There is wide interhospital variation in the removal rates of histologically normal appendices, and variation also in appendectomy rates among countries and among hospital catchment areas. Decision theory suggests that better patient workup and careful observation of doubtful cases result in improved discrimination between appendicitis patients and those with nonspecific abdominal pain. By improving the data base and weighing evidence with care, a surgeon can reduce his false-positives without risk of increasing his perforation rate. In some cases it may even be possible to reduce the false-positives while simultaneously reducing perforations. In a survey of New England hospitals there was no inevitable inverse relationship between normal removal rates and perforation rates. Two examples of improvement in performance are cited from the literature.