Abstract
This cross-sectional analysis provides up-to-date information about dietary patterns (DP) and their sociodemographic correlates in European elderly individuals. We studied 6686 enrollees aged 65+ (55% women) in the ongoing French population-based NutriNet-Santé e-cohort. Diet was assessed via three 24 h records. The sex-specific correlates of factor analysis derived DP were identified with multivariable linear regression. Using 22 pre-defined food groups, three DP were extracted. The "healthy" DP (fruit, vegetables, grains, nuts, fish) was positively associated with education, living alone, and being a former smoker (women), and negatively associated with being overweight, current smoker (men), age 75+ years, having hypertension, and obesity (women). The "western" DP (meat, appetizers, cheese, alcohol) was positively associated with BMI (men) and being a former/current smoker; it was negatively associated with age 75+ years (women) and living alone. The "traditional" DP (bread, potatoes, milk, vegetables, butter, stock) was positively associated with age and negatively associated with being a former/current smoker, education (men), and residing in an urban/semi-urban area. The findings support the diversity of DP among the elderly, highlighting sex-specific differences. The "healthy" DP explained the largest amount of variance in intake. Future studies could replicate the models in longitudinal and international contexts.