Vitamin D supplements have long been recommended for older people (>65 years) to treat or prevent osteoporosis,1 with some early evidence suggesting benefits for musculoskeletal health, including increasing bone mineral density and preventing falls and fractures.2 However, more recent systematic reviews have reported no effect of vitamin D supplementation on bone mineral density,3 falls,4, 5, 6, 7 or fractures.7, 8, 9, 10 Findings from our trial sequential analyses6, 9 testing the hypothesis of a 15% relative risk reduction in falls or fractures showed that doing further trials of vitamin D, with or without calcium supplementation, which are similar to existing trials, are unlikely to alter the conclusion of these recent systematic reviews. However, correspondents questioned the use of this efficacy threshold and also suggested that inadequate vitamin D doses might explain these null results,11, 12 although some randomised controlled trials have reported increased risk of falls or fractures with high-dose intermittent vitamin D.13, 14, 15 Since the last major systematic reviews of vitamin D and musculoskeletal health were published in 2012–17,3, 4, 5, 6, 7, 8, 9, 10 45 randomised controlled trials of vitamin D monotherapy (n=20131) have reported on bone mineral density, falls, and fractures, increasing the number of trial participants with these outcomes by 40–85%. Most new trials have also used substantially higher doses of vitamin D than earlier trials. Consequently, the currently available set of randomised controlled trials has much greater power for meta-analysis and trial sequential analysis, and allows a detailed exploration of potentially important clinical factors in subgroup analyses, including comparisons of high and low doses of vitamin D. A comprehensive update of previous systematic reviews, meta-analyses, and trial sequential analyses, which includes the key clinical and major surrogate endpoints, is warranted. An advantage of assessing these outcomes concurrently is that an effect might be found for some endpoints whereas no effect is found for others, which could have clinical and biological relevance. Trial sequential analyses of vitamin D and bone mineral density have also not been reported previously.
Research in context
Evidence before this study
We used findings from literature searches in previously published meta-analyses, which we updated by searching PubMed, Embase, and Cochrane Central on Sept 14, 2017, and Feb 26, 2018, using the search term “vitamin D”, without any language restrictions. A full list of keywords is shown in the appendix. Evidence from older systematic reviews suggested vitamin D supplements might have benefits for musculoskeletal health, but more recent systematic reviews have reported no effect of vitamin D supplementation on fractures, falls, or bone mineral density. Some authors have suggested that inadequate vitamin D doses might explain these null results. At least 30 trials of vitamin D have been published since these systematic reviews, which nearly doubles the available trial results for vitamin D for these outcomes.
Added value of this study
Our meta-analyses and trial sequential analyses show that in a large number of clinical trials, vitamin D supplementation does not have clinically relevant effects on fractures, falls, and bone mineral density, and this conclusion is unlikely to be altered by future trials with similar designs. Effects of high doses of vitamin D were similar to effects of low doses, and none of the other potential modifiers of vitamin D effects were found to influence efficacy for any outcome.
Implications of all the available evidence
There is little justification for the use of vitamin D supplements to maintain or improve musculoskeletal health (except for the prevention or treatment of rickets and osteomalacia in high-risk groups), and clinical guidelines should reflect these conclusions.
Vitamin D supplements have often been co-administered with calcium supplements. Recent systematic reviews have suggested that the evidence for benefits of calcium supplements in preventing fractures, with or without vitamin D, is weak and inconsistent,10, 16 with any effect on bone mineral density or fracture likely to be small and of doubtful clinical relevance.10, 16, 17 Additionally, uncommon but important side-effects of calcium supplements18, 19, 20, 21 have been identified, which contribute to an unfavourable risk–benefit profile. No large trials of co-administered calcium and vitamin D supplements have become available with fracture or falls as the primary endpoint since the previous systematic reviews.
We did this systematic review, meta-analyses, and trial sequential analyses of randomised controlled trials in adults to study the effect of vitamin D supplements on the clinical musculoskeletal outcomes of fractures and falls, and the commonly used surrogate endpoint of bone mineral density. To align with recent findings on calcium supplements, and the recent design of vitamin D randomised controlled trials, we focused on randomised controlled trials that used vitamin D as monotherapy, and included randomised controlled trials that compared high doses of vitamin D with low doses.