Myths and facts about autism

Myth: Vaccines cause autism.

Fact: They don't. This is among the most thoroughly tested questions in modern medicine.

Large studies covering millions of children across multiple countries have found no link between any vaccine — including MMR — and autism. The 1998 paper that started the myth was retracted for data manipulation and its author lost his medical license. Early signs of autism are often first noticed around the age routine vaccines happen, which creates a coincidence of timing, not causation. Skipping vaccines does not lower autism likelihood; it only removes protection from serious diseases.

Evidence: Hviid et al., Ann Intern Med 2019 followed 657,461 Danish children and found no increased autism rate after MMR, including in children with a sibling history of autism. Taylor et al., Vaccine 2014 pooled five cohort studies covering 1,256,407 children: odds ratio 0.99 (95% CI 0.92–1.06) — no association, with thimerosal and MMR analysed separately and also null. The 1998 paper was fully retracted by The Lancet in 2010 after the GMC found its claims of consecutive referral and ethics approval to be false; its author was struck off the UK medical register the same year.

Myth: Autism is caused by parenting — too little affection, too much screen time.

Fact: Autism is a neurodevelopmental difference with strong genetic underpinnings. Nothing you did or didn't do caused it.

The "refrigerator mother" theory was discredited half a century ago. Twin and family studies show autism is among the most heritable of developmental conditions, with differences in brain development beginning before birth. No study has shown that screens cause autism — reviews that report an association grade their own evidence as very low, because the studies are largely cross-sectional and cannot separate cause from effect. Replacing screen time with responsive back-and-forth interaction is still good advice for every toddler's language development, for reasons that have nothing to do with autism risk.

Evidence: Bai et al., JAMA Psychiatry 2019 analysed 2,001,631 people across Denmark, Finland, Sweden, Israel and Western Australia and put heritability at roughly 80%, with no support for maternal effects. Stoner et al., NEJM 2014 found focal patches of disrupted cortical layering in 10 of 11 autistic children — a disruption that forms prenatally, before any parenting has happened. On screens: a 2024 systematic review in J Autism Dev Disord covering 30 studies and 356,666 participants reports an association but rates the evidence very low, citing cross-sectional designs, unresolved confounding, and reverse causality.

Myth: He makes eye contact and is affectionate — so it can't be autism.

Fact: Many autistic children make some eye contact, smile, hug, and deeply love their families.

Autism is a pattern across social communication, flexibility, and sensory experience — not the absence of affection. Inconsistent or effortful eye contact, or connection mainly on the child's own terms, is fully compatible with autism. No single behavior rules autism in or out; that's what structured screening is for.

Evidence: Rutgers et al., J Child Psychol Psychiatry 2004, a meta-analysis of sixteen attachment studies, found secure attachment in 53% of autistic children assessed with the Strange Situation procedure, and concluded that attachment security is compatible with autism. The difference from non-autistic comparison children disappeared entirely in samples of children with higher mental development. That autism is defined as a pattern rather than any single behavior is the structure of the DSM-5-TR criteria themselves.

Myth: Let's wait and see — many kids grow out of it. Boys just talk late.

Fact: "Wait and see" is the single most costly piece of common advice.

Some late talkers do catch up — but there is no reliable way to know in advance which ones, and the signs of autism go well beyond late speech. Pediatric guidance is unambiguous: persistent parental concern warrants screening and referral, not waiting. Evaluation and early intervention are free or covered in most systems and harm no one if the concern proves unfounded; waiting a year, if the concern was founded, has a real developmental cost.

Evidence: The American Academy of Pediatrics clinical report (Hyman, Levy & Myers, Pediatrics 2020;145(1):e20193447) recommends standardized autism screening at 18 and 24 months alongside ongoing surveillance — not watchful waiting. On whether an early picture holds: Pierce et al., JAMA Pediatrics 2019 re-evaluated 1,269 toddlers first assessed between 12 and 36 months; diagnostic stability was 0.84 overall and only 2% of those initially identified later looked typically developing. On what acting early buys: Whitehouse et al., JAMA Pediatrics 2021 randomised 103 infants showing early signs; at age three, 6.7% of the intervention group met criteria for autism versus 20.5% of those receiving usual care. Read that last figure with care — it was a secondary outcome in a single trial, and Project AIM (Sandbank et al., Psychol Bull 2020), the largest meta-analysis of early autism intervention, finds the overall evidence base weaker than the field's confidence in it.

Myth: Speaking two languages at home confuses children and causes speech delays or autism-like signs.

Fact: Bilingualism does not cause language delay or autism.

Bilingual children may split vocabulary across languages early on, but total vocabulary and milestones stay on track. Research shows autistic children in bilingual homes are not disadvantaged by hearing both languages. Never drop your family's language "just in case" — connection in the language of your heart matters.

Evidence: Hambly & Fombonne, J Autism Dev Disord 2012 compared bilingually and monolingually exposed autistic children and found no additional language delay from bilingual exposure. Drysdale, van der Meer & Kagohara, Rev J Autism Dev Disord 2015 reached the same conclusion across a systematic review, as did Gilhuber, Raulston & Galley, Autism 2023. No study in this literature supports dropping a home language.

Myth: An autism diagnosis is a tragedy and means a limited life.

Fact: Autism is a spectrum, and autistic people live full, meaningful lives across the entire range of support needs.

A diagnosis doesn't change who your child is — it explains their experience and unlocks support. Outcomes are not fixed at diagnosis, and early support can meaningfully improve communication. Listen to autistic adults: asked directly what research and services should prioritise, they name the things that change daily life, not a cure.

Evidence: Pellicano, Dinsmore & Charman, Autism 2014;18(7):756–70 surveyed and interviewed autistic adults, family members, practitioners and researchers, and found a clear mismatch between how autism research is funded and what the community actually asks for: priorities cluster around services, support and everyday life rather than biology or cure. On outcomes not being fixed: Whitehouse et al., JAMA Pediatrics 2021, tempered by Sandbank et al., Psychol Bull 2020 — support helps, and anyone quoting a precise figure for how much is going beyond the evidence.

Myth: There are treatments that cure autism — special diets, supplements, chelation, "detox."

Fact: There is no cure for autism, and products sold as cures range from useless to actively dangerous.

Bleach-based "protocols" (MMS/CD), chelation, and hyperbaric "cures" have injured children and are condemned by every serious medical body. Evidence-based support — speech-language therapy, developmental and behavioral interventions, occupational therapy, parent coaching — helps autistic children build communication and life skills. Anyone promising recovery from autism is selling something. See how to find legitimate support.

Evidence: On bleach: the FDA warns outright against Miracle Mineral Solution, which becomes industrial bleach when prepared as directed. On chelation: the Cochrane review (James et al., 2015) found no clinical-trial evidence of benefit and concluded that, given reported hypocalcaemia, renal impairment and death, the risks outweigh any proven benefit. On hyperbaric oxygen: the Cochrane review (Xiong et al., 2016) found no improvement in social interaction, communication, behavior or cognition. On elimination diets: Quan et al., Nutrition Reviews 2022;80(5):1237–46, the first meta-analysis of gluten- and casein-free diets in autism, found no reliable benefit. (Cochrane's own review of these diets was withdrawn in 2019 as out of date; sites still citing it are citing a document Cochrane has retired.)

Myth: Autism mainly affects boys — girls don't really get it.

Fact: Autism is diagnosed more often in boys, but girls are systematically under-identified — and identified later.

Autistic girls more often "mask": imitating peers, holding it together at school, and melting down at home. Their intense interests may look socially typical (animals, characters, reading) and get overlooked. If your daughter's differences show mainly in exhaustion after social situations, friendship struggles, and rigidity at home, take that pattern seriously even if teachers see "no problem."

Evidence: Loomes, Hull & Mandy, JAACAP 2017 pooled 54 studies: the male-to-female ratio is 4.20 to 1 among children already diagnosed, but drops to 3.25 to 1 in studies that screened the whole population regardless of diagnosis. The authors' conclusion is explicit — "girls who meet criteria for ASD are at disproportionate risk of not receiving a clinical diagnosis." On masking: Hull et al., J Autism Dev Disord 2017;47(8):2519–34 interviewed 92 autistic adults and described camouflaging as deliberate masking and compensation, with exhaustion as its most consistently reported cost.