The Body Adiposity Index (BAI) estimates a person’s percent body fat using hip circumference and height. A slight variation of the BAI from the FELS longitudinal study (BAIFELS) has more recently proposed to estimate percent body fat better than the BAI. Purpose: Assess criterion validity of the BAI and BAIFELS with air displacement plethysmography (ADP). Methods: Percent body fat was estimated for 1636 volunteer healthy adults (881M; 755F, age: 36.9 ±10.5 yrs; BMI: 25.4 ± 4.6 kg/m2) using ADP and BAI and BAIFELS. Differences in percent body fat between ADP, BAI, and BAIFELS were tested using an repeated measure ANOVA with post hoc analyses. Agreement between methods was tested using Bland-Altman plots. Results: Differences in percent body fat between each method were statistically significant (p<.05).Percent body fat assessed by ADP correlated moderately with BAI (r=0.534, p<0.001), BAIFELS (r=0.526, p<0.001),and BMI (r=0.476, p<0.001).BAI and BAIFELS underestimated percent body fat compared to ADP on average 2.1% and 1.1% respectively. BAI equations overestimated percent body fat for lean persons and overestimated percent body fat for overweight or obese persons.Agreement between BAI, BAIFELS was poor with wide limits of agreement. Conclusion: BAI and BAIFELS are easier and less expensive alternatives to estimate percent body fat in humans compared to ADP. However, health risk due to body fat will be better estimated using either ADP or waist circumference and waist-to-hip ration in addition to an BAI.BAI would be most suitable where resources for estimating percent body fat are most limited.
Journal of Fitness Research / Vol. 5, No. 3, pp.28-37