To evaluate the efficacy of lafutidine (20mg) , famotidine (40mg) and placebo in patients with mild reflux esophagitis (Grades A and B according to the Los Angeles classification) , a double-blind, multicenter, randomized clinical trial was performed for the first time in Japanese patients. In addition to each physician's evaluation, efficacy was evaluated by judging panels using images submitted by each physician. The healing rate after 8 weeks for lafutidine, famotidine and placebo were 67.7%, 56.6% and 41.2%, respectively. Lafutidine was significantly more effective than placebo (p=0.002, according to the judging panels) . Based on the evaluation of endoscopic images by the judging panels, 91 (27.1%) of 336 images submitted by each physician were judged to not be mucosal breaks. Judging panels are considered one of the ways to resolve the problem of the need to unify the criteria.