Automated Verification of Care Pathways Using Constraint Programming

Abstract

Bad construction of modeled care pathways can lead to satisfiability problems during the pathway execution. These problems can ultimately result in medical errors and need to be checked as formally as possible. Therefore, this study proposes a set of algorithms using a free open-source library dedicated to constraint programming allied with a DSL to encode and verify care pathways, checking four possible problems: states in deadlock, non-determinism, inaccessible steps and transitions with logically equivalent guard conditions. We then test our algorithms in 84 real care pathways used both in hospitals and surgeries. Using our algorithms, we were able to find 200 problems taking less than 1 second to complete the verification on most pathways.

Automated conflict resolution for patients with multiple morbidity being treated using more than one set of single condition clinical guidance: A case study

“…General practitioners (GPs) attempting to treat patients with multiple morbidities are required to reconcile the conflicts that might occur between medication or lifestyle advice recommended by single condition guidelines using only their experience and their knowledge of the patient and the symptoms they describe. This places an enormous responsibility on the often discretionary decision making behaviour of GPs that even before the COVID-19 pandemic were accomodating unprecedented demands on their services and are now coping with fundamental changes to their working environment [5]. The growing pressure being placed on primary care has potentially serious implications for consistency of care, and patient safety [6] and has highlighted the need to improve the integration of existing guidelines and provide more timely, reliable and consistent decision support mechanisms [7].…”