Artificial intelligence (AI) can help reduce unexpected deaths in hospitals by accurately identifying patients at high risk of health deterioration as per research published in CMAJ (Canadian Medical Association Journal) (1✔ ✔Trusted Source
Clinical evaluation of a machine learning–based early warning system for patient deterioration
Go to source
).
This AI-based system enables healthcare providers to intervene earlier, improving patient outcomes and saving lives.
Rapid deterioration among hospitalized patients is the primary cause of unplanned admission to the intensive care unit (ICU).
Previous research has attempted to use technology to identify these patients, but the evidence is mixed about the application of prediction tools to help vulnerable patients at the highest risk.
Researchers from Unity Health Toronto, ICES, and the University of Toronto studied the effectiveness of CHARTWatch, an AI-based early warning system used on the general internal medicine (GIM) ward at St. Michael’s Hospital after 3 years of development and testing.
Transforming Healthcare with AI
The study included 13 649 patients aged 55–80 years admitted to GIM (9626 in the pre-intervention period and 4023 using CHARTWatch) and 8470 admitted to subspecialty units that did not use CHARTWatch.
During the 19-month-long intervention period, 482 patients in GIM became high-risk, compared with 1656 patients who became high-risk in the 43-month-long pre-intervention period.
Advertisement
There were fewer nonpalliative deaths in the CHARTWatch group than in the pre-intervention group (1.6% v. 2.1%).
“As AI tools are increasingly being used in medicine, it is important that they are evaluated carefully to ensure that they are safe and effective,” says lead author Dr. Amol Verma, a clinician-scientist at St. Michael’s Hospital, Unity Health Toronto, and Temerty professor of AI research and education in medicine, University of Toronto, Toronto, Ontario.
Advertisement
“Our findings suggest that AI-based early warning systems are promising for reducing unexpected deaths in hospitals.”
Regular communications helped reduce deaths as CHARTWatch engaged clinicians with real-time alerts, twice-daily emails to nursing teams, and daily emails to the palliative care team.
The team also created a care pathway for high-risk patients with increased monitoring by nurses, enhanced communication between nurses and physicians, and prompts to encourage physicians to reassess patients.
“Ultimately, this study shows how AI systems can support nurses and doctors in providing high-quality care,” says Dr. Verma.
The authors hope that AI solutions like CHARTWatch can improve patient health and avoid premature deaths.
“This important study evaluates the outcomes associated with the complex deployment of the entire AI solution, which is critical to understanding the real-world impacts of this promising technology,” says coauthor Dr. Muhammad Mamdani, vice president of data science and advanced analytics at Unity Health Toronto and director of the University of Toronto Temerty Faculty of Medicine Centre for AI Research and Education in Medicine.
“We hope other institutions can learn from and improve upon Unity Health Toronto’s experiences to benefit the patients they serve. Unity Health Toronto is a collaborative leader already helping to spread our AI tools via innovative partnerships with more to come.”
Reference:
- Clinical evaluation of a machine learning–based early warning system for patient deterioration
– (http://dx.doi.org/10.1503/cmaj.240132)
Source-Eurekalert