Validation of an FFQ for evaluation of EPA and DHA intake.

OBJECTIVE To validate an FFQ for the assessment of dietary EPA and DHA against their relative concentrations in red blood cells (RBC). DESIGN Cross-sectional analysis of baseline data. Intakes of marine food products and EPA and DHA were estimated by FFQ on the basis of consumption of marine food products in the last month. Fatty acid composition of RBC membranes was quantified by GC. SETTING Saint-François d'Assise Hospital, Québec, Canada. SUBJECTS A total of sixty-five middle-aged women who participated in a randomized clinical trial. RESULTS Spearman's correlation coefficient between intake of EPA, DHA and EPA + DHA and their corresponding concentration in RBC was 0.46, 0.40 and 0.42, respectively (all P < 0.05). Multiple regression analysis of EPA+DHA intake and RBC EPA + DHA concentration indicated positive and significant correlations for oily fish (beta = 0.44, 95% CI 0.16, 0.72, P = 0.0027), total fish (beta = 0.42, 95% CI 0.19, 0.64, P = 0.0005) and marine food products (beta = 0.42, 95% CI 0.20, 0.64, P = 0.0003). No other marine food products significantly predicted RBC EPA + DHA concentration. CONCLUSIONS Although the present validation study was undertaken among middle-aged women with low consumption of marine food products (<3 servings/week), our FFQ provided estimates of EPA and DHA intakes that correlated fairly well with their RBC concentrations. However, the absence of correlations between EPA + DHA intakes from different marine species suggests that a minimum EPA + DHA intake is necessary to observe a relationship with RBC EPA + DHA concentrations.