Generalized Equations for Predicting Percent Body Fat from Anthropometric Measures Using a Criterion Five-Compartment...

Area of Science:

• Human physiology
• Body composition analysis
• Biostatistics

Background:

• Anthropometric equations are widely used for estimating body fat percentage (%BF) due to the impracticality of laboratory methods.
• Existing equations, often based on two-compartment models, are susceptible to errors stemming from assumptions about fat-free mass composition.

Purpose of the Study:

• To develop and validate a novel anthropometric equation for predicting %BF.
• To utilize a five-compartment (5C) model as the criterion measure for %BF estimation.

Main Methods:

• A cohort of healthy adults (n=357) underwent hydrostatic weighing, DEXA, and bioimpedance spectroscopy to establish a 5C %BF (%BF5C) criterion.
• Anthropometric measures including skinfolds and circumferences were collected.
• A new equation was developed using hierarchical variable selection and cross-validated on a separate subset of participants.

Main Results:

• The new equation demonstrated strong predictive power, explaining 77.5% of the variance in %BF5C (R² = 0.775, SEE = 4.0%).
• Key predictors included the sum of specific skinfolds (SSnew) and waist circumference.
• Cross-validation confirmed excellent agreement with %BF5C (ICC = 0.90, mean bias ± limits of agreement = 0.4% ± 8.6%, SEE = 2.5%).

Conclusions:

• The developed anthropometric equation offers improved accuracy and agreement for %BF estimation compared to existing methods.
• This new equation provides a practical and reliable tool for estimating %BF5C in healthy adults when laboratory assessments are not feasible.