Computerized clinical decision support systems for the early detection of sepsis among pediatric, neonatal, and maternal inpatients: scoping review

Background:
Sepsis is a severe condition associated with extensive morbidity and mortality worldwide. Pediatric, neonatal, and maternal patients represent a considerable proportion of the sepsis burden. Identifying sepsis cases as early as possible is a key pillar of sepsis management and has prompted the development of sepsis identification rules and algorithms embedded in computerized clinical decision support (CCDS) systems.

Objective:
This scoping review aimed to systematically describe studies reporting on the use and evaluation of CCDS systems for early detection of pediatric, neonatal, and maternal inpatients at risk of sepsis.

Methods:
The databases MEDLINE, Embase, CINAHL, The Cochrane database, LILACS, Scopus, Web of Science, OpenGrey, clinicaltrials.gov, and PQDT were searched using a search strategy incorporating terms for sepsis, clinical decision support, and early detection. Title, abstract, and full-text screening were performed by two independent reviewers, consulting a third reviewer as needed. One reviewer performed data charting with a sample of data checked by a second reviewer and discussions with the review team as necessary.

Results:
33 studies were included in this review: 13 pediatric, 18 neonatal, and 2 maternal. All studies were published after 2011, with 27 (82%) published in the last 5 years. Of the pediatric studies, the most common outcome investigated was accuracy of sepsis identification (9/13; 69%). The gold standard definition for sepsis used in these studies varied considerably with 5 different definitions used across 9 studies. In addition, the clinical criteria applied by pediatric CCDS systems to detect sepsis were diverse, with 18 criteria used across 13 studies in various combinations. Temperature was the most commonly included clinical criterion, present in all 7 peer-reviewed pediatric articles. In neonatal studies, 78% (14/18) investigated the Kaiser Permanente early onset sepsis risk calculator. The gestational age of included neonates varied between studies, with only 1 study investigating very preterm neonates born at <33 gestational weeks. Most studies investigating neonatal sepsis reported treatment and management outcomes, with antibiotics-related outcomes being the most common. Usability and cost-related outcomes were each reported in 2 pediatric or neonatal studies. Neither of the 2 studies investigating CCDS systems in maternal populations were peer-reviewed articles.

Conclusions:
This review found limited research investigating CCDS systems to support early detection of sepsis among pediatric, neonatal, and maternal populations, despite the overrepresentation of these vulnerable populations in the global sepsis burden. We have highlighted the need for a consensus definition for pediatric and neonatal sepsis, and the study of usability and cost-related outcomes as critical areas for future research.