SCAT5 baseline values, test-retest reliability, and reliable change metrics in high school athletes

Abstract

OBJECTIVE In the United States, more than 1 million sport-related concussions afflict children annually, with many cases undetected or unreported. The Sport Concussion Assessment Tool (SCAT) is widely used to detect concussions in high school, collegiate, and professional sports. The objective of this study was to establish baseline values for the SCAT version 5 (SCAT5) in high school athletes. METHODS Baseline SCAT5 evaluations were conducted in students (ages 14–19 years) from 19 high schools in central Illinois who were participating in various school-sponsored sports. The SCAT5 evaluations were retrospectively extracted from the electronic medical record system for analysis. Statistical analyses included the Wilcoxon rank-sum test for continuous variables and the chi-square test for categorical variables, considering significance at p < 0.05. Test-retest reliability at < 6 months, 10–14 months, and 16–20 months was computed using intraclass correlation and Spearman’s rho (ρ). Reliable change indices are provided using the Iverson formula. RESULTS A total of 2833 unique athletes were included, and the average age was 15.5 ± 1.14 (SD) years. There were 721 female (25.5%) and 2112 male (74.5%) athletes. Students ≥ 15 years old had more prior concussions (p < 0.001), and male athletes were more frequently hospitalized for head injury (p = 0.013). Female athletes exhibited a significantly higher prevalence of mood disorders (14.7% vs 4.6%, p < 0.001), whereas attention-deficit/hyperactivity disorder was more common in male athletes (5.2% vs 13.2%, p < 0.001). Symptom number and severity were significantly greater in female athletes (3.17 ± 4.39 vs 2.08 ± 3.49, p < 0.001; 5.47 ± 9.21 vs 3.52 ± 7.26, p < 0.001, respectively), with mood-related symptoms representing the largest differences. Female athletes and students ≥ 15 years old performed better on most cognitive assessments. Female athletes and students < 15 years old performed better on the modified Balance Error Scoring System (p < 0.001). Test-retest reliability was poor to moderate for most assessment components. Reliable change index cutoff values differed slightly by sex, with female athletes often having a greater cutoff value. CONCLUSIONS This study underscores the variability of SCAT5 baseline values influenced by age, sex, and medical history among adolescent athletes. It provides a robust dataset, delineating baseline values stratified by sex and age within this demographic. Additionally, the results provide enhanced guidance to clinicians for interpretation of change and reliability of baselines.