The biomarker-based validity of a brief-type diet history questionnaire for estimating eicosapentaenoic acid and docosahexaenoic acid intakes in pregnant Japanese women

Objectives: Maternal docosahexaenoic acid (DHA) intakes is important for brain development in fetuses. Accurate assessment of EPA and DHA intakes is required in clinical settings to identify women with deficiency of these nutrients and provide an appropriate intervention for them. We examined the validity and reproducibility of a brief-type self-administered diet history questionnaire (BDHQ) for evaluating EPA and DHA intakes of pregnant Japanese women, to establish an easily administered dietary assessment tool. Methods: A total of 105 women in the second trimester and 102 women in the third trimester were studied at a university hospital in Tokyo, between November 2010 and February 2012. The reference values for the validation study were plasma concentrations of EPA and DHA. For the reproducibility study, 54 women completed the BDHQ twice, within a 4-week period in the second trimester. Results: Energy-adjusted intakes of EPA, DHA, and EPA+DHA were significantly associated with the corresponding plasma concentrations (r_s=0.354, r_s=0.305, and r_s=0.327 in the second trimester; r_s=0.391, r_s=0.316, and r_s=0.358 in the third trimester, respectively). Intraclass correlation coefficients for the two-time BDHQ were 0.543 (EPA), 0.611 (DHA), and 0.581 (EPA+DHA). In the Bland-Altman plots, the intakes of EPA, DHA, and EPA+DHA in the two-time BDHQ showed that the values for most participants were in the accepted range of agreement. Conclusions: BDHQ has an acceptable validity level for assessing EPA and DHA intakes among Japanese women in the second and third trimesters.