NCT03484793

Reduce Medication Errors by Translating AESOP Model Into CPOE Systems

Using Big Data and Deep Neural Network to Prevent Medication Errors

• Status: Completed
• Phase: N/A
• Study type: Interventional
• Enrollment: 37 (actual)
• Sponsor: Taipei Medical University · Academic / Other
• Sex: All
• Age: —
• Healthy volunteers: Not accepted

Summary

Medication errors are common, life-threatening, costly but preventable. Information technology and automated systems are highly efficient for preventing medication errors and therefore widely employed in hospital settings. In this study, investigators would perform a cluster randomized controlled trial of a clinical reminding system that uses DNN and Probabilistic models to detect and notify physicians of inappropriate prescriptions, giving them the opportunity to correct these gaps and increase prescriptions completeness. This study aim is to assess whether or not this system would improve prescription notation for a broad array of patient conditions.

Detailed description

This paper focuses on "Big data" in the knowledge base, using "Data minig" study of DM (Disease-Medication) and MM (Medication-Medication) of relevance to develop associated decision resources system-"the intelligent safety system" (Advanced Electronic Safety of Prescriptions,AESOP Model), and test the system in the clinical environment in hospital can assist physicians when open orders reduce medication errors, the system is named "AESOP Model".

Conditions

• Hypertension

Interventions

Type	Name	Description
OTHER	AESOP service system	Investigators develop an electronic reminder in CPOE system which notifies physicians when there appears to be an inappropriate prescription. At the time, a physician saves a typed prescription, our system analyzes the patient's medications, diseases and uses the knowledge base to determine whether a medication is uncommonly prescribed to all diseases in a given prescription. If the system detects the common associations of medications and diseases in a given prescription, it considers an appropriate prescription, and, if not, an actionable reminder is shown onscreen. To the right of each suggested uncommon medication is a reason why the reminder is appearing. Physicians can accept the reminder or ignore the reminder.

Timeline

• Start date: 2017-05-01
• Primary completion: 2017-12-31
• Completion: 2018-02-28
• First posted: 2018-04-02
• Last updated: 2018-04-03

Locations

1 site across 1 country: Taiwan