Using EHR Data to Detect Prescribing Errors in Rapidly Discontinued Medication Orders.

Area of Science:

• Health Informatics
• Medication Safety
• Clinical Pharmacy

Background:

• Previous prospective methods for detecting prescribing errors in electronic medication orders were effective but not feasible for routine use.
• Electronic medication order systems offer opportunities for error detection and system improvement.

Purpose of the Study:

• To assess a retrospective method for detecting prescribing errors in rapidly discontinued electronic medication orders.
• To characterize the types of prescribing errors identified and their potential clinical significance.
• To identify opportunities for improving prescribing systems and processes.

Main Methods:

• A retrospective review of electronic medication orders discontinued within 120 minutes of submission.
• Categorization of orders as errors, nonerrors, or indeterminate.
• Analysis of error characteristics including time to discontinuation, error type, prescriber, and clinical significance.

Main Results:

• 147 prescribing errors were identified in 305 reviewed medication orders.
• The retrospective method was most effective for orders discontinued within 90 minutes.
• Duplicate orders were the most frequent error type, with physicians in training exhibiting the highest error rate.

Conclusions:

• Retrospective analysis of rapidly discontinued electronic medication orders can identify prescribing errors without disrupting clinical workflow.
• This method provides valuable data for understanding error patterns and improving medication safety.
• Further validation in diverse clinical settings is recommended for widespread implementation.