A Logic Programming Based Adverse Event Reporting and Learning System

Changes are taking place in the way patients, physicians, administrators, legislators and society in general view healthcare, including its quality and safety. The conclusion that more people may die as a result of medical errors than from injuries sustained in motor vehicle accidents is alarming. An adverse event reporting system may help to improve patient safety and the quality of the healthcare institution. However, the accumulation of potentially relevant data in databases contributes little to healthcare services improvement. It is crucial to apply models to identify the underlying system failures, the root causes that led to the event and enhance the sharing of knowledge and experience. In the real world complete information is hard to obtain, so systems should have the ability to reason with incomplete information. We developed a model to classify the adverse events root causes in the medical imaging field where our logic programming approach allows the representation of incomplete information. In this paper we present a model for the adverse events root causes classification in the medical imaging field and an adverse event reporting and learning system that applies the developed model. This system is deployed in two Portuguese healthcare institutions with promising results. The conceptualized logic model offered the means for knowledge extraction, providing the identification of the most significant causes and suggestions of changes in the healthcare organization policies and procedures.