Reliability of body composition measurement techniques is essential to the accurate reporting of intervention outcomes. However, the between-day precision error of commonly used techniques, as well as the reference multi-compartment model, in a population-representative sample are currently unknown. To quantify technical and biological precision error of body composition techniques in comparison to the referent 4-compartment (4C) model. Men and women (1:1 ratio; 18-85 years old; n=90) completed 2 consecutive-day body composition testing sessions, including individual components of the referent 4C model. Testing was undertaken in accordance with best practice guidance for each technique, including standardized presentation and a consistent time of day. Repeat measurements were conducted on day 1 for technical precision, and between-day measurements were conducted for biological precision quantification. On average, all measurements met acceptable error limits and presented typically low technical and biological error [<2% fat-free mass (FFM) and<3% fat mass (FM) precision error]. For technical precision of FFM, all techniques met a priori cut points (80%; CV=0.45-0.81%). For FM, all techniques were equivalent to the best-rating method on average (CV=0.78-1.35%), except air displacement plethysmography (CV=2.13%). For biological precision, only 3-compartment (3C) and 4C equations sufficiently met the a priori determined cut point for estimates for FFM (CV=0.77-0.79%), and only DXA met the 80% cut point (CV=1.17%) for FM. The primary purpose of a study design is imperative when deciding on body composition assessment techniques used for longitudinal measurements. If reliable longitudinal assessments of FFM are central, a 3C or 4C model may be indicated. If FM is a primary outcome, DXA may be preferable. However, considering the low error rates presented within the current study across a broad age span of healthy adults with implementation of best-practice guidelines, any technique assessed here may be used, provided that strict protocols are adhered to.
Read full abstract