Vitamin D protects against cognitive decline in animals but evidence in humans has been inconsistent. Fibroblast growth factor 23 (FGF23) is a hormone that inhibits vitamin D activation yet few studies examined whether FGF23 is associated with cognitive impairment. The objective of this study was to examine associations of 25(OH)D and FGF23 with incident cognitive impairment in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study, a cohort of black and white adults ≥45 years old. FGF23 and 25(OH)D were measured in 474 incident impairment cases and 561 controls. In multivariable-adjusted models, there were no significant associations of FGF23 with incident cognitive impairment. In analyses using clinically-relevant categories of 25(OH)D (< 20 ng/ml, 20-29.9 ng/ml, ≥30 ng/ml), there was no statistically significant association of lower 25(OH)D concentrations with odds of incident cognitive impairment in models adjusted for demographic, clinical, and laboratory variables and season of blood draw (tertile 1 [≥30 ng/ml] reference; tertile 2 [20-29.9 ng/ml], odds ratio [OR] 0.96, 95%CI 0.67, 1.38; tertile 3 [<20 ng/ml] OR 1.26, 95%CI 0.83, 1.91). When 25(OH)D was modeled as race-specific tertiles, there were no significant associations of 25(OH)D with incident cognitive impairment in whites, whereas lower 25(OH)D was associated with higher odds in blacks (tertile 1 [>23 ng/ml] reference; tertile 2 [15-23 ng/ml], OR 2.96, 95%CI 1.48,5.94; tertile 3 [<15 ng/ml] OR 2.40, 95%CI 1.07,5.40) in the fully adjusted model. In this cohort of older adults, lower race-specific tertiles of 25(OH)D were associated with higher incidence of cognitive impairment in black individuals but not white individuals. These data suggest that treating low 25(OH)D may be a novel strategy for addressing racial disparities in neurocognitive outcomes.