Written by Joshua Belfer
Spoon Feed
No single symptom or exam finding definitively diagnoses pediatric concussion, but mental fog, light/noise sensitivity, nausea, and oculomotor abnormalities meaningfully increase likelihood, while absence of headache decreases it.
The findings that matter (and the ones that don’t)
Pediatric concussion is common and often challenging to diagnose clinically, especially in the ED, where imaging is usually normal and symptoms are subjective. This systematic review and meta-analysis of 23 observational studies, part of JAMA’s “The Rational Clinical Examination” series, evaluated the diagnostic accuracy of history and physical exam findings in children with suspected concussion.
Symptoms most strongly associated with concussion included mental fog (LR ~12), noise sensitivity (LR 6.9), nausea (LR 6.7), and light sensitivity (LR 6.4), all with high specificity. Oculomotor findings—abnormal near-point convergence (LR 7.0), smooth pursuits (LR 6.5), and saccades (LR 4.8)—were also helpful but insensitive (<0.40). Headache was common (LR 3.1), but notably, absence of headache reduced likelihood (LR 0.20). Importantly, no single finding had sufficient sensitivity or specificity to definitively rule in or out concussion.
Limitations include that all included studies were case-control (risk of overestimating diagnostic accuracy), most were sport-related injuries, and younger children were underrepresented.
How will this change my practice?
This paper reinforces something we already feel at the bedside but sharpens how we apply it. I’m not hanging my hat on any single symptom, but I am going to weigh certain findings more heavily. If a kid has mental fog, light sensitivity, or nausea, I’m much more confident calling this a concussion. On exam, I’m going to be more intentional about quick oculomotor testing—near-point convergence and saccades are high-yield and take seconds. On the flip side, if a child truly has no headache, that meaningfully lowers my suspicion (though doesn’t rule it out).
Bottom line: Concussion remains a clinical diagnosis built from multiple imperfect pieces, but now I have a clearer sense of which pieces matter most and how much they should shift my thinking.
Source
Does This Child Have a Concussion?: The Rational Clinical Examination Systematic Review. JAMA. 2026 Apr 6. doi: 10.1001/jama.2026.1233. Epub ahead of print. PMID: 41941197.
