Pediatric Sepsis in Community Emergency Care Settings: Guideline Concordance and Outcomes

Abstract

Objectives: Bundled pediatric sepsis care has been associated with improved outcomes in tertiary pediatric emergency departments. Sepsis care at nontertiary sites where most children seek emergency care is not well described. We sought to describe the rate of guideline-concordant care, and we hypothesized that guideline-concordant care in community pediatric emergency care settings would be associated with decreased hospital length of stay (LOS).

Method: This retrospective cohort study of children with severe sepsis presenting to pediatric community emergency and urgent care sites included children 60 days to 17 years with severe sepsis. The primary predictor was concordance with the American College of Critical Care Medicine 2017 pediatric sepsis resuscitation bundle, including timely recognition, vascular access, intravenous fluids, antibiotics, vasoactive agents as needed.

Results: From January 1, 2015, to December 31, 2017, 90 patients with severe sepsis met inclusion criteria; 22 (24%) received guideline-concordant care. Children receiving concordant care had a median hospital LOS of 95.3 hours (50.9-163.8 hours), with nonconcordant care, LOS was 88.3 hours (57.3-193.2 hours). In adjusted analysis, guideline-concordant care was not associated with hospital LOS (incident rate ratio, 0.99 [0.64-1.52]). The elements that drove overall concordance were timely recognition, achieved in only half of cases, vascular access, and timely antibiotics.

Conclusions: Emergency care for pediatric sepsis in the community settings studied was concordant with guidelines in only 24% of the cases. Future study is needed to evaluate additional drivers of outcomes and ways to improve sepsis care in community emergency care settings.

FIGURE 1.

Study flow diagram of 264 patients who were identified from sepsis registry between January 1,2015, and December 31,2017. Excluded patients are described as listed. Final sample size of included patients with severe sepsis is 90.

FIGURE 2.

Concordance with the sepsis care bundle and median time to intervention. Dot and whiskers represent median and IQR.

FIGURE 3.

Pareto diagram of guideline elements. The primary Y-axis depicts total rate of nonconcordance to each component of guideline concordant care. The secondary Y-axis depicts the cumulative percentage that each guideline element contributes to overall nonconcordance (solid black line). The dotted line represents that overall nonconcordance was associated with 3 guideline elements (recognition, intravenous access, and antibiotics).