Abstract
We aimed to determine whether the 14-item Mediterranean diet adherence screener (MEDAS) is suitable in Taiwan and associate the MEDAS score with the risk of prediabetes. In this cross-sectional study 346 patients were recruited between 2014 and 2019 at Taipei Tzu Chi Hospital. The MEDAS score was obtained with a 14-item MEDAS used in the PREDIMED trial. The blood glucose level is measured by fasting glucose and HbA(1c). The results of the screener were analyzed for internal consistency and compared with the blood glucose level using multivariate regression models. The MEDAS score was significantly (p = 0.001) and inversely associated with both measures of blood glucose level. Adjusted data (95% CI) showed that each additional point in the MEDAS score decreases the risk of prediabetes with abnormal fasting glucose (> 100 mg/dL) level by 60% and the risk of prediabetes with abnormal HbA(1c) (> 5.7%) by 22.4%. Consuming at least 3 servings of legumes each week was significantly (p = 0.007) related to a lower risk of prediabetes under logistic regression. A higher score on the 14-item MEDAS screener was significantly associated with a lower risk of prediabetes.