Comparison of two knowledge bases on the detection of drug-drug interactions.

This paper describes a drug ordering decision support system that helps with the prevention of adverse drug events by detecting drug-drug interactions in drug orders. The architecture of the system was devised in order to facilitate its use attached to physician order entry systems. The described model focuses in issues related to knowledge base maintenance and integration with external systems. Finally, a retrospective study was performed. Two knowledge bases, developed by different academic centers, were used to detect drug-drug interactions in a dataset with 37,237 drug prescriptions. The study concludes that the proposed knowledge base architecture enables content from other knowledge sources to be easily transferred and adapted to its structure. The study also suggests a method that can be used on the evaluation and refinement of the content of drug knowledge bases.

Design, implementation and evaluation of a clinical decision support system to prevent adverse drug events.

Adverse drug events are known to be a major health problem worldwide. It is estimated that the annual costs related to these events in the United States are greater than the total costs with cardiovascular disease care. Decision support systems that assist drug ordering have demonstrated to be a powerful tool to prevent prescription errors and adverse drug events. On the other hand, some issues related to the development, implementation, configuration, and evaluation of these decision support systems still need further research. This paper presents the development and evaluation of a decision support system prototype that helps with the prevention of adverse drug events by detecting drug-drug interactions in drug orders. The structure of the system tries to solve some of the problems described by the literature, such as integration with hospital information systems, adaptability to local needs, and knowledge base maintenance. The proposed model has shown to be an effective method for representing drug-drug interactions. The prototype was evaluated by a retrospective study using a dataset with 37.237 prescriptions. The system was able to detect 10.044 (27.0%) orders containing one or more drug-drug interactions. Among these interactions, 6.4% had high severity. In a future study, it is intended to apply the developed system in a real-time on-line environment, evaluating the benefits achieved in terms of improvement in medical practice and patient outcomes.