In 15,427 Czech and Polish adults aged 45-69 followed for about 20 years (3,792 deaths), higher MIND diet adherence initially tracked with lower all-cause and cardiovascular mortality and better cognition after adjusting for age and sex. Full adjustment erased the mortality signal: high versus low adherence gave HR 1.01 (95% CI 0.93-1.10) for all-cause and HR 1.01 (0.88-1.16) for CVD death. Among cognitive domains, only verbal fluency stayed significantly better (coefficient 0.058, p = 0.015).
The attenuation is the story. Age-and-sex-adjusted associations for healthy dietary patterns often shrink once education, smoking, alcohol, and physical activity enter the model, which suggests MIND scores partly mark a broader healthy-living profile rather than carrying independent protection. That fits the weakest points of this design: a single baseline food frequency questionnaire used across two decades, which blurs exposure as diets change, and cognition measured only cross-sectionally at baseline, so no decline was tracked, even though MIND was built to slow decline. The population also differs from the US cohorts where MIND was derived; its berry and leafy green components may map poorly onto Central and Eastern European eating habits, and mortality there is shaped heavily by alcohol patterns and cardiovascular risk. Residual confounding can cut either way, so the null should not be read as proof of no effect. Editorially, this is a useful corrective rather than a reversal. It tempers enthusiasm for MIND as a mortality tool, while leaving its modest cognitive association plausible and its component foods sensible to eat anyway.