North Oaks Health System Harnessing Predictive Modeling to Save Lives from Sepsis

Sepsis is a medical emergency that can lead to tissue damage, organ failure and even death. It is a complication of an existing infection left untreated or not treated appropriately or in a timely fashion. Each year, sepsis leads to death for a quarter of a million Americans and causes two percent of all hospital admissions nationwide, according to the Centers for Disease Control and Prevention (CDC). Anyone, anywhere can have an infection that becomes septic, but children, senior adults and chronically ill individuals are most at risk.

Clinicians and computer analysts at North Oaks Health System have innovated a tool that harnesses the power of Epic’s electronic health record technology and predictive modeling to save lives from sepsis. In just one year, the sepsis mortality rate at North Oaks has dropped by nearly 18 percent compared with the previous year since the implementation of this model, according to Dr. Herbert Robinson, chief health information officer for North Oaks Health System.

“If left untreated, septic patients will die. So treatment must begin within the first few hours of onset to offer the greatest chance of survival,” Robinson explains. “That’s why from the time patients arrive at our hospitals, we evaluate them carefully and then constantly monitor their clinical course to detect any developing signs or symptoms of sepsis. We do this so that appropriate and timely interventions can occur.”

The team at North Oaks is using a computerized tool, called a Clinical Care Advisory (CCA), to help notify caregivers of patients at risk for sepsis based on information in the patient’s medical record. The CCA runs on a “predictive model" in the patient’s electronic health record looking for signs and symptoms of sepsis.

“By embedding machine learning into the existing workflow, minutes can be saved,” explains Seth Hain, Epic’s director of analytics and machine learning. “As data flows into the chart, the algorithm looks for patterns indicating the early onset of sepsis. Then the results are put directly in front of clinicians, giving them the cue to intervene earlier and help patients.”

Once a patient arrives in the emergency room, the CCA sweeps the patient’s health information every 15 minutes, monitoring more than 80 different data points, to generate a sepsis risk score. If the patient’s score reaches a certain threshold, the clinicians get a warning that the patient may be becoming septic. The patient is then evaluated more closely, and the CCA recommends the best treatments for sepsis. This process continues throughout the patient’s hospital stay, allowing doctors and nurses to make decisions to implement life-saving treatments at the earliest possible time.

“Patients are now receiving antibiotics about 25 percent faster — or 30 minutes sooner — than they were before CCA. Combined with the clinical expertise of our doctors and nurses, our approach is saving lives,” Robinson affirms.

“Our goal is to make this new tool more accessible to community health systems like North Oaks by ensuring the setup and statistical validation are efficient and easy to use,” says Hain. “North Oaks was able to set up the model quickly, allowing them to focus on the outcomes.”

The most frequently identified germs that cause infections that lead to sepsis include staphylococcus, E. coli and some types of streptococcus, according to the CDC. Symptoms can include one or any combination of the following:

• Confusion or disorientation
• Shortness of breath
• High heart rate
• Fever or shivering or feeling very cold
• Extreme pain or discomfort
• Clammy or sweaty skin.

Robinson encourages anyone who suspects sepsis or has an infection that’s not getting better or is getting worse to seek immediate medical attention.