Evaluating dispensing error detection rates in a hospital pharmacy.
OBJECTIVES: The filling of unit dose orders and checking for filling errors are two essential distributive responsibilities of a hospital pharmacy. Previous studies have shown that nonpharmacists, usually technicians, are capable of assuming these distributive tasks traditionally performed by hospital pharmacists. The study tested whether nonpharmacists, in this case licensed practical nurses/medication nurses, were as competent as pharmacists in checking for errors in unit dose cassettes prepared for hospital patients. METHODS: A university teaching hospital was used for the study. Artificial errors (n = 812) were introduced into the drug distribution system during a 4-month period in 1995. Included in the study were seven staff pharmacists and nine medication nurses (licensed practical nurses) involved in the decentralized drug distribution system. The primary measure was the ratio of errors detected to the number of artificial errors introduced into the system. This primary measure is different from those used in prior studies that do not separate dispensing errors and checking errors. RESULTS: Overall, pharmacists were significantly more accurate in detecting errors (87.7% vs. 82.1%). In one category of serious errors, that of wrong strength, the difference between pharmacists and licensed practical nurses was even greater (93.3% vs. 83.3%). CONCLUSIONS: This study's results do not support conclusions of prior studies that nonpharmacists can match the error detection accuracy of pharmacists. It demonstrates the importance of considering the types of errors under examination and of using appropriate measures of error checkers when drawing conclusions on relative competence.