Plant-Based Diets and Cognitive Outcomes: A Systematic Review and Meta-analysis

Although plant-rich dietary patterns like the Mediterranean and Mediterranean-DASH Intervention for Neurodegenerative Delay diets have been linked to cognitive benefits, the role of predominantly plant-based diets is less understood. This systematic review aimed to evaluate the association between plant-based diets and cognitive outcomes. A literature search was conducted in Medline and Embase using keywords related to plant-based diets (e.g., "vegetarian diet") and cognitive outcomes (e.g., "dementia"). Studies of any design were eligible. Reviewers independently screened studies, extracted data, and assessed quality using the Newcastle-Ottawa Scale. Meta-analyses were conducted on prospective studies that examined the same dietary exposure and cognitive outcome, using fixed-effects regression models. Twenty-two studies were included, with considerable variability in methodologies and outcomes. Plant-based diets were defined either categorically (e.g., vegetarian compared with nonvegetarian), or using indices of adherence, such as the healthful plant-based diet index (hPDI), with higher scores reflecting higher adherence. Two meta-analyses, each based on 2 high-quality prospective cohort studies, examined associations between plant-based diet indices and cognitive outcomes. For cognitive impairment, pooled odds ratios (95% confidence interval) for highest compared with lowest quartiles were 0.61 (0.55, 0.68; I² = 97.1%) for plant-based diet index (PDI) and 0.68 (0.62, 0.75; I² = 84.3%) for hPDI. For dementia, pooled hazard ratios were 1.03 (0.91, 1.17; I² = 0%) for PDI, 0.85 (0.75, 0.97; I² = 0%) for hPDI, and 1.17 (1.03, 1.33; I² = 60.3%) for unhealthful PDI. These findings suggest that dietary patterns emphasizing healthful plant-based foods and limiting less healthful plant foods and animal products are associated with lower odds of cognitive impairment and risk of dementia. However, findings across individual studies were inconsistent, highlighting the need for further high-quality research. This review was registered at PROSPERO as CRD42022380055.