Clariyon’s artificial intelligence (AI)-powered sepsis care agent has shown potential in improving a physician’s ability to provide quality care to patients in the hospital setting.
In a clinical trial (NCT07581340), which was held across two academic emergency departments within the University of California, San Diego’s (UCSD) health ecosystem, researchers pitted the standard process in sepsis care against targeted recommendations from Clariyon’s large language model (LLM).
Go deeper with GlobalData
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
The study’s goal was to determine whether the technology’s near-real-time feedback could boost 66 physicians’ compliance with the severe sepsis and septic shock management bundle (SEP-1) protocol, which lays out the best practices for identifying and treating patients with sepsis within the first three and six hours of diagnosis. Participating physicians treated 301 patients in this study.
During the trial, the results of which are now published in the academic journal, JAMA Network Open, 82.9% of physicians randomised to the AI intervention arm achieved SEP-1 compliance compared to 70.1% in the control, standard protocol group – marking a statistically significant 13% absolute improvement (p=0.02).
Human expert reviewers also widely agreed with feedback given by Clariyon’s AI tool, with the technology and human conclusions matching in 92% of the recorded cases during this study.
While the technology did have an impact on overall physician SEP-1 compliance, this did not lead to a significant impact on intensive care unit admissions or the 30-day mortality rate of patients.
Enhancing clinical workflows in sepsis care
Clariyon debuts these results as the US spends more than $15bn per year on quality reporting, and the average physician spends around 785 hours per year performing administrative tasks manually.
According to the healthcare automation specialist, replacing the manual extraction of specific and relevant information from complex medical records, known as clinical data abstraction, with AI-driven agents can help healthcare systems diverge away from slow, costly retrospective reporting to focus on real-time, proactive care.
“This research validates the immense potential of agentic AI in acute clinical settings,” noted Clariyon’s CMO, Dr Mike McCurdy. “By automating the extraction of these complex SEP-1 metrics in near real-time, our platform benefits hospitals by alleviating administrative effort to save money and, most importantly, actively drives proactive clinical interventions that improve evidence-based bundle adherence,” McCurdy added.
Clariyon is one of many companies in the healthcare technology space looking to implement AI into the clinical workflow, with solutions focused on imaging interpretation and the streamlining of care becoming increasingly common.
A recent GlobalData report predicts that, across the medical device, pharmaceutical and healthcare provider niches, the value of the healthcare AI market will swell at a compound annual growth rate (CAGR) of around 37% from $11.9bn in 2024 to $57.4bn in 2029.
GlobalData is the parent company of Clinical Trials Arena.
