Test-Retest Reliability and Visual Perturbation Performance Costs During 2 Reactive Agility Tasks.

Context: High secondary injury rates after orthopedic surgeries have motivated concern toward the construct validity of return-to-sport test batteries, as it is evident that common strength and functional assessments fail to elicit pertinent behaviors like visual search and reactive decision making. This study aimed to establish the test-retest reliability of 2 reactive agility tasks and evaluate the impact of visual perturbation on physical performance.

Methods: Fourteen physically active individuals completed 2 agility tasks with reaction time (ie, 4 corner agility), working memory, and pathfinding (ie, color recall) components. Participants completed both tasks 4 times in 2 sessions scheduled 7 days apart. Outcomes included performance metrics of reaction time, time to target, number of targets, and total time assessed with reactive training timing gates. To assess test-retest reliability, we used intraclass correlation coefficients (ICCs), standard error of measurement (SEM), and minimal detectable change (MDC). Stroboscopic goggles induced visual perturbation during the fourth trial of each task. To assess the effect of visual perturbation, we used paired t tests and calculated performance costs.

Results: The 4-corner agility task demonstrated excellent reliability with respect to reaction time (ICC3,1 = .907, SEM = 0.13, MDC = 0.35 s); time to light (ICC3,1 = .935, SEM = 0.07, MDC = 0.18 s); and number of lights (ICC3,1 = .800, SEM = 0.24, MDC = 0.66 lights). The color recall task demonstrated good-to-excellent test-retest reliability for time to lights (ICC3,1 = .818-.953, SEM = 0.07-0.27, MDC = 0.19-0.74 s); test time (ICC3,1 = .969, SEM = 5.43, MDC = 15.04 s); and errors (ICC3,1 = .882, SEM = 0.19, MDC = 0.53 errors). Visual perturbation resulted in increased time to target (P = .022-.011), number of targets (P = .039), and total test time (P = .013) representing moderate magnitude degradation of performance (d = 0.55-0.87, performance costs = 5%-12%).

Conclusions: Both tasks demonstrated acceptable test-retest reliability. Performance degraded on both tasks with the presence of visual perturbation. These results suggest standardized reactive agility tasks are reliable and could be developed as components of dynamic RTS testing.