Maternal vitamin D status during pregnancy has drawn attention as a possible lever for offspring brain development. Observational cohort data link higher second-trimester 25-hydroxyvitamin D [25(OH)D] with modestly higher child IQ — roughly one IQ point per 10 ng/mL increment in Melough et al. (2021)’s CANDLE cohort. The picture beyond that single study is messier: large observational cohorts replicate small positive associations, severe deficiency tracks with elevated autism risk, but randomized supplementation trials have been mostly null for cognitive endpoints. The signal is real, the effect size is small, and the public-health case is strongest for correcting frank deficiency rather than chasing optimal levels in already-replete women.
The CANDLE Cohort: Where the Number Comes From
The CANDLE study followed 1,019 mother-child pairs in the southeastern United States. Researchers measured maternal plasma 25(OH)D in the second trimester and tested children’s IQ between ages 4 and 6 using the Stanford-Binet 5. After adjustment for maternal IQ, age, education, race, household income, and child characteristics, each 10 ng/mL increase in maternal 25(OH)D was associated with a 1.17-point higher Full Scale IQ, a 1.17-point higher Verbal IQ, and a 1.03-point higher Nonverbal IQ.
Two contextual numbers matter for interpretation. Mean maternal 25(OH)D was 21.6 ng/mL — below the 30 ng/mL threshold many clinicians treat as sufficient — and 45.6% of women were classified as deficient at <20 ng/mL. Black women had substantially lower mean concentrations than White women (19.8 vs 25.9 ng/mL), reflecting darker skin’s reduced cutaneous vitamin D synthesis. The relationship between maternal vitamin D and child IQ did not differ by race, which means the per-unit effect size held across groups even though baseline levels did not.
A 1-point gain per 10 ng/mL is a real but small effect. Moving a deficient woman from 15 ng/mL to a clinically sufficient 30 ng/mL would predict roughly 1.5 IQ points in her child. That is a population-scale signal, not a transformative individual one — but neither is it noise.
What Other Cohorts Show
The Avon Longitudinal Study of Parents and Children (ALSPAC) examined 7,065 mother-child pairs in Sass et al. (2017). Children of vitamin D-deficient mothers (<50 nmol/L, equivalent to <20 ng/mL) were more likely to score in the lowest quartile for gross-motor development at 30 months (odds ratio 1.20), fine-motor development at 30 months (OR 1.23), and social development at 42 months (OR 1.20). The IQ at age 8, however, did not show a significant association after full adjustment. Early developmental milestones moved; later cognitive testing did not.
The Generation R cohort in Rotterdam contributes the autism-related signal. Vinkhuyzen et al. (2017) followed 4,334 mother-child pairs with mid-gestation maternal 25(OH)D and neonatal cord blood measurements. Mid-gestation deficiency was associated with more than double the risk of clinically diagnosed autism spectrum disorder (OR 2.42, 95% CI 1.09 to 5.07). The finding survived restriction to children of European ancestry, which addresses some confounding by skin pigmentation.
These three cohorts — CANDLE, ALSPAC, and Generation R — point in the same direction: prenatal vitamin D status correlates with neurodevelopmental outcomes in offspring, with effect sizes that are small for cognition and larger for the tail-end risk of severe outcomes like autism diagnosis.
Why the Trial Evidence Looks Weaker
The 2022 systematic review by Upadhyaya et al. screened 3,729 studies and included 29 in the final analysis. The authors found “a small amount of evidence” for prenatal vitamin D deficiency and autism, with positive findings for ADHD and schizophrenia emerging only when studies used larger samples and stricter deficiency thresholds. Their conclusion was cautious: associations exist but are inconsistent, and observational designs cannot rule out residual confounding by maternal nutrition more broadly, sun exposure, socioeconomic status, or the same lifestyle factors that determine both 25(OH)D levels and child outcomes.
The 2024 umbrella review by Chien et al. aggregated 16 systematic reviews covering 250,569 women. Their summary is more optimistic on supplementation: vitamin D supplementation in pregnancy was associated with reduced offspring ADHD and autism spectrum disorder risk in the pooled review-of-reviews. The signal for raw cognitive endpoints (IQ scores, neurodevelopmental scales) was less clean, and individual high-dose RCTs in pregnancy have generally not moved standardized cognitive scores in offspring.
This is the standard observational-vs-randomized gap in nutrition science. Cohort studies pick up associations driven partly by what vitamin D tracks — overall maternal health, dietary quality, socioeconomic position. Randomized trials remove that confounding but typically dose women whose baseline status is already adequate, leaving little room for the supplement to do work. The honest read is that severe deficiency probably matters; the deficiency-to-sufficiency margin is where the evidence becomes inconsistent.
What This Means for Prenatal Care
Three practical points fall out of the literature.
First, screening matters in at-risk groups. Women with darker skin, limited sun exposure, obesity, vegan or low-dairy diets, and those living at higher latitudes are at substantially greater risk of frank deficiency. CANDLE’s 45.6% deficiency rate is not unusual in U.S. obstetric populations. For these women, the case for measuring 25(OH)D and correcting deficiency is strong on multiple grounds — bone health, preeclampsia risk, gestational diabetes, and the offspring cognitive signal — without needing the IQ effect to be large in isolation.
Second, ACOG’s standing recommendation of 600 IU/day in pregnancy is consistent with the lower bound of evidence. Higher doses (1,000-2,000 IU/day) are commonly used for women with documented deficiency and are within the safe range. There is no evidence that pushing 25(OH)D well above the sufficient threshold (30 ng/mL) confers additional cognitive benefit, and observational dose-response curves in CANDLE flatten at the upper end.
Third, vitamin D is one nutrient among many. Pregnancy diet effects on offspring cognition operate through folate, iodine, iron, n-3 fatty acids, and choline alongside vitamin D — see the broader review in our pregnancy diet and brain development summary. A vitamin D-only intervention layered onto an otherwise inadequate diet will not carry the load.
How This Sits in the Broader Literature
The cognitive returns to prenatal vitamin D are similar in scale to the returns documented for breastfeeding once confounding is properly handled — see our writeup of breastfeeding and intelligence evidence, where the PROBIT cluster-randomized trial estimated approximately 5.9 IQ points at age 6 with substantial attenuation at 16-year followup. Prenatal vitamin D’s ~1 point per 10 ng/mL is smaller per increment but applies to a population where deficiency is common and correctable.
The largest leverage on child cognitive outcomes still comes from postnatal environment — caregiving quality, school exposure, language input, and household resources. Poverty and brain development and postnatal nutrition both produce effect sizes that dwarf any single prenatal nutrient. Vitamin D belongs in the prenatal toolkit; it is not the toolkit.
Conclusion
Maternal vitamin D status during pregnancy is associated with small but consistent improvements in child cognitive outcomes and with reduced risk of autism diagnosis at the deficient end of the distribution. The cognitive effect is real but modest — about 1 IQ point per 10 ng/mL in the best-controlled cohort. Severe deficiency, common in dark-skinned women and those with limited sun exposure, is worth identifying and correcting; routine supplementation at the ACOG-recommended 600 IU/day is reasonable for most pregnancies. Treating prenatal vitamin D as one of several nutrient inputs — alongside folate, iodine, iron, and n-3 fatty acids — fits the evidence better than treating it as a standalone IQ intervention.
Frequently Asked Questions
Does taking vitamin D during pregnancy raise my child’s IQ?
On observational data, yes, by a small amount — about 1 IQ point per 10 ng/mL increase in maternal 25(OH)D. Randomized trials of supplementation have been less consistent, particularly when starting from already-adequate vitamin D status. The strongest case is for correcting frank deficiency, not for pushing levels above sufficient.
How much vitamin D should I take in pregnancy?
ACOG recommends 600 IU/day as a general baseline. Women with documented deficiency are commonly given 1,000-2,000 IU/day, well within the safe range. Going above that requires clinical guidance; there is no evidence that high doses improve cognitive outcomes beyond correcting deficiency.
Is the autism risk from vitamin D deficiency real?
The Generation R cohort found a roughly twofold increase in clinically diagnosed autism in offspring of mid-gestation deficient women. This finding has been partially replicated and partially not. Severe deficiency appears to matter more than mild deficiency; the absolute risk increase remains small.
Do dark-skinned women need more vitamin D?
Often, yes. Reduced cutaneous synthesis means dietary intake or supplementation has to do more of the work. CANDLE found Black women averaged 19.8 ng/mL versus 25.9 ng/mL in White women. Screening 25(OH)D and adjusting supplementation for women at risk is reasonable practice.
Does vitamin D supplementation help if I already have sufficient levels?
The evidence for additional cognitive benefit beyond a sufficient threshold (typically 30 ng/mL) is weak. Observational dose-response curves flatten at the upper end. Maintaining sufficiency is the goal; chasing higher numbers in already-replete women has not been shown to improve offspring outcomes.
References
- Chien, M. C., Huang, C. Y., Wang, J. H., Shih, C. L., & Wu, P. (2024). Effects of vitamin D in pregnancy on maternal and offspring health-related outcomes: An umbrella review of systematic review and meta-analyses. Nutrition & Diabetes, 14. https://doi.org/10.1038/s41387-024-00296-0
- Melough, M. M., Murphy, L. E., Graff, J. C., Derefinko, K. J., LeWinn, K. Z., Bush, N. R., Enquobahrie, D. A., Loftus, C. T., Kocak, M., Sathyanarayana, S., & Tylavsky, F. A. (2021). Maternal Plasma 25-Hydroxyvitamin D during Gestation Is Positively Associated with Neurocognitive Development in Offspring at Age 4–6 Years. The Journal of Nutrition, 151(1), 132-139. https://doi.org/10.1093/jn/nxaa309
- Sass, L., Bjarnadóttir, E., Stokholm, J., Chawes, B., Vinding, R. K., Mora-Jensen, A. R. C., & Bisgaard, H. (2017). Maternal vitamin D status in pregnancy and offspring neurodevelopmental outcomes — the Avon Longitudinal Study of Parents and Children. British Journal of Nutrition, 117(12), 1816-1829. https://doi.org/10.1017/s0007114517001398
- Upadhyaya, S., Ståhlberg, T., Silwal, S., Arrhenius, B., & Sourander, A. (2022). Maternal Vitamin D Levels during Pregnancy and Offspring Psychiatric Outcomes: A Systematic Review. International Journal of Molecular Sciences, 24(1), 63. https://doi.org/10.3390/ijms24010063
- Vinkhuyzen, A. A. E., Eyles, D. W., Burne, T. H. J., Blanken, L. M. E., Kruithof, C. J., Verhulst, F., Jaddoe, V. W., Tiemeier, H., & McGrath, J. J. (2017). Gestational vitamin D deficiency and autism spectrum disorder. BJPsych Open, 3(2), 85-90. https://doi.org/10.1192/bjpo.bp.116.004077