Dietary intake variations from pre-conception to gestational period according to the degree of industrial processing: A Brazilian cohort

Objective: To estimate food intake changes from pre-conception to gestational period according to the degree of food processing. Methods: Prospective cohort conducted in a public health care center in Rio de Janeiro with Brazilian pregnant women (n = 189). A food frequency questionnaire was applied at the first (5th-13th) and third (30th-36th) gestational trimesters. The food items were classified according to characteristics of food processing in four groups: unprocessed/minimally processed foods; sugar/fat; processed foods and ultra processed foods. The variation of food intake according to the degree of processing between the preconception and gestational period was compared using paired Student's t-test. Linear regression models were performed to estimate the association of mother's characteristics on the variation Of food group contribution to the total energy intake between periods. Results: Total energy intake was 2415 (SD = 813) in the pre-conception and 2379 (750) kcal in the gestational period. We excluded women who had implausible total energy intake (<600 and >6000 kcal/day). The contribution of unprocessed/minimally processed food group to total energy intake during pregnancy when compared to the pre-conception period was higher [50.5 (14.1) vs. 48.8 (12.4), p-value = 0.048], while the caloric share of ultra-processed food group was lower [41.3 (14.6) vs. 43.1 (12.5), p-value = 0.032]. We observed a negative association of age (p-value = 0.009) and a positive association of pre-pregnancy BMI (p-value = 0.060) with the variation of ultra-processed food intake. Conclusions: Ultra-processed food intake decreased, while minimally/unprocessed food intake slightly increased from the pre-conception to gestational period. These results indicate potential for a larger improvement in the women's diet quality and that nutritional counseling interventions in pregnant women are still needed. (C) 2016 Elsevier Ltd. All rights reserved.