Diagnostic and Predictive Ability of Reaction Time Assessments for Adolescent Concussion

Abstract

Background:Concussed individuals demonstrate significantly slower reaction time than that of healthy controls, particularly in the acute post-injury timeframe. In addition, post-injury reaction time may predict concussion recovery timing. Given that multiple measurement methods exist to assess reaction time post-injury, clinicians treating concussion may desire more information regarding the clinical utility of these methods.Purpose:To determine the clinical utility of two reaction time assessments and identify the ability of each to predict persistent post-concussion symptoms (PPCS), defined as symptoms lasting >28 days post-concussion.Methods:Concussed participants, within 14 days of injury, and healthy controls aged 12 to 18 years completed measures of clinical and smartphone-recorded reaction time. For clinical reaction time testing, participants completed 8 trials of a drop-stick protocol in which participants caught a weighted measuring stick dropped from a standard height between the thumb and palm of the dominant hand. Reaction time was calculated as the distance the stick fell before being caught. Three clinical reaction time outcomes were calculated: mean time, fastest trial, and coefficient of variance among all 8 trials.Smartphone reaction time was assessed through a smartphone application (Reaction Test Pro), which measured the speed at which patients responded to a simple on-screen stimulus. Participants were instructed to tap the phone as quickly as possible when an on-screen button changed color for 30 trials, an approximate test time of 2-3 minutes. Smartphone reaction time was output automatically by the application as the mean of all completed trials.Results:Concussion participants (n=54; 15.1±1.68 years of age; 50% female) assessed 7.0±3.05 days post-injury demonstrated worse performance than healthy controls (n=17; 16.3±1.13 years of age, 47% female) on all measures of reaction time (Table 1). Smartphone-recorded reaction time provided the largest mean difference between groups to distinguish concussed vs control participants. After controlling for age and concussion history, measures of mean clinical and smartphone reaction time were significant predictors of developing PPCS (Table 2).Conclusion:Smartphone-recorded reaction time within 2 weeks of concussion provide the greatest mean difference between concussed and control participants and predicted PPCS development in adolescents. Given the simplicity of smartphone-based measurements in comparison to other options for assessing reaction time, as well as the ubiquity and accessibility of smartphone technology, this technique may be preferred by physicians looking for direct measures of post-concussion reaction time. These assessments may also be used for early identification of patients at increased risk of developing PPCS.Table 1.Between-groups differences for reaction time and AUC values. Data represented as mean ± SD, or median [IQR].Table 2.Logistical regression predicting persistent post-concussion symptoms, controlling for age and concussion history.