The Bone-Health Gospel
Your doctor has probably told you to take vitamin D, and if you're over fifty, that advice likely came with a specific warning: without supplementation, your bones will weaken and your fracture risk will climb. The U.S. Preventive Services Task Force, the Endocrine Society, and the Institute of Medicine have all recommended vitamin D for skeletal health at various points, and roughly 40% of American adults now take a supplement. The global vitamin D market crossed $5 billion in 2025, with bone and joint health accounting for nearly 40% of those sales.
The underlying logic feels airtight: vitamin D helps the body absorb calcium, calcium builds bone, and therefore supplementing vitamin D should make bones stronger and fractures rarer. It's a clean syllogism, the kind that sounds too obvious to test, yet a team at Harvard tested it anyway in the largest randomized trial of vitamin D and fractures ever conducted, and the syllogism broke.
What VITAL Actually Found
The Vitamin D and Omega-3 Trial, known as VITAL, enrolled 25,871 men aged fifty and older and women aged fifty-five and older from across all fifty U.S. states. Participants were randomized to receive 2,000 IU of vitamin D3 per day or placebo. Critically, participants were not recruited because they had osteoporosis, low bone mass, or vitamin D deficiency. This was a test of the general-population recommendation: the advice given to millions of healthy adults.
Over a median follow-up of 5.3 years, 1,991 fractures occurred in 1,551 participants, and Meryl LeBoff and colleagues at Brigham and Women's Hospital tallied every one, confirming each through centralized medical-record review.
The results landed in the New England Journal of Medicine in July 2022 with numbers that weren't just negative but emphatically so: total fractures showed a hazard ratio of 0.98, nonvertebral fractures came in at 0.97, and hip fractures registered 1.01, meaning not one outcome reached statistical significance. The P-values ranged from 0.50 to 0.96, so far from the 0.05 threshold that the numbers weren't even suggestive, and no subgroup defined by age, sex, race, BMI, or baseline vitamin D levels showed a benefit.
"Our results do not support the use of vitamin D supplements to prevent fractures in generally healthy U.S. men and women," LeBoff told Medscape.
Eighty-One Trials Say the Same Thing
If the VITAL result were an outlier, it could be set aside, but it isn't: in 2018, Mark Bolland and colleagues at the University of Auckland published the largest meta-analysis of vitamin D and musculoskeletal health ever assembled, pooling 81 randomized controlled trials enrolling 53,537 participants and running both conventional meta-analyses and trial sequential analyses, a method that estimates whether the accumulated evidence is large enough to draw reliable conclusions.
Vitamin D supplementation had no effect on total fracture (relative risk 1.00, 95% CI 0.93–1.07), hip fracture (RR 1.11, 0.97–1.26), or falls (RR 0.97, 0.93–1.02), and higher doses performed no better than lower ones. The trial sequential analysis went further, finding not only that the evidence was conclusive but that future trials of similar design were unlikely to change the conclusion, because the sheer weight of accumulated null results had already crossed the threshold for futility. A separate 2017 JAMA meta-analysis by Zhao and colleagues, drawing on 51,145 participants, reached the same verdict: calcium, vitamin D, or both did not significantly reduce fracture incidence in community-dwelling older adults.
Where the Confidence Interval Splits
The strongest challenge to these conclusions comes from a nuance in how the evidence was compiled. Bolland's meta-analysis analyzed vitamin D alone. When vitamin D is combined with calcium, the picture shifts: a subanalysis found the combination reduced hip fractures by 16% (RR 0.84, 95% CI 0.74–0.96). Critics also noted Bolland excluded roughly 40% of high-quality trials testing vitamin D plus calcium. A re-analysis that included only studies using the recommended dosage of 800–1,000 IU daily with adherence above 50%, while excluding large trials of infrequent bolus dosing, found a significant 14% reduction in total fractures (RR 0.86, 0.75–0.98).
These objections aren't trivial. They suggest that vitamin D may have modest skeletal benefit when taken at the right dose, consistently, alongside calcium. But the objections also underscore how narrow the remaining window of benefit is: the effect appears only in specific analytical subsets, and it disappears in the largest, most rigorously designed trials that didn't preselect for deficiency. A 14% relative reduction in fractures among compliant users of a specific dose range, drawn from a re-analysis of a meta-analysis, is a genuine statistical finding. Whether it should sustain a $5 billion industry and near-universal clinical recommendation is a different question.
What We Didn't Prove
Neither VITAL nor the Bolland meta-analysis tested people with severe vitamin D deficiency, diagnosed osteoporosis, or those living in residential care, and LeBoff stated this explicitly: "These results do not apply to older adults in residential communities or those with severe vitamin D deficiency, low bone mass or osteoporosis." In institutionalized populations with documented deficiency, vitamin D supplementation to prevent rickets and osteomalacia remains appropriate and uncontested.
The VITAL trial also allowed participants to continue taking up to 800 IU of vitamin D on their own, and since 43% did, the trial effectively compared 2,800 IU (supplement plus personal use) against 800 IU (personal use plus placebo) rather than against zero, meaning that if vitamin D's benefit plateaus at low doses, the trial design could have obscured a real effect at the low end of the intake curve. The meta-analyses also cannot speak to vitamin D's non-skeletal effects, since evidence for its role in immune function, cancer prevention, and cardiovascular health is separate, contested, and still evolving. VITAL's parent trial found a 17% reduction in cancer mortality that emerged after two years, though the pre-specified primary cancer endpoint showed no benefit, which means dismissing all vitamin D supplementation based on the fracture data alone would be overreach.
The Bottom Line
For generally healthy adults with adequate dietary intake and any sun exposure, the evidence no longer supports taking vitamin D supplements to prevent broken bones. The largest randomized trial ever conducted found a hazard ratio of 0.98 for total fractures. A meta-analysis of 81 trials confirmed the null. The conventional wisdom persists because the biological logic is sound, because earlier observational studies showed correlations, and because a $5 billion industry has no incentive to update the narrative. The bones of the argument are solid. The bones of the evidence are not.
What You Can Do
Stop taking vitamin D solely to prevent fractures unless your doctor has documented a deficiency, you have osteoporosis, or you live in a care facility, because for the general healthy population, the evidence says the pills don't reduce fracture risk.
If you do take vitamin D, pair it with calcium, since the combination retains some evidence of modest hip-fracture reduction (16% in pooled analyses) that vitamin D alone does not show, and aim for 800–1,000 IU of vitamin D daily with at least 1,000 mg of dietary or supplemental calcium. Get your vitamin D level tested before supplementing, not after: blood levels of 25-hydroxyvitamin D below 20 ng/mL indicate true deficiency that warrants treatment, but above that threshold, supplementation hasn't demonstrated skeletal benefit in randomized trials.
Prioritize weight-bearing exercise over supplements, because walking, jogging, stair climbing, and resistance training have consistent evidence for maintaining bone density and reducing fall risk, and a 2019 Cochrane review found exercise reduced fall-related fractures by 20%, an effect no vitamin D trial has matched in the general population.