Rapid deterioration among hospitalized patients is the leading cause of unplanned admission to the intensive care unit (ICU).
But CHARTWatch acted as an early warning system to improve patients' health and alert healthcare workers to reduce unexpected deaths, the team said in the article published in CMAJ (Canadian Medical Association Journal).
"As AI tools become increasingly used in medicine, it is important that they are carefully evaluated to ensure they are safe and effective," said lead author Dr. Amol Verma, a clinician at St. Michael's Hospital. Unity Health Toronto. , Canada.
"Our findings suggest that AI-based early warning systems show promise in reducing unexpected deaths in hospitals," Verma said.
The effectiveness of CHARTWatch was evaluated in 13,649 patients aged 55 to 80 years who were admitted to general internal medicine (GIM) (about 9,626 in the pre-intervention period and 4,023 used CHARTWatch). About 8,470 admitted to subspecialty units did not use CHARTWatch.
Regular communications helped reduce deaths, as CHARTWatch engaged doctors with real-time alerts, twice-daily emails to nursing teams, and daily emails to the palliative care team, the researchers said.
A care pathway was also created for high-risk patients, prompting greater follow-up by nurses and better communication between nurses and doctors. This encouraged doctors to reevaluate patients.
Verma said the artificial intelligence system can be used to help nurses and doctors provide high-quality care.
Co-author Dr. Muhammad Mamdani, director of the University of Toronto, said the study evaluates the outcomes associated with the complex deployment of the entire AI solution.
It's important to understand the real-world impacts of this promising technology, Mamdani said.
But CHARTWatch acted as an early warning system to improve patients' health and alert healthcare workers to reduce unexpected deaths, the team said in the article published in CMAJ (Canadian Medical Association Journal).
"As AI tools become increasingly used in medicine, it is important that they are carefully evaluated to ensure they are safe and effective," said lead author Dr. Amol Verma, a clinician at St. Michael's Hospital. Unity Health Toronto. , Canada.
"Our findings suggest that AI-based early warning systems show promise in reducing unexpected deaths in hospitals," Verma said.
The effectiveness of CHARTWatch was evaluated in 13,649 patients aged 55 to 80 years who were admitted to general internal medicine (GIM) (about 9,626 in the pre-intervention period and 4,023 used CHARTWatch). About 8,470 admitted to subspecialty units did not use CHARTWatch.
Regular communications helped reduce deaths, as CHARTWatch engaged doctors with real-time alerts, twice-daily emails to nursing teams, and daily emails to the palliative care team, the researchers said.
A care pathway was also created for high-risk patients, prompting greater follow-up by nurses and better communication between nurses and doctors. This encouraged doctors to reevaluate patients.
Verma said the artificial intelligence system can be used to help nurses and doctors provide high-quality care.
Co-author Dr. Muhammad Mamdani, director of the University of Toronto, said the study evaluates the outcomes associated with the complex deployment of the entire AI solution.
It's important to understand the real-world impacts of this promising technology, Mamdani said.
