Carilion Clinic Deploys Risk Scores for C. Diff Infection

Sept. 15, 2020
Roanoke, Va.-based system piloted Wolters Kluwer’s Sentri7 clinical surveillance solution enhanced with algorithm to identify patients at risk for C. diff infection

Patients with Clostridioides difficile (C. diff) infection (CDI) are associated with longer lengths of stay, higher readmission rates, and higher mortality. Carilion Clinic has been piloting an artificial intelligence-based solution from Wolters Kluwer that identifies at-risk patients earlier to help address modifiable risk factors, such as high-risk antimicrobials and acid suppressants.

“I don't think there is a health system that has fully figured out CDI,” said Nathan Everson, PharmD, an infectious diseases clinical pharmacist at Roanoke, Va.-based integrated health system Carilion Clinic. “I think for every health system, it is on their hit list for their infection control program. The problem is multi-factorial. Lots of things can affect a patient’s normal flora and predispose them to a CDI. It is something we constantly struggle with.” He noted that besides poor outcomes, health systems can potentially face millions of dollars in penalties from CMS. There were 786 hospitals penalized  under the hospital-acquired condition (HAC) reduction program in 2020.

Wolters Kluwer’s Sentri7 clinical surveillance solution has been enhanced with AI technology to identify patients at risk for C. diff infection. Sentri7 pulls data from multiple facets of the EHR such as laboratory, demographics, pharmacy records, and vital signs. The solution uses an algorithm to produce CDI risk scores for individual patients, helping clinicians to proactively alter patient care to reduce risks for infection and improve outcomes. The company says Sentri7’s predictive CDI algorithm tracks patients’ risk levels and automatically updates the score if a patient’s condition changes.

Steve Mok, PharmD, Wolters Kluwer’s manager of pharmacy services, said the company has been working on this project for a year and a half. “We had some resources available internally on our data science team to see how we can leverage AI to see how we can improve patient care practices. The reason we decided to go after C. diff is because it is a real clinical problem and we recognize the challenge of the health systems’ experience. It is one thing to be able to do a risk prediction, but we also want to be able to give the power back into the clinicians’ hands to do something about it,” he said. “C. diff is one of those cases where there is long-established literature that there are things that a provider can do to modify a patient’s risk factors. Some of them we cannot change such as age or previous experience with C. diff. But other things are low-hanging fruit, such as patients being put on a proton pump inhibitor (PPP) like Prilosec or antibiotics that are no longer necessary. Nathan and I are both trained in infectious disease, so we are passionate about getting rid of antibiotics when they are no longer needed.”

Everson said Wolters Kluwer developed some rules for Carilion to look at using their own data, and Carilion uses that CDI score to target specific interventions. “We have hundreds of rules for different issues and problems with patients. The stewardship team is looking at those throughout the week and calling to do interventions and document interventions. “A lot of it has to do with getting rid of medications that we know are going to predispose someone to CDI. Our hospital has 800 beds, and only so many people looking at stewardship on a given day, so this gives us the ability to focus in on the more high-risk patients. It helps us optimize our time.”

At Carilion decisions are made by pharmacists in conjunction with the physicians in the hospital. “We will reach out to them and ask if a proton pump inhibitor is absolutely necessary and tell them that the patient is high risk of C. diff,” Everson said. “A lot of times we can get rid of it, so the interventions are pretty well accepted.”

Everson noted that Carilion is still relatively early in its deployment. “We had a small group of people in the pilot, and now we have rolled it out to all the pharmacists. This helps us prioritize people with a rule we feel pretty good about.”

He said the infection protection group is definitely interested in tracking the impact of the intervention going forward. “One issue is that they haven’t done anything that has not been COVID-related in six months,” Everson said, but they will be interested in tracking the impact on CDI rates.

“We thought this was a great idea, but there was probably some skepticism about how good it would be,” Everson said. “But knowing that the top 20 percentile of our risk scores capture greater than 50 percent of our CDI cases was actually nice to see. Getting that data back that it is performing well at our institution has helped our pharmacists include this in their workflow.”

Sponsored Recommendations

How Digital Co-Pilots for patients help navigate care journeys to lower costs, increase profits, and improve patient outcomes

Discover how digital care journey platforms act as 'co-pilots' for patients, improving outcomes and reducing costs, while boosting profitability and patient satisfaction in this...

5 Strategies to Enhance Population Health with the ACG System

Explore five key ACG System features designed to amplify your population health program. Learn how to apply insights for targeted, effective care, improve overall health outcomes...

A 4-step plan for denial prevention

Denial prevention is a top priority in today’s revenue cycle. It’s also one area where most organizations fall behind. The good news? The technology and tactics to prevent denials...

Healthcare Industry Predictions 2024 and Beyond

The next five years are all about mastering generative AI — is the healthcare industry ready?