Abstract
PURPOSE: Adolescent obesity, as defined by BMI, is amongst the highest in the world in Kuwait. This study aimed to determine the extent to which BMI might be underestimating obesity as defined by excessive fatness in Kuwaiti female adolescents and young adults. METHODS: A total of 400 apparently healthy Kuwaiti female university students (mean age 18.0 years, SD 0.6) were recruited. Excessive fatness was defined as body fat percentage ≥30, measured using the Tanita model TBF-310 Bio-impedance system with the manufacturer's equation. Obesity was defined as recommended by the WHO in adult participants - those aged ≥19.1 years - as BMI≥30 kg/m(2). In the adolescent participants (age <19.1 years) obesity was defined as recommended by the WHO as a BMI-for-age Z score of ≥2.00. The accuracy of BMI-defined obesity to identify excessively fat individuals was determined by estimating the prevalence of obesity using high BMI and prevalence of excessive fatness, and by calculating sensitivity and specificity and predictive values. RESULTS: Median BMI was 27.8 kg/m(2) (range 15.1-51.2) and median body fat percentage was 32.0 (range 5.0-54.0). The prevalence of excessive fatness was 62% (247/400 individuals were excessively fat), while the prevalence of obesity according to BMI was 42% (169/400 individuals were obese according to their BMI). The sensitivity of BMI to identify the excessively fat individuals was moderate (66%) but specificity was high (96%). The positive predictive value of BMI was 96% and the negative predictive value was 64%. CONCLUSION: BMI-based measures substantially underestimate the prevalence of excessive fatness in Kuwaiti adolescent females. Obesity is even more prevalent, and requires more urgent attention, than is apparent from BMI-based measures used in most research and national surveys. BMI may also be too crude for use as an exposure or outcome variable in many epidemiological studies of Arab adolescent girls and adult women.