Failure of low-cost audits with feedback to reduce laboratory test utilization.

The medical literature contains more than 20 studies of education as a cost-control measure.'2 Most of these educational programs have involved multiple modalities such as written manuals, lectures, audits with feedback, and case presentations. Also, most of them have involved a heavy investment of faculty time and therefore have been quite expensive to develop and to maintain. Indeed, in the only study that compared benefits and costs, Schroeder et al. found that the money saved by their ambitious educational program was completely consumed by its costs.3 Another problem not adequately addressed in this literature is the unknown mechanism by which these programs achieve their savings. Only two studies of education as a cost-control measure at-