Thousands of studies, still no answers. Here’s why, and what comes next
STUDY: Jorde R, Journal of Steroid Biochemistry and Molecular Biology 2026;256:106895
STUDY TYPE: Narrative review
FUNDING: Independent
Background
Vitamin D has generated over 110,000 PubMed publications, including massive randomized controlled trials enrolling up to 25,000 participants. And yet we still don’t know what blood level counts as deficient, how much supplementation people need, or whether vitamin D prevents anything beyond rickets and softening of the bones (osteomalacia).
The Study
This narrative review by a leading vitamin D research, Rolf Jorde, examines the problem.
- Giving vitamin D to people who don’t need it shows no effect, and almost every large trial enrolled people who were already vitamin D sufficient.
- Mean baseline 25-hydroxyvitamin D (25OHD) levels across the five largest trials ranged from 63 to 77 nmol/L, well above the highest deficiency threshold of 50 nmol/L. That cut-off rested on a single autopsy study with a non-standardized assay.
The 2024 Endocrine Society guidelines acknowledged this failure directly, withdrawing their previous sufficiency threshold of 75 nmol/L and conceding there’s no clinical trial evidence to define deficiency at all.
Despite this limitation, we do see some small signals with supplementation. These are from pooled analyses, with risk-reductions that are too small for individual trials to detect (eg, around 10-15%). Vitamin D may prevent:
- Transition from prediabetes to diabetes
- Autoimmune disease
- Cancer
Practice Implications
- Dr. Jorde’s analysis applies to the mental health studies as well. It’s difficult to see a clear benefit unless the levels are below 20 nmol/L, at which point supplementation makes sense to improve bone health.
- Supplementing vitamin D when the levels are above 30 nmol/L may help, or may not, we just don’t know.
— Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
