Aim: To investigate whether body adiposity index (BAI; hip/height1.5–18), pediatric BAI (BAIp; hip/height0.8–38), and other hip/heightP ratios are useful in obese children.Method: Ninety obese Japanese children, 55 boys and 35 girls, who visited our University Clinic, were enrolled. The age was 9.92 ± 2.6 (mean ± SD) years, and the percentage overweight (POW) was 51.6 ± 18.8%. We set the power value of the hip/heightP 0, 0.5, 0.8, 1, 1.5, and 2 and studied the association with overweight indices, biochemical data, and fat area measured by computed tomography. Waist, waist/height ratio, and waist/hip ratio were also evaluated.Results: Hip/height and hip/height0.8 (BAIp) were more closely correlated with POW, body mass index percentile, and percentage body fat than hip/height1.5 (BAI). The correlation coefficient of hip/height with POW (r = 0.855) was the highest among the studied hip/heightP indices. The approximate line to predict POW was 411 × hip/height−207. The waist/height was also highly correlated with POW (r = 0.879). Hip and hip/height0.5 were more closely correlated with visceral fat area than hip/height, BAIp, and hip/height1.5. Hip and hip/height0.5 were significantly correlated with insulin. Only hip was also significantly associated with dyslipidemia. All hip/heightP indices were not significantly correlated with alanine aminotransferase (ALT). Waist was significantly correlated with serum lipids, ALT, and insulin.Conclusion: Hip/height and BAIp are better markers for overweight (adiposity) in obese children than BAI. However, hip/height, BAIp, and BAI are not useful to predict metabolic complications. Waist appears to be the best index for obese children overall at this time.