European Journal of Nutrition & Food Safety

Background: Analysis of athlete body composition often forms part of team/squad selection and inaccurate measurements could have huge implications for teams and individuals. Currently there is no standard protocol for assessing athlete body composition with regard to the timing of measurements in relation to training sessions, yet assessment days are often based around training sessions. The appropriateness of BodPod™ for assessing athletic populations has previously been identified as an area for further investigation Fields et al. [1], though it is considered a valid tool for assessing female athlete body composition Ballard et al. [2]. There is very little published research into the influence of exercise on the validity of skinfold caliper (SFC). The aim of this study was to investigate the effects of exercise on the accuracy and reliability of BodPod™ and SFC in assessing body composition of athletes and the inter-rater variability in SFC measurements, to inform the development of a standard protocol. 
Methods: The study was approved by the University of Nottingham (UoN) Medical School Ethics Committee. Members of University sports clubs were invited to participate in the study through the distribution of posters. Potential participants were screened against inclusion and exclusion criteria. The final sample comprised two males (triathlon and judo) and four females (rowing). Participants attended three test days at UoN, at which body fat percentage was assessed using BodPod™ and 4-site skinfold measurements - bicep, tricep, subscapular, suprailiac (all skinfold measurements were taken by two investigators), before and after a 45-minute cardiovascular exercise session. Intensity of the exercise was between 3-5 on the Borg RPE scale 1982 and activities included running, cycling, rowing and cross-training (to match the participants’ typical training sessions). Averages of the data collected from the three sessions were used for analysis. Data was compared using t-tests to identify any differences between methods, between measurements taken before and after exercise, and inter-rater differences. Only P values <0.05 were considered significant. Pearson’s correlation coefficients were used to assess the correlation between the two methods. Only P values < 0.01 were considered significant. A Bland-Altman plot was used to assess agreement between the two methods.
Results: BodPod™ and skinfold measurement were not significantly different before and after the exercise. The two assessment methods were significantly correlated before (r=0.945, P<0.01) and after (r=0.945, P<0.01) exercise. The Bland-Altman plot indicated that the difference between assessment methods (measurement bias) was 0.9%. The limits of agreement (average difference ±1.96 SD of the difference) were -6.9% and 5.1%. Significant inter-rater differences in skinfold measurements were seen before (t=2.922, P<0.05) and after (t=2.810, P<0.05) exercise.
Discussion: The results suggest that the accuracy and reliability of both BodPod™ and SFC are unaffected by cardiovascular exercise. These results are slightly unexpected, as the influence of body temperature and moisture has previously been demonstrated to impact upon the accuracy of the BodPod™ [3]. Whenever possible the same assessor should measure an athlete over a period of time to minimise the risks of inter-rater differences. The small sample size and short time-frame greatly limit the ability to draw firm conclusions. Further research is required, and this should include different types of exercise, subjects from a wider variety of sports and of differing body composition. In particular, research is needed into the development of SFC equations for female athletes.
Conclusion: The data from this pilot study indicate that assessment days involving the use of BodPod™ or SFC can be set around cardiovascular training sessions without compromising the accuracy of the measurements.