On-scene predictors of injury severity, deterioration, and death following road injury: a systematic scoping review of information available to responders and its use in triage tools

Background: Road injury including motor vehicle collisions (MVCs) remain a leading cause of death and disability worldwide. Early recognition of injury severity at the scene is critical to optimise emergency response, yet existing triage systems under- and over-triage substantial numbers of patients. This scoping review aimed to identify physiological, mechanistic, and contextual indicators available at the scene that predict injury severity, deterioration, or death, and to map these indicators across clinical responders, professional bystanders, and lay responders. Methods: We conducted a systematic scoping review in accordance with PRISMA-ScR guidelines. Nine electronic databases and the Cochrane Library were searched from inception to 11-Jul-24, supplemented by reference chaining and grey literature searches. Eligible studies included observational cohorts, trials, modelling studies, and validation papers involving road injured patients, reporting on-scene variables and subsequent injury severity, deterioration, or mortality outcomes. Data were thematically synthesised into physiological, mechanistic, and contextual domains. Results: A total of 14,089 articles were identified. Following screening, 402 articles were included for full-text review with an additional 9 identified via hand and reference list searches. 64 papers met eligibility criteria and were included. Physiological indicators such as hypotension, abnormal Glasgow Coma Scale (GCS) scores, and Shock Index derivatives remain critical predictors, but require equipment often unavailable to non-clinical responders. Mechanistic indicators, including intrusion depth and vehicle rollover, were valuable adjuncts but inconsistently defined. Contextual and patient factors, including age, sex and comorbidity, were recognised as important but inconsistently captured. Standardisation of on-scene data collection, simplification of triage tools for bystanders, and novel technologies offer opportunities for improving early recognition. Conclusion: Broader recognition, standardisation of nomenclature and recording of physiological, mechanistic, and contextual indicators is essential to strengthen early post-collision care. Enabling all responder groups to identify critical injuries through structured information capture will help strengthen the early links in the Road Injury Chain of Survival and improve outcomes for MVC patients.