Investigation of validity, reliability, and practice effects of the Immediate Postconcussion Assessment and Cognitive Test (ImPACT) and traditional paper-pencil neuropsychological tests Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Kontos, Daniel L.
    • Affiliation: College of Arts and Sciences, Department of Exercise and Sport Science
  • The purpose of this study was to determine: 1) if an athlete’s age significantly affects neuropsychological test performance, 2) if an athlete’s performance remains consistent across serial neuropsychological tests, and 3) the concurrent validity of the Immediate Postconcussion and Cognitive Test (ImPACT) scores when compared to traditional paper-pencil test scores of similar cognitive domains. A healthy sample of 20 college and 20 high school athletes completed both ImPACT and traditional paper-pencil neuropsychological test batteries on three separate occasions. Means and standard deviations, 2x3 mixed model ANOVAs (age x session), reliability (ICC2,1) and precision (SEM) values, and linear regressions were calculated on outcome measures for both test batteries. The ANOVAs revealed significant main effects of age for the Trail Making Test Form B (TMT-B) total time and ImPACT processing speed composite score with college athletes performing better than high school students on both measures. The ANOVAs also revealed significant main effects of session for the Brief Visuospatial Memory Test – Revised (BVMT-R) total recalled (immediate and delayed), the TMT-B total time, Stroop Test total score, and ImPACT processing speed composite score. Reliability measures ranged from 0.12 to 0.72 with the majority of the outcome measures achieving a moderate level of reliability across testing iii sessions. Linear regressions revealed that ImPACT test scores had low levels of shared variance with select paper-pencil neuropsychological tests. Coefficients of determination for these linear regressions left much of the variance unexplained (52-88%). Only the ImPACT Three Letters average counted correctly reached a moderate level (R2=0.481). This study demonstrates the need of the clinician to understand the differences in neuropsychological test performance for athletes of different age groups and across serial neuropsychological tests. It is also recommended that caution be exhibited when evaluating ImPACT test results of athletes 15-17 and 19-21 as the concurrent validity has not been conclusively proven.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Guskiewicz, Kevin M.
  • Open access

This work has no parents.