Vector Researchers Use OHDP to Identify Mortality Predictors for LTC Residents with COVID-19

Vector Institute Faculty Member Bo Wang and his team utilized machine learning techniques on data from the Ontario Health Data Platform (OHDP) to identify mortality predictors for long-term care residents infected with COVID-19. Collaborating with researchers at ICES and UHN’s Peter Munk Cardiac Centre, the study analyzed de-identified health records of over 60,000 Ontario residents tested during the first two pandemic waves. Published in the Journal of the American Geriatrics Society, the research confirmed known risk factors like age and comorbidities but notably highlighted functional status as a critical predictor of outcomes. This marks the first application of machine learning to assess patient risk factors for COVID-19 at the population level in Canada. The study underscores the severe impact of the pandemic on long-term care facilities and demonstrates the utility of the OHDP, a platform established by the Province of Ontario to enhance health data access for researchers. The findings aim to improve detection, planning, and response strategies for future health crises affecting vulnerable populations.