Developing an Algorithm for Pediatric Sepsis Surveillance

An algorithm that uses clinical data was successful in providing an efficient tool for pediatric sepsis surveillance.

Investigators from the University of Pennsylvania Perelman School of Medicine and the Children’s Hospital of Philadelphia conducted a retrospective observational study at a single academic children’s hospital to test the surveillance algorithm.

Their findings were published in Pediatric Critical Care Medicine and presented as a late breaker at the 49^th Critical Care Congress.

The investigators acknowledged that a method to identify the occurrence of sepsis that is not affected by a change in diagnosis or claims-based coding practices does not exist. To address this area of need, the team derived an algorithm using routine clinical data to identify infection and concurrent organ dysfunction and applied it to study longitudinal trends in the epidemiology of sepsis.

To evaluate the algorithm, the investigators looked at hospital encounters to determine trends in incidence and mortality from January 2011 through January 2019.

“Among 93,987 hospital encounters and 1065 episodes of suspected sepsis in the derivation period, the surveillance algorithm yielded sensitivity 78% (95% CI, 72—84%), specificity 76% (95% CI, 74–79%), positive predictive value 41% (95% CI, 36–46%), and negative predictive value 94% (95% CI, 92–96%),” the authors of the study reported.

During the validation period, it was observed that the algorithm yielded sensitivity 84%

(95% CI, 77—92%), specificity of 65% (95% CI, 59–70%), positive predictive value 43% (95% CI, 35–50%), and negative predictive value 93% (95% CI, 90–97%).

The authors also noted that most “false-positives” should be deemed clinically relevant sepsis following manual review.

Overall, the hospital-wide incidence of sepsis was 0.69% (95% CI, 0.67—0.71%), and inpatient incidence was 2.8% (95% CI, 2.7–2.9%).

It is also reported that risk-adjusted sepsis incidence increased over time without bias from changing diagnosis or coding practices (adjusted incidence rate ratio per year 1.07; 95% CI, 1.06—1.08; P < 0.001).

Mortality was 6.7% and did not change over time (adjusted odds ratio per year 0.98; 95% CI, 0.93—1.03; P = 0.38).

“An algorithm that uses routine clinical data available within the EHR can provide an objective, efficient, and reliable method for pediatric sepsis surveillance across emergency and inpatient hospital settings,” the authors concluded. “An increased sepsis incidence and stable mortality, free from influence of changes in diagnosis or billing practices, were evident.”

The study, Identification of Pediatric Sepsis for Epidemiologic Surveillance Using Electronic Clinical Data, was presented at the 49^th Critical Care Congress on Monday, February 17, 2020, in Orlando, Florida.