We collected in vitro (eg, half-maximal inhibitory concentration [IC 50 ], fraction unbound to plasma protein) and in vivo (eg, dose, maximum concentration, change in maximum concentration or area under the plasma concentration-time curve of the substrate digoxin) data for 50 Food and Drug Administration-approved, orally administered drug products containing 53 NMEs, from the University of Washington Metabolism and Transport Drug Interaction Database, Drugs@FDA, and PubMed. All methods yielded similar accuracy with small differences in false-negative (FN) and false-positive (FP) predictions. In addition, use of ratio of the theoretical maximum gastrointestinal concentration to IC 50 is sufficient for a reasonable prediction for these orally administered drugs as potential P-gp inhibitors based on our dataset. The FN and FP rates varied depending on the cut-off value for the ratio of the theoretical maximum gastrointestinal concentration/IC 50 . Possible reasons underlying FP and FN results from different methods should be taken into consideration to predict in vivo P-gp inhibition.