Relationship Between Serum Vitamins and Cognitive Impairment in the Elderly: A Study Based on the NHANES Database.

BACKGROUND: Vitamins, as a modifiable lifestyle factor, are increasingly recognized for their protective role in cognitive health. However, the synergy or interaction among vitamins remains unclear. OBJECTIVES: This work aimed to analyze the association of serum vitamin D, folic acid (FA), and vitamin B12 with cognitive impairment in the elderly. METHODS: Data included 2582 elderly participants aged 60 and older from the NHANES, 2011-2014. Weighted logistic regression, Bayesian kernel machine regression (BKMR), and weighted quantile sum (WQS) models were used to analyze the association of serum vitamins and their mixture with cognitive disorder risk. Interaction effects among vitamins were investigated. Sensitivity analyses accounting for vitamin supplements, depression, and sleep disorders, and subgroup analyses focusing on high vitamin B12 levels were performed. RESULTS: After adjusting for confounding factors, serum Vitamin D (OR = 0.695, 95% CI: 0.534-0.905, p = 0.003) and FA (OR = 0.777, 95% CI: 0.604-0.999, p = 0.034) levels were inversely correlated with cognitive disorder risk. Both remained robust after considering vitamin supplements and comorbidities in the sensitivity analyses. The BKMR model indicated a significant increase in cognitive impairment risk when the overall vitamin mixture level fell below the 50th percentile. A U-shaped association was detected between vitamin B12 and cognitive disorder risk. Vitamin B12 and FA had potential interaction effects. The WQS model revealed the largest contribution by FA (56.0%) to the overall protective effect on the cognitive disorder risk. The association with high vitamin B12 was predominantly observed in individuals with specific metabolic conditions, including kidney stones and hypertension. CONCLUSION: Optimizing vitamin D and FA levels remains the key to reducing cognitive impairment risk in FA-supplemented populations. Vitamin B12 management requires greater precision; its high level in specific metabolic patients may signal health risks. This study provides new evidence for precise nutritional intervention for the elderly.