Inter & Intra Tester Reliability of Motion Capture of Six Sports Task used for Return-to-Sport Testing

Abstract

Background:Motion capture has become increasingly recognized as an important evaluation tool for guiding rehabilitation planning and in determining when it is safe to return to sport. To ensure motion capture utility in sports evaluations, it is important to establish training protocols across test centers.Hypothesis/Purpose:The purpose of this study was to evaluate the inter and intra tester reliability of return-to-sport testing using motion capture at three different motion analysis centers.Methods:A single tester and athlete traveled to three different motion analysis centers. On the first day, the first tester demonstrated marker placement and the execution of the six sports related tasks. The second tester completed the same protocol the same afternoon. On the following day, both testers repeated the process on the same athlete. The six tasks included the drop vertical jump (DVJ), deceleration, heel touch, lateral shuffle, 45° cut, and single leg hop. Kinematic data included trunk flexion/lean, pelvic obliquity, hip flexion/rotation/adduction, and knee valgus/flexion. Additionally, kinetics were analyzed for knee valgus moments. For each testing day, the mean, median, standard deviation, and range were calculated. To compare between days, differences between means and paired t-tests were calculated (α=0.05).Results:Looking at inter tester reliability kinematics and kinetics, of the 48 calculated mean differences only two variables were statistically significant – DVJ moment (mean difference 0.22 Nm/kg, p=0.008) and lateral shuffle trunk flexion (mean difference 4.8 degrees, p=0.004). Looking at only the kinematic data, 95% (41/43) of the variables had mean differences that were four degrees or less. For intra tester reliability, of the 48 calculated mean differences three variables were statistically significant – DVJ knee flexion (mean difference 4.65 degrees, p=.041), DVJ trunk flexion (mean difference -3.7 degrees, p=0.03) and deceleration hip flexion (mean difference -5.71 degrees, p=0.034). Looking at only the kinematic data, all mean differences were less than 6 degrees, and 86% (37/43) were less than four degrees.Conclusion:This study demonstrates that after one training session, motion capture sports testing was remarkably consistent within and between testers at three different motion analysis centers. This strengthens the case that sports evaluations using motion capture is reproducible and can be replicated at novel centers with trained motion analysis personnel. To confirm inter and intra tester reliability, future studies should expand to more centers, additional testers, and athletes.Table 1.Mean difference within same tester (intra tester reliability) and between testeres (inter tester reliability)