Scientists have found little to justify using vitamin D supplements to maintain or improve musculoskeletal health, according to a study.
Taking the supplement would be useful only for high-risk groups who want to prevent rare conditions such as rickets and osteomalacia, which can occur due to vitamin D deficiency, researchers say.
A study, published in the Lancet Diabetes and Endocrinology journal, also found no differences in the effects of higher versus lower doses of vitamin D.
Vitamin D helps regulate the amount of calcium and phosphate in the body, which are needed to keep bones, teeth and muscles healthy.
The British Department of Health recommends all children under five should take a daily supplement, along with pregnant and breastfeeding women, and older people who are not often outdoors - for instance, if they are frail, housebound or live in a care home.
It suggests other adults should take a supplement during the autumn and winter, when they are less likely to get sunlight on their skin.
The study authors said clinical guidelines that recommend vitamin D supplementation for bone health should be changed to reflect the best available evidence.
They said their study was the largest meta-analysis ever carried out, with data from 81 randomised controlled trials.
Lead author Dr Mark Bolland, of the University of Auckland said: "Since the last major review of evidence in 2014, more than 30 randomised controlled trials on vitamin D and bone health have been published, nearly doubling the evidence base available.
"Our meta-analysis finds that vitamin D does not prevent fractures, falls or improve bone mineral density, whether at high or low dose.
"Clinical guidelines should be changed to reflect these findings.
"On the strength of existing evidence, we believe there is little justification for more trials of vitamin D supplements looking at musculoskeletal outcomes."
In secondary analyses looking at bone density, there were small differences for lumbar spine, femoral neck and for total body but none of these were clinically relevant.
