Blessing employs NLP for smarter sepsis detection

  • August 29, 2022
  • William Payne

A chain of hospitals in Illinois are using natural language processing (NLP) to analyse clinical notes to spot missed signs of emerging sepsis among patients. The adoption of NLP for acute patient monitoring is part of a move to smarter sepsis surveillance at the Blessing Health System in Quincy, Illinois, and forms part of the POC Advisor suite from Wolters Kluwer.

By analysing patient data in real-time, POC Advisor identifies both risk of sepsis and emerging symptoms. The system allows clinicians to respond more rapidly and guides them to more effective clinical decisions. The move expands Blessing’s adoption of Wolters Kluwer’s clinical surveillance suite by adding POC Advisor to its Sentri7 infection prevention and pharmacy surveillance solutions.

POC Advisor is in use at Blessing Hospital, the health system’s acute care hospital in Quincy and at its critical-access hospital in Pittsfield, the Illini Community Hospital.

Blessing employs a collaborative approach to sepsis symptom detection and management with various care settings across the health system.

This process has been largely retrospective and relied on a multi-disciplinary group of clinicians to review cases, track data, and outcomes, and stay up to date on evolving guidelines and protocols for sepsis management. With a sepsis coordinator on board using POC Advisor’s centralised patient monitor and advanced analytics capabilities, Blessing is able to streamline manual, time-intensive processes. This sepsis surveillance solution monitors patients in real-time to better identify and alert on sepsis cases allowing clinicians to intervene earlier.

Every hour sepsis goes untreated mortality increases by about 8%.2 Wolters Kluwer’s POC Advisor technology detects sepsis six hours earlier than common systemic inflammatory response syndrome (SIRS)-triggered electronic health record (EHR) alerts by analysing a broad spectrum of real-time data, monitoring trends, and assessing all data in the context of the patient’s clinical status and co-morbid conditions. Alerts routed directly to a central monitor, into the EHR, or to point of care devices equip the bedside teams to act fast.

At Blessing, many providers rely on documenting patient conditions in text notes; POC Advisor’s natural language processing (NLP) scours clinical notes for critical information missing from standardised EHR data that might indicate sepsis. Coupled with evidence-based recommendations for the treatment of sepsis, POC Advisor helps improve patient outcomes by ensuring the delivery of bundle care components.

“Early detection of sepsis is a challenge for many health systems, and we were determined to tackle this through more proactive monitoring,” said Mary Barthel, MD, Chief Quality and Safety Officer, Blessing Health System. “Using POC Advisor, we can now deliver real-time patient status updates to our clinicians so they can intervene immediately with an evidence-based approach to help save more lives and improve our sepsis bundle compliance. By combining advanced technology with our sepsis coordinator on the ground, we are better equipped to identify patients who develop sepsis earlier and optimise their care.”

Wolters Kluwer’s team of clinical and informaticist experts have optimised the use of POC Advisor for Blessing by developing team workflows, building and maintaining the clinical rules that drive sepsis alerts, and providing detailed case reviews to clinical teams.

“Clinicians are already overloaded with alerts. Our Wolters Kluwer team is helping healthcare providers like Blessing break through this noise so they have actionable, timely, and specific recommendations that allow them to deliver high-quality care that adheres to the most current evidence,” said Karen Kobelski, Vice President and General Manager of Clinical Surveillance Compliance & Data Solutions, Wolters Kluwer Health. “Blessing has made its sepsis programme a priority, and we’re honored to work hand in hand with the Blessing team to seamlessly integrate sepsis alerts into provider workflows to achieve the hospital’s goals.”