Grant: Operating Grant: SARS-CoV-2 variants supplement stream 1

Hospital care has been dramatically reorganized to respond to the COVID-19 pandemic, with an urgent need to preserve scarce resources like ventilators and personal protective equipment. Yet, there is minimal evidence about how to care for patients with COVID-19 in hospital and how to maintain high quality non-COVID care during the pandemic. Detailed clinical data, such as patient vital signs or medications, are needed to study COVID-19 and its individual- and system-wide effects. These data are available in hospital computer systems but have not been widely collected and shared for research. The General Medicine Inpatient Initiative (GEMINI) has collected detailed clinical data from >340,000 admissions at 7 hospitals in Ontario, making it the largest inpatient repository about adult medicine in Canada and a valuable data source to study COVID-19. We propose to create the COVID-19 Hospital Analytics Laboratory by extending GEMINI to the 30 largest hospitals in Ontario, representing 70% of acute medical/ICU beds in the province. We will collect detailed clinical data from all medical and ICU admissions, including COVID-19 and other medical illnesses. Data will be linked to ICES longitudinal population datasets and updated every 1-3 months to create a globally unique platform that enables advanced analytics and machine learning and includes a diverse sample of patients, meaning research insights will be widely applicable. This platform will address all 5 CIHR objectives by supporting a breadth of research focused on improving hospital care in the setting of COVID-19. Data will be accessible to global experts in artificial intelligence, operations engineering, population modeling, public health and clinical epidemiology. Our first two research priorities will be to improve understanding and prediction of clinical outcomes in patients hospitalized with COVID-19 and to study how the organization of hospital care in the setting of COVID-19 affects patient outcomes.