Abstract P190: Validity and Reproducibility of a Food Frequency Questionnaire to Assess Compliance to a Cardioprotective Mediterranean Diet

Abstract

Background: The cardioprotective effects of the Mediterranean diet have been well established in epidemiological studies. More recently, it has been shown to improve the cardiometabolic risk profile and endothelial function, as well as reduce markers of vascular inflammation, independent of weight loss. Since cardiovascular disease is the second cause of death in the Canadian and North American populations, randomized clinical trials evaluating the efficacy of the Mediterranean diet in high-risk primary and secondary prevention are warranted. In order to conduct such studies, validated dietary assessment methods specific to population and food habits are needed. Therefore, the aim of our study was to assess the reproducibility and the relative validity of a quantitative food frequency questionnaire (FFQ) to be used in clinical trials at the Montreal Heart Institute’s Prevention Center. Methods: Fifty (50) participants (54% (27 of 50) women) aged 19 to 86 years with and without coronary disease were recruited, and randomized in 3 groups in a crossover design where the sequence of administration of questionnaires differed in each group. The FFQ includes 157 food items and was designed to measure food intake over one month. It was administered twice 3 to 5 weeks apart to assess reproducibility and was compared to a 12-day dietary record carried out over a 1-month period to assess validity. Participants were asked not to modify their diet for the duration of the study. FFQs were self-administered and reviewed by a registered dietician. All questionnaires were analyzed using The Food Processor software. Results: Reproducibility (n = 49) assessed by the intraclass correlation coefficient (ICC) revealed good (ICC > 0.5) to very good (ICC > 0.7) correlations for energy and all nutrients (except vitamin A) with ICCs ranging from 0.48 (95% confidence interval (CI), 0.23 to 0.67) for vitamin A to 0.91 (95% CI, 0.85 to 0.95) for alcohol (mean 0.68). Preliminary results for validity (n = 15) revealed very good significant correlations for EPA, DHA, selenium and alcohol and good significant correlation for energy, total fat, saturated fatty acids, monounsaturated fatty acids, oleic acid, linolenic acid, trans fat, vitamin B3, vitamin D, iron and magnesium. Conclusions: Our FFQ demonstrates good to very good reproducibility. Furthermore, the FFQ appears valid for evaluating adherence to a Mediterranean diet, notably because it provides very good estimates of marine omega-3 fatty acid (EPA, DHA) intake. These results support the eventual use of our FFQ in clinical trials on the Mediterranean diet. Complete sample analysis will be available in March 2014.