Obesity prevalence in N orwegian adults assessed by body mass index, waist circumference and fat mass percentage

The prevalence of obesity presents serious public health challenges worldwide and is most commonly estimated by the categorization of body mass index (BMI). The purpose of this study was (i) to describe the sensitivity of BMI, waist circumference (WC) and fat mass percentage in the classification of overweight and obesity and (ii) to describe the prevalence of overweight and obesity using different measurement methods in a national sample of Norwegians. A total of 904 Norwegians (20-85 years) were randomly enrolled. Body weight, height, WC and skinfold thickness were measured; BMI and fat mass percentage (FP(skf) ) were calculated. The sensitivity and specificity varied from 77.0% to 86.9% and from 60.6% to 82.3%, respectively. Area under the curve was 0.837 (95% confidence interval [CI]: 0.801-0.872) for the ability of BMI to predict obesity by FP(skf) and 0.811 (95% CI: 0.770-0.852) for WC. A total of 12.7% (male: 12.1%, female: 13.4%) were obese based on BMI classifications, 29.1% (male: 36.5%, female: 20.0%) of the sample were obese based on FP(skf) and 24.3% of men and 37.8% of women were abdominally obese by WC. The agreement between selected measuring methods was fair to good, although the variation in obesity prevalence rates between the different assessments methods varied markedly.