The effect of vitamin D supplementation on skeletal, vascular, or cancer outcomes: a trial sequential meta-analysis

Background Vitamin D insufficiency is associated with many disorders, leading to calls for widespread supplementation. Some investigators suggest that more clinical trials to test the effect of vitamin D on disorders are needed. Methods We did a trial sequential meta-analysis of existing randomised controlled trials of vitamin D supplements, with or without calcium, to investigate the possible effect of future trials on current knowledge. We estimated the effects of vitamin D supplementation on myocardial infarction or ischaemic heart disease, stroke or cerebrovascular disease, cancer, total fracture, hip fracture, and mortality in trial sequential analyses using a risk reduction threshold of 5% for mortality and 15% for other endpoints. Findings The effect estimate for vitamin D supplementation with or without calcium for myocardial infarction or ischaemic heart disease (nine trials, 48 647 patients), stroke or cerebrovascular disease (eight trials 46 431 patients), cancer (seven trials, 48 167 patients), and total fracture (22 trials, 76 497 patients) lay within the futility boundary, indicating that vitamin D supplementation does not alter the relative risk of any of these endpoints by 15% or more. Vitamin D supplementation alone did not reduce hip fracture by 15% or more (12 trials, 27 834 patients). Vitamin D co-administered with calcium reduced hip fracture in institutionalised individuals (two trials, 3853 patients) but did not alter the relative risk of hip fracture by 15% or more in community-dwelling individuals (seven trials, 46 237 patients). There is uncertainty as to whether vitamin D with or without calcium reduces the risk of death (38 trials, 81 173). Interpretation Our findings suggest that vitamin D supplementation with or without calcium does not reduce skeletal or non-skeletal outcomes in unselected community-dwelling individuals by more than 15%. Future trials with similar designs are unlikely to alter these conclusions.