COVID-19 Long-Haul Project

About

virtual health graphic

A subset of COVID patients experience chronic symptoms that may or may not have been associated with their COVID disease. These COVID-19 “long-hauler” (CLH) symptoms can range from slight to severe and can linger for months beyond the recovery from the active disease. For those with chronic CLH symptoms, these are often documented across various encounters with different units in the health system that often are not conflated with each other (e.g. digestive and gynecologic symptoms). This makes identification of and coordinated support of CLH patients an acute and timely research and operational issue poised for rapid innovation.

To address this issue, we propose a multi-step, data-driven approach that would allow us to 1) identify, validate and cluster CLH-symptoms of patients within the health system; 2) develop an algorithmic decision-making model to identify patients who might be at risk for CLH; and 3) work with clinicians to identify opportunities to integrate model outputs into clinical practice. We will adopt best principles from the latest AI/machine learning approaches that allow us to leverage both structured and unstructured clinical data that has been validated via gold standard expert reasoning and data from the Parkview Post-COVID Clinic. Our model will provide an interpretable and transparent decision outcome that can be used by clinicians to make inferences about CLH patients and prevent them from being diagnosed improperly. Below we outline the major aims of the research and the associated outcomes.


Partners

Funding

Publications and presentations

Press

All projects