According to a new study, over 99 percent of medication doses leaving a hospital pharmacy were free from errors. However, given the huge volume of medications dispensed by hospital pharmacies, even the low 0.75 percent error rate found in the study translates into a large number of medication errors, many of which had the potential to harm patients. While the hospital pharmacy medication dispensing accuracy rate is impressive, according to the researchers who conducted the study, it is still inadequate. By directly observing the hospital pharmacy's medication dispensing process, researchers found that 3.6 percent (5,075) of 140,755 medication doses filled by the pharmacy over a 7-month period contained errors. The hospital pharmacist detected only 79 percent (4,016) of these errors during routine verification. Thus, 0.75 percent (1,059) of doses filled would have left the pharmacy with undetected errors.
Of these undetected errors, 23.5 percent (249) were potential adverse drug events (ADEs), of which 70 were serious and 2 were life-threatening. The most common potential ADEs were incorrect medications (36 percent), incorrect strength (35 percent), and incorrect dosage form (21 percent). A physician panel deemed 26 medication dispensing errors to be potentially life-threatening, such as a medication with adult dosage strength dispensed to the neonatal intensive care unit.
Of the four dispensing processes evaluated, automated dispensing cabinet fill had the highest number of errors (4.2 percent), whereas the cart fill process had the highest error rate (6 percent). The automated dispensing cabinet accounts for two-thirds of doses dispensed from the pharmacy, and the filling and checking of these medications is a high-volume and repetitive task, which can lead to a high number of errors. During cart fill, pharmacy staff work with multiple medications that are different for each patient, usually medications that are not routinely dispensed—factors that may contribute to errors. In contrast, first dose fill and controlled substances fill had lower error rates, perhaps due to their smaller volume. The study was supported in part by the Agency for Healthcare Research and Quality (HS14053).
See "How many hospital pharmacy medication dispensing errors go undetected?" by Jennifer L. Cina, Pharm.D., Tejal K. Gandhi, M.D., M.P.H., William Churchill, M.S., R.Ph., and others, in the February 2006 Journal on Quality and Patient Safety 32(2), pp. 73-80.