LAUSR

OBJECTIVES The purpose of this study was to evaluate the agreement and risk factors for underestimation and overestimation between nutritional status and self-perceived body image and to assess the prevalence and associated factors for dissatisfaction with body weight among Brazilian adolescents. METHODS Students aged 12-17 years participating in the Study of Cardiovascular Risk in Adolescents ("ERICA"), a multicenter, cross-sectional, school-based country-wide study, were included (n=71,740). Variables assessed as covariates were sex, age, skin color, socioeconomic status, and common mental disorders (screened by the General Health Questionnaire, GHQ-12). Multinomial logistic regression was used to explore the association between covariates and combinations between self-perceived body image and body mass index (agreement, underestimation and overestimation). The associations between dissatisfaction with body weight and exposure variables were investigated using multivariable Poisson regression models. RESULTS Approximately 66% students rightly matched their body mass index with self-perceived weight (kappa coefficient was 0.38 for boys and 0.35 for girls). Agreement was higher among younger students and adolescents from low income households. Male sex, older age, and GHQ-12 score ≥3 were associated with weight overestimation. Prevalence of dissatisfaction with body weight was 45.0% (95% CI: 44.0-46.0), and higher among girls, older adolescents, those with underweight or overweight/obesity, as well as those who were physically inactive and with GHQ-12 ≥3. CONCLUSIONS Most of the sample rightly self-perceived their body image according to body mass index. Students with body image misperception and those dissatisfied with their weight were more likely to present a positive screening for common mental disorders.