The Mediterranean diet, characterized by high consumption of fruits, vegetables, whole grains, legumes, nuts, fish, and moderate alcohol intake, has consistently shown a decreased risk of diabetes in Mediterranean and European populations. However, studies in U.S. populations have yielded inconsistent results. This inconsistency may be due to the racial and ethnic heterogeneity, higher body mass indexes (BMI), sedentary lifestyles, and different cultural emphasis on diet and lifestyle compared to the Mediterranean populations. This study aimed to assess the association between a Mediterranean-style eating pattern and incident diabetes in a large, diverse U.S. cohort, exploring potential variations across racial and BMI subgroups. Understanding this association is crucial for developing effective diabetes prevention strategies tailored to the U.S. population, which has a high prevalence of diabetes and significant racial disparities in risk.

Numerous observational and experimental studies have demonstrated the positive impact of a Mediterranean-style eating pattern on cardiovascular health, reducing the risk of cardiovascular events and mortality. The American Diabetes Association also recommends this diet as a strategy to prevent cardiovascular complications in individuals with diabetes. However, evidence regarding its role in diabetes prevention among U.S. adults is limited and inconsistent. Previous U.S. studies have shown varying associations between Mediterranean diet adherence and diabetes risk, with some showing a protective effect and others showing no association. These inconsistencies highlight the need for further research in diverse U.S. populations to determine the generalizability of the benefits observed in Mediterranean and European cohorts.

This prospective cohort study utilized data from the Atherosclerosis Risk in Communities (ARIC) study, a large, population-based study that originally enrolled 15,792 middle-aged adults (45–65 years) from four U.S. communities. A total of 11,991 participants without prevalent cardiovascular disease, diabetes, or cancer at baseline (1987–1989) were included in the analysis. Dietary intake was assessed using a 66-item semi-quantitative food frequency questionnaire (FFQ) administered at visits 1 and 3. Alternate Mediterranean Diet (aMed) scores were calculated based on the self-reported dietary intake, with higher scores indicating greater adherence to a Mediterranean-style pattern. Participants were followed from visit 1 through December 31, 2016, for incident diabetes, which was ascertained through self-reported physician diagnosis, medication use, and blood glucose measurements. Cox proportional hazards regression models were used to analyze the association between aMed scores and incident diabetes, adjusted for various covariates, including energy intake, age, sex, race, study center, education, smoking status, physical activity, fasting glucose, hypertension status, LDL cholesterol, and BMI. Interaction terms were included to assess potential effect modification by race and BMI. The proportional hazards assumption was checked via log-log plots and Schoenfeld’s residual tests. Three covariate structures were used in the Cox regression analyses (Model 1: adjusted for energy intake, and socio-demographic factors; Model 2: Model 1 plus behavioral variables associated with diabetes risk; and Model 3: Model 2 plus clinical mediators of diabetes).

Over a median follow-up of 22 years, there were 4,024 incident cases of diabetes. Higher aMed scores were significantly associated with a lower risk of incident diabetes in Model 1 [HR (95% CI) for quintile 5 vs. 1: 0.83 (0.73–0.94); p-trend < 0.001], indicating a 17% reduction in risk. This association was attenuated but remained statistically significant after adjusting for behavioral factors (Model 2) and clinical mediators (Model 3), although the strength of association decreased. A one-point increase in the aMed score was associated with a 4% decrease in diabetes risk across all three models. The association between aMed score and diabetes risk was stronger among Black participants (interaction p < 0.001) compared to white participants and weaker among obese individuals compared to those with normal BMI (interaction p < 0.01). Higher intakes of nuts, legumes, and moderate alcohol consumption, along with lower red and processed meat consumption were associated with lower risk of incident diabetes.

This study provides further evidence supporting the potential benefits of a Mediterranean-style eating pattern for diabetes prevention in a U.S. population. The findings highlight the importance of considering racial and BMI subgroups when assessing dietary patterns' impact on health outcomes. The stronger association in Black participants suggests potential benefits of targeted interventions within this high-risk group. The attenuation of the association in obese individuals underscores the critical role of weight management in diabetes prevention. The inconsistent findings across different models suggest that the observed association between Mediterranean diet adherence and diabetes risk may be partially mediated by several modifiable risk factors. Therefore, interventions targeting multiple modifiable risk factors simultaneously might yield greater effectiveness in diabetes prevention.

A Mediterranean-style eating pattern, rich in fruits, vegetables, whole grains, legumes, nuts, and fish, and moderate in alcohol, was associated with a reduced risk of diabetes, especially among Black and normal-weight individuals. The study emphasizes the need for tailored interventions promoting healthy eating habits within high-risk populations and highlights the critical role of weight management in diabetes prevention. Future research should investigate the effectiveness of a calorically restrictive Mediterranean-style diet, leading to clinically meaningful weight loss, in overweight and obese individuals.

This study's reliance on self-reported dietary data through FFQ might introduce measurement error and potential bias. The aMed score used didn't fully capture all aspects of a traditional Mediterranean diet, particularly olive oil consumption. The study's cross-sectional nature of dietary assessment at baseline limits the ability to determine causality. Residual confounding may still exist despite adjustments for several covariates. The lack of information on aspects of Mediterranean lifestyle beyond diet limits a complete understanding of the observed association.