Background: Diabetes is a known risk factor for impaired cognitive function. Higher adherence to healthy dietary patterns, such as the Dietary Approaches to Stop Hypertension (DASH), is associated with lower risk of cognitive impairment in healthy individuals. Whether the association of DASH with cognitive function differs by diabetes status is not well known. Objectives: In the German National Cohort (NAKO), we assessed the cross-sectional association of adherence to DASH with cognitive performance in individuals with and without diabetes. Methods: The study sample comprised n=5,986 individuals with diabetes and n=93,128 individuals without diabetes from the NAKO baseline examination (51.3% women, age 48.7 (±12.2) years). Dietary intake was assessed with a multiple source method combining up to four 24h dietary recalls with a self-administered, validated Food Frequency Questionnaire. Cognitive performance was assessed by a brief cognitive test battery (semantic fluency, 12-word list recall task (verbal memory), Stroop test (cognitive interference), and digit span backwards (working memory capacity)). Linear regression models stratified by diabetes status were used to relate DASH (effects per 5-point increment) to cognitive performance (as 1-point increments), adjusting for age, sex, body mass index, smoking, socioeconomic status, physical activity, alcohol, daily energy intake, German language ability, and prevalence of neurological or psychiatric diseases. Results: In multivariable-adjusted models, better adherence to DASH was associated with better performance in semantic fluency (ß Diabetes : 0.38 95% CI: 0.21; 0.54, ß No Diabetes : 0.34 95% CI: 0.31; 0.38) and verbal memory (ß Diabetes : 0.06 95% CI: 0.02; 0.11; ß No Diabetes : 0.05 95% CI: 0.04; 0.06) irrespective of diabetes status. By contrast, better performance in cognitive interference with better adherence to DASH was only observed in individuals without diabetes (ß Diabetes : 0.01 95% CI: -0.36; 0.38, ß No Diabetes : -0.28 95% CI: -0.34; -0.22, p Interaction > 0.05 ). For working memory capacity, no clear association with DASH adherence was observed (ß Diabetes : 0.01 95% CI: -0.02; 0.04, ß No Diabetes : 0.01 95% CI: 0.00; 0.02). Conclusion: Better adherence to the DASH diet was associated with better semantic fluency and verbal memory in individuals irrespective of diabetes status, while the protective association of DASH with cognitive interference was only seen in individuals without diabetes.
Stuermer et al. (2026) studied this question.