Can Predicate Logic and Algorithms Improve Patient Outcomes?
Research by Wojtek Michalowski demonstrates the potential for using advanced decision-support tools for managing patients with multiple illnesses. It comes at a time of great discussion in the healthcare field around the themes of technology-assisted decision-making and use of disease-specific guidelines in complex patient cases.
“From a medical perspective, it’s a super important issue,” says Michalowski, professor of health informatics at the Telfer School. “There’s strong evidence that use of guidelines improves patient care. But for patients with comorbid conditions which have multiple diseases, these guidelines cannot be used directly. Concurrent use of multiple guidelines poses a risk of undesired health effects, including adverse drug-drug or drug-disease interactions.”
So Michalowski and his team are collaborating with hospital physicians to develop algorithms that will allow for the automatic execution of multiple guidelines; “along with automatic mitigation when there are adverse interactions.”
The ultimate goal of this research, which is funded by the Natural Sciences and Engineering Research Council of Canada (NSERC), can be illustrated by the following simple scenario. A complex patient with three comorbid diseases is being managed in a medical ward. The clinical practice guidelines (CPGs) for these diseases are automatically retrieved; patient data recorded in the Electronic Health Record is consulted; and the algorithms developed by Michalowski and his team automatically revise the CPGs into a patient-specific and fully customized guideline – enabling the physician to arrive at a consistent therapy for the patient.
Some studies show that individuals who have two or more comorbid conditions represent more than half the population 65 years or older. “For these patients, a lack of methods to facilitate the concurrent application of the guidelines severely limits their use in clinical practice,” says Michalowski. “How to fix this is one of the grand challenges of clinical decision support.”
Professor Michalowski will present this research at the prestigious American Medical Informatics Association (AMIA) Annual Symposium in Washington, D.C., Nov. 15-19. He and his colleagues were nominated by the Scientific Committee of AMIA 2014 for a Distinguished Paper Award, one of only 15 out of 249 accepted papers to receive this honor.