Abstract
AIM: To assess whether the Mediterranean diet (MD) is associated with lower levels of type 2 diabetes (T2D) in a non-Mediterranean population. METHODS: Cross-sectional analysis of follow-ups 1 (FU1, 2009-2012, n = 4398, 45.7% men, 57.7 ± 10.5 years), 2 (FU2, 2014-2017, n = 3154, 45.0% men, 61.7 ± 9.9 years), and 3 (FU3, 2018-2021, n = 2394, 45.2% men, 65.0 ± 9.6 years) of the Colaus|PsyCoLaus study (Lausanne, Switzerland). Two MD scores (Trichopoulou, noted MD1, and Sofi, noted MD2) were calculated using participants' dietary data. T2D was defined as a fasting plasma glucose ≥7 mmol/L and/or the presence of an antidiabetic drug treatment. RESULTS: Participants with the highest MD adherence had a higher educational level, a lower BMI, were less frequent smokers, presented less frequently with hypertension, and were more frequent alcohol consumers. After multivariable adjustment, no differences were found between participants with and without T2D regarding MD scores: 3.93 ± 0.07 vs. 3.97 ± 0.02; 4.08 ± 0.10 vs. 3.98 ± 0.03, and 3.83 ± 0.11 vs. 3.97 ± 0.03, respectively, for the MD1 score in FU1, FU2, and FU3. In addition, no association was found between adherence to MD and T2D: odds ratio (and 95% confidence interval) for medium and high relative to low adherence to MD1: 0.87 (0.68-1.10) and 0.89 (0.64-1.24) in FU1, 1.04 (0.76-1.42) and 1.07 (0.68-1.67) in FU2, and 0.73 (0.53-1.03) and 0.61 (0.37-1.02) in FU3, respectively. Corresponding results for MD2 were 0.90 (0.70-1.15) and 1.03 (0.69-1.53) in FU1, 1.16 (0.82-1.63) and 1.40 (0.81-2.41) in FU2, and 0.93 (0.65-1.34) and 0.55 (0.28-1.08) in FU3. CONCLUSION: We found no association between Mediterranean diet adherence and T2D in a non-Mediterranean population.