Avoiding medication conflicts for patients with multimorbidities

Abstract

Clinical pathways are care plans which detail essential steps in the care of patients with a specific clinical problem, usually a chronic disease. A pathway includes recommendations of medications prescribed at different stages of the care plan. For patients with three or more chronic diseases (known as multimorbidities) the multiple pathways have to be applied together. One common problem for such patients is the adverse interaction between medications given for different diseases. This paper proposes a solution for avoiding medication conflicts for patients with multimorbidities through the use of formal methods. We introduce the notion of a pharmaceutical graph to capture the medications associated to different stages of a pathway. We then explore the use of an optimising SMT solver (Z3) to quickly find the set of medications with the minimal number and severity of conflicts which is assumed to be the safest. We evaluate the approach on a well known case of an elderly patient with five multimorbidities.