The aim of this study was to compare plasma concentrations of vitamin A in 710 women after delivery, with a simplified Food Frequency Questionnaire (FFQ). Vitamin A was determined in plasma, by HPLC. The FFQ included 55 foods, and was based on the typical diet of the low socioeconomic families in the Southeast of Brazil. For calculation purposes, first were identified all potential sources of retinol and carotenoids; the portion sizes being assigned to each food item on the basis of a previous study in the same area. Daily, weekly and monthly frequencies of consumption of foods with high, moderate and low concentrations of vitamin A were grouped according to the IVACG recommendations (1) in usual pattern of food consumption (UPF) scores. Pregnant women were allocated by these scores in 3 different categories (high, moderate and low) according to their risk of developing vitamin A deficiency. The categories of risk were based on the safe level of vitamin A intake of 800 RE/day (5600 RE/week) (7) and the basal requirement of 370 RE/day for pregnant women (8). 1.3% of the women had plasma concentrations of vitamin A < or = 0.70 mumol/l, 10.2% between 0.71-1.05 mumol/l and 88.5% > 1.05 mumol/l. According to the score categories, 49.5% of the women were at low risk for vitamin A deficiency, and 50.5% were at moderate risk. There was a weak statistically significant correlation between the plasma concentrations of vitamin A and the score categories (r = 0.11; p = 0.005). Our results indicate that the FFQ does not provide a very precise information on vitamin A status, probably because of the day to day variation in vitamin A intake, and the insensitivity of plasma concentrations to consumption of vitamin A. However, by using a simple score based on the FFQ, we were able to predict that the majority of the mothers were not at high risk for vitamin A deficiency.