Agreement and validity of existing anthropometric calibration models to estimate whole body density in male professional football players

Introduction
To date no calibration models exist to estimate whole body density (Db) of professional footballers, as such the development of
practical anthropometric calibration models to make sound body composition judgements is imperative. The aim of this study was
to investigate the agreement and validity of estimating Db from 15 existing calibration models through comparison to a criterion
method of hydrostatic weighing (HW) in male professional footballers.
Materials and Methods
Data were gathered from a total of n=206 professional football participants (x±s; age=24.1±5.4-years, body mass=78.8±8.4 kg,
stature=180.1±7.0 cm, Db 1.075±0.01 g.ml-1). Bland and Altman limits of agreement (LoA) approaches were used to determine
bias and random variation derived from the calibration models.
Results
Bias and random errors for the published calibration models ranged from -0.005 to +0.015 g.ml-1 and 1.012 to 1.090 g.ml-1 respectively. A priori criterion (±3.8% p=<0.05 (g.ml-1)) was set as acceptable limits for the LoA method of which 13 calibration models
found that (on average) estimated Db derived from HW was greater than Db derived from the models.
Discussion
A rank order of LoA identified the best model to use, however, LoA were not narrow enough for measurements to be of practical
use and in most instances, selected models are not appropriate for estimating Db in male professional footballers