In this randomized, controlled, unblinded trial, we sought to test Repeat Back's (RB) effect on comprehension following informed consent discussions. RB has been suggested as a method to improve patient comprehension when obtaining informed consent. In this technique, patients are asked to recount what they have been told in the informed consent discussion. Despite preliminary data, this practice has not been evaluated in any large scale study. This study was conducted in 7 Veterans Health Administration Medical Centers where informed consent is obtained using iMedConsent, the VA's computer based platform. Patients scheduled for elective surgeries were randomized to RB (a module added to the iMedConsent package) or standard iMedConsent (no RB). Comprehension was tested after the informed consent using procedure-specific questionnaires. Time stamps in the iMedConsent program estimated the time spent completing the informed consent process. Provider and patient satisfaction were measured using 5-point Likert scales. Statistical comparisons of groups were performed using t-tests and chi tests. A total of 575 patients were enrolled. In the RB group, providers spent 2.6 minutes longer (P < 0.0001) obtaining informed consent. The mean comprehension score was significantly higher in the RB group (71.4%) versus the no RB group (68.2%, P = 0.03); the effect was greatest in carotid endarterectomy patients (RB = 73.4% vs. no RB = 67.7%, P = 0.02). Quality of decision making was rated similarly. Providers were neutral to slightly favorable regarding RB. RB implemented within an electronic informed consent system improved patient comprehension. The additional time required was acceptable to providers. RB should be considered as an enhancement to surgical informed consent.This clinical trial was registered at http://www.clinicaltrials.gov (Identifier NCT00288899).