An interdisciplinary method of classifying and monitoring medication errors.

A system is described for reporting medication errors according to practitioner and type and using these data as part of a departmental quality-assessment program. Types of errors committed in prescribing, in dispensing, and in administering drugs in a teaching hospital were listed. Prescribers' and pharmacists' reports of errors committed. Data on dispensing errors by 16 staff pharmacists were collected over a 12-month period in 1982-83, and data on prescribing by resident physicians were collected during three separate one-month periods. For each pharmacist, data on dispensing errors were compiled monthly. Pharmacists were counseled if they had a large number of errors in one category, a high total number of errors, or a substantial number of errors in a single category over several months. Likewise, physicians who committed more errors than their department's mean were counseled by the clinical chairman. After 12 months, each pharmacist's monthly error rate was compared with the departmental average. The number of dispensing errors reported during the study period (average 60.7 per month for the department) was lower than expected. Serious prescribing errors were minimal. In spite of probable underreporting of errors in this voluntary system, these methods of classifying medication errors by type and practitioner group and of recording error rates were useful tools for monitoring and improving performance of pharmacists and prescribers.