A Comparison of Collaborative Filtering Methods for Medication Reconciliation

Medication Reconciliation has emerged as a major patient safety goal in the management of medication errors and prevention of adverse drug events. The medication reconciliation process supports the task of detecting and correcting potential mistakes in a patient’s medication list so that physicians can make correct, consistent, timely and safe prescribing decisions. Maintaining an accurate list of a patient’s medications is a very challenging task for which the current solution is a process driven approach. In prior work, we proposed a promising data driven approach through the use of collaborative filtering algorithms to improve the accuracy of the medication list. This is analogous to the framework used by online retailers to recommend relevant products to customers. In this paper, we extend our original framework to include other types of patient information, develop some new collaborative filtering approaches and test them using medication data from a long-term care clinic. The results are encouraging and suggest several promising directions for the future, including embedding these methods in current medication reconciliation processes and evaluating them in actual clinical settings.