Background Hospital-based clinical decision tools are used to support clinicians when a child presents to the Emergency Department with a head injury. These are effective in informing decisions regarding computed tomography scanning. However, there is no pre-hospital decision tool to reduce unnecessary conveyance to hospital for head-injured children. This study aims to determine what elements of in-hospital tools for the assessment and management of children with head injury can be adapted for use in pre-hospital care.

Methods Systematic mapping review and narrative synthesis of published journal articles and grey literature. Searches were conducted using MEDLINE, EMBASE, PsycINFO, CINAHL and AMED. We systematically identified all in-hospital clinical decision support tools and extracted from these the clinical criteria used in decision making. We complemented this with a narrative synthesis.

Results Following deduplication 874 articles were identified. After screening titles and abstracts, 697 articles were excluded, leaving 177 full text articles. Of these, 95 articles were excluded, yielding 82 studies. A further 14 studies were identified in the literature, totalling 96 analysed studies. 25 relevant in-hospital clinical decision tools were identified from these studies, which included 67 different clinical criteria, and these were grouped into 18 categories.

Conclusion Factors that increase the likelihood of neurosurgical intervention, and should be considered for use in a clinical decision tool designed to support paramedics in the assessment and management of children with head injury, are: signs of skull fracture; a large, boggy or non-frontal scalp haematoma (particularly in infants); focal neurological deficit; Glasgow Coma Scale score less than 15; prolonged or worsening headache; prolonged loss of consciousness; post traumatic seizure; amnesia in older children; non-accidental injury; drug or alcohol use; less than one year old. Clinical criteria that require further investigation include mechanism of injury; clotting impairment/anticoagulation; vertigo; length of time of unconsciousness; number of vomits. It appears that any clinical predictor present in isolation is unlikely to indicate clinically significant traumatic brain injury. There are likely to be additional clinical criteria that are relevant to paramedic assessment and practice, which are not included in any of these tools. None of the existing hospital-based clinical decision support tools can be directly implemented into paramedic practice as they are.