Severity and probability of harm of medication errors intercepted by an emergency department pharmacist.

OBJECTIVES  The objective of this study was to evaluate the severity and probability of harm of medication errors (MEs) intercepted by an emergency department pharmacist. The phases of the medication-use process where MEs were most likely to be intercepted were determined. METHODS  The emergency department was staffed with a full-time pharmacist during the 7-month study period. The MEs that were intercepted by the pharmacist were recorded in a database. Each ME in the database was independently scored for severity and probability of harm by two pharmacists and one physician investigator who were not involved in the data collection process. KEY FINDINGS  There were 237 ME interceptions by the pharmacist during the study period. The final classification of MEs by severity was as follows: minor (n = 42; 18%), significant (n = 160; 67%) and serious (n = 35; 15%). The final classification of MEs by probability of harm was as follows: none (n = 13; 6%), very low (n = 96; 41%), low (n = 84; 35%), medium (n = 41; 17%) and high (n = 3; 1%). Inter-rater reliability for classification was as follows: error severity (agreement = 75.5%, kappa = 0.35) and probability of harm (agreement = 76.8%, kappa = 0.42). The MEs were most likely to be intercepted during the prescribing phase of the medication-use process (n = 236; 90.1%). CONCLUSIONS  A high proportion of MEs intercepted by the emergency department pharmacist are considered to be significant or serious. However, a smaller percentage of these errors are likely to result in patient harm.