To develop a targeted FFQ (TFFQ) and evaluate its effectiveness for ranking minority women by their intake of multiple nutrients important to a healthy pregnancy.

In Stage 1, food groups derived from 24 h recall dietary data were entered into a stepwise multiple regression model to identify ‘discriminating’ foods for the TFFQ; in Stage 2, the TFFQ and a 24 h recall were administered to pregnant women. The TFFQ data were used to create a Habitual Multi-Nutrient Intake Scale (H-MNIS) for ranking the women. The effectiveness of the ranking was tested against women’s reported nutrient intakes on the 24 h recall.

Southern Israel.

In Stage 1, 519 Bedouin Arab adults; in Stage 2, 404 pregnant Bedouin women.

Of the ninety-three food groups included in the regression analysis, twenty-eight explained >80 % of the between-person variability in the intake of the target nutrients. Whole-wheat bread was a main discriminator for protein, Fe, Zn, Ca and folate. Fish, poultry and canned tuna explained ≥88 % of the between-person variation in DHA and EPA intakes. The mean daily intakes of pregnant Bedouin women from the 24 h recall data differed significantly across H-MNIS quartiles (P < 0·01) for all but three target nutrients (folate, EPA, DHA); and the P for trend was significant (P < 0·02) for all but EPA.

The TFFQ and H-MNIS represent rapid assessment tools for studying maternal nutrition on a multi-nutrient basis, and effectively ranked pregnant Bedouin women into exposure groups by their intake of the target nutrients.