Concordância entre métodos de avaliação de consumo de sódio e potássio em participantes do Estudo Longitudinal de Saúde do Adulto - ELSA- Brasil

Abstract

This dissertation is about the evaluation methods for consumption of sodium and potassium. It is structured in two manuscripts. The first part aimed to validate the consumption of sodium and potassium evaluated by a food frequency questionnaire (FFQ) applying the triads method to a subsample of participants of the Longitudinal Study of Adult Health (ELSA-Brasil). The consumption of those nutrients was estimated by a semi-quantitative FFQ with 114 food items, the urinary excretion of 12 hours during the night and three food records (FR) of 24 hours. Correlation coefficients were obtained between each of the methods. The coefficient of validity (CV) and the 95% confidence intervals were calculated using the sampling “bootstrap”. Pearson correlation was performed in order to corroborate the triads method’s assumptions. The coefficients of validity for sodium were considered moderate FFQ IR (0,37), FR IR (0,56) and B IR (0,21) whereas for potassium, they revealed both moderate (FFQ IR: 0,60; B IR: 0,42) and elevated (FR IR: 0,79). It has been concluded that the FFQ-ELSA-Brasil is relatively valid in order to estimate the intake of potassium, but it does not apply to sodium. The second manuscript aimed to evaluate the concordance of sodium and potassium consumption estimated by two methods in the Longitudinal Study of Adult Health (ELSA-Brasil). Data from 12,593 participants obtained through food frequency questionnaire (FFQ) and urinary excretion 12 hours nightly (UrEx), was analyzed. Means and consumption quintiles were compared. Weak correlations and significant differences between means of sodium (FFQ: 4.5±1.7g; UrEx: 4.3±2.1g) and potassium (FFQ: 4.8±1.8g; UrEx: 2.4±1.0g) were found. Percentage of disagreement between methods ranged from 41.2 to 44.4% and exact concordance from 22.6 to 23.7%. No concordance was observed between methods.