A new sepsis watch program that capitalizes on predictive analytics to improve clinical workflows cut sepsis mortality rates by nearly one-third, researchers say.
The key to the HIMSS Electronic Medical Record Adoption Model (EMRAM) is that it gauges clinician use of EMR tech, patients’ engagement and clinical results to enhance organizational results.
Earlier this month, experts at the Centers for Medicare & Medicaid Services said at a quality conference that they will be taking a closer look at sepsis rates in nursing homes while continuing broader efforts to improve infection control.
Sepsis is a bloodstream infection caused by bacteria. Urinary tract infections are the most common source of it in the elderly.
Nursing home residents are seven times more likely than non-nursing home residents to suffer from severe sepsis, according to the National Institutes of Health. The agency said that mortality due to severe sepsis in seniors is up to one-and-a-half times higher than in younger populations.
Sepsis cases remain a key source of lawsuits against nursing homes.
That’s a big reason why Duke Health’s tech-led Sepsis Watch program is being so heavily watched. At Duke, the program is hospital-based, but it is designed to improve outcomes across patient populations.
Machine learning within the system’s EMR program is the framework for the predictive analytics model, which can be customized for each individual patient, officials said.
EMRAM’s revalidation helped Duke build Sepsis Watch, which it calls the world’s first machine learning model for sepsis screening. It now claims a 93% screening accuracy rate. Meanwhile, Duke’s false sepsis diagnoses have fallen by almost two-thirds.
Before the algorithm-driven system was devised, Duke Health had more than five times as many false alerts as positive ones. In addition, only 6.8% of the patients it did tag as having sepsis actually did.
As McKnight’s has previously reported, CMS is working with the Centers for Disease Control and Prevention and the Sepsis Alliance to develop more education on best practices for preventing, recognizing and treating sepsis.
“In the nursing home, where we really want to focus is on making sure that the staff is well-trained on the identification of early sepsis before it becomes really dramatic,” said Colleen Frey, director of the CMS Division of Community and Population Health, at the quality conference April 10. “As you know, sepsis is kind of a runaway train if you don’t stop it early.”
She also said nursing homes that work with family members to explain sepsis might also see results, given that family members involved in care or visiting frequently might be first to notice changes in a patient’s conditions.
