There is abundant evidence that clinical chemistry laboratory tests are over-ordered in North America, but there does not seem to be an effective corrective strategy that has a prolonged effect. The goal of this study was to design one that had a prolonged effect. Using a pre- and post-intervention survey study design, the authors observed the effect of physician education followed by a ban on test-panel ordering of common clinical chemistry tests, reinforced by written reminders to physicians not heeding the ban, on ordering patterns (tests per specimen), and total numbers of these tests ordered. Panels of > 16 common biochemistry tests per specimen were reduced from 15% to 6% of orders for inpatients and from 44% to 11% for outpatients 1 year after the implementation of the ban on test-panel ordering. However, the ban had little effect on the ordering rates for panels of 7 common tests. Educational exercises (newsletters and lectures) had no effect. The authors conclude that a ban on test-panel (profile) ordering reinforced by continuing reminders to nonconforming physicians is an effective means of reducing clinical chemistry test usage over the long term. A 38% reduction of common biochemistry tests ordered was achieved. However, overall costs savings were modest. Nevertheless, the authors conclude that the cost-effective use of the clinical pathology laboratory by careful selection of tests in an essential part of a medical trainee's education.