Warfarin therapy in the Republic of Ireland : the potential role of neural networks in optimising therapy

Abstract

Anticoagulant therapy is used to reduce the embolic risk associated with many diseases e.g. atrial fibrillation, deep vein thrombosis and aortic valve replacement. The decision to anticoagulate a patient is made individually, taking into account the potential benefits and risks. Warfarin is the only oral anticoagulant licensed in the Republic of Ireland. It has a very narrow therapeutic index and for safe rational anticoagulant therapy, an understanding of the interrelationships which influence an individual’s response to the drug is essential. The objective of this study was to investigate the potential role that neural networks (NNs) may play in oral anticoagulant therapy. NNs are computer based mathematical modelling systems that mimic the computational ability of the human brain. In order to improve the quality of patient care, we considered the possibility of using NNs to mimic the prescribing behaviour of the health care practitioners (HCP) in the anticoagulant clinic (AC) of the Adelaide and Meath Hospital incorporating the National Children’s Hospital (AMNCH). An additional role for NNs to predict patients’ International Normalised Ratio (INR) measurements upon return to the AC was also investigated.