AI could help improve hospital infection prevention
Click Here to Manage Email Alerts
Key takeaways:
- AI can help hospitals improve infection prevention.
- It can also assist clinicians adhere to antimicrobial stewardship guidelines.
Artificial intelligence could help hospitals improve infection prevention efforts and make their outbreak responses faster and more accurate, according to experts.
Two pre-conference presentations delivered before the European Congress of Clinical Microbiology and Infectious Diseases suggest that increasing and improving employee interaction tracking methods and refining AI systems to properly track data can catch outbreaks in real time.
AI is slowly being adopted for a wide array of health care services, including detecting stroke, speeding up administrative duties and helping to identify people at high risk for HIV and three other sexually transmitted infections.
The ability to use AI systems to organize and monitor data, including the integration of robotic technology for certain tasks, can be used to detect hospital outbreaks, if not help prevent them, according to researchers involved with both presentations.
“We should look to offload repetitive tasks to AI systems, such as environmental cleaning and mask compliance auditing,” Richard Drew, MD, FRCPI, FRCPath, DipHIC, a microbiologist at Rotunda Hospital in Dublin, Ireland, said in a press release. “AI can also offer significant opportunities in terms of big data analytics of certain patient groups.”
The key, he said, is recognizing where AI will waste money — as opposed to saving it — in individual institutions by identifying issues unique to their facility, from ensuring staff is properly wearing face masks, to keeping the air and environment in a facility clean, to switching patients from IV antibiotics to oral therapy for individual patients.
In his presentation, Drew referred to an analysis of more than 30 studies on using facial recognition AI technology to assess if medical staff was wearing masks properly, which the systems successfully did.
He also noted that AI systems in robots have prevented the need for manually cleaning every corner of a facility because the robot can monitor the environment and target cleaning to where it is needed. He said AI systems can streamline antimicrobial stewardship methods to assist clinicians when deciding on changing a patient’s medication.
In a second presentation, Jonas Marschall, MD, professor of medicine in the division of infectious diseases at Washington University School of Medicine in St. Louis, described a vancomycin-resistant Enterococcus faecium (VRE) outbreak at Bern University Hospital in Switzerland from late 2017 to mid-2020.
Initial investigations showed that most patients with VRE were colonized rather than infected, that isolating patients with VRE was expensive, and VRE screening required significant staff effort and investment. But a reanalysis by Marschall and colleagues found that AI could have helped correctly predict risk factors for patient colonization.
Marschall said that by improving systems that track all employee, patient and visitor interactions with individual patients, beyond just the records kept by nursing personnel, AI systems can analyze data continuously, tracking colonizations, infections and potential risk factors to help medical personnel prevent them.
“This approach could even help with individual patients infected with a multidrug-resistant organism or small clusters of such patients because it could identify surrounding patients/employees and rooms/devices that would need to be addressed, either by screening or by targeted intervention,” Marschall said. “The beauty of AI in outbreak management — and where its greatest power lies — is to make real-time or near real-time operational decisions easier, quicker and more precise.”
References:
- Drew R, et al. Presentation 3790-1. Presented at: Pre-ECCMID Day on Infection Control and Prevention; Feb. 28, 2024; online.
- Marschall J, et al. Presentation 3790-3. Presented at: Pre-ECCMID Day on Infection Control and Prevention; Feb. 28, 2024; online.