Because of the workings of health care systems, new, important, and cost-effective treatments sometimes do not become routine care while well-marketed products of equivocal value achieve widespread adoption. Should policymakers attempt to influence clinical behavior and correct for these inefficiencies? Implementation methods achieve a certain level of behavioral change but cost money to enact. These factors can be combined with the cost-effectiveness of treatments to estimate an overall policy cost-effectiveness. In general, policy cost- effectiveness is always less attractive than treatment cost- effectiveness. Consequently trying to improve the uptake of underused cost-effective care or reduce the overuse of new and expensive treatments may not always make economic sense. in this article, we present a method for calculating policy cost- effectiveness and illustrate it with examples from a recent trial, conducted during 1997 and 1998, of educational outreach by community pharmacists to influence physician prescribing in England.