Vitamin D has been intensively studied for its association with human health, but the scope of such association and the causal role of vitamin D remain controversial. We aim to comprehensively investigate the links between vitamin D and human health through both epidemiological and Mendelian randomization (MR) analyses. We examined the epidemiological associations between serum 25-hydroxyvitamin D (25(OH)D) concentration and 90 diseases/traits in 326,409 UK Biobank (UKBB) Europeans. The causal relations between 25(OH)D and 106 diseases/traits were investigated by performing MR analysis using genome-wide significant 25(OH)D-associated variants (N = 143) from the largest UKBB GWAS to date. In epidemiological analysis, we found 25(OH)D was associated with 45 diseases/traits across cardiovascular/metabolic diseases, psychiatric/neurological diseases, autoimmune/inflammatory diseases, cancer, musculoskeletal diseases, and quantitative traits. In MR-analysis, we presented evidence suggesting potential causal role of 25(OH)D in increasing height (β = .064, 95% confidence interval [CI] = 0.019-0.11) and preventing the risk of ovarian cancer (odds ratio [OR] = 0.96, 95% CI = 0.93-0.99), multiple sclerosis (OR = 0.96, 95% CI = 0.94-0.98), leg fracture (OR = 0.60, 95% CI = 0.45-0.80) and femur fracture (OR = 0.53, 95% CI = 0.32-0.84). These findings confirmed associations of vitamin D with a broad spectrum of diseases/traits and supported the potential causal role of vitamin D in promoting health.
Keywords: Mendelian randomization; epidemiological association; vitamin D.
© 2020 Wiley Periodicals LLC.