Grant: Project Grant - PA: Patient-Oriented Research: Early-Career Investigator

Delirium is a sudden onset of confusion that affects up to 1 in 3 hospitalized adults. It is a leading cause of preventable harm in hospitals and leads to longer hospital stays, greater costs, and greater risk of death. Experiencing delirium can be profoundly distressing to patients, their caregivers, and their healthcare providers. Up to 40% of delirium cases can be prevented through a bundle of proven interventions including improving sleep, mobility, cognitive stimulation, and nutrition. However, these interventions are not implemented well in hospitals because they are too labour-intensive and difficult to sustain over time. In brief, although we know what strategies are proven to prevent delirium, we do not know who needs them or how to implement them in real-world settings. We aim to overcome these barriers by bringing together innovative artificial intelligence and implementation science methods to create a Delirium Prevention Toolkit. The project will occur in 11 diverse Ontario hospitals. First, we will use machine learning to develop a real-time Delirium Risk Prediction Tool to estimate each patient's risk of developing delirium at the time they are admitted to hospital, so prevention strategies can be targeted to those who need them the most. Second, we will use implementation science methods to develop a Prevention Toolkit that provides strategies for how delirium prevention interventions can be delivered to patients in real-world settings. We have assembled a world-class team with expertise in machine learning, implementation science, and delirium prevention, a highly engaged caregiver partner, and strong partnerships with key stakeholders across the health system. This project will produce a Delirium Prevention Toolkit that is ready for implementation and evaluation in a future randomized trial to assess the impact on delirium rates, patient-important outcomes, and costs of care in hospitals across Canada.