ABSTRACT
Clinical Decision Support Systems (CDSSs) are implemented in clinical settings in order to improve patient outcomes and/or clinical practices. However, they are still not widely accepted by healthcare professionals due to over-alerting. The aim of the "Patient Safety through Intelligent Procedures in medication" (PSIP) project is to develop and demonstrate innovative tools so as to generate and provide relevant knowledge to healthcare professionals and patients for Adverse Drug Event (ADE) prevention by means of Information and Communication Technologies (ICT). PSIP employs a Knowledge Base (KB) as the core of its CDSS. This KB encapsulates signals capable of automatically detecting potential ADEs and contextualizing the CDSS output to the patient and healthcare professionals. To exploit the KB, a Global Knowledge Platform (GKP) has been created comprising of a KB system, a Connectivity Platform and appropriate user interface modules. The GKP has been tested to demonstrate integration of the KB in different work situations and it has been deployed in three different medical applications. The first is a Web application; the second involves a commercial French EHR (Electronic Health Record) and the third is a Danish CPOE (Computerised Physician Order Entry) system. This paper presents recent progress as regards the exploitation of the PSIP KB and the results obtained in the three different medical applications.
Subject(s)
Decision Support Systems, Clinical/organization & administration , Drug-Related Side Effects and Adverse Reactions/prevention & control , Internet , Knowledge Bases , Medical Order Entry Systems/organization & administration , Medical Records Systems, Computerized/organization & administration , Humans , User-Computer InterfaceABSTRACT
Our main objective is to detect adverse drug events (ADEs) in former hospital stays. As ADEs are rare, that supposes to screen thousands of electronic health records (EHRs). For that purpose, we need to define a data model that has two main objectives: (1) being able to describe hospital stays from various hospitals (2) being tuned so as to prepare the data mining process: as ADEs are not flagged in the datasets, the data model must be optimized for ADE detection. The article presents the phases of the design and the data model that results from this work. It is compatible with many hospitals. It deals with diagnoses, drug prescriptions, lab results and administrative information. It allows for data mining and ADE detection in EHRs.
