Child and Adolescent Public Health Epidemiology Group, Department of Public Health Sciences, Karolinska Institute, Norrbacka, Stockholm, Sweden. finn.rasmussen@ki.se
OBJECTIVE: To assess the size of biases in self-reported height, weight in a large sample of adolescents with special attention to possible effects of body dissatisfaction and to assess how such biases may influence estimates of overweight and obesity. DESIGN: Cross-sectional study. SETTING: Unselected population from Southwestern parts of Stockholm County, Sweden. SUBJECTS: Two-thousand seven hundred and twenty-six boys and girls, 15 years of age. METHODS: Data were collected by a questionnaire answered by adolescents and a physical examination made by trained study nurses. A validated physical appearance scale and body silhouettes were embedded into the questionnaire. RESULTS: Obese boys under-reported their weight (5.2 kg) (95% confidence intervals (CI) 3.7; 6.6) more than obese girls (3.8 kg) (95% CI 1.8; 5.8). Agreement between self-reported and measured body mass index (BMI)-categories (obese, overweight and non-overweight/obese) as estimated by weighted kappa was 0.77 (95% CI 0.72; 0.82) for girls and 0.74 (95% CI 0.70; 0.79) for boys. For obese girls and boys sensitivity of self-reports were 0.65 (95% CI 0.47; 0.79) and 0.52 (95% CI 0.38; 0.66). Boys with low scores on the physical appearance scale under-reported their weight and BMI more than those with high scores. Boys and girls who wished to be leaner under-reported their weight and BMI more than subjects who were satisfied with their body size (P