We send one common response letter to the three authors because the contents of the comments are similar and all of them come from the Department of Nutrition, Gillings School of Global Public Health.
We agree with the letters of the three authors in the possibility of misclassification due to the fact that most people consume very near to the median for each of the Mediterranean diet food group. It is something very well described in the field of Nutrition Epidemiology.1 We have to assume this fact that it is common to every method of classification. However, the aim of the study was to analyze the role of an overall dietary pattern and not of a specific food item. Hence, it is necessary to calculate a score (with its specific limitations) that measures the adherence. To solve this problem, we used categories of adherence as the main exposure (categories of Mediterranean dietary score). When comparing extreme groups, it is very unlikely to suffer from that problem and it is difficult to conceive that all subjects are very close to the median in all scores. In any case and following the authors’ comments, we have analyzed the mean consumption of each Mediterranean diet component within each category of the Mediterranean dietary score (Table 1). As you can see in the table, the score managed to classify subjects in a gradient of nutrient intakes. The nutrient-intake differences between extreme categories are large, offering a guarantee of correct classification. Finally, if some misclassification exists, this would be nondifferential, leading the results toward the null. So, the association observed between the adherence to the Mediterranean dietary score and quality of life had been even higher than that reported in our paper.
Moreover, misclassification regarding the adherence to the Mediterranean dietary pattern is not probable among participants of the SUN Project. The association between the adherence to the Mediterranean diet and some outcomes, such as weight change within the SUN project, has showed that there are no differences in the results according to the method used to ascertain the adherence to the Mediterranean diet2 (not only through Trichopoulou score but also with other hypothesis-oriented methods).
In addition, the Mediterranean dietary score used by Trichopoulou has been widely used and universally accepted to assess the adherence to the Mediterranean diet among very different populations and its association with different outcomes and conditions.
To assess empirically derived patterns and to analyze the association between their adherence and quality of life are interesting activities that we have performed, and the results of which are under consideration in a scientific journal.
Willett WC . Issues in analysis and presentation of dietary data. In: Willett W (ed.) Nutritional Epidemiology 2nd edn. Oxford University Press: New York, 1998 pp 321–346.
Beunza JJ, Toledo E, Hu FB, Bes-Rastrollo M, Serrano-Martínez M, Sánchez-Villegas A et al. Adherence to the Mediterranean diet, long-term weight change, and incident overweight or obesity: the Seguimiento Universidad de Navarra (SUN) cohort. Am J Clin Nutr 2010; 92: 1484–1493.
The authors declare no conflict of interest.