Data from: Validity of bioelectrical impedance analysis in predicting total body water and adiposity among Senegalese school-aged children

Introduction: Childhood obesity is currently a serious public health
challenge in developing countries. Therefore, an accurate assessment of
adiposity is required. The objective of this study was to validate BIA
prediction equations for the assessment of total body water and adiposity
or percentage of body fat for the first time in Senegalese school-aged
children. Methods: One-hundred-fifty-one (151) pupils who were 8-11 years
old were randomly selected from four public schools in Dakar. The body
composition measured by deuterium dilution method (DDM) was used as the
reference method and compared to that predicted by BIA using a
multi-frequency analyser. Stepwise backward multiple linear regression was
performed to calculate TBW and %BF in a subsample, which were then
validated it in the rest of the sample. The Bland and Altman approach was
used to assess the agreement between the two methods (bias and limits of
agreement). Results: FFM was higher in boys (24.6±6.9 kg) compared to
girls (21.2±3.3 kg; P<0.001), and %BF was lower in boys (5.1±4.6
kg) compared to girls (5.3±3.8 kg; P<0.05). Overall, 11.3% of
children were obese (%BF ≥25% in boys, and ≥30% in girls) and 4.6% were
overweight/obese according to their BMI-z score (>+1). The
equations developed were as follows: TBW=0.376(Height²/Z50)-0.470 (sex)
+0.076(weight) +0.065(height)-2.28. %BF= -1.10(height²/Z50)
+3.14(sex)+1.57(weight)-4.347. These specific equations showed good
precision and a low and non-significant mean bias (0.11 kg, P=0.279; and
0.19 kg, P=0.764) for TBW and %BF, respectively. Conclusion: The newly
developed equations can be used as an accurate and alternative tool for
screening for obesity among African school-aged children in various
settings