PRISMATA - PRedictors of mISsed trauMA Team Activations

Background: Traumatic injury is a leading cause of death among younger Canadians and drives millions of emergency visits annually. Structured Trauma Team Activation (TTA) protocols reduce morbidity and mortality, yet criteria and implementation vary nationwide, increasing risks of missed or delayed activations. Under-triage, when eligible patients do not receive timely TTA, poses significant risk for adverse events and challenges trauma systems. Our study aim was to examine demographic, physiologic, clinical, and triage-related predictors of under-triage within a large regional trauma system in British Columbia. Methods: We conducted a retrospective cohort study of anonymized patient data from the Island Health Trauma Registry, including all adults treated by trauma services from Jan 2023–Dec 2024. Records without TTA data and visits with 50% missing data were excluded. We compared accurately triaged and under-triaged patients, defined as a missed or delayed by 30-minute TTA, using descriptive statistics and identified predictors with multivariable logistic regression. Results: Of 2,308 trauma cases, 165 (7.1%) were under-triaged. No significant differences were observed between correctly triaged and under-triaged patients by sex, age, or mode of arrival. Differences were noted in mechanism of injury, number of injuries, triage documentation features, and vital signs. Under-triaged patients typically had a trauma-related CEDIS presenting complaint, longer triage narratives that included reference to explicit activation criteria, and physiologic indicators warranting higher-level activation (W = 3.10, p = 0.002). The logistic regression model showed modest explanatory power (R² = .138), high accuracy (93%), and very low sensitivity (1.9%), suggesting many false negatives. This likely suppressed the apparent influence of some predictors; improved case identification may strengthen future models. Vital-sign abnormalities were the strongest predictors of triage error. Systolic blood pressure < 90 mmHg (OR 10.46, 95% CI 5.01–21.84, p = .001), and abnormal respiratory rates, Rates < 10 or > 30, (OR 2.82, 95% CI 1.43–5.57, p = 0.003) were most strongly predictive of under triage . Exploratory analyses showed 63.5% and 58.0% higher rates of under-triage rates among assaults and pedestrian-struck injuries, respectively; while falls and vehicle collisions, were 11.5% and 30.9%, respectively, top be under triaged (H = 4.83, p = 0.028). Correctly triaged patients were 16.9% more likely to be admitted to trauma services (H = 4.03, p = 0.045). Overall, under-triage occurred in a small but important proportion of cases and was most strongly associated with physiologic derangement rather than demographic or structural factors. Implications and lessons learned: Regional trauma registries address an important gap in understanding and evaluating trauma services across Canada. In this study, demographic factors did not predict under-triage; instead, physiologic abnormalities, high-risk mechanisms, and triage-communication features were most influential. Acuity was often recognized without initiating activation, suggesting inconsistent application of criteria. Pedestrian-struck injuries were frequently missed. Findings highlight the value of routine triage data for TTA surveillance and automation, the need to strengthen adherence to TTA criteria, and opportunities to enhance decision-support and triage education. Implications: 1. Regional trauma registries address the gaps created without a national registry 2. Missed clinical indicators for TTA activation at triage suggest additional training is needed 3. Variability in activation for pedestrian struck and assault related presentations suggests additional research is needed.