Search

Cross-sectional population-based study conducted in the city of Porto Alegre, south Brazil, to evaluate the use of 24-hour dietary records in the assessment of sodium intake. Subjects were 188 children and adolescents aged 6 to 17 years. Overnight urine was collected for all subjects, and 24-hour urine was collected for 16% of the subjects. Daily sodium intake ranged from 8.5 to 10 g. Mean amount of sodium excreted in 24-hour urine was 83% of the mean intake calculated by dietary records. The Pearson correlation coefficient between the two methods of assessment of urinary excretion of sodium was r = 0.71 (P < 0.001). However, Bland-Altman analysis revealed that the 95% limits of agreement varied widely, thus providing evidence of the large difference between the two methods. Bland-Altman analysis could be considered an important improvement over the use of correlation coefficient to evaluate the agreement between two methods. Estimation of sodium intake by means of dietary records was inaccurate when compared with urine collections (r = −0.41; P < 0.05 (24h) and r = 0.09; P = 0.175 (overnight). [Copyright &y& Elsevier]