Surveillance tool has mixed impact on hospital outbreak response
An automated tool successfully reduced the size of outbreaks of healthcare-associated pathogens by 64% in 82 US hospitals prior to the COVID-19 pandemic.
Source: Getty Images
By - An automated, algorithm-driven surveillance tool successfully acted as an early warning system to detect and respond to outbreaks of healthcare-associated pathogens in a cohort of US hospitals pre-pandemic, but had little impact once COVID-19 emerged, according to a recent NEJM Evidence study.
The research was the result of a collaboration between Harvard Pilgrim Health Care Institute, HCA Healthcare, UCI Health and the Centers for Disease Control and Prevention (CDC) aimed at reducing the risk of healthcare-associated infections and improving patient care.
The research team indicated that these infections present significant patient safety risks, necessitating approaches to detect and contain pathogens in hospitals.
"Despite significant progress in reducing healthcare-associated infection outbreaks, including of antimicrobial-resistant pathogens, they remain an industry challenge and can present as clusters that signal potential for transmission to patients,” said Joseph Perz, DrPH, MA, senior advisor for Public Health Programs in CDC’s Division of Healthcare Quality Promotion, and committee member for the CDC’s Council for Outbreak Response: Healthcare-Associated Infections, in a press release. “The CLUSTER trial provides evidence that early detection powered by automation tools and quick action can prevent outbreaks from growing."
However, while these approaches might prevent an outbreak from growing, the researchers noted that currently, it is unknown whether the use of an automated statistical surveillance system could enhance efforts to reduce the size of an outbreak.
To test this, the research team developed an automated outbreak detection model designed to use clinical laboratory and microbiology data – including information about organisms grown from patients’ clinical cultures.
Using statistical analysis, this data was assessed to detect increases in the presence of 100 bacterial and fungal species potentially indicative of increased pathogen transmission. Any detected increases were designed to trigger automatic, real-time notifications for hospital staff to deploy their outbreak response and infection prevention protocols.
To evaluate the efficacy of the tool, researchers conducted a randomized clinical trial from 2019 to 2022 in 82 hospitals within the HCA Healthcare system. Half of these hospitals implemented the automated tool, while the rest served as study controls and were instructed to care for patients and conduct outbreak response as they usually would.
All hospitals in the cohort followed outbreak response protocols if and when an outbreak was detected by their respective infection prevention programs. The number of additional infection cases occurring after outbreak detection was measured.
From there, differences in case numbers between the baseline period — February 2017 to January 2019 — and the intervention period — July 2019 to January 2022 — were compared.
The analysis revealed that the hospitals directed to deploy the automated detection tool saw a 64 percent reduction in the size of potential outbreaks during the pre-pandemic period. However, upon the emergence of COVID-19, these trends shifted.
Across the combined pre-pandemic and pandemic trial periods, the automated tool was observed to have no overall effect on outbreak size. The study authors postulated that during the pandemic, hospital personnel were unable to respond as effectively to automated alerts, even though the tool notified each hospital about three possible outbreaks on average per year.
The early success of the model is driving researchers to continue to assess its potential.
“This ongoing collaboration continues to leverage our scale – both the number of our hospitals and our advanced data ecosystem – as we work to rapidly answer clinical questions that benefit patients everywhere,” stated Kenneth Sands, MD, MPH, chief epidemiologist at HCA Healthcare. “We are using this detection tool in the hospitals where we tested it and are evaluating implementation more widely across our system.”
“Outbreaks in hospitals are often missed or detected late, after preventable infections have occurred. This study provides a practical and standardized approach to identify early transmission and halt events that could become an outbreak in hospitals,” explained lead investigator Meghan A. Baker, MD, ScD, Harvard Medical School assistant professor of population medicine at the Harvard Pilgrim Health Care Institute.
This research is the latest to explore how advanced analytics technologies could bolster outbreak response.
In October, researchers from Harvard Medical School and the University of Oxford detailed the development of an AI tool capable of forecasting how a virus could evolve to escape the immune system.
The model, called EVEscape, uses evolutionary and biological information to predict potential new virus variants and mutations.
Researchers emphasized that in doing so, the tool could help inform vaccine and therapeutic development for rapidly evolving viruses like SARS-CoV-2.
