Research Group on Quality, Safety and Bioactivity of Plant Foods, Campus de Espinardo, Centro de Edafologia y Biologia Aplicada del Segura-Consejo Superior de Investigaciones Científicas (CEBAS-CSIC), P.O. Box 164, 30100 Murcia, Spain.
Department of Life and Health Sciences, University of Nicosia, 1700 Nicosia, Cyprus.
Nutrients. 2020 Sep 27;12(10):2960. doi: 10.3390/nu12102960.
This study provides comprehensive validation of the 14-item Mediterranean Diet Adherence Screener (14-MEDAS) in an adult population from Greece (GR), Portugal (PT), Italy (IT), Spain (SP), Cyprus (CY), Republic of North Macedonia (NMK), and Bulgaria (BG). A moderate association between the 14-MEDAS and the reference food diary was estimated for the entire population (Pearson = 0.573, -value < 0.001; Intraclass Correlation Coefficient (ICC) = 0.692, -value < 0.001) with the strongest correlation found in GR, followed by PT, IT, SP, and CY. These results were supported by kappa statistics in GR, PT, IT, and SP with ≥50% of food items exhibiting a fair or better agreement. Bland-Altman analyses showed an overestimation of the 14-MEDAS score in the whole population (0.79 ± 1.81, 95%Confidence Interval (CI) 0.61, 0.96), but this value was variable across countries, with GR, NMK, and BG exhibiting the lowest bias. Taking all analyses together, the validation achieved slightly better results in the Mediterranean countries but a definitive validation ranking order was not evident. Considering growing evidence of the shift from Mediterranean Diet (MD) adherence and of the importance of culture in making food choices it is crucial that we further improve validation protocols with specific applications to compare MD adherence across countries.
这项研究在来自希腊（GR）、葡萄牙（PT）、意大利（IT）、西班牙（SP）、塞浦路斯（CY）、北马其顿共和国（NMK）和保加利亚（BG）的成年人群中，对 14 项地中海饮食依从性筛查器（14-MEDAS）进行了全面验证。14-MEDAS 与参考食物日记之间存在中度关联，整个人群的 Pearson 相关系数为 0.573（P 值 < 0.001）；内类相关系数（ICC）为 0.692（P 值 < 0.001），GR 的相关性最强，其次是 PT、IT、SP 和 CY。GR、PT、IT 和 SP 的kappa 统计数据支持这些结果，其中≥50%的食物项目具有良好或更好的一致性。Bland-Altman 分析显示，整个人群的 14-MEDAS 评分存在高估（0.79 ± 1.81，95%置信区间（CI）为 0.61，0.96），但这一数值在各国之间存在差异，GR、NMK 和 BG 的偏差最小。综合所有分析结果，在地中海国家的验证结果略好，但没有明确的验证排名顺序。考虑到地中海饮食（MD）依从性的转变以及文化在食物选择中的重要性的证据越来越多，我们必须进一步改进具有特定应用的验证方案，以比较各国的 MD 依从性。
