The objective of most epidemiological studies is to classify/rank people according to their relative quantity of consumption. In the specific case of quantifying fish consumption, in order to assess mercury intake rate, the method used must estimate the individual quantity consumed accurately, or from the self-referred quantity it must be possible to estimate the real consumption-termed calibration. Therefore, the objective of this study was to validate and calibrate the 24-h recall method for assessing the amount of fish consumed to estimate the mercury intake in riverine populations from Baixada Cuiabana, in Pantanal Mato-grossense, Brazil. The weighed dietary method was used as the reference standard, and the validity of the 24-h recall method was assessed using correlation coefficients (r), regression coefficients (beta), and the intraclass correlation coefficient (ICC). The results showed that the ICC, which assesses the agreement level between measures, was a more adequate measure than the correlation coefficient (r). Therefore, the calibration equation was calculated applying a multiple linear regression model using the consumption mean of the second and third days, which presented the highest ICC, stratified by sex and adjusted by age and season variables.
Copyright 2001 Academic Press.