Interobserver reliability and diagnostic accuracy of prehospital triage for identifying traumatic brain injury in paediatric patients: a systematic review

Abstract

Abstract Purpose The consistency and accuracy of paediatric TBI triage tools can be affected by different factors, such as patients’ characteristics and the level of knowledge and skill of the caregiver. This systematic review included all the available data on the level of agreement between paramedics and ED physicians about the reliability of tools to identify paediatric TBI and the diagnostic accuracy of several such tools in prehospital settings when used by paramedics. Methods MEDLINE (OVID), EMBASE (OVID), Cochrane Library (OVID), and CINAHL Plus (EBSCO) databases were searched from inception to 27 October 2022. Quality, bias, and applicability were assessed using COSMIN for interobserver reliability studies and QUADAS-2 tool for diagnostic accuracy studies. Narrative synthesis was employed because data were unsuitable for meta-analysis. Results Initial searches identified 660 papers in total. Five met the inclusion criteria. Two studies showed moderate agreement between paramedics and ED physicians for GCS assessment. The PTS overtriage rate was 10% and the undertriage rate was 62%, while the triage tape had an overtriage rate of 18% and an undertriage rate of 68%. Pre-hospital GCS had 86.67% sensitivity and 71.43% specificity [95% CI]: 0.74–0.96 for neurosurgically significant TBI. Conclusion Low level of GCS agreement and poor diagnostic accuracy may cause further harm to the patient; thus, further studies are recommended to improve the prehospital management of children with head injuries.