The Use of Analytics and Algorithms for Sepsis Management

For the last several years, Iatric Systems Visual SmartBoard™ has been helping healthcare organizations detect the early signs for sepsis. Identifying sepsis requires examining multiple clinical measures, which can be difficult when the information is buried in multiple reports that could be many hours old. If a physician isn’t able to put all the pieces together or doesn’t have access to all the information in a timely manner, he or she can be blindsided.

More than 1.5 million cases of sepsis occur in the United States each year with a mortality rate of up to 50 percent. Even with ideal treatment, the mortality rate from sepsis or septic shock is around 40 percent. 

We have taken our knowledge of clinical workflows, and combined that with algorithms to monitor a patient’s sepsis score in real time, and placed into our new Analytics on Demand™ for Sepsis Management product. These algorithms will allow the care team to monitor the patient course of treatment and can be altered when a track toward sepsis is identified.

The algorithms used by the Sepsis Management module are based on the criteria put forth by the Surviving Sepsis Campaign and are broken into three criteria scores (see below). A septic patient meeting two or more of the conditions in Criteria 1, one or more of the conditions in Criteria 2, and one or more of the conditions in Criteria 3, would be considered Positive for sepsis. Other combinations of course exist as well, and will result in a patient being Negative, Possible, Probable, or Positive for sepsis. 

Criteria Level 1: SIRS – Systemic Inflammatory Response Syndrome (two or more of the following)

• Temperature greater than or equal to 101°F, or temperature less than or equal to 96.8°F
• Heart Rate greater than 90 beats/minute
• Respiratory Rate greater than 20 breaths/minute
• White Blood Cell count greater than or equal to 12,000/mm3, or, WBC count less than or equal to 4,000/mm3
• Glucose greater than 140 ml/dL in a non-diabetic patients
• Altered mental status
• Chills with Rigors

Criteria Level 2: Organ Dysfunction (one or more of the following)

• Systolic Blood Pressure (SBP) < 90 mmHg or Mean Arterial Pressure (MAP) <65 mmHg
• Creatinine greater than 2.0 mg/dl
• Oliguria: (urine output) < 400ml/24h
• Bilirubin > 2.0 mg/dl
• Platelet count < 100,000 uL
• Lactate (Lactice Acid) > 2 mmol/L
• Coagulopathy: INR > 1.5 or APTT > 60 secs

Criteria Level 3: Infection (one or more of the following)

• Documented infection via positive Microbiology Culture
• Documented infection via Problem List or EDM Impressions
• Antibiotic therapy, non-prophylaxis

In addition to monitoring the patient course toward sepsis, Analytics on Demand for Sepsis Management uses analytics to mine information — looking at locations, facilities, and attending and admitting MDs. This information is used to perform “what if” scenarios such as:

• What doctors have a high percentage of patients with sepsis?
• Was it due to the procedure or specific location of the operation?

These are just a few of the ways we are helping our customers with sepsis management. If you would like more information on this topic, please attend our upcoming webcast on March 22^th, Using Analytics for Value-Based Care.