A machine learning-based clinical decision support system to identify prescriptions with a high risk of medication error.
Machine learning can improve the accuracy of clinical decision support (CDS) tools. This single-site study used data from the electronic health record (EHR) and clinical pharmacist review to test the accuracy of a hybrid CDS system to identify prescriptions with high risk of medication error. The machine-learning based approach was more accurate than existing techniques such as the traditional CDS system and can improve the reliability of prescription checks in an inpatient setting.