Impact of simulation-based learning on medication error rates in critically ill patients.

Purpose: To compare medication administration error rates before and after the provision of educational sessions using either traditional didactic lecture or simulation-based training.
Methods: A single-center, parallel, controlled, prospective study conducted in adult coronary critical care (CCU) and medical intensive care units (MICU). Twenty-four nurses were observed administering medications. Documentation included drug name, dose, route, time and technique during observation and active medication orders in the patient's chart. A direct observation method was completed at baseline and repeated twice after the interventions. Data obtained during observation were analyzed for medication administration error rates. Interventions were two types of educational sessions with content developed from baseline medication administration error data: simulation-based training for CCU nurses and a didactic lecture for MICU nurses. Quizzes completed before and after the interventions were used to assess knowledge.
Results: A total of 880 doses (402 CCU, 478 MICU) were observed. After the simulation-based educational intervention in the CCU, medication administration error rates decreased from 30.8 to 4.0% (p < 0.001) in the initial post-intervention observation and were sustained in the final post-intervention observation (30.8 to 6.2%; p < 0.001). The error rate in the MICU after the didactic lecture intervention was not significantly different from the baseline and increased in the final post-intervention observation from 20.8 to 36.7% (p = 0.002). Mean quiz scores were significantly improved after education sessions in both ICUs.
Conclusions: Simulation-based learning provides a significant advantage to patient care through the reduction of medication administration errors compared to lecture style education.