Why People Are Autistic: Mostly Genes, Not Parenting, Vaccines or Diet

Research across millions of families points one way: most of the difference between who is autistic and who is not comes from genes, many of them common variants found throughout the population. Older parents and being born early are linked to a higher likelihood of a child being autistic, mostly by small amounts: shifts in probability, not something parents did wrong. Parenting, vaccines and diet do not cause autism.

Many parents of autistic children have asked themselves whether it was something they did: a late pregnancy, a painkiller, a vaccine, the way they spoke to their baby. In the mid-twentieth century, influential clinicians told mothers the answer was yes. The research since says no. This page sets out what it shows, uncertainties included, and a view many autistic people hold firmly: how autistic people are supported matters far more than why they are autistic.

Autism runs in families, and genes explain most of it

The clearest evidence comes from comparing relatives. Identical twins share all their genes; other brothers and sisters share fewer, and cousins fewer still. If autism tracks genetic closeness more tightly than a shared home, genes are doing most of the work.

A study using national registers included 2,001,631 children born between 1998 and 2011 in Denmark, Finland, Sweden, Israel and Western Australia, followed up to age 16; 22,156 were diagnosed autistic. The median heritability estimate was 80.8%, with country estimates from about 51% in Finland to 87% in Israel. The authors concluded that variation in autism “is mostly owing to inherited genetic influences, with no support for contribution from maternal effects”^[1]. WHO notes the same pattern for “families with an autistic member and particularly for identical twins”^[2].

Other designs and places agree (table below). A Swedish reanalysis estimated 83%, revising an earlier estimate of 50% from the same data, calculated a different way^[3]; a meta-analysis of seven twin studies found 64–91%^[4]; twins in England and Wales gave 56–95%, depending on the assessment^[5]. A 2024 Swedish study estimated 87.0% in males and 75.7% in females, which its authors suggest may partly explain why more boys are diagnosed^[6]; see our article on autistic women and girls.

Heritability is easy to misread. It describes a population, not a person: most of the differences between people in whether they are autistic are linked to differences in their genes. It does not mean 80% of anyone’s autism is genetic, or that a parent has an 80% chance of passing it on. Nor is the remainder simply “the environment”: the 2024 Swedish study said its leftover term captured factors such as “maternal effects, de novo variants, rare genetic variants not additively inherited, or gene-environment interactions”^[6], some of them genetic.

What these studies mostly do not find is a meaningful effect of the home environment siblings share: 0.2% in the pooled Nordic data, although in Finland and Western Australia, where estimates were far less precise, the same model put it at 14.5% and 25.2%, and the authors note “possible modest differences in the sources of ASD risk between countries”^[1]. In the 2024 Swedish study the shared environment was “close to 0 and ... not statistically significant”^[6]. The twin meta-analysis judged earlier reports of significant shared effects “likely a statistical artefact”^[4]. Autism does cluster in families: in an international network that followed 1,605 babies with an autistic older sibling to age 3, 20.2% were themselves classified as autistic^[7].

Many common gene variants with tiny effects, plus a smaller number of rare ones

There is no single “autism gene”. Most of the inherited contribution comes from common variants: small, widespread differences in DNA, each nudging the likelihood of being autistic by a tiny amount^[8]. A 2019 genome-wide study of 18,381 autistic people and 27,969 comparison participants, boosted by a Danish population resource, confirmed that “common genetic variants contribute substantially” and found the first five locations in the genome robustly linked to autism^[9].

Rare variants matter too, including de novo changes that appear for the first time in the child rather than being inherited. Sequencing 35,584 people, 11,986 of them autistic, identified 102 genes in which rare variants are linked to autism, most active early in brain development^[10]; a 2022 analysis of 63,237 people found 72 at a very strict statistical threshold^[11]. A de novo change can matter a great deal for one person, but across the population such changes explained just 2.6% of the variation in a Swedish analysis^[8].

“Almost all genetic risk factors” for autism “can be found in the general population”, where they are linked to ordinary variation in social behaviour^[12], suggesting that autism sits within the everyday range of human genetic variation. Genetics also helps explain why autistic people differ from one another: a 2025 study found two only modestly related genetic profiles, one linked to earlier diagnosis, the other to later diagnosis and to stronger genetic overlap with ADHD and mental-health conditions^[13].

Older parents and early birth are linked to a higher likelihood, mostly by small amounts

WHO lists factors that “appear to occur more frequently in children with autism, or their parents”, including advanced parental age, maternal diabetes, prenatal exposure to air pollutants or certain heavy metals, prematurity, severe birth complications and low birth weight, and says more research is needed on “how they interact with genetic variations”^[2]. These are population associations, not proof that any factor made a particular child autistic.

Among 5,766,794 children born 1985–2004 in Denmark, Israel, Norway, Sweden and Western Australia, children of fathers aged 50 or over had 1.66 times the risk of an autism diagnosis of those with fathers in their twenties, after adjustment; children of mothers aged 40–49 had 1.15 times the risk, and of mothers under 20, 1.18 times, compared with mothers in their twenties. In absolute terms, 0.87% of children of the oldest fathers had been diagnosed by the end of follow-up, against 0.51% for fathers aged 20–29: a difference of well under one percentage point^[14]. A meta-analysis of 27 studies found 1.41 times the odds for the oldest mothers and 1.55 times for the oldest fathers^[15].

Why is not settled: older fathers may pass on more new genetic changes, but the same study found a higher likelihood among young mothers and couples with large age gaps, and its authors concluded that “multiple mechanisms” are involved^[14].

Being born early shows a clearer gradient. Among all 4,061,795 singleton babies born in Sweden in 1973–2013, 1.4% of those born at full term had an autism diagnosis, against 6.1% of those born extremely preterm (22–27 weeks). After adjustment, extremely preterm birth was associated with 3.72 times the prevalence in boys and 4.19 times in girls, compared with full term. The authors judged the links “largely independent of covariates and shared familial factors, consistent with a potential causal relationship”^[16]: the strongest conclusion in this section, and still an inference from observational data.

Apart from births at the very earliest gestations, the absolute differences are small: fractions of a percentage point for parental age, and 1.9% against 1.4% for babies born late preterm (34–36 weeks) compared with full term^[16]. Nothing here suggests that a parent who had a child later in life, or whose baby arrived early, did anything wrong.

Parenting, vaccines and diet do not cause autism

For decades, mothers carried the blame. Leo Kanner, who defined autism as a clinical condition, portrayed parents of autistic children as “highly educated, intelligent, and emotionally distant”; researchers who compared his clinic records with his published cases concluded the stereotype arose “through a process of confirmation bias”^[17]. The child psychologist Bruno Bettelheim, author of The Empty Fortress (1967), held that the cold “refrigerator mother” “was the actual cause of the child’s autistic behaviour” and proposed “parentectomy”: removing children to institutions such as his own^[18].

None of it survived scrutiny. A 2020 history concludes that “Bettelheim’s ideas about the origin of autism were wrong, that most of his patients were not autistic in the first place, and that his therapy was flawed”^[18]; a 2016 review calls it the “discredited and harmful ‘refrigerator mother hypothesis’”^[19]. Yet the National Autistic Society notes that “even now some parents say they feel blamed for their children being autistic, particularly when trying to get the right support put in place. This is wrong and shouldn’t happen.”^[20] WHO is direct: “Autism is not caused by bad parenting or vaccines and vaccine ingredients, and is not linked to diet.”^[2]

In a Danish study of 657,461 children born 1999–2010, those given the MMR vaccine were no more likely to be diagnosed autistic than unvaccinated children (hazard ratio 0.93, 95% CI 0.85–1.02), including children with an autistic sibling^[21]. The full evidence is in our article on vaccines and autism.

Paracetamol in pregnancy is the newest claim to reach parents. In September 2025 US federal officials announced a link between paracetamol (acetaminophen) in pregnancy and autism, and the US Food and Drug Administration began a label change while stating that “a causal relationship has not been established and there are contrary studies in the scientific literature”^[22]. The UK’s MHRA found “no evidence that taking paracetamol during pregnancy causes autism in children”^[23], the European Medicines Agency reached the same conclusion and left its advice unchanged^[24], and WHO said “no consistent association has been established”^[25]. Among 2,480,797 children born in Sweden in 1995–2019, 1.53% of those exposed in pregnancy and 1.33% of those not exposed had an autism diagnosis by age 10; the small whole-cohort association (hazard ratio 1.05) was absent when siblings were compared (hazard ratio 0.98, 95% CI 0.93–1.04), which the authors said “may have been attributable to familial confounding”^[26]. A 2025 umbrella review found that nine systematic reviews “reported a possible to strong association” but rated confidence in them low to critically low, and concluded that “existing evidence does not clearly link maternal paracetamol use during pregnancy with autism or ADHD in offspring”^[27]. The MHRA advises pregnant women to “speak to their healthcare professional if they have questions about any medication during pregnancy”^[23].

Rising diagnoses do not, by themselves, point to a new cause. In Sweden, autistic traits measured in children born 1993–2002 stayed stable while registered diagnoses rose; the authors concluded that administrative changes, “rather than secular factors affecting the pathogenesis”, were important^[28]. See how many autistic people there are worldwide.

Asked what research should focus on, autistic people put support first

The NHS describes autism as “a difference in how your brain develops”, says “It’s not an illness”, and notes “It is often called a type of neurodivergence”^[29]. Many autistic people see autism as part of natural human diversity, and are wary of causation research for that reason. The Autistic Self Advocacy Network (ASAN) states that autism is “a genetic disability”, but says: “ASAN does not think autism is a good or bad thing. We think it is just another way to be a person.” It warns that “this ‘cause research’ is just another kind of ‘cure’ research” and that “genetic research takes time and money away from other kinds of research that would probably help autistic people more”, while seeing possible benefit in understanding conditions autistic people often have, “like seizures or stomach issues”^[30].

In a UK survey in 2012–13 of 1,517 people, 122 of them autistic adults, autistic adults rated “How can public services best meet the needs of autistic people?” at 4.59 out of 5, but genetic causes at 3.45. Then, 56% of UK autism research grant spending (2007–2011) went on biology, brain and cognition and 5% on services; some autistic adults called this “neurotypical priorities regarding us – not autistic people’s priorities”^[31].

When Autistica asked over 1,000 people in a 2016 James Lind Alliance partnership, none of the top ten concerned causes; they covered mental health, communication, social care, anxiety, education, families, adult diagnosis, employment, sensory processing and services^[32]. A 2021 review of seven priority-setting studies found research “that will lead to real-world changes in the daily lives of the autism community” was very important to respondents^[33].

What the evidence asks of us

• Say plainly that parents are not to blame. Clinicians, schools and support services should repeat what WHO states^[2], at diagnosis and afterwards, because some parents still feel blamed when they seek support^[20]. See also our article on the health of parents and carers.
• Answer new claims with the best-designed evidence. Sibling comparisons are why a small raw association between paracetamol and autism is not read as cause and effect^[26][27]; public bodies should explain evidence of that quality clearly and promptly^[23][24][25].
• Fund what autistic people and families ask for. The authors of the UK study called for “greater involvement of the autism community both in priority setting and in research more broadly”^[31]; community priorities centre on mental health, services and daily life^[32][33].
• Put autistic people at the centre of genetic research. ASAN calls for “field-wide ethical standards that require the leadership of autistic scientists and involvement of self-advocates at every stage of study development”, with protections for privacy and consent^[30].
• Aim genetic findings at health needs autistic people want met, such as the seizures and stomach problems ASAN names^[30].

More evidence is collected in our Autism & Health research library.

Language: this piece uses identity-first language (“autistic people”), the preference of most autistic-led organisations, and uses clinical terms such as “autism spectrum disorder” only when naming a study or diagnostic category. “Risk” appears only where it is a study’s own measure.

Why we're publishing this

Health Insurance UK is a commercial health-insurance resource, not a charity and not a campaign. We compiled this evidence because the question of what causes autism is too often answered with blame or misinformation, and the families and autistic people deserve a clear, sourced account. The case for better support belongs to the autistic community and the organisations it leads; our part is to stand alongside it and say plainly what the research shows.

How to read this data

Heritability figures come from twin and family studies using national registers, mostly Nordic, with data from Israel, Western Australia and the UK. They describe populations, not any one family, and vary with country, period, diagnostic practice and model. The large genetic studies draw mainly on a few high-income countries. Parental-age and preterm findings are observational associations, reported in each study’s own measure; the parental-age percentages are unadjusted and not lifetime figures. The UK funding figures (2007–2011) and survey (2012–13) are dated.

Use this data

Free to cite with attribution to Health Insurance UK. Charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research for general information. It is not medical advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else's health, please speak to a qualified health professional.

Sources

1. Bai D, Yip BHK, Windham GC, et al. Association of genetic and environmental factors with autism in a 5-country cohort. JAMA Psychiatry, 2019. jamanetwork.com/journals/jamapsychiatry/fullarticle/2737582
2. World Health Organization. Autism: questions and answers, 17 September 2025. WHO, 2025. who.int/news-room/questions-and-answers/item/autism-spectrum-disorders-(asd)
3. Sandin S, Lichtenstein P, Kuja-Halkola R, et al. The heritability of autism spectrum disorder. JAMA, 2017. jamanetwork.com/journals/jama/fullarticle/2654804
4. Tick B, Bolton P, Happé F, Rutter M, Rijsdijk F. Heritability of autism spectrum disorders: a meta-analysis of twin studies. Journal of Child Psychology and Psychiatry, 2016. europepmc.org/article/MED/26709141
5. Colvert E, Tick B, McEwen F, et al. Heritability of autism spectrum disorder in a UK population-based twin sample. JAMA Psychiatry, 2015. jamanetwork.com/journals/jamapsychiatry/fullarticle/2173394
6. Sandin S, Yip BHK, Yin W, et al. Examining sex differences in autism heritability. JAMA Psychiatry, 2024. jamanetwork.com/journals/jamapsychiatry/fullarticle/2817601
7. Ozonoff S, Young GS, Bradshaw J, et al. Familial recurrence of autism: updates from the Baby Siblings Research Consortium. Pediatrics, 2024. europepmc.org/article/MED/39011552
8. Gaugler T, Klei L, Sanders SJ, et al. Most genetic risk for autism resides with common variation. Nature Genetics, 2014. europepmc.org/article/MED/25038753
9. Grove J, Ripke S, Als TD, et al. Identification of common genetic risk variants for autism spectrum disorder. Nature Genetics, 2019. europepmc.org/article/MED/30804558
10. Satterstrom FK, Kosmicki JA, Wang J, et al. Large-scale exome sequencing study implicates both developmental and functional changes in the neurobiology of autism. Cell, 2020. europepmc.org/article/MED/31981491
11. Fu JM, Satterstrom FK, Peng M, et al. Rare coding variation provides insight into the genetic architecture and phenotypic context of autism. Nature Genetics, 2022. europepmc.org/article/MED/35982160
12. Robinson EB, St Pourcain B, Anttila V, et al. Genetic risk for autism spectrum disorders and neuropsychiatric variation in the general population. Nature Genetics, 2016. europepmc.org/article/MED/26998691
13. Zhang X, Grove J, Gu Y, et al. Polygenic and developmental profiles of autism differ by age at diagnosis. Nature, 2025. europepmc.org/article/MED/41034588
14. Sandin S, Schendel D, Magnusson P, et al. Autism risk associated with parental age and with increasing difference in age between the parents. Molecular Psychiatry, 2016. europepmc.org/article/PMC/PMC5414073
15. Wu S, Wu F, Ding Y, Hou J, Bi J, Zhang Z. Advanced parental age and autism risk in children: a systematic review and meta-analysis. Acta Psychiatrica Scandinavica, 2017. europepmc.org/article/MED/27858958
16. Crump C, Sundquist J, Sundquist K. Preterm or early term birth and risk of autism. Pediatrics, 2021. europepmc.org/article/MED/34380775
17. Sterwald C, Baker J. Frosted intellectuals: how Dr. Leo Kanner constructed the autistic family. Perspectives in Biology and Medicine, 2019. europepmc.org/article/MED/31761801
18. van Rosmalen L, van der Veer R, van der Horst FCP. The nature of love: Harlow, Bowlby and Bettelheim on affectionless mothers. History of Psychiatry, 2020. europepmc.org/article/PMC/PMC7433398
19. Mandy W, Lai MC. Annual Research Review: The role of the environment in the developmental psychopathology of autism spectrum condition. Journal of Child Psychology and Psychiatry, 2016. europepmc.org/article/MED/26782158
20. National Autistic Society. Myths about the causes of autism (accessed September 2026). National Autistic Society, 2026. autism.org.uk/advice-and-guidance/about-autism/myths-about-the-causes-of-autism
21. Hviid A, Hansen JV, Frisch M, Melbye M. Measles, mumps, rubella vaccination and autism: a nationwide cohort study. Annals of Internal Medicine, 2019. europepmc.org/article/MED/30831578
22. U.S. Food and Drug Administration. FDA responds to evidence of possible association between autism and acetaminophen use during pregnancy (press release, 22 September 2025). FDA, 2025. fda.gov/news-events/press-announcements/fda-responds-evidence-possible-association-between-autism-and-acetaminophen-use-during-pregnancy
23. Medicines and Healthcare products Regulatory Agency. MHRA confirms taking paracetamol during pregnancy remains safe and there is no evidence it causes autism in children (22 September 2025). GOV.UK, 2025. gov.uk/government/news/mhra-confirms-taking-paracetamol-during-pregnancy-remains-safe-and-there-is-no-evidence-it-causes-autism-in-children
24. European Medicines Agency. Use of paracetamol during pregnancy unchanged in the EU (23 September 2025; updated 20 January 2026). EMA, 2025. ema.europa.eu/en/news/use-paracetamol-during-pregnancy-unchanged-eu
25. World Health Organization. WHO statement on autism-related issues, 24 September 2025. WHO, 2025. who.int/news/item/24-09-2025-who-statement-on-autism-related-issues
26. Ahlqvist VH, Sjöqvist H, Dalman C, et al. Acetaminophen use during pregnancy and children’s risk of autism, ADHD, and intellectual disability. JAMA, 2024. europepmc.org/article/MED/38592388
27. Sheikh J, Allotey J, Sobhy S, et al. Maternal paracetamol (acetaminophen) use during pregnancy and risk of autism spectrum disorder and attention deficit/hyperactivity disorder in offspring: umbrella review of systematic reviews. BMJ, 2025. europepmc.org/article/MED/41207796
28. Lundström S, Reichenberg A, Anckarsäter H, Lichtenstein P, Gillberg C. Autism phenotype versus registered diagnosis in Swedish children: prevalence trends over 10 years in general population samples. BMJ, 2015. europepmc.org/article/MED/25922345
29. NHS. What is autism? (page last reviewed 6 May 2026). NHS website, 2026. nhs.uk/conditions/autism/what-is-autism/
30. Autistic Self Advocacy Network. ASAN statement on genetic research and autism. Autistic Self Advocacy Network, 2022. autisticadvocacy.org/wp-content/uploads/2022/03/genetic-statement.pdf
31. Pellicano E, Dinsmore A, Charman T. What should autism research focus upon? Community views and priorities from the United Kingdom. Autism, 2014. europepmc.org/article/MED/24789871
32. Autistica. Your priorities for autism research (James Lind Alliance Priority Setting Partnership, 2016; accessed September 2026). Autistica, 2016. autistica.org.uk/our-research/your-research-priorities
33. Roche L, Adams D, Clark M. Research priorities of the autism community: a systematic review of key stakeholder perspectives. Autism, 2021. europepmc.org/article/MED/33143455