Abstract
Although some researches has already found associations between meals consumption and nutritional status, and also observed that changes in the composition of a single meal is enough to change the postprandial and inflammatory response in individuals, there is no consensus on the importance of studying meals associations with health outcomes, considering that the daily food intake could be enough. In this context, combining hypothesis driven methods, data driven methods and hybrid methods, allowing a holistic view of diet and health relationship, this study aims to evaluate the impact of each meal (breakfast, lunch and dinner) in the overall dietary quality, considering lifestyle, nutritional status, blood pressure, blood glucose, and lipid and inflammatory profile of adults living in São Paulo. Structural equation models will perform the association between the overall dietary quality, estimated by the Health Eating Index adapted to the Brazilian population, with breakfast, lunch and dinner patterns, estimated by factor analysis, a multivariate data analysis, combining statistical methods of factor analysis and multiple regression. Finally, to compare the impact of each meal (breakfast, lunch and dinner) in the overall dietary quality will be applied a classification algorithm (decision tree). (AU)
|