International Journal of Body Composition Research vol:7 issue:4 pages:123-129
The purpose of this study was to further document the extent of the effect of wearing deviating clothing
schemes on the accuracy of body composition measurement using air-displacement plethysmography
(ADP) compared to the clothing prescribed by the manufacturer. Thirty-one men (21.9±1.7 yrs) and 31
women (20.7±1.8 yrs) were measured in the reference (REF) condition, wearing an extra t-shirt (TS), wearing
an extra t-shirt + long track-suit trousers (TT), and again in the REF condition respectively. Technical
error of test-retest under REF conditions was excellent for percentage body fat (0.57%) (PF). Both deviating
conditions resulted in significantly lower mean PF estimations (P<0.05) compared to the REF condition in
men (REF:11.8±6.9%; TS:7.7±7.3%; TT: 0.1±8.2%) and women (REF:25.4±6.0%; TS:22.5±6.0%; TT:
15.2±6.7%). Variance tended to increase as more clothing was worn, and a significant trend in bias was
found for body density in the TT condition with underestimations increasing with increasing body density.
Together with the existing literature the present study clearly demonstrates that mean bias increases as the
amount of non-standard clothing worn in the BOD-POD increases, together with measurement variation.
The observed mean bias, increased variance and significant trend in bias as body density increases, result
in faulty results that are of little if any practical use to the health professional and his client when assessing
body composition status. It is therefore not recommended to use ADP to assess body composition if for
any reason clients are unable or unwilling to adhere to the prescribed clothing conditions.