Skeletal age prediction models by maturity status in male soccer players.

This study focus is to develop a new model to estimate skeletal age (SA) as a function of the state of biological maturation in male soccer players, and to propose cut-off points to classify the state of biological maturation based on the percentage of adult height (PAH). SA was determined in 747 Portuguese male soccer players, using the Tanner-Whitehouse (TW) 3 method, and PAH was predicted by TW3 (P-TW3) and Khamis-Roche (P-KR) methods. Subsequently, the sensitivity and specificity of the P-TW3 were estimated to classify late, on-time and early maturers to obtain cut-off points, by age; and to develop specific equations for each maturation stage. Both the model using P-TW3 and the model using P-KR showed a SA predictive capacity of 93%. The average differences were similar to zero. P-TW3 cutoff points were established by ROC curve analysis to identify late and early maturers according to their SA. Following, predictive models were developed to estimate SA according to maturity status. The predictive capacity of the models was 87.3% in late maturers, 92.3% in on-time maturers and 93.5% in early maturers. The prediction models are a reliable and cost-effective method to estimate SA in male soccer players.