Clinical decision support tools save 20 lives in 6 months at Saint Luke's Health System

This case study examines a platform that helped identify 150 percent more sepsis patients, decreased mortality by 30 percent, and correctly identified and treated sepsis, increasing billing income by $70,000 per month.
By Bill Siwicki
03:30 PM
Share
outside view of Saint Luke's Health System in Kansas City
Saint Luke’s Health System is an 11-hospital network based in Kansas City, Missouri. Credit: Saint Luke’s

Saint Luke’s Health System is an 11-hospital network based in Kansas City, Missouri, that accepts transfer patients from various rural and community-based hospitals throughout the area. Saint Luke’s noticed increased mortality from sepsis among transfer patients due to this time-sensitive and deadly condition not being diagnosed early enough.

With multiple hospitals using multiple EHR technologies sending patients into the Saint Luke’s system, it created an information black hole for these sepsis patients. This inevitably led to missteps of care collaboration and coordination.  

“Quick identification of sepsis is a challenge to even the most experienced clinician,” said Joann Paul, vice president of quality and patient safety at Saint Luke’s Health System. “The early signs are often subtle, mimic other conditions, and are very easily lost when transferring a patient. Our goal is to very quickly identify sepsis, treat it rapidly using evidenced-based protocols, and ultimately save lives.”

"Often the transferring hospitals have different EHR systems, which can make continuity of care and communication challenging."

Gina Lawson, DO, Saint Luke’s Physician Group

PROPOSAL

Saint Luke’s worked with Redivus Health to implement a sepsis screening program that operated outside the EHR to proactively address this problem. The nurse transfer team began using the Redivus clinical decision support platform in 2017. The goals of the pilot program were to identify sepsis patients at the transferring facilities, start treatment sooner and improve patient outcomes.

The clinical decision support platform enabled transfer nurses to identify significantly more (200+) sepsis cases by providing simple intuitive guidance, and it allowed for patients to receive time-critical treatment before being transported to a higher level of care.

MARKETPLACE

There is a big market for clinical decision support technology. CDS vendors include Argo Healthcare Solutions, Automated Clinical Guidelines, Epic, Information Builders, National Decision Support Company, Pepid, Talis Clinical and Wolters Kluwer.

"We are working to integrate the outcome of executing the questionnaire so the documentation of the result populates our EHR."

Debe Gash, Saint Luke’s Health System

MEETING THE CHALLENGE

As the project moves forward, Redivus will work to integrate with the Saint Luke’s EHR so the Redivus detailed documentation of sepsis identification and treatment can seamlessly close the loop of care.

“During transfer calls – except for STEMI, stroke or trauma – the transfer nurse conducts a screen through the sepsis app guided workflow,” said Dr. Gina Lawson, chief medical officer at Saint Luke’s Physician Group. “The result of the screen is communicated to both the transferring doctor and the doctor receiving the transfer patient on behalf of Saint Luke’s.”

*Patient ID: When a patient is identified through the software, it then alerts providers and next steps are suggested so treatment may begin promptly, Lawson added.

“The application operates separate of the EHR,” said Debe Gash, chief digital officer at Saint Luke’s Health System. “However, we are working to integrate the outcome of executing the questionnaire so the documentation of the result populates our EHR and provides further decision support once the patient is admitted to the hospital. This application can operate standalone or in an integrated fashion with any EHR.”

*CDS takes action: If an emergency department physician calls a transfer nurse at Saint Luke’s, for example, that nurse is told the patient has mental status changes. The transfer nurse then asks the ED physician for vital signs, CBC, or any lab they may have for the patient. As the transfer nurse brings the transfer doctor online, she or he tells the physician the app has triggered for sepsis.

“The transfer doctor not only gets regular reports, but also asks if a UA, CXR, Blood c and lactate can be obtained while the patient is still there in the local emergency department,” Lawson explained. “The sepsis protocol is initiated before the patient is transferred. We are able to start fluids and antibiotics hours earlier than before.”

"Quick identification of sepsis is a challenge to even the most experienced clinician."

Joann Paul, Saint Luke’s Health System

RESULTS

In the 6-month pilot program,. CDS helped identify 150 percent more sepsis patients; it decreased mortality by 30 percent, which led to approximately 20 lives saved; and it correctly identified and treated sepsis, increasing billing by $70,000 per month.

“The presentation of sepsis is often subtle,” Paul remarked. “The tool takes the patient’s presentation, including vital signs and lab work, and places it in an algorithm that detects the probability of sepsis very quickly. It also alerts the provider to begin the bundle and keeps track of when more intervention should occur.”

Sepsis identification and prompt treatment saves lives, she added.

“The identification of sepsis is much more difficult than figuring out when a stroke happens or when a heart attack occurs,” she said. “The beginning of sepsis is not often visible, and requires multiple pieces of data coming together. The faster sepsis is detected, the faster treatment can begin, and the lower the mortality rate. Each hour of delay in treatment increases mortality by 4 percent.”

ADVICE FOR OTHERS

“I would advise to plan early for implementation of the technology, particularly around how to operationalize it into the workflows of clinicians,” Paul said. “Adding another data point to a very busy handoff may not drive the actual utilization well. Keeping the focus on the potential impact of correct treatment to the patient will help drive improvement.”

She also suggests recruiting a clinical expert or leader who can help with the utilization and the “why” behind the application.

Twitter: @SiwickiHealthIT
Email the writer: bill.siwicki@himssmedia.com