OBJECTIVE: This study estimated the existence of risk factors associated with eating disorders in students from Salvador, Bahia, Brazil, using ethnicity/race as a factor of heterogeneity. METHODS: The study subjects were 626 female students aged 15 to 30 years attending public and private high schools and universities in Salvador, Bahia. The instruments Eating Attitudes Test-26, Body Shape Questionnaire, and Beck Depression Inventory were used for screening the students for risk of eating disorders. The students reported their ethnicity/race using the list provided by the Instituto Brasileiro de Geografia e Estatística. The statistical treatment included descriptive and bivariate analysis (Pearson χ2) and multivariate logistic regression. RESULTS: Students self-identified as Asian or Native Brazilian were 3.6 times more likely to have disordered eating behaviors and 4.8 times more likely to be dissatisfied with their bodies than those self-identified as Caucasians. Students self-identified as biracial (European/African Brazilian) were 2.5 more likely to present that same dissatisfaction than those self-identified as Caucasians. The comorbidity 'depression' should be considered but was not significantly associated with race. CONCLUSION: Non-Caucasian females in Salvador, Bahia, Brazil are at risk of developing eating disorders. Other studies associating quantitative and qualitative methods would allow more robust analyses on the relationship between eating disorders and race/ethnicity.
Anorexia; Bulimia; Body image; Ethnicity and health; Eating disorders
