Automatic Mitigation of Adverse Interactions in Pairs of Clinical Practice Guidelines Using Constraint Logic Programming

FieldValue
dc.contributor.authorWilk, Szymon
dc.contributor.authorMichalowski, Wojtek
dc.contributor.authorMichalowski, Martin
dc.contributor.authorFarion, Ken
dc.contributor.authorMainegra Hing, Marisela
dc.contributor.authorMohapatra, Subhra
dc.date.accessioned2012-10-10T14:04:01Z
dc.date.available2012-10-10T14:04:01Z
dc.date.created2012
dc.date.issued2012-10-10
dc.identifier.otherWP.2012.07
dc.identifier.urihttp://hdl.handle.net/10393/23379
dc.identifier.urihttp://dx.doi.org/10.20381/ruor-2625
dc.descriptionLe texte intégral de ce document de travail n'est pas disponible en ligne. Pour plus de renseignements sur ce document, veuillez communiquer avec la Direction de la recherche de l'École de gestion Telfer à l'adresse recherche@telfer.uottawa.ca. // The full text of this working paper is not available online. For more information regarding this working paper, please contact the Telfer School of Management Research Office at research@telfer.uottawa.ca.
dc.description.abstractIn the paper we propose a new method to automatically mitigate (identify and address) adverse interactions (drug-drug or drug-disease) that occur when a patient with comorbid diseases is managed according to two concurrently applied clinical practice guidelines (CPGs). The lack of formal methods and computer-based tools to facilitate concurrent application of CPGs severely limits their use in clinical practice and the development of such methods is one of the grand challenges for clinical decision support. The proposed method responds to this challenge. We introduce and formally define logical models of CPGs and other related concepts, and develop the mitigation algorithm that operates on these concepts. In the algorithm we combine domain knowledge encoded as interaction and revision operators with the constraint logic programming (CLP) paradigm. The operators characterize adverse interactions and describe revisions to logical models required to address them, while CLP allows us to efficiently solve these models – a solution represents a feasible therapy that may be applied to a patient. The mitigation algorithm accepts two CPGs and available (likely incomplete) patient information. It starts by translating CPGs into logical models and then checks for adverse interactions (either by examining the structure of logical models or by applying interaction operators). Identified interactions are flagged as potential sources of infeasibility (PSI) that cause the logical models to not to have a solution (indicating that a feasible therapy cannot be established). If a PSI exists, the algorithm attempts to address it using revision operators. Finally, the algorithm reports whether mitigation has been successful (a feasible therapy exists) or not, gives a solution representing possible therapy, points at the PSI, and reports the revision operator that addresses the associated interactions. Thus, we consider the mitigation algorithm as an alerting tool that warns the physician about issues with the concurrent application of pairs of the CPGs and that can be implemented as a procedure within a CPG execution engine or as a component of a clinical decision support system. We illustrate our method in the context of two clinical scenarios involving a patient with a chronic disease (duodenal ulcer) who experiences an episode of an acute disease (transient ischemic attack).
dc.titleAutomatic Mitigation of Adverse Interactions in Pairs of Clinical Practice Guidelines Using Constraint Logic Programming
CollectionTelfer - Documents de travail // Telfer - Working Papers

Files
There are no files associated with this item.