[Stability of body composition in boys and girls 12 through 15 years of age].

Although underwater weighing has been regarded as a representative of the criterion reference methods for determining body composition, this technique is cumbersome for screening. The purpose of the present study was to examine the stability of body composition assessed not only by underwater weighting but also by existing bioelectrical impedance equations and skinfold thickness equations during a 4-week period. Ten male and 10 female junior high school pupils, aged 12 to 15 years, volunteered to serve as subjects. Body density (Db) was estimated by underwater weighing (UW), skinfold thickness (ST) method, and bioelectrical impedance (BI) method. The UW was considered criterion reference method. The underwater weight was the heaviest value that was reproduced twice at least among 5 to 10 measurements. Residual lung volume was estimated "in water," with the subject in the sitting position, by a closed-circuit helium-dilution method. The ST equations used for determination of Db were those developed by Nagamine et al., while in the BI method equations proposed by Nakadomo et al. were applied. The Db for boys by UW did not differ significantly during the 4-week period; however, in girls it was significantly lower in the 2nd test (1.0479 +/- 0.018 g/ml) than in the 1st test (1.0500 +/- 0.019 g/ml). Likewise, the Db estimated by ST and BI methods remained unchanged in boys, while in girls it was significantly lower in the 2nd test (BI: 1.0445 +/- 0.010 vs. 1.0389 +/- 0.013 g/ml). Test-retest reliability of all the three methods for measurements conducted 4 weeks apart were statistically significant (e.g. UW: r = 0.99, ST: r = 0.94, BI: r = 0.89 in girls). In conclusion, the validity of Db estimated by these two methods was found to be insufficient when absolute values are required. However, body composition estimates assessed by ST and BI methods for boys and girls may be stable on a relative basis.