Improving prescription-writing skills in a family practice residency.

The objective of this study was to evaluate a novel method of physician education that uses copies of prescriptions written by study participants to provide feedback on their prescription-writing skills. A prospective, blind, completely random before-after design was used, with observation, intervention, and postintervention periods, each lasting four months. The study group comprised 20 family practice residents, stratified by year of training and randomly assigned to treatment or control groups. Copies of all prescriptions written by participating residents were evaluated for prescription-writing errors. During the intervention phase, prescription copies with errors were returned to the treatment group residents with a tactful comment pointing out the error and suggesting changes. Over the three time periods, there was a continuous decline in the rate of prescription-writing errors in the treatment group, but not in the control group. A significant number of physicians who received the feedback improved their prescription writing (p less than 0.01), but logistic regression failed to show a difference attributable to the intervention. In conclusion, written feedback on copies of prescriptions produced a small decrease in the number of prescription-writing errors. Further, long-term study may show a magnified effect of this teaching method.