Using Clinical Trial Randomization to Ensure Uniform Medication Error Coverage in EHR Medication Verification Activities

Abstract

Objectives: The objective of this project was to define a process to create and assign EHR patients that would (1) allow for the uniform coverage of 10 medication error types across three medication verification laboratories and (2) create a unique student-patient schedule that would limit the impact shared patient information outside of the laboratory. Method: EHR usage for institutional medication verification laboratories necessitated the creation of individual patient records to allow student access and processing of patient medication orders. Ensuring coverage of 10 medication error types necessitated the creation of 4 medication error cohorts using random number generation. 4 template patients were created for each medication cohort containing 5 active medication order sets with a specified number of errors. A modified split-plot design was used to assign student-patient schedules minimizing overlapping assignment of template patients. Results: 152 unique patient assignment schedules were generated with 608 individual student-patient assignments. While each patient template was used for 30-40 students using this technique, the use of post-assignment naming limited identifiable patient duplication to <15 students. This ensured all students received exposure to all 10 medication error types. Student-Patient Assignments were distributed on barcoded activity rubrics used by facilitators to review student work during EHR laboratory sessions. Implications: This framework could be extended to allow assignment of patients based on laboratory schedule, medication, and/or blooms taxonomy to allow for randomization by scheduled laboratory time, uniform coverage of medications, and/or degree of difficulty.