Validity of self-reported body mass index of Japanese children and adolescents.

ABSTRACT

BACKGROUND: Self-reported body mass index (BMI) has been widely used in epidemiological studies, but no study thus far has validated the self-reported BMI of Japanese children and adolescents. The aim of the present study was to assess the accuracy of self-reported BMI and its use for classification into weight categories. METHODS: Fifth- and eighth-grade students (n= 358) from all schools in Shunan City, Japan, from 2006 to 2010 were included. BMI was calculated from both self-reported and measured data. Pearson's correlation, Kappa statistics and the number of overweight children, and Bland-Altman plots with 95% limits of agreement were calculated to assess the agreement. RESULTS: Pearson's correlation between self-reported and measured data ranged between 0.946 and 0.987 for height, 0.978 and 0.992 for weight, and 0.930 and 0.964 for BMI, depending on gender or age. Kappa was excellent 0.949 for boys, 0.867 for girls, 0.897 for 10-11-year-old children, 0.928 for 13-14-year-old adolescents. The number of overweight children based on self-reported BMI was not significantly different from that based on measured BMI. The 95% limits of agreement were -3.62 and 2.90 cm for height, -2.80 and 2.38 kg for weight, and -1.42 and 1.44 kg/m(2) for BMI. CONCLUSIONS: Self-reported BMI was generally reliable for predicting the number of overweight Japanese children and adolescents. There is slight discrepancy, however, between self-reported BMI and measured BMI, and self-reported BMI should be used with caution, keeping the variance and concordance rate in mind.