Previous observational studies suggested that vitamin D may control the absorption of iron (Fe) by inhibition of hepcidin, but the causal relevance of these associations is uncertain. Using placebo-controlled randomisation, we assessed the effects of supplementation with vitamin D on biochemical markers of Fe status and erythropoiesis in community-dwelling older people living in the UK. The BEST-D trial, designed to establish the optimum dose of vitamin D3 for future trials, had 305 participants, aged 65 years or older, randomly allocated to 4000 IU vitamin D3 (n 102), 2000 IU vitamin D3 (n 102) or matching placebo (n 101). We estimated the effect of vitamin D allocation on plasma levels of hepcidin, soluble transferrin receptor (sTfR), ferritin, Fe, transferrin, saturated transferrin (TSAT%) and the sTfR–ferritin index. Despite increases in 25-hydroxy-vitamin D, neither dose had significant effects on biochemical markers of Fe status or erythropoiesis. Geometric mean concentrations were similar in vitamin D3 arms v. placebo for hepcidin (20·7 [se 0·90] v. 20·5 [1·21] ng/ml), sTfR (0·69 [0·010] v. 0·70 [0·015] µg/ml), ferritin (97·1 [2·81] v. 97·8 [4·10] µg/l) and sTfR–ferritin ratio (0·36 [0·006] v. 0·36 [0·009]), respectively, while arithmetic mean levels were similar for Fe (16·7 [0·38] v. 17·3 [0·54] µmol/l), transferrin (2·56 [0·014] v. 2·60 [0·021] g/dl) and TSAT% (26·5 [0·60] v. 27·5 [0·85]). The proportions of participants with ferritin < 15 µg/l and TSAT < 16 % were unaltered by vitamin D3 suggesting that 12 months of daily supplementation with moderately high doses of vitamin D3 are unlikely to alter the Fe status of older adults.