This is an interesting study that explored the body composition of various Australian athletes. In sport, optimizing body composition is important due to factors such as: weight class restrictions, efficiency/economy of movement, energy expenditure, force production and range of motion to name a few. In many cases body composition is estimated using equations as this is typically the most practical although not the most accurate. There are also inherited flaws in which equations can only be used with certain populations (think sample that it was validated with). This study came about as no equation had yet to be developed on a large number of athletes in a variety of sports. The goal of the study was to:
Provide descriptive data (which could be used as normative data) on the percentage of body fat in male athletes in a variety of sports.
Develop a population specific equation for prediction body density from anthropometric variables
Cross validate with existing equations.
Method – How did the authors conduct the test:
207 male members participated in this study, all of whom were a part of the South Australian senior representative squads. The range of sports in which athletes competed was vast; and comprised 25 different sports.
Anthropometric markers were taken in the form of body mass, skinfolds, diameters and circumferences.
Various statistical tests were run: one way ANOVA, multiple regression analysis for the development of equations as well cross validation of equations to other literature.
Results – What did the authors find:
For the descriptive data found, the overall mean body fat % was 10.0. For the squash group in particular the results are as follows:
Participants: 9
Age (yrs): 22.6 ± 6.8
Height (cm): 177.5 ± 4.1
Body Mass (kg): 71.90 ± 8.31
Body Density (g.cm-3): 1.07333 ± 0.0086
Body Fat (%): 11.2 ± 3.7
Fat Mass (kg): 8.13 ± 2.89
Fat Free Mass (kg): 63.77 ± 7.39
As it pertained to the regression equations, the best overall equation yielded an R of 0.79 which could be considered strong. The standard error of the estimate was 0.00537g.cm-3 for the body density calculation which was equivalent to 2.3% body fat when transformed.
When cross validating other equations, the authors found significant differences as it pertained to the accuracy of the calculations (higher errors of estimate) in 8 of 11 equations that we compared and concluded that those models were inaccurate for the population in the study. This in large was due to the equations being developed on untrained persons, thus under predicting body density and over predicting body fat.
In short, the way in which the authors calculated body fat % tended to be more accurate in athletic populations than previous equations. One must remember that these calculations are not entirely accurate as they have inherent flaws based on the populations in which they were validated in. If one were using skinfolds to measure body composition, it may be best to sum skinfolds rather than transform into a body fat percentage. If one is after the most accurate measure of body composition, then it would be best to use a DEXA scanner.
If you enjoyed this blog and you would like to stay updated on all things squash, make sure to follow Global Squash Coach on your preferred social media medium.
Facebook - Like, Comment and Share with friends, families and associates.
Global Squash Group - join the Facebook group to interact with the Global Squash Coach community.
Squash Blogs - join the Facebook group and get access to a worldwide community of Squash bloggers.
Instagram - Follow and Re-post!
YouTube - Like and Subscribe!
Twitter - Follow, Like and Re-Tweet!
LinkedIn - Follow and Like
Yours Truly,
Dominic Benacquista - Global Squash Coach
Ps. I have created a Question and Answer section on the site. I ask that you submit any questions relating to squash performance or this blog there.
Reference:
Withers, R. T., Craig, N. P., Bourdon, P. C., & Norton, K. I. (1987). Relative body fat and anthropometric prediction of body density of male athletes. European Journal of Applied Physiology and Occupational Physiology, 56(2), 191-200.