Background  A role for vitamin D deficiency in Parkinson disease (PD) has recently been proposed.

Objective  To compare the prevalence of vitamin D deficiency in a research database cohort of patients with PD with the prevalence in age-matched healthy controls and patients with Alzheimer disease (AD).

Design  Survey study and blinded comparison of plasma 25-hydroxyvitamin D (25[OH]D) concentrations of stored samples in a clinical research database at Emory University School of Medicine.

Setting  Referral center (PD and AD patients), primary care clinics, and community setting (controls).

Participants  Participants were recruited into the study between May 1992 and March 2007. Every fifth consecutively enrolled PD patient was selected from the clinical research database. Unrelated AD (n = 97) and control (n = 99) participants were randomly selected from the database after matching for age, sex, race, APOE genotype, and geographic location.

Main Outcome Measures  Prevalence of suboptimal vitamin D and mean 25(OH)D concentrations.

Results  Significantly more patients with PD (55%) had insufficient vitamin D than did controls (36%) or patients with AD (41%; P = .02, ²test). The mean (SD) 25(OH)D concentration in the PD cohort was significantly lower than in the AD and control cohorts (31.9 [13.6] ng/mL vs 34.8 [15.4] ng/mL and 37.0 [14.5] ng/mL, respectively; P = .03).

Conclusions  This report of 25(OH)D concentrations in a predominantly white PD cohort demonstrates a significantly higher prevalence of hypovitaminosis in PD vs both healthy controls and patients with AD. These data support a possible role of vitamin D insufficiency in PD. Further studies are needed to determine the factors contributing to these differences and elucidate the potential role of vitamin D in pathogenesis and clinical course of PD.