1 Departament of Nutrition, School of Public Health, University of São Paulo, Brazil. 2 Department of Epidemiology and Biostatistics, Institute of Collective Health, Federal Fluminense University, Brazil.
Introduction: Breakfast has been widely studied, however no consensus about breakfast definition is found in the scientific literature. Meal definitions according to what the respondent named (self-reported) and according to the period of the day that it was consumed are the most used. The aim of this study was to evaluate the nutritional profile of breakfast, considering two different meal definitions. Method: Dietary data from the cross-sectional Health Survey of Sao Paulo-2008 (Brazil) was used. The final sample consisted of 547 adults that had consumed breakfast accordingly to two criterions. Food intake was measured by 24-hour recalls administered by previously trained interviewers. Breakfast classification was done considering two definitions: (1) self-report and (2) hour of consumption (meals consumed between 6am and 10am). Nutrients intake (contents of energy, carbohydrate, total fat, saturated fat, protein, animal protein, fiber, vitamin C, total folate, iron, sodium, added sugar and calcium) comparison between meal classifications was done using the non-parametric Wilcoxon test due to lack of normality distribution. Results: Median nutrient profile was not statistically different between the two breakfast classifications studied (p>0.05) and median intake of fiber, vitamin C and calcium presented the same value for both definition. Conclusions: Results of this study suggest that both definitions are similar, but it is necessary to compare others breakfast definitions to standardize an appropriate definition making studies comparable.