Variability and accuracy of body fat estimates using skinfolds and DXA in elite Gaelic football players: a validation study and development of novel prediction equations

In elite Gaelic football, body composition is commonly assessed by methods, such as skinfold equations and dual-energy X-ray absorptiometry (DXA) with estimates of body fat used to evaluate and monitor nutrition and training programmes. The aim of this study was to assess the accuracy and variability of the existing skinfold equations in predicting body fat % (BF%) and fat mass (FM) as compared to DXA in elite Gaelic football players and to derive novel sport-specific skinfold equations. Body composition in thirty-two elite outfield Gaelic football players (23.4 ± 3.3 years; 83.6 ± 7.2 kg; 183.5 ± 6.4 cm) was assessed via skinfold thickness at 7 sites and DXA. A standardised testing protocol was followed for both methods of assessment. Male-specific prediction equations commonly used in Gaelic football research to estimate BF% (Durnin and Womersley, Evans, Lohman, Reilly and Withers) and FM (Stewart) were selected. Using the skinfold data collected, Gaelic football prediction equations were developed relative to DXA. All the existing skinfold equations underestimated BF% (8.9–11.8%) and FM (6.8 kg) by varying magnitudes relative to DXA (15.2%; 11.8 kg). Gaelic football prediction equations for BF% and FM were designed and accounted for 69% and 76% variance in DXA derived BF% and FM, respectively. The existing skinfold equations demonstrated large variation in estimates of BF% and FM in elite Gaelic football players as compared to DXA, causing the potential for misleading comparisons and inappropriate intervention. The Gaelic football equations offer sport-specific estimates of BF% and FM and greater consistency as compared to the selected equations when evaluating nutrition and training strategies.