No abstract

Ensuring the accuracy of dietary assessment instruments is paramount for interpreting diet-disease relationships. The present study assessed the relative and construct validity of the 14-point Mediterranean Diet Adherence Screener (MEDAS) used in the Prevención con Dieta Mediterránea (PREDIMED) study, a primary prevention nutrition-intervention trial. A validated FFQ and the MEDAS were administered to 7146 participants of the PREDIMED study. The MEDAS-derived PREDIMED score correlated significantly with the corresponding FFQ PREDIMED score (r = 0.52; intraclass correlation coefficient = 0.51) and in the anticipated directions with the dietary intakes reported on the FFQ. Using Bland Altman's analysis, the average MEDAS Mediterranean diet score estimate was 105% of the FFQ PREDIMED score estimate. Limits of agreement ranged between 57 and 153%. Multiple linear regression analyses revealed that a higher PREDIMED score related directly (P < 0.001) to HDL-cholesterol (HDL-C) and inversely (P < 0.038) to BMI, waist circumference, TG, the TG:HDL-C ratio, fasting glucose, and the cholesterol:HDL-C ratio. The 10-y estimated coronary artery disease risk decreased as the PREDIMED score increased (P < 0.001). The MEDAS is a valid instrument for rapid estimation of adherence to the Mediterranean diet and may be useful in clinical practice.