Developing a New Curvilinear Allometric Model to Improve the Fit and Validity of the 20-m Shuttle Run Test as a Predictor of Cardiorespiratory Fitness in Adults and Youth.

BACKGROUND AND OBJECTIVES: Doubts have been raised concerning the validity of the 20-m shuttle-run test (20 mSRT) as a predictor of cardiorespiratory fitness (CRF) in youth based on Léger's equation/model. An alternative allometric model has been published recently that is thought to provide, not only a superior fit (criterion validity) but also a more biologically and physiologically interpretable model (construct validity). The purposes of this study were to explore whether allometry can provide a more valid predictor of CRF using 20 mSRT compared with Léger's equation/model.

METHODS: We fitted and compared Léger's original model and an alternative allometric model using two cross-sectional datasets (youth, n = 306; adult n = 105) that contained measurements of CRF ([Formula: see text]/[Formula: see text]) and 20 mSRT performance. Quality-of-fit was assessed using explained variance (R2 ) and Bland and Altman's limits of agreement.

RESULTS: The allometric models provided superior fits for the youth (explained variance R2  = 71.9%) and adult (R2  = 77.7%) datasets compared with Léger's equation using their original fixed (R2  = 35.2%) or re-estimated parameter models (R2  = 65.9%), confirming that the allometric models demonstrate acceptable criterion validity. However, the allometric models also identified a non-linear "J-shaped" increase in energy cost ([Formula: see text]/[Formula: see text]) with faster final shuttle-run speeds (fitted speed exponent = 1.52; 95% CI 1.38-1.65).

CONCLUSION: Not only do allometric models provide more accurate predictions of CRF ([Formula: see text]/[Formula: see text]; ml kg-1  min-1 ) for both youth and adults (evidence of criterion validity), the "J-shaped" rise in energy demand with increasing final shuttle-run speed also provides the evidence of construct validity, resulting in a more plausible, physiologically sound, and interpretable model.