PurposeTo use a randomized controlled design to explore the effects of evidence-based medicine (EBM) education on physician assistant (PA) students' EBM knowledge, self-efficacy, and evidence-seeking behavior in a simulated clinical situation and to present a model of EBM competence.MethodsSixty-one didactic-year PA students from one Midwestern University (2 sequential cohorts) were randomized to receive the standard PA curriculum plus EBM training (intervention) or the standard PA curriculum only (control). Evidence-based medicine knowledge was measured with a validated Fresno test. Self-efficacy was measured with a validated Likert scale. Clinical application of EBM skills was measured with an objective structured clinical examination (OSCE).ResultsEvidence-based medicine education led to significant improvements on the Fresno and self-efficacy tests, both within and between groups. On the OSCE, the intervention group performed no better than the control group. Higher Fresno pretest scores were significantly related to decreasing improvements in the posttest scores: R = -0.634.ConclusionTeaching EBM to PA students improved their EBM knowledge and self-efficacy but not their clinical application. Future research should focus on enhancing EBM evaluation and application in the clinical setting.