LEG Lead & Administrators Meeting

Amol Verma is a a 2020 AMS Healthcare Fellow in Compassion and Artificial Intelligence. He is a physician, scientist, and Assistant Professor in General Internal Medicine at St. Michael’s Hospital and the University of Toronto. He is working to study and improve hospital care using electronic clinical data. Amol co-leads GEMINI, Canada’s largest hospital clinical data analytics network, which includes >30 hospitals across Ontario. He is a Provincial Clinical Lead for Quality Improvement in General Internal Medicine with Ontario Health, and the Vice-Chair of the Researcher Council of the Digital Research Alliance of Canada. Amol completed medical training at the University of Toronto, a Masters degree at the University of Oxford as a Rhodes Scholar, and research fellowships through the Royal College of Physicians and Surgeons of Canada Clinician Investigator Program and the Canadian Frailty Network. He received the 2022 Royal College of Physicians and Surgeons of Canada Early Career Leadership Award and the 2022 Canadian Institutes of Health Research’s early career Trailblazer Award in Population and Public Health Research.

Amol’s AMS Healthcare Compassion and Artificial Intelligence Fellowship projects included developing a delirium identification tool and implementing CHARTwatch, a hospital early warning system. Delirium is acute confusion that affects 20-30% of hospitalized adults and leads to 2-fold greater mortality, ~$11,000 greater average hospital costs, and 8 days longer average hospital stays. Up to 40% of delirium cases are preventable by improving patient nutrition, mobilization, sleep, and cognitive stimulation and avoiding harmful medications. However, these interventions are resource intensive and have not been properly implemented, in part because there is no sustainable method to accurately measure delirium rates. Hospital administrative data capture only 25% of delirium cases. Amol co-led a multidisciplinary collaboration with clinicians and engineers to develop the GEMINI AI-based Delirium Identification Tool that measures delirium with 90% accuracy, allowing reliable unit-level estimates of delirium prevalence. This tool is now being tested for implementation across several hospitals as part of a quality improvement project that has been approved by the CEO Table of the Toronto Academic Health Science Network and will be rolled out over the next few years.

Amol also led the implementation of CHARTwatch an AI-based early warning system that he developed with his team at St. Michael’s Hospital. CHARTwatch uses electronic medical record data to alert clinicians in real-time about patients who are at high risk of death or critical illness in the next 48 hours. Amol led a prospective study comparing CHARTwatch to >3,000 predictions by nurses and physicians about their patients and found that combining AI and clinician prediction improved the detection of future deaths and ICU transfers by 16% compared to clinicians alone. CHARTwatch was launched in September 2020 and has received positive anecdotal feedback from clinicians and preliminary data suggest a 10-15% reduction in mortality related to its implementation. We are currently conducting rigorous analyses to evaluate the tool’s impact on mortality and clinical care through quantitative and qualitative analyses. This research is exploring how AI tools affect clinical outcomes, patient-clinician relationships and clinical decision-making, which are at the very core of compassionate care.