Abstract
Objective: The King-Devick (KD) is an objective rapid number naming test used to diagnose concussion. The purpose of this study was to assess test-retest reliability, sensitivity/specificity and AUC of the King-Devick test (K-D).
Methods: Athletes completed baseline testing including the K-D. When an athlete presented acutely with a suspected concussion (sideline/within 24 hours) from February 2020 to April 2022, a battery of tests were performed. If a concussion was clinically diagnosed, a control was identified to perform testing, matched on comorbid conditions (attention deficit disorder, learning disability, headache disorder, mood disorder), gender, team, season, and baseline scores.
Results: Baseline and post-injury data were collected on 89 concussed athletes and 89 matched controls. Athletes were predominantly male (n=112, 55%), with the largest sport distributions amongst football (n=78, 39%), volleyball (n=24, 12%), softball (n=22, 11%), and baseball (n=11, 6%). The ICC was 0.70 (95% CI: 0.56–0.79; average of 281.9 days between assessments); sensitivity/specificity was 79%/60% (any increase), 67%/83% (3-second increase), 58%/90% (5-second increase); AUC was 0.84 (0.71–0.89).
Conclusion: The K-D has fair test-retest reliability and good AUC. Any increase in time is sensitive for a concussion but not specific. A greater than 5-second increase in time is specific for concussion although not sensitive. Understanding performance characteristics of concussion tests is important in the diagnosis of concussion.
Summary Points:
- The K-D has test-retest reliability and good AUC.
- Any increase in time is sensitive for a concussion but not specific.
- A greater than 5-second increase in time is specific for concussion although not sensitive.
- Understanding performance characteristics of concussion tests is important in the diagnosis of concussion.