Diagnostic agreement does not establish triage accuracy in paediatric HEMS trauma

Dear Editor,

We read with interest the recent study by Roehl and colleagues examining diagnostic accuracy and outcomes among children with trauma attended by helicopter emergency medical services (HEMS) [1]. The authors report substantial agreement between prehospital and in-hospital diagnoses, with an overall Cohen’s κ of 0.64, and conclude that their findings support the clinical value of HEMS physicians in correctly triaging paediatric trauma patients. We believe that the distinction between diagnostic agreement and triage accuracy deserves closer consideration.

Cohen’s κ quantifies chance-corrected agreement between classifications, but it does not indicate whether clinically important diagnoses are being missed. This distinction is especially relevant in paediatric trauma, where diagnostic categories vary considerably in prevalence and clinical consequence. Agreement on the absence of a relatively uncommon injury contributes to the overall κ alongside failure to recognise a potentially important multisystem injury, although these two types of agreement or disagreement are not clinically equivalent.

This issue is reflected in the study’s own findings. Despite the overall κ being interpreted as substantial, polytrauma was underdiagnosed in the prehospital setting, whereas traumatic brain injury and extremity and pelvic injuries tended to be overdiagnosed [1]. The direction of these errors may be more informative for clinical decision-making than the aggregate agreement statistic itself. In prehospital trauma care, failure to recognise multisystem or time-critical injury may influence destination choice, trauma-team activation and the urgency of definitive care.

Diagnosis-specific measures would therefore strengthen interpretation of these data. Sensitivity, specificity, positive and negative predictive values, or at least positive and negative agreement for the major diagnostic categories would allow readers to determine whether the overall κ is accompanied by an acceptable false-negative rate. Polytrauma is particularly relevant because it was the category specifically identified as being underdiagnosed.

A further point concerns the interpretation of diagnostic concordance as evidence of correct triage. Agreement between a prehospital diagnostic impression and the subsequent hospital diagnosis does not directly measure whether the patient was appropriately triaged. Demonstrating triage accuracy would require an outcome linked to the triage decision itself, such as appropriateness of destination, trauma-team activation, requirement for immediate specialist intervention, or clinically important undertriage and overtriage.

Roehl and colleagues provide useful evidence that prehospital and hospital diagnostic impressions frequently correspond in paediatric HEMS trauma. However, we suggest that their results most directly demonstrate diagnostic agreement, rather than triage accuracy. Reporting diagnosis-specific false-negative performance, particularly for polytrauma and other time-critical injuries, would considerably strengthen the clinical interpretation of this important study.

Author contributions

Ercan Yılmaz and Zeynep Yamancan Yılmaz conceived the correspondence and critically evaluated the methodological and clinical interpretation of the original study. Ercan Yılmaz drafted the manuscript. Zeynep Yamancan Yılmaz critically revised the manuscript for important intellectual content. Both authors read and approved the final manuscript.

Funding

No funding was received for this work.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Clinical trial number

Not applicable.

Authors’ information

Not applicable.

Competing interests

The authors declare no competing interests.