A Comparative Meta-Analysis of the Effects of Concussion on a Computerized Neurocognitive Test and Self-Reported Symptoms

Abstract

Context: Meta-analyses examining construct-specific cognitive impairment concurrently with self-reported symptoms postconcussion are sparse.

Objective: To review the literature on the effects of concussion on construct-specific neurocognitive declines and to compare them with self-reported symptoms before 1 week and between 1 and 3 weeks postconcussion.

Data sources: Relevant studies in PubMed, CINAHL, and PsycINFO published from January 1, 1999 through November 30, 2015.

Study selection: Studies were included if participants completed the Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) before and after concussion and if test performance and Postconcussion Symptom Scale (PCSS) scores were reported at both times.

Data extraction: After reviewing the full texts, we extracted data from 17 studies consisting of 29 independent samples; therefore, this meta-analysis consisted of 1777 unique participants.

Data synthesis: The Hedges g effect size (ES) was estimated. A random-effects or fixed-effects model was used based on heterogeneity findings. When heterogeneity was present, we used meta-regression to assess unexplained between-studies variance. Within the first week of injury, the ESs were small to moderate for cognitive declines, ranging from -0.43 (95% confidence interval [CI] = -0.52, -0.35) to -0.67 (95% CI = -0.77, -0.58), and large for the PCSS score (Hedges g = -0.81; 95% CI = -0.91, -0.71). After 1 week, the ESs for cognitive declines (Hedges g range = -0.25 [95% CI = -0.35, -0.15] to -0.37 [95% CI = -0.55, -0.19]) and PCSS score (Hedges g = -0.38; 95% CI = -0.53, -0.22) were also small. Within 2 weeks of injury, PCSS score and time since injury weakly moderated the cognitive ES.

Conclusions: When a neurocognitive test was administered within 1 week of injury, the ES was larger for self-reported symptoms than for ImPACT scores generated at the same session. After 1 week of injury, the ESs for ImPACT and PCSS scores were comparable. If the athlete reports symptoms within 1 week of injury, administering a cognitive test does not appear to offer additional information to the clinician. However, if the athlete does not report symptoms postconcussion, cognitive testing may inform the clinical management of the injury.

Keywords: ImPACT; effect size; mild traumatic brain injuries.

Figure.

Study-selection process.

Appendix 1.

Effect sizes for A, verbal memory, B, visual memory, C, processing speed, D, reaction time, and, E, Postconcussion Symptom Scale score within 1 week postconcussion. Continued on next page.

Appendix 1.

Effect sizes for A, verbal memory, B, visual memory, C, processing speed, D, reaction time, and, E, Postconcussion Symptom Scale score within 1 week postconcussion. Continued on next page.

Appendix 1.

Continued from previous page.

Appendix 2.

Effect sizes for A, verbal memory, B, visual memory, C, processing speed, D, reaction time, and, E, Postconcussion Symptom Scale score between 1 and 3 weeks postconcussion.