We are heartened by the correspondents' endorsement of the main point of our editorial, that randomized controlled trials (RCTs) are required to investigate the health benefits of vitamin D. We do not agree, however, that there is convincing existing evidence that vitamin D improves the range of disease outcomes they list. For example, the oft-repeated claim that vitamin D prevents cancer is based on an analysis of the secondary end point of a trial that included only 50 events, did not include a vitamin D treatment group, and found no difference in the effect of calcium and vitamin D compared with calcium alone.1,2 As we indicated, many RCTs with surrogate end points have not demonstrated efficacy of vitamin D. If, as the correspondents suggest, this is because insufficient vitamin D was administered rather than that the intervention is ineffective, trials need to be conducted that evaluate “adequate” doses. Such trials should also address the safety of the intervention, of which hypercalcemia is only 1 component. Recent evidence that falls and fractures are increased by annual dosing of vitamin D that raises levels of serum 25(OH)D into the range advocated by some authors3 suggests that there are levels of 25(OH)D above which adverse health effects occur, as is the case for other hormones such as cortisol and thyroxine.
Grey A, Bolland M. Vitamin D Supplementation and Fracture Risk—Reply. Arch Intern Med. 2011;171(3):265-266. doi:10.1001/archinternmed.2010.532