Body mass index and reciprocal of ponderal index as predictors of overweight and obesity

The Body Mass Index (BMI) has been utilized in clinical and epidemiological studies to identify overweight and obesity. Rabelo & Araujo (2001) suggested that the use of the Reciprocal of Ponderal Index (RIP) would be more appropriated. The objective of this study was to compare the identifi cation of overweight by for ponderal indices and estimated body fat (F%). The study enrolled 89 male subjects (21.9 ± 2.3 years, 177.7 ± 7.2 cm, 69.5 ± 5.9 kg, 8.0 ± 0.2 F% and 22.0 ± 1.8 kg/m2 BMI and 43.3 ± 1.63 m/ kg1/3 RIP) and 70 female subjects (21.4 ± 4.7 years, 162.3 ± 5.4 cm, 54.9 ± 6.4 kg, 22.9 ± 4.5 F% and 20.8 ± 2.1 kg/m2 BMI and 41.3 ± 4,1 m/kg1/3 RIP). F% was estimated by Jackson and Pollock (1978) and Jackson et al. (1980) equations. Both ponderal indices exhibited similar results. Correlation coeffi cients for males were 0.60 for BMI and 0.57 for RIP in relation to F%. Correlations for females were 0.63 and 0.66, respectively. The identifi cation of overweight by ponderal indices and by F% in males demonstrated 85% agreement and comparable numbers of false-positive and false-negative cases. In females, agreement was 70 % with 30% of false-negative cases. We concluded that the use of ponderal indices to identify overweight in males for fi tness programs has a fair level of agreement with F%, but that for females, ponderal indices should be used with caution. In this sample, using either BMI or RIP returns similar results.