Methods for Collecting Paired Observations From Emergency Medical Services and Emergency Department Providers for Pediatric Cervical Spine Injury Risk Factors.

OBJECTIVES: Cervical spine injuries (CSIs) after blunt trauma in children are rare, but cause substantial morbidity and mortality. Emergency medical services (EMS) and emergency department (ED) providers routinely use spinal precautions and cervical spine imaging, respectively, during the management of children experiencing blunt trauma. These practices lack evidence, and there is concern that they may be harmful. A pediatric CSI risk assessment tool is needed to inform EMS and ED provider decision making. Creating this tool requires prospective data collection from EMS and ED providers at the time of patient evaluation. The purpose of this article is to describe the methods used to prospectively capture paired EMS and ED provider observations of children cared for after blunt trauma. Given the rarity of prospective observational research with EMS, the novel use of Research Electronic Data Capture (REDCap) in this study, and the potential to inform future studies, we are publishing our methodology in advance of outcome data related to the risk assessment tool.

METHODS: The study was conducted at four tertiary children's hospitals as a prerequisite for a planned larger study to derive a CSI risk assessment tool. We created a web-based, branch-logic questionnaire using the REDCap data collection system. The survey was administered via tablet computer to ED providers evaluating children with blunt trauma and, if applicable, to EMS providers who provided patient care at the scene. We collected information regarding factors determined a priori to be plausibly associated with CSI in children. Eligible children presenting to the ED after blunt trauma with at least one of the following one of the following were included: prehospital EMS spinal precautions, ED trauma team evaluation, or cervical spine imaging in the ED. Exclusions included penetrating trauma, language barrier, or state's custody. Enrollment occurred when research coordinators (RCs) were available, generally 12-16 hours/day. RCs approached EMS providers prior to departing the ED and ED providers after they completed their patient assessments. An ED provider survey was required for enrollment. Enrolled children were followed for 28 days to determine the presence of CSI (primary outcome) by subsequent imaging or by patient/family telephone follow-up for those without imaging.

RESULTS: Over 18 months, we prospectively enrolled 4,144 of 5,764 (71.9%) eligible children, including 74 of 110 (67.3%) children diagnosed with CSI. Enrollment during RC hours was 85.9%. Fifty-three enrolled children were withdrawn from the study. Of those in the final study cohort, 36.5% arrived by EMS scene response in spinal precautions. The remaining 63.5% arrived by EMS scene response without spinal precautions or by private vehicle or interfacility transfer. EMS scene response providers completed surveys for 60.2% of enrolled children arriving in spinal precautions. RCs missed the EMS providers for 37.1% of children; however, EMS declined participation for only 2.6%.

CONCLUSIONS: Our method of data collection demonstrates the ability to prospectively capture paired observations from EMS and ED personnel for children undergoing evaluation after blunt trauma. While this methodology will be used to implement and evaluate a CSI tool in future studies, it may also be adapted to studies requiring prospective data collection from EMS and ED personnel.