Validade de medidas autorrelatadas na estimativa de sobrepeso e obesidade na população geral

• Ano de defesa: 2017
• Autor(a) principal: Ferriani, Lara Onofre
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Saúde Coletiva; Centro de Ciências da Saúde; UFES; Programa de Pós-Graduação em Saúde Coletiva
• Palavras-chave em Português: Nutritional status; Anthropometry; Obesity; Body mass index; Validity; Selfreport; Estado nutricional; Validade; Autorrelato; Nutrição - Avaliação; Antropometria; Obesidade; Índice de massa corporal; Saúde Coletiva; 614
• Link de acesso: http://repositorio.ufes.br/handle/10/10111

Obesity has become a major public health problem, as it reaches pandemic proportions and is associated with a number of chronic diseases. For the continuous monitoring of populations’ nutritional status, the use of self-reported measures of weight and height, and derived calculation of the Body Mass Index (BMI), has become a common alternative. Thus, the objective of this study is to validate and correct self-reported measures of weight and height in order to estimate prevalence rates of overweight and obesity in the general adult population. The data analyzed were collected in the São Paulo Megacity Mental Health Survey that was divided in two phases; in the first, a total of 5.037 individuals were interviewed at home, where information of weight and height were self-reported. In the second phase, a subsample of 766 individuals had the weight and height measured in the hospital scope. Statistical analysis were performed using Stata version 13.0 software, with a 5% significance level. The validity of the self-reported measures was evaluated and then different linear regression models were constructed for the correction of the self-reported data. The concordance between the measured, self-reported and corrected measures of weight, height and BMI was evaluated through intraclass correlation coefficients (ICC) and kappa coefficient. The reliability found between the measured and self-reported measures was high for both sexes and for the total sample (ICC weight 0.951 / CI 0.938-0.961; ICC height 0.870 / CI 0.597-0.939; ICC BMI 0.865 / CI 0.677- 0.928). However, the estimation of the prevalence of overweight by self-reported measures was underestimated by 13%. The kappa coefficient comparing the self-reported data with the measured one is classified as moderate agreement (0.574), whereas for the corrected measures the agreement found changes from the classification category considered substantial (0.653 / 0.665), demonstrating that the corrected measure is closer to the actual measure. A high prevalence of overweight and obesity was found, considering the estimates from the measured measures (38.0% and 22.6%, respectively), as well as from those self-reported (30.9% and 11.7%) and corrected ones (37.4-39.0% and 17.8-18.6%). Therefore, the corrected measure seems to be a considerable alternative when using self-reported measures, since the corrected measure can represent more accurately the actual measure and thus reduce the underestimation of the prevalence rates of overweight resulting from the use of these measures.