Researchers from the University of Toronto and Unity Health Toronto, ICES, have revealed new details about an AI-powered software that could help reduce the risk of death in patients admitted to the intensive care unit (ICU).
According to research, which was published in the Canadian Medical Association Journal, the researchers evaluated the effectiveness of CHARTWatch, an early warning system used at St. Michael’s Hospital wards following 3 years of development and testing.
The study included 13,649 participants between the ages of 55 and 80 admitted to General Internal Medicine (GIM). Of these, 9,626 were in the pre-intervention period, 4,023 used CHARTWatch, and 8,470 patients were admitted to units without CHARTWatch.
During the intervention period, which lasted for 19 months, only 482 patients became high-risk in GIM compared to 1,656 high-risk patients within 43 months of the pre-intervention period.
Reduced Non-palliative Death
The researchers also noted there were fewer nonpalliative deaths (1.6%) for patients in the CHARTWatch category than 2.1% for those in the pre-intervention category.
According to the research findings, CHARTWatch helped reduce the death rates by facilitating regular communications and engaging clinicians with real-time alerts. It also sent emails to the nursing teams twice daily and a daily email to the palliative care units.
The palliative care teams also created a streamlined care pathway for high-risk patients, employing increased monitoring, enhanced communications between nursing teams and physicians, and critical prompts to encourage physicians to re-examine patients.
Lead author Dr. Amol Verma, said that the research findings suggest that AI-powered early warning systems in hospital wards are promising in helping reduce the likelihood of patients’ health conditions deteriorating rapidly, reducing the cases of unexpected deaths of hospitalized individuals.
