Relative validity of a food frequency questionnaire for adolescents from a capital in the Northeastern region of Brazil.

The present study was conducted to evaluate the validity of the Food Frequency Questionnaire (FFQ) used in the RPS Birth Cohort Consortium (Ribeirão Preto, Pelotas, and São Luís) to assess dietary intake of adolescents from São Luís, Maranhão. The research was developed with 152 adolescents aged 18 and 19 years. For the validation of the FFQ, the average of three 24-hour recalls (24HRs) was used as the reference method. The mean and standard deviation of energy and nutrient intake extracted from the surveys were estimated. The paired Student's t-test was used to verify the differences between the instruments. Pearson correlation coefficient, intraclass correlation coefficient (ICC), weighted Kappa, and the Bland-Altman plot were calculated in order to measure the agreement. The study adopted a level of significance <5%. Compared with the three 24HRs, the FFQ overestimated the consumption of most nutrients. Energy-adjusted and de-attenuated concordance Pearson correlation coefficients ranged from 0.06 to 0.43, and correlations were significant for iron, calcium, riboflavin, sodium, saturated fat, niacin, and vitamin C. The energy-adjusted and de-attenuated ICCs ranged from 0.01 to 0.31, and the weighted Kappa ranged from 0.01 to 0.46. The analyses of agreement were significant for vitamin C, fiber, calcium, riboflavin, niacin, sodium, lipids, and iron. In conclusion, the FFQ presented acceptable relative validity for lipids, saturated fatty acids, fiber, calcium, iron, riboflavin, niacin, vitamin C, and sodium. This instrument will be useful in studies about food consumption of adolescents in São Luís, Maranhão.