Development of a Concise Lower Extremity Physical Performance Test Set for Return to Sport Decision-Making in Pediatric Populations

Abstract

Objectives: Physical performance tests (PPT's) are used to assess lower extremity function in pediatric populations. Common PPT's include the lower quarter Y-balance, stork balance, stork balance on BOSU, single leg squat (SLS), SLS on BOSU, clockwise and counterclockwise quadrant single leg hop (SLH), forward SLH, timed SLH, and triple crossover SLH. Each of these tests assesses distinct performance characteristics, but administration of the full 10-item test battery is not practical. The aims of this analysis were to 1) define the primary underlying components of physical performance assessed by these 10 PPT's, and 2) derive a reduced item set of PPTs that eiciently and accurately measures performance on each underlying component. Methods: This study included healthy, uninjured volunteers (n=61) between the ages 6 and 17 [mean age = 10.7 ±3.2 years; 33 females (54.1%)]. Aer a brief warm-up, subjects performed the 10 PPT's in the same order (listed above), however we randomly assigned the starting test to avoid practice/fatigue eects. Subjects completed 2 trials on each leg for each test with the exception of the SLH tests, which were performed 3 times. We developed a composite score for each test by averaging trials across sides. Prior to item reduction, all Y-balance, and hop tests were normalized to leg length. Item reduction was performed using principal components analysis (PCA) with oblique rotation (Promax) on all 10 tests. We used the Kaiser criterion (eigenvalue > 1) to determine the optimal number of components. Items with loadings > 0.55 were considered for the reduced test item set. Cross-loaded items with < 0.25 absolute dierence in loading between components were dropped. If two or more factors loading on the same component were highly correlated (r > 0.7), we dropped the item(s) with the lowest factor loading. Results: The PCA identified 2 components. Component 1 (neuromuscular control/balance) included all balance, single leg squat and quadrant hop test items. The 2 quadrant hop tests were highly correlated (r=0.94), had the second and third lowest factor loadings on the component (<0.78), and were dropped. The SLS test had the lowest factor loading (0.69) and was also dropped. The Stork, Stork BOSU, and SLS BOSU had similar factor loadings (0.79, 0.90, and 0.88, respectively) and did not meet the correlation criterion for removal (all < 0.61). Component 2 (Power) included the forward SLH and Crossover SLH. Each demonstrated high factor loadings (0.94 and 0.79, respectively) and only moderate correlation (r=0.56). The Timed SLH and lower quarter Y-balance did not meet the loading criterion and were dropped. The final 5-item PPT set had a cumulative response variance of 76.0%. The internal consistency (Cronbach α) of the 3-item Component 1 (0.80), 2-item Component 2 (0.72) and overall 5-item set (0.70) were all satisfactory (i.e., > 0.7). Conclusion: The 10 PPT's measure 2 primary components of lower extremity performance: neuromuscular control/balance and power. These components are most appropriately assessed with the stork balance, stork BOSU, SLS BOSU, forward SLH and triple crossover SLH tests. Of these, the Stork BOSU and forward SLH are most capable of