Vitamin D3 for Life

Live Enlightened

Physiology

Vitamin D in Early Development: Programming and Later Outcomes: Full Evidence Review

Previous related pageVitamin D in Early Development: Programming and Later Outcomes: Study SummaryNext related pageVitamin D in Early Development: Programming and Later Outcomes: Study Summary
Study at a glance

What was studied?

Researchers studied vitamin D exposure before birth and later development. Human trials measured growth, bone scans and fractures. Rat and mouse experiments examined changes in pancreatic cells and developing blood and immune cells.

What did the studies find?

Changing supplements raised blood measurements in human trials, but did not consistently improve later outcomes. Some childhood bone scans were favourable; other growth and bone outcomes showed no clear difference. Animal experiments found changes in particular tissues or cell populations.

What does this mean?

A blood change, an animal cell finding and a benefit to a child are different results. An effect measured at one age does not prove that it lasts throughout life.

What are the limits?

We could not obtain the full 2026 MAVIDOS report on small-scale bone structure. Its abstract reports no detected group difference, but we could not fully assess its methods or results. Other limits include selected follow-up groups and differences between animal experiments and people.

Read the full article below ↓

Vitamin D exposure before birth may influence development, but the meaning of that influence depends on what was measured. Placental handling, an experimental cell phenotype, a child’s bone scan and a later fracture are different findings. This review compares five independent study programmes addressing developmental programming and offspring outcomes, with placental physiology as supporting context. It asks which effects persist, which are demonstrated in people, and where evidence remains incomplete. For the shorter account, read Vitamin D in Early Development: Programming and Later Outcomes: Study Summary.

From maternal supply to developmental exposure

Vitamin D3, the substance used in many supplements, differs from 25-hydroxyvitamin D, commonly written as 25(OH)D. The latter is the circulating substance generally measured when assessing vitamin D status. It also differs from 1,25-dihydroxyvitamin D, the active hormone. A higher maternal blood concentration does not directly measure receptor activity in a fetal organ or establish a future clinical benefit. The separate blood markers are explained in Vitamin D Testing: 25-OH-D vs 1,25-OH₂-D, while Vitamin D Activation describes conversion between vitamin D forms.

Ashley and colleagues studied vitamin D handling in human placental tissue. Their experiments included perfusion of five term placentas, tissue experiments and an observational gene-expression analysis. The findings support active placental uptake and handling rather than a model in which a maternal blood result simply passes unchanged to the fetus. However, labelled active hormone was detectable in only one placenta. This is not robust evidence of identical conversion across all five preparations. Some tissue experiments used exposures that should not be equated with ordinary maternal supplementation.

This provides physiological context, not a trial of early pregnancy development. A term placenta examined outside the body cannot represent every stage of pregnancy. Observational gene-expression findings cannot establish that changing a mother’s supplement would cause a particular childhood outcome. Plausible mechanisms must not be used to imply benefits that the experiments did not measure. The general conversion and inactivation framework is set out in Vitamin D Enzyme Pathways (CYP27B1 / CYP24A1). Vitamin D Receptors explains the cellular response machinery; neither overview establishes a benefit from supplementation during pregnancy.

What developmental programming means here

Developmental programming means that an early exposure is associated with a later change in structure, regulation or function. It does not automatically mean an irreversible change, a specific epigenetic mechanism or a future disease. An experiment may show persistence after an exposure ends without demonstrating persistence throughout life. Vitamin D and Lasting Gene Regulation: Full Evidence Review examines the distinction between a lasting molecular response and a measurement made during continuing exposure.

Human and animal studies make different contributions. Randomized pregnancy trials can compare assigned exposures and later outcomes, but follow-up may be selective and the original trial may not have been designed for each later measurement. Animal experiments can examine mechanisms more closely, but their diets, timing and procedures may differ substantially from human circumstances. A mechanistic result alone cannot establish a preventive treatment for people. Vitamin D, Cell Fate and Tissue Structure: Full Evidence Review considers how cell maturation and tissue measurements differ; those experiments provide context rather than evidence for a pregnancy treatment.

For current UK guidance, see NHS Pregnancy vitamins and supplements. Trial findings do not establish a personal supplement dose or blood target.

Changed status without improved infant growth

The Maternal Vitamin D for Infant Growth trial, or MDIG, randomized 1,300 pregnant women in Dhaka, Bangladesh. Supplementation began at 17 to 24 weeks of pregnancy. Five groups compared different prenatal regimens, with one also continuing assigned supplementation after birth. All groups received calcium, iron and folate. That shared care is part of the experiment and matters when applying its findings elsewhere.

The intervention raised maternal, cord and infant 25(OH)D concentrations and reduced maternal parathyroid hormone, a hormone involved in mineral regulation. Nevertheless, the trial did not detect improvement in the principal infant length-for-age outcome at one year. Participants could respond in blood measurements without a corresponding improvement in measured growth. A biochemical response and developmental benefit are not interchangeable.

This does not establish that vitamin D is unnecessary during development. The comparison concerned particular strategies beginning during pregnancy, with shared calcium and other care. It did not test every earlier exposure window, nutritional combination or developmental endpoint. A trial of infant length cannot answer every question about later bone, immunity or adult metabolic disease.

A published correction addressed coding of maternal parity and the recorded sex of 14 infants. The corrected report did not change the overall conclusion. This review uses the corrected interpretation and avoids reconstructing detailed subgroup or safety claims from incomplete supplementary access. Relevant corrections form part of the publication record.

MAVIDOS: different ages and bone measurements

The UK MAVIDOS trial randomized 1,134 pregnant women to vitamin D3 or placebo beginning at 14 to 17 weeks. Participants had baseline 25(OH)D within a specified range, excluding severe deficiency. The primary neonatal bone-mineral-content analysis included 665 infants with usable measurements. It did not show a statistically clear overall benefit.

A favourable result among winter births was secondary. Several interactions were examined, so the seasonal finding needs that context. It should not replace the overall result or become a recommendation based solely on season of birth. Motion, missing scans and differences between participants with and without usable measurements also affect interpretation.

A later MAVIDOS follow-up assessed children at approximately six to seven years. In this selected Southampton sample, 447 children had usable scans and 384 entered the fully adjusted analysis. Investigators reported modest favourable differences in several whole-body-less-head bone measurements. These exclude the skull, whose contribution can complicate interpretation of a growing skeleton. Results were not identical for every skeletal site or adjustment model.

This is a meaningful positive finding, but the follow-up was exploratory and not specified in the original trial’s analysis plan. Families who attended differed from those who did not. Some behaviours after birth, including breastfeeding duration and child supplement use, also differed between groups. Participants and assessors remained blinded, yet blinding cannot remove every consequence of selective follow-up. Multiple related outcomes were assessed without adjustment for multiple comparisons.

The conclusion is therefore more specific than either calling the whole trial negative or claiming permanent stronger bones. The neonatal outcome was not clearly improved overall; a selected later sample showed favourable scan findings. These reports did not demonstrate adult benefit or fracture prevention. Estimated bone density is also not the same measurement as bone’s small-scale structure. Bone Mineralisation, Turnover and Fractures Measure Different Outcomes explains why these skeletal endpoints should remain separate.

A relevant report could not be fully assessed

A 2026 MAVIDOS paper examined bone microarchitecture, meaning the small-scale structure of bone. Its published abstract reports no detected difference between the prenatal supplementation groups in the measurements studied. We could not obtain the full paper and material supplements for this review. We therefore could not fully assess participant selection, analysis choices, detailed results or robustness.

This missing information matters because the earlier childhood follow-up used a different measurement and reported some favourable findings. The abstract-level result should neither be ignored nor treated as a fully appraised refutation. Different skeletal sites, methods and selected samples can produce different answers. Without the complete report, this review cannot fully reconcile them.

This is the same MAVIDOS trial, not an additional independent study. Its abstract is not counted as fully assessed evidence. The available reports do not support a general claim that prenatal supplementation improves every aspect of childhood bone or provides lasting protection from fractures.

A separate trial followed children into adolescence

The Danish COPSAC trial provides an independent comparison at a later follow-up age. Women were assigned additional vitamin D3 or placebo during pregnancy, while both groups received the recommended background amount. The intervention began later than MAVIDOS. Differences in timing, comparator and population mean the trials are not identical replications.

In the 2026 report, 416 offspring had scans at age 13. Investigators did not detect a clear difference in the reported bone mineral measurements. The estimated whole-body bone mineral content difference was 6.7 grams, with a 95% confidence interval from −28.3 to 41.6 grams. That interval includes differences in either direction; it does not establish exact equivalence.

Clinical follow-up included 550 children. The fracture analysis also did not establish protection. Its hazard ratio was 0.85, with a 95% confidence interval of 0.57 to 1.25. A hazard ratio compares event occurrence over time between groups. Here, the uncertainty includes both a potentially lower fracture rate and no clear reduction. A lower numerical estimate should not be presented as proven protection.

The bone follow-up extended an originally secondary outcome, and the fracture analysis was post hoc. The trial had been powered for respiratory outcomes, not bone or fracture benefit. Missing scans, puberty, different scanners and loss of blinding after early childhood further qualify interpretation. Age 13 is not adulthood: the study cannot establish the final adult skeleton or lifetime osteoporosis risk.

Together, MAVIDOS and COPSAC show why age and endpoint matter. A favourable childhood scan in one trial does not establish persistence into adolescence in another population. Conversely, no detected difference in a later trial does not erase every earlier finding. These results support an outcome-specific account rather than a single claim of lasting benefit or universal absence of effect.

Rat pancreatic development: persistence with limits

Schavinski and colleagues examined maternal vitamin D deficiency in rats before conception, during pregnancy and during lactation. Offspring received a standard diet after weaning and were assessed at 90 days. Because exposure continued through lactation, the experiment cannot isolate a prenatal-only effect.

Male offspring showed lower selected insulin measurements and pancreatic beta-cell mass, alongside changes in gene expression. Beta cells produce insulin. The findings illustrate a later tissue phenotype after an early nutritional exposure, but glucose regulation was not uniformly impaired across every test. Females did not show the same pattern, although describing them as entirely unaffected would be inaccurate.

A mineral-rescue group helped examine whether changes could be explained solely by mineral availability. However, the rescue diet changed more than calcium alone, and most male findings were not removed. That supports further investigation, not a single isolated cause. Small numbers for individual endpoints and unclear independence of all litter-level observations also limit certainty.

The study did not demonstrate the proposed methylation change at the examined gene promoter. Methylation is a chemical mark on DNA that can be related to gene regulation. A persistent phenotype is not, by itself, proof of that epigenetic mechanism. Nor does a rat pancreatic phenotype show that maternal supplementation prevents diabetes in children.

Mouse immune-cell programming: highly experimental evidence

Oh and colleagues examined another form of developmental persistence. They transferred fetal-liver cell preparations from offspring of vitamin-D-deficient or sufficient mothers into adult mouse recipients. The preparations contained several stem and progenitor cell populations capable of producing later blood and immune-cell lineages. They were not identical purified stem-cell samples.

Glucose intolerance, reduced responsiveness to insulin and some immune-cell characteristics persisted in recipients whose diets supplied vitamin D. This supports the possibility that an early environment can alter developing cell populations. It does not establish normal human childhood development. Transplantation into prepared recipients is artificial, and differences in transferred cell composition complicate attribution to one cell type.

Several regulatory changes accompanied the phenotype. The experiment did not establish that one specific DNA methylation change caused all later findings. Physiological effects also differed between organs and tests. Claims that every tissue was affected in the same way, or that one inflammatory signal explained the entire outcome, would exceed the evidence.

A small human cord-blood component supplied additional observations, but it was not a clinical follow-up of children. Associations involving cord cells and laboratory responses cannot show that those children would develop type 2 diabetes. This distinction matters when a publication’s title uses disease language: experimental phenotype, proposed mechanism and diagnosed human disease are separate levels of evidence.

What can reasonably be concluded

The assessed studies support developmental sensitivity to vitamin D exposure in particular experimental systems and show that prenatal supplementation can alter circulating status. They do not support uniform improvement in later health. Human growth, neonatal mineral content, childhood scans, adolescent bone measurements and fractures produced different findings and different degrees of uncertainty.

This is a selected narrative review, not an exhaustive assessment of every organ or publication. The full 2026 MAVIDOS microarchitecture paper was unavailable. Some supplements supporting other reports were not completely inspected, so detailed claims depending on them are excluded. Kidney-development and other tissue-specific avenues were not fully appraised. Their absence from the assessed set is not evidence that no relevant research exists or that those organs are unaffected.

Identify the exposure, species, developmental window, measured outcome and follow-up period before deciding what a result means. Persistent experimental changes justify further research. Selected childhood benefits deserve accurate reporting. Null and incomplete findings must remain visible. None of these, alone, establishes an individual pregnancy regimen or a promise of lifelong protection.

Sources

Related reading

Original sources

Source

Pregnancy vitamins and supplements

For current UK guidance, see NHS Pregnancy vitamins and supplements.

Publication details

For current UK guidance, see NHS Pregnancy vitamins and supplements. Trial findings do not establish a personal supplement dose or blood target.

Read more →
Source

Vitamin D Supplementation in Pregnancy and Lactation and Infant Growth

Published in Roth et al. New England Journal of Medicine (2018).

Publication details

Vitamin D Supplementation in Pregnancy and Lactation and Infant Growth. Roth et al. New England Journal of Medicine. 2018;379:535 to 546. DOI:10.1056/NEJMoa1800927. Principal assessed; interpretation incorporates the 2021 correction, DOI:10.1056/NEJMx210014. Archived numerical details may predate the correction.

Read more →
Source

Maternal gestational vitamin D supplementation and offspring bone health (MAVIDOS): a multicentre, double-blind, randomised placebo-controlled trial

Published in Cooper et al. Lancet Diabetes & Endocrinology (2016).

Publication details

Maternal gestational vitamin D supplementation and offspring bone health (MAVIDOS): a multicentre, double-blind, randomised placebo-controlled trial. Cooper et al. Lancet Diabetes & Endocrinology. 2016;4:393 to 402. DOI:10.1016/S2213-8587(16)00044-9. Principal and relevant supplementary material assessed.

Read more →
Source

Pregnancy vitamin D supplementation and offspring bone mineral density in childhood follow-up of a randomized controlled trial

Published in Moon et al. American Journal of Clinical Nutrition (2024).

Publication details

Pregnancy vitamin D supplementation and offspring bone mineral density in childhood follow-up of a randomized controlled trial. Moon et al. American Journal of Clinical Nutrition. 2024;120:1134 to 1142. DOI:10.1016/j.ajcnut.2024.09.014. Principal and eight supplementary tables assessed; same MAVIDOS trial.

Read more →
Source

Vitamin D supplementation during pregnancy and offspring bone microarchitecture: A post hoc analysis of the MAVIDOS randomised controlled trial

Published in Maden et al. Bone (2026).

Publication details

Vitamin D supplementation during pregnancy and offspring bone microarchitecture: A post hoc analysis of the MAVIDOS randomised controlled trial. Maden et al. Bone. 2026;207:117843. DOI:10.1016/j.bone.2026.117843. ABSTRACT ONLY: full principal and material supplements unavailable; not counted as fully assessed evidence.

Read more →
Source

Prenatal high-dose vitamin D supplementation and offspring bone mineral content and density at age 13 years: a secondary analysis of a randomised clinical trial

Published in Brustad et al. Lancet Regional Health: Europe (2026).

Publication details

Prenatal high-dose vitamin D supplementation and offspring bone mineral content and density at age 13 years: a secondary analysis of a randomised clinical trial. Brustad et al. Lancet Regional Health: Europe. 2026;71:101870. DOI:10.1016/j.lanepe.2026.101870. Principal, material supplementary tables and relevant protocol history assessed; earlier COPSAC reports are the same trial.

Read more →
Source

Maternal Vitamin D Deficiency Impairs the Development of β Cells in Offspring Rats in a Sex-Dependent Manner

Recent studies have shown that maternal vitamin D deficiency (VDD) causes long-term metabolic changes in offspring.

Publication details

Maternal Vitamin D Deficiency Impairs the Development of β Cells in Offspring Rats in a Sex-Dependent Manner. Schavinski et al. International Journal of Molecular Sciences. 2024;25:4136. DOI:10.3390/ijms25084136. Principal, tables, figure descriptions and diet appendix assessed.

Read more →