Prescription Errors and the Impact of Computerized Prescription Order Entry System in a Community-based Hospital
Adverse drug events occur often in hospitals. They can be prevented to a large extent by minimizing the human errors of prescription writing. To evaluate the efficacy of a computerized prescription order entry (CPOE) system with the help of ancillary support in minimizing prescription errors. Retrospective study carried out in a community-based urban teaching hospital in south Brooklyn, NY from January 2004 to January 2005. Errors were categorized into inappropriate dosage adjustment for creatinine clearance, duplication, incorrect orders, allergy verification, and incomplete orders. The pharmacists identified the type of error, the severity of error, the class of drug involved, and the department that made the error. A total of 466,311 prescriptions were entered in the period of 1 year. There were 3513 errors during this period (7.53 errors per 1000 prescriptions). More than half of these errors were made by the internal medicine specialty. In our study, 50% of the errors were severe errors (overdosing medications with narrow therapeutic index or over-riding allergies), 46.28% were moderate errors (overdosing, wrong dosing, duplicate orders, or prescribing multiple antibiotics), and 3.71% were not harmful errors (wrong dosing or incomplete orders). The errors were also categorized according to the class of medication. Errors in antibiotic prescription accounted for 53.9% of all errors. The pharmacist detected all these prescription errors as the prescriptions were reviewed in the CPOE system. Prescription errors are common medical errors seen in hospitals. The CPOE system has prevented and alerted the prescriber and pharmacist to dosage errors and allergies. Involvement of the pharmacist in reviewing the prescription and alerting the physician has minimized prescription errors to a great degree in our hospital setting. The incidence of prescription errors before the CPOE has been reported to range from 3 to 99 per 1000 prescriptions. The disparity could be due to the definition of medical errors, which has changed over the years, and also number of prescriptions included in the study and the study design.