Body composition in elderly people: effect of criterion estimates on predictive equations
- Wright State Univ. School of Medicine, Dayton, OH (USA)
The purposes of this study were to determine whether there are significant differences between two- and four-compartment model estimates of body composition, whether these differences are associated with aqueous and mineral fractions of the fat-free mass (FFM); and whether the differences are retained in equations for predicting body composition from anthropometry and bioelectric resistance. Body composition was estimated in 98 men and women aged 65-94 y by using a four-compartment model based on hydrodensitometry, {sup 3}H{sub 2}O dilution, and dual-photon absorptiometry. These estimates were significantly different from those obtained by using Siri's two-compartment model. The differences were associated significantly (P less than 0.0001) with variation in the aqueous fraction of FFM. Equations for predicting body composition from anthropometry and resistance, when calibrated against two-compartment model estimates, retained these systematic errors. Equations predicting body composition in elderly people should be calibrated against estimates from multicompartment models that consider variability in FFM composition.
- OSTI ID:
- 5544685
- Journal Information:
- American Journal of Clinical Nutrition; (USA), Vol. 53:6; ISSN 0002-9165
- Country of Publication:
- United States
- Language:
- English
Similar Records
In vivo neutron activation facility at Brookhaven National Laboratory
Evaluation of an in vivo prompt gamma neutron activation facility for body composition studies in critically ill intensive care patients: results on 41 normals
Related Subjects
AGED ADULTS
BODY COMPOSITION
MATHEMATICAL MODELS
ELECTRIC CONDUCTIVITY
EQUATIONS
FORECASTING
ISOTOPE DILUTION
REGRESSION ANALYSIS
TRITIUM COMPOUNDS
ADULTS
AGE GROUPS
ELECTRICAL PROPERTIES
HYDROGEN COMPOUNDS
ISOTOPE APPLICATIONS
MATHEMATICS
PHYSICAL PROPERTIES
STATISTICS
TRACER TECHNIQUES
550201* - Biochemistry- Tracer Techniques