Baseline Neurocognitive Testing May Be Misleading in Concussion Evaluation

High base rates of failure are consistently observed for baseline neurocognitive testing for the management of sports-related concussion, particularly in child athletes, according to the findings from a retrospective cross-sectional study published in JAMA Neurology.

Athletes between age 10 and 21 attending a clinical referral center were included in the analysis (n=7897). All participants had completed baseline neurocognitive testing for the management of concussion. The computerized Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) test was used at baseline. Investigators sought to determine base rates of failure on ImPACT validity indicators and to compare these rates within and across all age groups.

For the default ImPACT embedded validity indicator (EVI), the base rate of failure was 6.4% (n=505), whereas the ImPACT Red Flags, Higgins et al logistic regression equation, and Schatz and Glatts EVI were 31.8% (n=2509), 34.9% (n=2759), and 47.6% (n=3757), respectively. Patients who were age 10 demonstrated the highest cumulative base rate of failure (83.6%) vs 21-year-old participants who had the lowest (29.2%) (risk ratio 2.86; 95% CI, 2.60-3.16; P <.001).

According to the investigators, the lack of objective criterion measures as well as the high baseline rate of failure signals suggest a possible confounder in the model used for measurement. In addition, the researchers suggest a need for EVI recalibration to adjust for potential associations between examinee age and assessment context.

The findings indicate that greater differential age-specific approaches to concussion management are needed, and clinicians should consider “the age-specific baseline rate of failure when making return-to-play decisions based on postinjury evaluations.”

Reference

Abeare CA, Messa I, Zuccato BG, Merker B, Erdodi L. Prevalence of invalid performance on baseline testing for sport-related concussion by age and validity indicator [published online March 12, 2018]. JAMA Neurol. doi:10.1001/jamaneurol.2018.0031