Reliability and validity of a motion-based reaction time assessment using a mobile device

Abstract

Information processing speed is often altered following a concussion. Few portable assessments exist to evaluate simple reaction time (SRT) in hospitals and clinics. We evaluated the use of a SRT application for mobile device measurement. 27 healthy adults (age = 30.7 ± 11.5 years) completed SRT tests using a mobile device with Sway, an application for SRT testing. Participants completed computerized SRT tests using the Computerized Test of Information Processing (CTIP). Test-retest reliability was assessed using intraclass correlation coefficients (ICC) between Sway trials. Pearson correlations and Bland-Altman analyses were used to assess criterion validity between Sway and CTIP means. ICC comparisons between Sway tests were all statistically significant. ICCs ranged from 0.84-0.90, with <i>p</i>-values <.001. A one-way analysis of variance (ANOVA) revealed no significant differences between trials (<i>F</i>_3,104 = 1.35, <i>p</i> = .26. Pearson correlation between Sway and CTIP outcomes yielded a significant correlation (<i>r</i> = 0.59, <i>p</i> = .001). The mean difference between measurement methods was 43.7 ms, with limits of agreement between -140.8-53.4 ms. High ICC indicates Sway is a reliable method to assess SRT. A strong correlation and clinically acceptable agreement between Sway and the computer-based test indicates that Sway is suited for rapid administration of SRT testing in healthy individuals. Future research using Sway to assess altered information processing in a population of individuals after concussion is warranted.