Before the first breath: pregnancy and autism
- Richard Moore
- 5 days ago
- 14 min read
A science-forward look at what research is learning about pregnancy biology and autism.
Billions of neurones had been generated, neural connections were forming, and early circuits were beginning to organise, all before your child took their first breath. Brain development does not stop at birth, but what happens in those months before it is extraordinary, and it is where some of the most interesting science in autism research is currently being done.
That is why so many parents, in the quiet moments after an autism diagnosis, find themselves drawn back to the pregnancy. Researchers are examining parts of that question, though the evidence cannot identify what caused an individual child's autism. What it can do is describe what is happening in prenatal biology and which factors appear to influence neurodevelopment more broadly. This piece is an attempt to do that honestly.
It is not a guide to preventing autism. Nothing here should be read that way. But if you are curious about the biology, this is what the science currently shows.
Start with genetics, because genetics changes everything
Any serious account of autism and pregnancy has to begin here. A population study of more than 37,000 Swedish families estimated the heritability of autism at approximately 83% (Sandin et al., 2017). Published in JAMA, it is one of the most cited findings in the field.
Heritability is a population-level statistic. It does not mean 83% of a specific child's autism was caused by genes, and it does not point to a single identifiable autism gene. What it tells us is that genetic differences account for the large majority of variation in autism rates across families and populations. Autism is genetically complex, shaped by many common and rare variants often acting in combination, and that genetic architecture is present from the earliest moments of development.
This matters for how we understand everything that follows. When researchers study the relationship between pregnancy factors and autism outcomes, they are studying influences that operate against a background of substantial genetic contribution. The prenatal environment does not write the story from scratch. But it does interact with the genetic story as it unfolds.
What the womb actually does
Genes are not static instructions. They respond to their environment, and during pregnancy that response is happening at a scale and speed that will never be repeated in a human life.
The science of epigenetics, the study of how environmental signals switch genes on or off without altering the deoxyribonucleic acid (DNA) sequence itself, has shown that conditions during pregnancy can influence aspects of gene regulation and development. Nutrients, immune signalling, hormonal environment, microbial exposure: all of these feed into an extraordinarily dynamic system. Neurones are being generated at rates that will never be matched again. Neural circuits are forming. The architecture of a brain is being laid down. The specific relationship of all this to autism remains incompletely understood, but the biological activity of the prenatal period is not in question.
Researchers are now asking specific questions about which aspects of that environment influence neurodevelopment and through what mechanisms. The answers so far are partial and sometimes contested, but the questions being asked are increasingly precise, and the science pursuing them is serious.
The folate story: more complex than the supplement label suggests
Folic acid has the clearest established role of any nutrient in prenatal neurodevelopment. Its connection to autism research begins with its role in methylation, a process that acts like a regulatory system for your DNA. Tiny chemical markers attach to specific sections of the genetic code, switching them active or silent. Folate sits at the centre of the one-carbon metabolism pathway that drives this process, making it critical to cell division and early brain development.
Observational research has consistently found lower autism rates among children whose mothers supplemented with folic acid around conception and in early pregnancy. An umbrella review of 23 systematic reviews and meta-analyses found maternal supplementation associated with lower odds of autism in offspring (odds ratio 0.66; 95% confidence interval 0.55–0.79) as well as reduced odds of attention deficit hyperactivity disorder (ADHD) and behavioural problems (Yu et al., 2025). A systematic review for the United States Preventive Services Task Force found no evidence that folic acid supplementation increases autism risk (Viswanathan et al., 2023). The overall association points toward lower odds, but the underlying evidence is observational, methodologically variable and unable to establish causation.
Vitamin B12 sits alongside folate in the same methylation pathway. It is essential for DNA synthesis, neurological function, and the maintenance of the myelin sheath around nerve cells (the insulating coating that allows signals to travel quickly and without interference). A 2025 review found that inadequate maternal B12 may alter neurodevelopmental trajectories and affect gut microbiota in ways relevant to autism, based primarily on observational and mechanistic evidence; the authors were clear that the link is plausible and worth monitoring, but not yet conclusive (Zwierz et al., 2025).
Not all folate questions are the same
The debate about folate variation involves several distinct questions that are worth separating, because they do not form one connected pathway.
The first concerns common MTHFR gene variants. MTHFR (methylenetetrahydrofolate reductase) is an enzyme involved in folate processing. Common variants are widespread and are frequently flagged by consumer genetic tests. Current clinical guidance is clear that common MTHFR variants do not prevent folic acid from working: blood folate concentrations increase with supplementation regardless of genotype, and these variants are not considered a clinical reason to switch to methylfolate or avoid the standard intake (CDC, 2025a; CDC, 2025b). Folic acid remains the form with established evidence for preventing neural tube defects (Viswanathan et al., 2023).
The second concerns higher folate status or higher-dose supplementation, which is a separate question from common MTHFR variants and from standard recommended intakes. Large cohort studies found that women with red cell folate above the clinical reference range (a level exceeded by 57.6% of women after mandatory food fortification) were associated with 48% higher odds of gestational diabetes (a form of diabetes that develops during pregnancy), with researchers proposing placental hormones may play a role (Jankovic-Karasoulos et al., 2025). A systematic review of 12 human studies found that high folic acid intake may be associated with increased gestational diabetes risk, with the combination of high folate and low vitamin B12 amplifying it further (Gomez-Cabrera et al., 2025). A further cohort study linked higher-dose supplementation to increased odds of vaginal bleeding in early pregnancy (Wang et al., 2026). These findings concern higher folate status or higher-dose supplementation and metabolic pregnancy outcomes rather than autism specifically. They do not establish that standard folic acid intake is harmful or that common MTHFR variants cause toxicity. They support following recommended doses and avoiding an unsupported approach of taking more than recommended. A 2025 scoping review also noted that individuals carrying the MTHFR C677T variant have a reduced capacity to convert folic acid to its active form, which may lead to greater unmetabolised folic acid accumulation at intakes above the tolerable upper level, and called for continued research into the maternal health implications (Hecker et al., 2025).
A note on a frequently cited study: the one published human study to directly measure unmetabolised folic acid in pregnancy and examine autistic traits in offspring found no association (Husebye et al., 2022). This study was conducted in children exposed to antiseizure medication during pregnancy, which is an important context when considering how its findings apply more broadly.
The third question is separate from both MTHFR variants and maternal dosing. Some autistic children have been found to carry proteins called folate receptor alpha autoantibodies (FRAAs). These are immune molecules, not genetic variants, and should not be conflated with common MTHFR polymorphisms. FRAAs can block the transport of folate into the brain, potentially leading to cerebral folate deficiency syndrome, where folate circulates adequately in the body but the brain is not receiving enough. Studies have found FRAAs in approximately one third to two thirds of autistic children depending on the population studied (Phunsawat et al., 2022; Bobrowski-Khoury et al., 2021). Small studies have investigated folinic acid, a form of folate that reaches the brain via a different route, in selected autistic children who have tested positive for FRAAs, and a 2021 review reported improvements in language and social interaction in some of these children (Bobrowski-Khoury et al., 2021). Current evidence is not sufficient to support routine FRAA testing or folinic acid treatment for autistic children, and this approach is not currently recommended as standard autism care. Any investigation or treatment for suspected cerebral folate deficiency requires appropriate medical assessment.
FRAAs have also been identified in a significant proportion of children with PANS (Paediatric Acute-onset Neuropsychiatric Syndrome) and PANDAS (Paediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections), and research into the autoantibody connection across these conditions is ongoing (Wells et al., 2024). Rare, severe disorders of folate metabolism form a further distinct category and may require specialist assessment (CDC, 2025a).
Broader nutrition: what research is beginning to map
Beyond folate and B12, a major research summit at Tufts University in 2022 identified several other nutritional areas where evidence is building (Maitin-Shepard et al., 2024). Choline was flagged as notably under-studied relative to its biological importance. It shares the one-carbon metabolism pathway with folate, is essential for the production of acetylcholine (a neurotransmitter involved in memory, muscle function and early brain development), and plays a direct role in neural tube closure and brain cell membrane formation. Many pregnant women do not reach recommended intakes, and its relationship to neurodevelopmental outcomes is only beginning to be examined in depth.
Iron deficiency is common in pregnancy and affects myelination (the process of building the insulating sheaths around nerve fibres) as well as oxygen delivery to developing brain tissue. Large epidemiological datasets have explored the relationship between maternal iron status and neurodevelopmental outcomes, though findings remain inconsistent. Ultra-processed food consumption, overall dietary quality, and environmental contaminants in food packaging were also identified as areas requiring further investigation (Maitin-Shepard et al., 2024). These are research priorities, not established causes, but they reflect a growing recognition that the nutritional environment of pregnancy is relevant to how the prenatal brain develops.
The immune system: the research everyone should know about
Maternal immune activation (MIA) is one of the most actively studied areas in autism research right now, and what it is beginning to reveal about pregnancy and brain development is worth following closely.
When the immune system is heightened during pregnancy, through infection, inflammatory conditions, or significant physiological stress, the body produces cytokines: chemical signalling proteins that coordinate immune responses. The cytokines most studied in relation to autism are interleukin-6 and interleukin-17. Researchers have been investigating whether these molecules can cross the placenta and influence foetal brain development, particularly through their effects on microglia (the brain's resident immune cells, which carry out maintenance, surveillance and the selective pruning of neural connections). Laboratory evidence suggests this may be possible. Whether and how often this occurs in human pregnancies, and to what effect, is the active question.
The human associational evidence is substantial. A review in Nature Reviews Neurology found that maternal inflammatory states including obesity, asthma, autoimmune conditions, infection, and psychosocial stress are associated with increased risk of neurodevelopmental disorders including autism (Han et al., 2021a). A systematic review drawing on 32 meta-analyses and 26 additional studies found associations between maternal conditions such as gestational diabetes, pre-eclampsia (a pregnancy complication involving dangerous elevations in blood pressure), depression, and pollution exposure, and neurodevelopmental outcomes in offspring (Han et al., 2021b). A 2023 review examining neuroinflammation and oxidative stress (cellular damage caused by an excess of unstable molecules) identified MIA as one of the most significant environmental factors of interest during the prenatal period (Usui et al., 2023).
These are associations in large populations, not verdicts about individual pregnancies. An autoimmune condition, a difficult few months, or a pregnancy complicated by illness does not mean a child's autism was caused by it. Mental health support during pregnancy matters because the wellbeing of the pregnant person matters. Research cannot tell us that ordinary worry, distress or a mental health condition causes autism, and no parent should read this evidence as though it can.
The gut microbiome: a newer piece of the puzzle
The maternal gut microbiome has entered autism research relatively recently, but it is moving fast. The gut is not simply a digestive organ. It produces neurotransmitters and neurotransmitter precursors, the chemical ingredients used to build brain-signalling molecules like serotonin and dopamine, and it communicates continuously with the brain via the vagus nerve and through immune molecules and short-chain fatty acids produced during fibre fermentation.
During pregnancy the microbiome shifts naturally. When those shifts move into dysbiosis, a broad and inconsistently defined term describing changes in microbial composition or function relative to a comparison population, researchers have proposed possible relationships with foetal immune development. A 2025 narrative review (a type of evidence synthesis that is more hypothesis-generating than hypothesis-testing) found pregnancy-related dysbiosis linked to adverse neurodevelopmental outcomes in offspring, including increased risk of autism and ADHD, based largely on observational and preclinical evidence (Biagioli et al., 2025). The Tufts summit also identified the maternal microbiome as a priority area for future autism research (Maitin-Shepard et al., 2024). There is currently no universally accepted test or threshold for identifying dysbiosis, and clinical interventions based on this evidence do not yet exist. But this is an active research frontier, and the pace of scientific interest in the maternal gut as a bridge between maternal environment and foetal neurodevelopment is unlikely to slow.
Where the evidence is thinner
Not every area covered in public conversation has equal support in the research.
Omega-3 fatty acids are important for brain and retinal development during pregnancy. But a systematic review of 33 studies found the evidence for omega-3 supplementation specifically reducing autism risk insufficient and inconsistent (Nevins et al., 2021), and a 2024 umbrella review found no evidence for omega-3 in autism treatment (Talib et al., 2024). Its general importance in pregnancy is not in question; its specific relationship to autism risk is.
Vitamin D deficiency is common and associated with a range of pregnancy complications. Whether low maternal vitamin D specifically influences autism risk is a question the research has raised without yet answering clearly. It remains an area of investigation rather than established connection.
Making sense of all of this
The prenatal period is fundamental to brain development. Researchers are examining how genetic, immune, nutritional and environmental factors interact during it. That does not establish a single straightforward pregnancy-autism connection, and it does not allow autism to be attributed to a parent's behaviour, diet, illness or supplement choices.
What the science can do is describe what is happening in prenatal biology and identify the factors researchers are examining. Maternal health and nutrition matter for pregnancy generally. Their specific relationships to autism, through which mechanisms and with what effect, remain under active investigation. The associations observed in large populations cannot be traced backwards to identify what caused an individual child's autism.
For parents sitting with those questions: the evidence does not answer them, and it was never going to. What it does do is point toward a picture of extraordinary biological complexity in those months before the first breath, one that deserves serious, curious attention rather than the weight of personal blame.
Personalised support for family nutrition
Jessica is our lead paediatric nutritional therapy practitioner, specialises in complex paediatric neurological conditions, with a particular focus on autism, ADHD, avoidant restrictive food intake disorder (ARFID) and selective eating, and neuroimmune conditions including Paediatric Acute-onset Neuropsychiatric Syndrome (PANS) and Paediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections (PANDAS). She developed our Hello World programme.
Clare, our paediatric specialist Nutritional Therapy Practitioner and qualified Fertility Practitioner, developed our Little Seed programme, which focuses on early life nutrition, reflux, feeding challenges, and food allergies and intolerances.
Both practitioners are trained in ARFID and neurodevelopmental and eating disorder-related eating difficulties (NEDDE). You can find out more about our work here.
Clare has also developed The Nutrition First Aid Kit, a children's nutrition membership covering a new topic each month, starting with gut health.
We spoke with Clare in our last blog about why she built it and what she hopes it gives the families who join. Read that conversation here.
🧩 Connect with us
To find out more about our paediatric programmes go to www.younutritionclinic.com/childs-health
For research updates, practical tips, and ongoing inspiration, follow us on Instagram:
Stay curious. Stay hopeful. Support your child's health.
Disclaimer: This article is for informational and educational purposes only and does not constitute medical advice. Always consult with a qualified, registered medical doctor (MD) for diagnosis and treatment decisions.
References
Biagioli, V., Matera, M., Ramenghi, L. A., Falsaperla, R., & Striano, P. (2025). Microbiome and pregnancy dysbiosis: A narrative review on offspring health. Nutrients, 17(6). https://doi.org/10.3390/nu17061033
Bobrowski-Khoury, N., Ramaekers, V. T., Sequeira, J. M., & Quadros, E. V. (2021). Folate receptor alpha autoantibodies in autism spectrum disorders: Diagnosis, treatment and prevention. Journal of Personalized Medicine, 11(8), Article 710. https://doi.org/10.3390/jpm11080710
Centers for Disease Control and Prevention. (2025a). MTHFR gene variant and folic acid facts. https://www.cdc.gov/folic-acid/data-research/mthfr/index.html
Centers for Disease Control and Prevention. (2025b). Folic acid: Facts for clinicians. https://www.cdc.gov/folic-acid/hcp/clinical-overview/index.html
Gomez-Cabrera, A. S., Gonzalez-Santiago, A. E., Castaneda-Arellano, R., Corona-Meraz, F. I., Baptista-Rosas, R. C., & Sanchez-Parada, M. G. (2025). Folic acid supplementation and risk of gestational diabetes mellitus: A systematic review of the literature. International Journal of Molecular Sciences, 26(16), Article 7977. https://doi.org/10.3390/ijms26167977
Han, V. X., Patel, S., Jones, H. F., & Dale, R. C. (2021a). Maternal immune activation and neuroinflammation in human neurodevelopmental disorders. Nature Reviews Neurology, 17(9), 564–579. https://doi.org/10.1038/s41582-021-00530-8
Han, V. X., Patel, S., Jones, H. F., Nielsen, T. C., Mohammad, S. S., Hofer, M. J., Gold, W., Brilot, F., Lain, S. J., Nassar, N., & Dale, R. C. (2021b). Maternal acute and chronic inflammation in pregnancy is associated with common neurodevelopmental disorders: A systematic review. Translational Psychiatry, 11(1), Article 71. https://doi.org/10.1038/s41398-021-01198-w
Hecker, J., Layton, R., & Parker, R. W. (2025). Adverse effects of excessive folic acid consumption and its implications for individuals with the methylenetetrahydrofolate reductase C677T genotype. Cureus, 17(2), Article e79374. https://doi.org/10.7759/cureus.79374
Husebye, E. S. N., Wendel, A. W. K., Gilhus, N. E., Riedel, B., & Bjork, M. H. (2022). Plasma unmetabolized folic acid in pregnancy and risk of autistic traits and language impairment in antiseizure medication-exposed children of women with epilepsy. The American Journal of Clinical Nutrition, 115(5), 1432–1440. https://doi.org/10.1093/ajcn/nqab436
Jankovic-Karasoulos, T., Smith, M. D., Leemaqz, S., Mittinty, M., Williamson, J., McCullough, D., Arthurs, A. L., Dekker, G. A., & Roberts, C. T. (2025). Maternal folate excess, placental hormones, and gestational diabetes mellitus: Findings from prospective cohorts before and after mandatory folic acid food fortification. Nutrients, 17(17), Article 2863.
Maitin-Shepard, M., O'Tierney-Ginn, P., Kraneveld, A. D., Lyall, K., Fallin, D., Arora, M., Fasano, A., Mueller, N. T., Wang, X., Caulfield, L. E., Dickerson, A. S., Diaz Heijtz, R., Tarui, T., Blumberg, J. B., Holingue, C., Schmidt, R. J., Garssen, J., Almendinger, K., Lin, P.-I. D., & Mozaffarian, D. (2024). Food, nutrition, and autism: From soil to fork. The American Journal of Clinical Nutrition, 120(1), 240–256. https://doi.org/10.1016/j.ajcnut.2024.04.020
Nevins, J. E. H., Donovan, S. M., Snetselaar, L., Dewey, K. G., Novotny, R., Stang, J., Taveras, E. M., Kleinman, R. E., Bailey, R. L., Raghavan, R., Scinto-Madonich, S. R., Venkatramanan, S., Butera, G., Terry, N., Altman, J., Adler, M., Obbagy, J. E., Stoody, E. E., & de Jesus, J. (2021). Omega-3 fatty acid dietary supplements consumed during pregnancy and lactation and child neurodevelopment: A systematic review. The Journal of Nutrition, 151(11), 3483–3494. https://doi.org/10.1093/jn/nxab238
Phunsawat, P., Chiangjong, W., Chutipongtanate, S., Dumrongwongsiri, O., Thommachot, P., Butdawong, W., & Chuthapisith, J. (2022). Folate receptor alpha autoantibodies in children with autism spectrum disorder. Biomarkers, 27(8), 715–719. https://doi.org/10.1080/1354750X.2022.2125579
Sandin, S., Lichtenstein, P., Kuja-Halkola, R., Hultman, C., Larsson, H., & Reichenberg, A. (2017). The heritability of autism spectrum disorder. JAMA, 318(12), 1182–1184. https://doi.org/10.1001/jama.2017.12141
Talib, M., Rachdi, M., Papazova, A., & Nicolis, H. (2024). The role of dietary patterns and nutritional supplements in the management of mental disorders in children and adolescents: An umbrella review of meta-analyses. Canadian Journal of Psychiatry, 69(8), 567–589. https://doi.org/10.1177/07067437241248070
Usui, N., Kobayashi, H., & Shimada, S. (2023). Neuroinflammation and oxidative stress in the pathogenesis of autism spectrum disorder. International Journal of Molecular Sciences, 24(6), Article 5487. https://doi.org/10.3390/ijms24065487
Viswanathan, M., Urrutia, R. P., Hudson, K. N., Middleton, J. C., & Kahwati, L. C. (2023). Folic acid supplementation to prevent neural tube defects: Updated evidence report and systematic review for the United States Preventive Services Task Force. JAMA, 330(5), 460–466. https://doi.org/10.1001/jama.2023.9864
Wang, H., Leemaqz, S., Smith, M. D., McCullough, D., Arthurs, A. L., Jankovic-Karasoulos, T., Dekker, G. A., & Roberts, C. T. (2026). High folic acid supplementation is associated with vaginal bleeding in early pregnancy in a fetal sex-specific manner: Findings from two prospective cohort studies. Reproductive Health, 23(1). https://doi.org/10.1186/s12978-026-02344-7
Wells, L., O'Hara, N., Frye, R. E., Hullavard, N., & Smith, E. (2024). Folate receptor alpha autoantibodies in the Pediatric Acute-Onset Neuropsychiatric Syndrome (PANS) and Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections (PANDAS) population. Journal of Personalized Medicine, 14(2), Article 166. https://doi.org/10.3390/jpm14020166
Yu, M., Hu, Y., Hou, L., Wu, X., Chen, X., Yan, R., Dong, J., & Wu, J. (2025). The effect of maternal folic acid supplementation on neurodevelopmental disorders in offspring: An umbrella review of systematic reviews and meta-analyses. Nutrients, 17(21). https://doi.org/10.3390/nu17213443
Zwierz, M., Suprunowicz, M., Mrozek, K., Pietruszkiewicz, J., Oracz, A. J., Konarzewska, B., & Waszkiewicz, N. (2025). Vitamin B12 and autism spectrum disorder: A review of current evidence. Nutrients, 17(7), Article 1220. https://doi.org/10.3390/nu17071220
Comments