Desenvolvimento e validação de um instrumento de rastreamento de sintomas de transtornos alimentares em crianças e adolescentes

Abstract

Introduction: Eating disorders are characterized by a persistent disturbance in eating or not eating behavior. Today, anorexia nervosa, bulimia nervosa, and binge eating disorder are the most largely literature. Early diagnosis generates a great success rate. However, today there is a shortage of tools that track eating disorders symptoms, particularly in Brazilian preschoolers. The objective of this study is to develop and verify the evidence of validity of an instrument for tracking symptoms of eating disorders in children and teenagers. Methods: Cross-sectional study that included participants between 5 and 12 years old of both sexes. Data collection was carried out for convenience in schools, sports clubs, and on-line. The validation processes included the following evaluations: content validity (quantitative and qualitative analysis of the judges), Item response theory (severity and discrimination of items and test information curve), construct validity (exploratory and confirmatory factor analysis), concurrent validity (variables: age, BMI and sex; KruskalWallis test) and reliability (omega and Spermann-Brown a test). Results: Regarding content validity, the judges answered that they agree with language clarity (83,3%), relevance in practice (75%), and theoretical relevance (91,7%). The instrument presented an adequate performance about severity and discrimination and the theory response's information curve to the item test. The model proposed in the AFE had a total percentage of variation explained was 69%, and the KMO value was 0,42. The factorial loads of some criteria were evaluated in the confirmatory analysis, and the results of the reliability presented values lower than expected. Female children had higher levels of eating disorders than male children and teenagers. Conclusions: The preliminary version of the instrument for assessing the level of eating disorders for children and teenagers developed performed well part of the psychometric evaluated. However, it was possible to diagnose with the other results that some facets for assessing the level of eating disorder were not included in the developed instrument.