{
  "id": "autism-spectrum-disorder",
  "title": "Autism spectrum disorder",
  "lang": "en",
  "source_lang": "en",
  "summary": "A lifelong neurodevelopmental condition defined by differences in social communication and by restricted, repetitive behaviors.",
  "category": "health",
  "subcategory": "diseases",
  "type": "concept",
  "aliases": [
    "autism",
    "autistic spectrum disorder",
    "autism spectrum condition",
    "ASD",
    "自闭症"
  ],
  "generated_at": "2026-10-10",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/autism-spectrum-disorder/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/autism-spectrum-disorder/",
      "json": "https://aiwiki.page/zh/autism-spectrum-disorder.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Autism spectrum disorder (ASD), usually called autism, is a neurodevelopmental condition that begins in early childhood and lasts throughout life. It is defined by persistent differences in social communication and interaction, together with restricted, repetitive patterns of behavior, interests, or activities. Because its presentation varies widely in language ability, intellectual profile, sensory sensitivity, and support needs, clinicians and researchers treat it as a spectrum rather than a single condition. Its prevalence is now estimated at roughly 1 in 31 eight-year-olds in the United States, according to the surveillance network of the U.S. public health agency, although that figure comes from a set of monitoring sites rather than a nationally representative sample.\n\nAutism matters for several reasons. It is among the most heavily studied developmental conditions, and its causes have shifted from early psychological theories to a model that combines strong genetic influence with prenatal and early-life environmental factors. Diagnosis has become more common and earlier, which raises questions about how much of the rise reflects real change and how much reflects better recognition. Autism also touches education, employment, disability rights, and mental health, and it has given rise to the neurodiversity movement, which reframes many traits as differences rather than deficits.\n\nThis entry covers the definition and diagnostic criteria, the core and associated features of autism, its epidemiology, diagnostic methods, genetic and environmental risk, neurobiology, the history of research, interventions, adult outcomes, social and legal dimensions, and persistent controversies and open questions."
    },
    {
      "heading": "Definition and diagnostic criteria",
      "text": "Autism spectrum disorder (ASD) is classified as a neurodevelopmental disorder in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5, 2013), and its text revision (DSM-5-TR, 2022). DSM-5 replaced the separate DSM-IV diagnoses of autistic disorder, Asperger’s disorder, childhood disintegrative disorder, and pervasive developmental disorder not otherwise specified (PDD-NOS) with one spectrum diagnosis. The revision was developed by the Neurodevelopmental Disorders Work Group, chaired by Susan Swedo; the association cited inconsistent application of the earlier categories across clinical settings as a reason for consolidation. “Spectrum” accommodates variation in presentation and severity within the diagnosis rather than preserving those categories as distinct subtypes. (psychiatry.org)\n\nThe DSM framework requires two core domains. Criterion A comprises persistent difficulties in social communication and interaction across multiple contexts, with evidence in all three specified areas: social-emotional reciprocity; nonverbal communication used in interaction; and developing, maintaining, and understanding relationships. Illustrative manifestations include difficulty sustaining reciprocal conversation, coordinating gestures with speech, or adjusting behavior to different social situations. DSM-5-TR explicitly added “all” to the wording of criterion A, clarifying the intended threshold rather than introducing a new requirement. Current presentation and developmental history can both establish the relevant features. (cdc.gov)\n\nCriterion B requires at least two of four categories of restricted or repetitive behavior: repetitive movements, object use, or speech; insistence on sameness or inflexible routines; unusually intense or narrowly focused interests; and heightened or reduced sensory responsiveness, or unusual sensory interests. Repeatedly arranging objects, marked distress when routines change, and strong reactions to particular textures are illustrative examples, not independently sufficient grounds for diagnosis. Both core domains are necessary: social-communication difficulties alone do not establish ASD. (cdc.gov)\n\nThree further conditions govern the diagnostic threshold. Features must originate in early development, although recognition may occur later, when social demands exceed the person’s capacities or learned compensatory strategies no longer conceal difficulties. They must produce clinically significant impairment in important areas of current functioning. The presentation must also not be better explained by intellectual disability or global developmental delay alone. Autism and intellectual disability can coexist, but the social-communication difficulties must exceed those expected at the person’s general developmental level. Thus, developmental onset is distinct from age at diagnosis, and isolated traits are distinct from the full disorder. (psychiatry.org)\n\nDSM severity is recorded separately for the two core domains: level 1 indicates a need for support, level 2 substantial support, and level 3 very substantial support. Different levels can therefore be assigned to social communication and repetitive behavior in the same person. Additional specifiers record accompanying intellectual and language impairment, an associated medical or genetic condition or environmental factor, other neurodevelopmental or mental disorders, and catatonia. These descriptors supplement the spectrum diagnosis rather than recreate the former subtypes. (psychiatryonline.org)\n\nThe World Health Organization’s International Classification of Diseases, eleventh revision (ICD-11), likewise classifies autism under neurodevelopmental disorders, using code 6A02. Its clinical requirements identify persistent difficulties in initiating and sustaining reciprocal social communication alongside restricted, repetitive, and inflexible patterns of behavior. ICD-11 subdivisions distinguish accompanying intellectual-developmental disorder and degrees of functional-language impairment, rather than using DSM’s three support levels as their principal coding structure. WHO published the detailed clinical diagnostic requirements in 2024; ICD-11 had been adopted in May 2019 and became effective as a basis for health reporting in January 2022. Consequently, the classification system, edition, and recorded specifiers are relevant when comparing diagnostic groups across countries or research studies. (iris.who.int)"
    },
    {
      "heading": "History of the concept and terminology",
      "text": "The term autism initially referred to a feature of schizophrenia, rather than a distinct developmental condition. In 1911, Swiss psychiatrist Eugen Bleuler used it to describe withdrawal into an inner world of thought. Leo Kanner gave the term a different clinical application in his 1943 paper “Autistic Disturbances of Affective Contact,” published in Nervous Child. His case series described 11 children with early difficulties in relating to people, unusual language development, repetitive activities, and resistance to changes in routine. Kanner distinguished this pattern from schizophrenia by emphasizing its presence from early life rather than withdrawal following previously typical development. His account combined observations of difficulties with examples of preserved abilities, including striking memory for words, numbers, or sequences. (sparkforautism.org)\n\nIn Vienna, Hans Asperger’s 1944 report described children under the historical label “autistic psychopathy.” Their difficulties with social integration overlapped with Kanner’s observations, but Asperger placed greater emphasis on articulate speech, unusual specialized interests, and the coexistence of social difficulties with original thinking or particular abilities. These descriptions did not establish equivalent diagnostic categories, nor did they correspond precisely to the later division between autistic disorder and Asperger syndrome. Asperger’s paper reached a wider English-language readership through Lorna Wing’s 1981 clinical account and Uta Frith’s annotated English translation, published in 1991. (media.autism.org.uk)\n\nMid-century explanations frequently interpreted autism through parent–child relationships. Although Kanner’s original paper proposed an inborn difficulty in establishing affective contact, it also described parents as emotionally distant. Bruno Bettelheim subsequently popularized the “refrigerator mother” hypothesis, attributing autistic withdrawal to parental rejection or emotional coldness; his The Empty Fortress appeared in 1967. This explanation was abandoned as unsupported, rather than incorporated into the developmental account of autism. Biological research provided an important alternative: Susan Folstein and Michael Rutter’s 1977 study of 21 same-sex twin pairs reported concordance for autism in 36% of identical-twin pairs and none of the fraternal-twin pairs. These historically important findings supported hereditary influences, although the small study did not establish a complete causal model. (sparkforautism.org)\n\nThe spectrum concept developed through studies that extended beyond narrowly defined early infantile autism. In 1979, Wing and Judith Gould examined children younger than 15 in an area of London. They identified a subgroup resembling Kanner’s cases, but found that difficulties in reciprocal social interaction, language abnormalities, and repetitive behavior also occurred together in children outside that subgroup. Their work supported classification by patterns of social interaction rather than treating Kanner’s presentation as an isolated category. Wing’s subsequent discussion of Asperger syndrome helped connect presentations involving fluent speech with the broader continuum of social and communication difficulties. (pubmed.ncbi.nlm.nih.gov)\n\nDiagnostic terminology changed alongside these conceptual developments. DSM-III, published in 1980, separated infantile autism from the schizophrenia category and placed it within pervasive developmental disorders. DSM-IV distinguished autistic disorder, Asperger’s disorder, childhood disintegrative disorder, and pervasive developmental disorder not otherwise specified. In 2013, DSM-5 brought these four diagnoses together under autism spectrum disorder. The change replaced separate subtype labels with a framework intended to represent variation within a shared neurodevelopmental condition; the detailed criteria are discussed in the preceding section. (sparkforautism.org)\n\nClinical nomenclature does not determine how individuals describe themselves. Identity-first language—“autistic person”—and person-first language—“person with autism”—remain distinct preferences. A UK survey by Lorcan Kenny and colleagues, published online in 2015 and in a 2016 journal issue, collected responses from 3,470 people. “Autism” and “on the autism spectrum” received broad endorsement across groups, but autistic adults more frequently endorsed “autistic,” whereas professionals more frequently endorsed “person with autism.” The study found no universally preferred formulation. Such differences concern identity and the relationship between a person and a diagnostic label, rather than competing diagnostic definitions; professional, research, family, and self-advocate usage therefore cannot be assumed to coincide. (pubmed.ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Core features and associated traits",
      "text": "Social-communication differences involve both initiating interaction and coordinating responses to other people. Eye contact may be reduced, inconsistent, or poorly integrated with gestures and speech; facial expression and vocal intonation may not convey the cues a listener expects. Joint attention—sharing a focus on an object or event with another person—may develop atypically, as when a child seldom points to show something interesting or brings an object to another person to share enjoyment. Reciprocal conversation can be difficult even when vocabulary is extensive: a speaker may give a detailed monologue without recognizing opportunities for the other person to contribute. These characteristics vary between individuals and do not occur uniformly in every autistic person. (nimh.nih.gov)\n\nRestricted and repetitive patterns include movements such as hand-flapping, rocking, and spinning, alongside repeated actions with objects, such as lining up toys or concentrating on a toy vehicle’s wheels. Insistence on sameness can appear as a need to follow a particular route, perform activities in a fixed sequence, or maintain familiar meals and daily routines. Small changes or transitions between activities may provoke substantial distress. Highly focused interests are distinguished by their intensity or narrowness rather than by any particular subject: numbers, moving objects, or detailed factual knowledge may become sustained preoccupations. Repetitive movement, ritualized routines, and focused interests can occur in different combinations. (cdc.gov)\n\nSensory differences include unusually strong or weak responses to sound, light, touch, temperature, smells, and food textures, as well as active interest in particular sensations. Examples include distress at specific sounds or clothing textures, apparent reduced responsiveness to pain or temperature, and prolonged fascination with lights or movement. Such responses entered the DSM-5’s restricted-and-repetitive-behavior domain in 2013. In a 2022 study, Anne V. Kirby and colleagues examined records for 25,627 autistic children aged four or eight identified through United States surveillance during 2006–2014; 74% had documented sensory features. This figure describes features recorded in health and educational records, not a standardized sensory examination of every child. (publications.aap.org)\n\nLanguage development ranges from little or no spoken language to fluent, grammatically complex speech. Some children acquire vocabulary unevenly, developing extensive knowledge of words associated with an interest while having difficulty understanding conversation. Others can decode written words earlier than they can understand the text. Fluent speakers may nevertheless have difficulties with pragmatics—the use of language in social context—including turn-taking, maintaining a shared topic, and interpreting vocal tone or body language. Echolalia is the repetition of previously heard speech: immediate repetition follows an utterance closely, whereas delayed repetition reproduces it later. A remembered question such as “Do you want something to drink?” may be used to request a drink, illustrating that repeated language can carry communicative meaning. (nidcd.nih.gov)\n\nCognitive ability also varies widely, and strengths in memory or particular academic skills may coexist with difficulties in other areas. In the CDC’s 2025 report on 2022 surveillance, cognitive information was available for 5,292 eight-year-old children identified with autism, representing 61.4% of the identified group. Among those with available information, 39.6% were classified as having intellectual disability, using an intelligence-quotient threshold of 70 or below; 24.2% scored between 71 and 85, and 36.1% above 85. These surveillance categories do not describe the full range of everyday functioning or establish a universal cognitive distribution across the spectrum. (cdc.gov)\n\nAssociated traits include delayed motor development, difficulties with coordination, sleep disturbance, and gastrointestinal complaints. Motor differences can affect gross- and fine-motor tasks; sleep problems include difficulty falling asleep and repeated waking. Constipation, diarrhea, and abdominal pain are reported, although they are not defining features of autism. Communication differences can affect how discomfort becomes apparent: gastrointestinal pain may present through agitation or disrupted sleep rather than a specific verbal complaint. These associated conditions can contribute independently to daily difficulties and are distinct from the core social-communication and repetitive-behavior features. (cdc.gov)"
    },
    {
      "heading": "Epidemiology and prevalence",
      "text": "The United States Autism and Developmental Disabilities Monitoring (ADDM) Network reported an identified prevalence of 32.2 per 1,000 eight-year-old children—approximately 1 in 31—in surveillance year 2022. Published in April 2025, the estimate combined 16 surveillance sites rather than a nationally representative sample. This public-health surveillance system identifies children through existing health and education records: qualifying evidence includes a documented autism diagnosis, autism special-education eligibility, or an autism diagnostic code. The estimate therefore measures recorded identification within defined communities, not the results of examining every child independently. (cdc.gov)\n\nSite-specific estimates ranged from 9.7 per 1,000 children in Laredo, Texas, to 53.1 per 1,000 in the California surveillance area, equivalent to 0.97% and 5.31%. These figures describe particular catchment areas, not statewide populations. Differences in access to evaluation, educational identification, and available records affect ascertainment. For example, restricting the case definition to documented diagnostic statements produced an overall estimate of 22.5 per 1,000, compared with 32.2 under the broader definition. Geographic variation consequently cannot be interpreted directly as variation in underlying biological risk. (cdc.gov)\n\nRecorded prevalence was 49.2 per 1,000 boys and 14.3 per 1,000 girls in 2022, a ratio of approximately 3.4 to 1. (cdc.gov) The size of this difference depends partly on how cases are found. A 2017 meta-analysis by Rachel Loomes, Laura Hull, and William Mandy included 54 studies and found a male-to-female odds ratio of 3.25 in studies that screened general populations, compared with 4.56 in studies relying on previously diagnosed cases. The authors interpreted this difference as evidence that girls meeting autism criteria are disproportionately likely to remain undiagnosed. These findings indicate that diagnostic ascertainment contributes to the observed gap, without establishing how much of the remaining difference reflects biological factors. (pubmed.ncbi.nlm.nih.gov)\n\nSuccessive ADDM estimates rose from 6.7 per 1,000 eight-year-olds in 2000 to 23.0 in 2018, 27.6 in 2020, and 32.2 in 2022; participating sites also changed over this period. (cdc.gov) Changes in diagnostic definitions, recognition, referral, service access, and reporting can increase measured prevalence without an equivalent increase in newly arising cases. Their contribution has been quantified in some settings: Stefan Hansen, Diana Schendel, and Erik Parner’s Danish registry study, published in 2015, estimated that changes to diagnostic criteria in 1994 and the inclusion of outpatient diagnoses in 1995 explained approximately 60% of the observed increase among children born in 1980–1991. That estimate applies to the particular population and reporting changes studied, rather than providing a universal explanation for rising prevalence. (jamanetwork.com)\n\nInternational estimates use different populations and methods. Jinan Zeidan and colleagues’ 2022 systematic review identified 99 estimates from 71 studies across 34 countries, with a median prevalence of 100 per 10,000, or 1%; this was a median of study estimates, not a population-weighted worldwide rate. The review documented substantial variation in case ascertainment, diagnostic criteria, and age ranges. (onlinelibrary.wiley.com) The World Health Organization’s September 2025 fact sheet instead reported an estimated global prevalence of approximately 1 in 127 people for 2021. This all-age estimate is not directly comparable with ADDM surveillance of eight-year-olds, and prevalence remains unknown in many low- and middle-income countries. (who.int)"
    },
    {
      "heading": "Diagnosis, screening, and assessment",
      "text": "Autism diagnosis is clinical: it combines developmental history, interviews with parents or other informants, and direct observation of social communication and behavior. No laboratory test or biomarker independently establishes the diagnosis. Assessment examines when differences first appeared, how they vary between settings, and their effects on everyday functioning. School reports and earlier health records can supplement interviews; discrepancies between reported behavior and a clinic observation may prompt further observation at home or school. Genetic testing can identify an associated condition, and electroencephalography can investigate suspected seizures, but neither replaces behavioral assessment. (newsinhealth.nih.gov)\n\nScreening identifies children who require closer evaluation rather than confirming autism. In the United States, the American Academy of Pediatrics recommends autism-specific screening at 18 and 24 months, alongside continuing developmental surveillance. Experienced professionals can make a reliable diagnosis by approximately age two. (cdc.gov) The Modified Checklist for Autism in Toddlers, Revised with Follow-Up (M-CHAT-R/F), developed by Diana Robins, Deborah Fein, and Marianne Barton, is a two-stage parent-report instrument validated for ages 16–30 months. Its initial questionnaire contains 20 items. Scores of 3–7 lead to a structured follow-up interview; a follow-up score of at least 2 constitutes a positive screen. Scores of 8–20 permit referral without that interview. The follow-up clarifies ambiguous answers and reduces false positives, but some screen-positive children have other developmental delays rather than autism. Persistent developmental concerns remain relevant even after a negative screen. (mchatscreen.com)\n\nStandardized diagnostic instruments organize evidence without substituting for clinical judgment. The Autism Diagnostic Observation Schedule, particularly its second edition (ADOS-2), uses structured social, play, and conversational activities to elicit observable behavior. Its modules are selected according to expressive language and chronological age, allowing assessment from minimally verbal toddlers to verbally fluent adults. Activities may examine response to name, pretend play, and reciprocal interaction. (wpspublish.com) The Autism Diagnostic Interview–Revised (ADI-R) complements observation with a systematic caregiver interview about development and autism-related behavior. Neither an instrument’s cutoff nor a single clinic encounter is sufficient on its own; diagnostic conclusions integrate the findings with developmental and contextual evidence. (depts.washington.edu)\n\nEarlier identification has increased, but substantial delays remain. The U.S. ADDM report published in 2025, using 2022 records, found a median earliest documented diagnosis of 47 months among 5,887 autistic eight-year-olds with a recorded diagnostic statement. Site medians ranged from 36 months in California to 69.5 months in Laredo, Texas. Among children with available evaluations, only 50.3% had been evaluated by 36 months. More recently born cohorts were more likely to have been identified by age four, although this does not establish a uniform decline in diagnostic age across populations. (cdc.gov) Research has documented racial, ethnic, and socioeconomic inequalities in diagnostic access; reported differences vary by study, with some recent evidence indicating narrowing racial gaps. (pubmed.ncbi.nlm.nih.gov)\n\nDifferential assessment considers intellectual disability, language disorder, attention-deficit/hyperactivity disorder, anxiety disorders, hearing impairment, and conditions involving developmental regression, including Rett syndrome. These conditions may resemble aspects of autism or coexist with it. Assessment therefore also characterizes intellectual ability, language, adaptive functioning, and physical and mental health rather than treating the diagnostic label as a complete description of the person. (nice.org.uk) In adults, early developmental history, school records, and accounts from relatives can help distinguish lifelong features from difficulties arising later. (nice.org.uk)\n\nAutistic camouflaging can obscure observable traits through rehearsed conversational scripts, imitation of others’ behavior, or deliberate management of eye contact. Laura Hull and colleagues’ 2017 qualitative study of 92 autistic adults described both masking and compensatory strategies, illustrating why apparently fluent interaction may not reveal its underlying effort. (doi.org) NICE guidance specifically notes that autism may be under-recognized in girls and that coping strategies or supportive environments can mask signs in older children. Adult assessment consequently includes self-reported difficulties and developmental evidence, not only the behavior visible during an appointment. (nice.org.uk)"
    },
    {
      "heading": "Genetic architecture",
      "text": "Twin and family studies indicate that genetic differences account for a large proportion of population variation in susceptibility to autism. A 2016 meta-analysis by Beata Tick and colleagues estimated heritability at 64–91%, depending on the statistical model and assumptions about prevalence. In 2019, Dan Bai and colleagues examined 2,001,631 people from Denmark, Finland, Sweden, Israel, and Western Australia, including 22,156 diagnosed with autism; the median country-specific heritability estimate was 80.8% (95% confidence interval, 73.2–85.5%). These estimates concern variation in an underlying statistical liability within populations: they do not mean that a fixed percentage of an individual’s autism is “genetic,” nor that genetic susceptibility invariably results in autism. (pubmed.ncbi.nlm.nih.gov)\n\nThe genetic architecture combines many common alleles with small individual effects and rarer variants that can have substantially larger effects. In a 2019 genome-wide association study, Jakob Grove and colleagues compared 18,381 autistic people with 27,969 controls and identified five genome-wide-significant loci. Such studies detect statistical associations across the genome, rather than a single gene that determines the diagnosis. Common variants collectively contribute to polygenic susceptibility, including in people who also carry rare, high-impact variants. A 2022 family-based study involving 37,375 individuals found additive contributions from rare variants and polygenic background, demonstrating that these are overlapping components rather than mutually exclusive explanations. (pubmed.ncbi.nlm.nih.gov)\n\nRare variants include protein-truncating changes, amino-acid substitutions, and copy-number variants—deletions or duplications of stretches of DNA that alter gene dosage. Some are inherited; others are de novo mutations, identified in a child but absent from the parental samples tested, helping explain autism-associated variants in families without a previous diagnosis. Large sequencing studies show substantial genetic heterogeneity. F. Kyle Satterstrom and colleagues identified 102 associated genes in 2020 at a false discovery rate of 10% or less. In 2022, Jack Fu and colleagues analyzed 63,237 individuals and identified 185 genes at a 5% threshold, including 72 at the stricter 0.1% threshold. Combining autism and developmental-delay cohorts yielded 373 associated genes at the stricter threshold; that larger count therefore includes genes implicated across both conditions, not exclusively in autism. (pmc.ncbi.nlm.nih.gov)\n\nSeveral recurrent findings concern proteins involved in synaptic organization. Neuroligins, encoded by NLGN genes, and neurexins, encoded by NRXN genes, participate in adhesion and signaling across synapses; SHANK proteins provide postsynaptic scaffolds that organize receptors and associated signaling machinery. Stéphane Jamain and colleagues reported autism-associated NLGN3 and NLGN4 mutations in 2003, and Christelle Durand and colleagues reported SHANK3 mutations in 2007. Experimental work has shown that some autism-associated SHANK3 variants impair synaptic transmission and neurexin–neuroligin-mediated signaling. Synaptic genes are not the only category implicated: sequencing studies also identify genes regulating gene expression and early neuronal development, indicating multiple molecular routes to overlapping developmental outcomes. (pmc.ncbi.nlm.nih.gov)\n\nDefined genetic syndromes provide particularly clear examples of these mechanisms. Fragile X syndrome usually results from expansion of a CGG repeat beyond 200 copies in FMR1, accompanied by abnormal DNA methylation and reduced expression of its protein product, FMRP; autism occurs in a subset of affected individuals. Tuberous sclerosis complex results from pathogenic variants in TSC1 or TSC2. Their protein products normally restrain mTOR signaling, a pathway regulating cellular growth and protein synthesis, so loss of their function disrupts this regulation. These syndromes explain only a minority of autism diagnoses, but connect identifiable genetic changes to specific biological pathways. Neither syndrome invariably entails autism, and most autistic people do not have either condition. (ncbi.nlm.nih.gov)\n\nGenotype–phenotype relationships are consequently probabilistic rather than one-to-one. The same associated gene can contribute to autism, developmental delay, or other neurodevelopmental presentations, while different variants within a gene can have different functional effects. Fu and colleagues found differences in mutation frequency between autism and developmental-delay cohorts even among shared associated genes. Family analyses also show that common genetic background contributes alongside rare variants, including in carriers of larger-effect changes. Thus, identifying an associated variant does not by itself specify an individual’s language development, intellectual ability, or support requirements. (escholarship.org)"
    },
    {
      "heading": "Environmental and prenatal risk factors",
      "text": "Environmental risk research examines chemical exposures, maternal health, and pregnancy or birth conditions alongside genetic susceptibility. A gene–environment interaction occurs when the association of an exposure with an outcome differs according to genetic background; finding separate genetic and environmental associations does not itself demonstrate interaction. One exploratory CHARGE analysis involving 158 autistic children and 147 controls reported an interaction between genomic duplication burden and ozone exposure, but its small sample limits generalization. Observational associations also require scrutiny for confounding: shared familial characteristics or maternal illness may account for an apparent exposure effect without establishing causation. (pmc.ncbi.nlm.nih.gov)\n\nPrenatal medication exposure provides some of the clearest quantitative evidence. In 2013, Jakob Christensen and colleagues reported a Danish cohort of 655,615 children, including 508 exposed to valproate during pregnancy. Estimated autism-spectrum cumulative incidence over 14 years was 4.42% among exposed children versus 1.52% among unexposed children; the adjusted hazard ratio was 2.9, with a 95% confidence interval of 1.7–4.9. Restricting comparisons to mothers with epilepsy reduced the spectrum-wide estimate and widened its uncertainty, illustrating the importance of the underlying treatment indication. Thalidomide has a much smaller historical evidence base: Kerstin Strömland and colleagues’ 1994 study identified at least four autistic individuals among 100 Swedish people with thalidomide embryopathy. Their congenital abnormalities suggested exposure approximately 20–24 days after conception. (jamanetwork.com)\n\nMaternal infection has been investigated both as an exposure and as a possible trigger of immune-system activation. Experimental models implicate inflammatory signaling rather than necessarily direct fetal infection: Gloria Choi and colleagues demonstrated in mice in 2016 that maternal T-helper-17 cells and the cytokine interleukin-17A were required for particular induced cortical and behavioral abnormalities. These findings do not establish the same causal pathway in human autism. In a 2022 Swedish cohort of 549,967 children, autism was diagnosed in 3.3% of children exposed to recorded maternal infection during pregnancy and 2.5% of unexposed children. The association disappeared in sibling comparisons, suggesting that shared familial factors explained the population-level association in that study. (pubmed.ncbi.nlm.nih.gov)\n\nMaternal diabetes and parental age are associated with autism identification, although neither association specifies a single mechanism. Anny Xiang and colleagues’ 2015 Southern California study found that gestational diabetes diagnosed by 26 weeks remained associated with autism after adjustment for demographic and medical factors: the hazard ratio was 1.42, with a 95% confidence interval of 1.15–1.74. Later-diagnosed gestational diabetes was not associated in that cohort. Sven Sandin and colleagues’ five-country study, published online in 2015, included 5,766,794 children. Relative to parents aged 20–29, adjusted relative risks were 1.15 for mothers aged 40–49 and 1.66 for fathers aged at least 50. Parental age combines biological and familial influences and is not simply an external environmental exposure. (pubmed.ncbi.nlm.nih.gov)\n\nPreterm birth, low birth weight, and complications involving oxygen deprivation have also been studied; these correlated conditions are difficult to separate from one another and from underlying maternal or fetal disorders. Casey Crump and colleagues’ 2021 Swedish study followed 4,061,795 singleton children. Recorded autism prevalence was 6.1% after extremely preterm birth at 22–27 weeks, 2.1% after birth before 37 weeks overall, and 1.4% after birth at 39–41 weeks. Associations were only slightly attenuated in sibling analyses, indicating that shared familial factors did not fully explain them in this cohort. (pmc.ncbi.nlm.nih.gov)\n\nThe Childhood Autism Risks from Genetics and the Environment (CHARGE) study, led by Irva Hertz-Picciotto, began fieldwork in California in 2003 and combined clinical assessment, exposure histories, and biological samples. A 2014 analysis associated residential proximity to organophosphate applications during pregnancy with approximately 60% higher odds of autism; proximity was an exposure proxy, not a measured absorbed dose. A 2013 CHARGE analysis also reported associations with modeled traffic-related air pollution. Metals research illustrates a different measurement limitation: a CHARGE blood-mercury study found no significant autism–control difference after accounting for exposure sources, but samples obtained at ages two to five could not reconstruct prenatal exposure. These designs distinguish signals requiring further investigation from evidence sufficient to attribute an individual child’s autism to an exposure. (nda.nih.gov)"
    },
    {
      "heading": "Neurobiology and brain development",
      "text": "Postmortem studies have identified differences in the cerebral cortex, cerebellum, limbic structures, and brainstem, but findings vary between individuals and do not establish a single defining autistic brain anatomy. A 2014 stereological investigation of 38 brain subdivisions reported significant differences concentrated in the striatum, amygdala, and cerebellum, rather than throughout all regions examined. A smaller 2008 study of five autistic brains and two controls found differences in neuronal shape and orientation in the medial superior olive, a brainstem nucleus involved in sound localization. These observations identify possible circuit-level differences but cannot establish their prevalence across the spectrum. (pmc.ncbi.nlm.nih.gov)\n\nCortical organization provides evidence concerning the timing of brain development. In 2014, Rich Stoner, Eric Courchesne, and colleagues examined postmortem tissue from children aged 2–15 years. They detected patches of atypical layer-specific gene expression in prefrontal cortex from 10 of 11 autistic children, compared with one of 11 controls. The patches measured approximately 5–7 mm and involved several cortical layers; reduced marker expression did not simply reflect fewer neurons. The authors proposed that the findings could reflect altered prenatal neuronal differentiation or migration, although their study did not determine the mechanism. The small sample and limited tissue coverage restrict generalization. (nejm.org)\n\nLongitudinal imaging distinguishes developmental trajectories from anatomical differences measured at a single age. In a 2017 study led by Heather Hazlett, infants with an older autistic sibling underwent magnetic resonance imaging at 6, 12, and 24 months. Infants subsequently diagnosed with autism showed accelerated cortical surface-area expansion between 6 and 12 months, preceding increased total brain-volume growth between 12 and 24 months. The later growth was associated with emerging social difficulties. This finding concerns a particular familial-risk cohort and does not imply that all autistic children have enlarged brains or follow the same growth trajectory. It also distinguishes surface-area expansion from cortical thickening, which are biologically different processes. (nature.com)\n\nAt the cellular level, proposed mechanisms include altered synaptic development, plasticity, and the balance between excitatory and inhibitory signaling. Experimental support includes a 2011 study in which optogenetically increasing relative excitation in mouse prefrontal cortex impaired social behavior and information processing. Such manipulation demonstrates a possible mechanism, not a universal explanation of human autism. Guomei Tang and colleagues reported in 2014 that childhood-to-adolescent reductions in dendritic spine density were attenuated in autistic postmortem temporal cortex. In associated mouse experiments, excessive mTOR signaling impaired autophagy—the cellular degradation and recycling process—and developmental synaptic pruning. These results connect intracellular signaling to circuit refinement, but do not establish that impaired pruning occurs in every autistic person. (nature.com)\n\nDifferences in connectivity can extend beyond individual synapses to distributed brain networks. Functional imaging measures relationships between regional signals rather than directly counting anatomical connections. A 2026 cross-species study examined 20 mouse models and human imaging data from 940 autistic and 1,036 non-autistic participants, identifying patterns of both reduced and increased functional connectivity. The opposing patterns were associated with different molecular pathways, illustrating why a simple account of uniformly weak long-distance connectivity is insufficient. These research-defined groupings are not established clinical subtypes. (nature.com)\n\nImmune-system signaling is another investigated mechanism. A 2022 analysis of 725 postmortem cortical samples from 49 autistic and 54 comparison individuals identified increased expression of immune- and glia-associated genes alongside reduced expression of synaptic genes; these associations do not establish whether immune changes initiate or follow altered development. In 2016, Gloria Choi and colleagues showed that experimentally induced maternal immune activation in mice engaged the cytokine interleukin-17A pathway, producing cortical abnormalities and altered offspring behavior. Genetic models involving autism-associated mutations likewise allow investigators to manipulate specific pathways and test causal effects within an experimental system. Mouse social-interaction, repetitive-behavior, and vocalization measures nevertheless represent selected behavioral components, not the full human condition, and causal findings in these models cannot by themselves establish causation in people. (pmc.ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Interventions and therapies",
      "text": "No medication has established efficacy for treating autism’s core features as a whole; pharmacological treatment instead addresses particular associated symptoms or co-occurring conditions. Developmental, behavioral, and communication interventions target functional goals that vary with the person’s communication abilities, developmental stage, sensory needs, and priorities. Examples include expressing needs, participating in reciprocal play, and completing daily routines. The United Kingdom’s National Institute for Health and Care Excellence (NICE) recommends developmentally adjusted, play-based social-communication interventions involving parents, carers, or teachers, with therapist modelling and video feedback to improve interaction. These approaches also modify how communication partners respond, rather than focusing exclusively on the autistic child’s behavior. (cdc.gov)\n\nApplied behavior analysis (ABA) uses learning principles, including reinforcement, prompting, and the systematic measurement of behavior, to teach skills. Structured discrete-trial teaching breaks skills into repeated teaching opportunities, whereas naturalistic approaches embed learning in everyday activities. Early intensive behavioral intervention (EIBI), an ABA-based comprehensive approach, has a substantial research literature, but its evidence is less conclusive than its widespread use might suggest. A 2018 Cochrane review included five studies involving 219 children, with interventions lasting 24–36 months; only one study used randomized allocation. It found low-certainty evidence of improvements in adaptive behavior, intellectual functioning, and receptive and expressive language, but no clear improvement in autism severity or problem behavior. Thus, evidence for particular developmental outcomes does not establish that every ABA program is effective, or that it changes all core features. (asha.org)\n\nDevelopmental and naturalistic behavioral models combine systematic teaching with child-led play, shared engagement, and naturally occurring reinforcement. The Early Start Denver Model (ESDM), associated with Sally Rogers and Geraldine Dawson, integrates developmental and behavioral methods with parent participation. In their randomized controlled trial, published in 2010, 48 children aged 18–30 months were assigned to ESDM or community intervention for two years. Mean cognitive standard-score gains were 17.6 points in the ESDM group and 7.0 points in the comparison group; adaptive-behavior trajectories also favored ESDM. These findings demonstrated benefit in a particular small trial, rather than a uniform response among autistic children. NICE’s broader social-communication framework similarly emphasizes interactive play, shared attention, and coaching communication partners. (pubmed.ncbi.nlm.nih.gov)\n\nSpeech-language therapy addresses understanding and producing language, conversational interaction, and functional everyday communication. Augmentative and alternative communication (AAC) supplements speech or provides another means of expression through gestures, signs, picture boards, written language, or speech-generating devices. Communication-partner training helps make these systems usable beyond therapy sessions; AAC may also support people who speak but cannot meet all their communication needs through speech alone. Occupational therapy addresses participation in activities such as dressing, eating, play, and other daily routines, using task practice and adaptations to activities or environments. These functional services are distinct from claims that a particular sensory intervention treats autism comprehensively. (asha.org)\n\nIntervention methods and goals are contested, especially when programs prioritize compliance or appearing non-autistic. Laura K. Anderson’s qualitative study, published online in 2022, interviewed seven autistic adults who had received ABA in childhood. Participants described some benefits alongside negative experiences, including traumatic memories and objections to being required to behave like non-autistic peers. A 2025 Canadian mixed-methods study by Nancy Sharon Marshall and colleagues likewise documented diverse experiences, including helpful skill development and concerns about coercion and suppression of autistic traits. Such studies identify experiences relevant to intervention design, but their selected samples cannot establish population-wide rates of benefit or harm. (pubmed.ncbi.nlm.nih.gov)\n\nEvaluation depends on what is measured, who measures it, and whether gains persist and generalize beyond the teaching setting. Micheal Sandbank and colleagues’ 2023 Project AIM meta-analysis found that estimated benefits became more limited when analyses excluded caregiver-reported outcomes and studies with high risk of detection bias. Improvements on practiced tasks therefore cannot automatically be interpreted as broad developmental change or improved quality of life. Adverse-event reporting was also inadequate: only 10% of reports mentioned adverse events, and only 28% of that subset described monitoring procedures. Long-term follow-up, independent assessment, everyday participation, and outcomes meaningful to autistic people remain important dimensions of evaluating both benefit and possible harm. (bmj.com)"
    },
    {
      "heading": "Treatment of co-occurring conditions",
      "text": "Treatment distinguishes co-occurring medical and psychiatric disorders from autism’s core developmental features. Pain, constipation, disrupted sleep, anxiety, and difficulties communicating needs can contribute to aggression or self-injury; assessment therefore examines these factors rather than automatically attributing a behavioral change to autism. The United Kingdom’s National Institute for Health and Care Excellence (NICE) guidelines place treatment of identified physical or mental disorders, alongside environmental adjustments, before medication directed at challenging behavior. (nice.org.uk)\n\nIn the United States, risperidone is approved for irritability associated with autistic disorder in children aged 5–16 years, and aripiprazole for the same indication at ages 6–17. The target symptoms include severe tantrums, aggression, and deliberate self-injury, not the core social-communication features. Both drugs affect dopamine and serotonin signaling, although their receptor actions differ. Their adverse effects include weight gain, sedation, and movement-related symptoms; risperidone can also elevate prolactin. (accessdata.fda.gov) In an eight-week randomized controlled trial published in 2002 by the Research Units on Pediatric Psychopharmacology (RUPP) Autism Network, 69% of children receiving risperidone met the response criterion, compared with 12% receiving placebo. Response required both a reduction of at least 25% on an irritability scale and a clinician rating of “much improved” or “very much improved.” These results concern short-term control of serious behavioral problems, rather than treatment of autism itself. (nejm.org)\n\nCo-occurring attention-deficit/hyperactivity disorder (ADHD), anxiety disorders, depression, and obsessive-compulsive disorder are managed through disorder-specific psychological and pharmacological approaches, with adaptations for communication, sensory needs, and medication tolerability. NICE describes adaptations to cognitive behavioral therapy such as concrete language, written or visual materials, explicit explanations, regular breaks, and involvement of a supporter with the person’s agreement. (nice.org.uk) Medication response cannot simply be assumed to match that in non-autistic populations. In the RUPP Network’s 2005 crossover trial of methylphenidate for hyperactivity in children with pervasive developmental disorders, 35 of 72 participants (49%) were classified as responders, while adverse effects caused 13 (18%) to discontinue treatment. The study used the diagnostic terminology current at that time. (pubmed.ncbi.nlm.nih.gov)\n\nEpilepsy occurs more frequently in autistic people, but seizure treatment follows general neurological principles rather than an autism-specific drug regimen. Accurate identification of seizure type and epilepsy syndrome guides medication selection; treatment also takes account of associated conditions, interactions, and adverse effects. (publications.aap.org) NICE’s epilepsy guidance favors a single antiseizure medication where possible, with alternative monotherapy or additional medication when seizure control is inadequate. Antiseizure treatment is directed at epilepsy: anticonvulsants are not recommended simply to modify autism’s core features. (nice.org.uk)\n\nSleep treatment addresses difficulty falling asleep, night waking, irregular schedules, and contributing problems such as reflux, seizures, anxiety, or medication effects. (publications.aap.org) The American Academy of Neurology’s 2020 guideline, led by Ashura Williams Buckley, places behavioral sleep strategies first and supports melatonin when these measures have been insufficient and contributing conditions have been addressed. Melatonin influences circadian timing and sleep initiation; treatment studies assess outcomes such as sleep-onset latency and total sleep time, rather than improvement in autism’s core features. The guideline emphasizes preparation quality, potential adverse effects, and the lack of long-term safety data. It found insufficient evidence for routine use of weighted blankets or specialized mattresses to improve sleep. (neurology.org)\n\nGastrointestinal and nutritional problems receive condition-specific assessment, including evaluation of constipation, reflux, restricted food intake, growth, and possible nutrient deficiencies. Communication differences may complicate recognition of discomfort, making behavioral changes relevant to the medical assessment. (publications.aap.org) Restrictive diets are not established treatments for autism’s core features, and NICE does not recommend gluten- or casein-free exclusion diets for that purpose. (nice.org.uk) Chelation likewise has no established role in treating autism. The U.S. Food and Drug Administration warns that unapproved chelation products marketed for autism can cause serious harm, including dehydration, kidney failure, and death. (fda.gov)"
    },
    {
      "heading": "Lifespan, adulthood, and health outcomes",
      "text": "Autism may first be recognized in adulthood, including middle and later life. Routes to assessment include recognition of similar traits after a child’s diagnosis and contact with mental-health services for longstanding difficulties. A 2019 qualitative study by Steven Stagg and Hannah Belcher interviewed nine recently diagnosed adults over 50: participants had previously received treatment for anxiety and depression, while their autistic characteristics had remained unrecognized. Participants described reassessing earlier experiences after diagnosis, although these accounts do not establish that diagnosis itself improves health outcomes. In older adults, reconstructing developmental history can also be difficult when parents are no longer alive or childhood records are unavailable. (staging.core.mdpi.com)\n\nAdult functioning varies substantially across daily living, employment, education, and relationships; childhood cognitive ability does not determine an individual’s eventual level of independence. Patricia Howlin and colleagues’ 2004 longitudinal study followed 68 people with childhood performance IQ scores of at least 50, from a mean age of seven to a mean age of 29. Under the study’s composite outcome classification, 22% had “good” or “very good” outcomes, whereas 58% had “poor” or “very poor” outcomes; most continued to depend substantially on family or services. Childhood IQ of at least 70 was associated with better group-level outcomes, but outcomes within the normal IQ range remained highly variable. These figures describe a small, historically selected cohort rather than contemporary population-wide probabilities. The classification primarily concerned conventional markers of independence and social functioning, not a direct measure of subjective quality of life. (pubmed.ncbi.nlm.nih.gov)\n\nThe transition out of school can interrupt established support arrangements. Paul Shattuck and colleagues’ 2011 analysis of a nationally representative US survey examined autistic young adults aged 19–23 after high school. An estimated 39.1% had received none of four measured service types—mental-health services, medical evaluation or assessment, speech therapy, and case management—during the preceding two years or since leaving school. Lower income and African American ethnicity were associated with higher adjusted odds of receiving none of these services. This measure did not encompass every possible form of support, but documented a substantial discontinuity during early adulthood. (pubmed.ncbi.nlm.nih.gov)\n\nHealth burdens extend beyond autism’s defining features. Lisa Croen and colleagues’ 2015 study identified 1,507 autistic adult members of Kaiser Permanente Northern California and found higher recorded rates of multiple psychiatric and physical conditions than among comparison adults, including depression, anxiety, epilepsy, sleep disorders, obesity, and diabetes. Healthcare access is a separate concern: adult studies describe barriers involving communication, sensory environments, appointment organization, and clinicians’ understanding of autism. These interacting features of the health system can make obtaining care difficult even when services formally exist; evidence on effective support during transfer from pediatric to adult healthcare remains limited. (pubmed.ncbi.nlm.nih.gov)\n\nPopulation studies also identify elevated premature mortality. Tatja Hirvikoski and colleagues’ Swedish register study, published in 2016, compared 27,122 diagnosed autistic people with 2,672,185 matched controls. During observation, 2.60% of the autistic group and 0.91% of controls died, with an all-cause mortality odds ratio of 2.56. Cause-of-death patterns differed according to intellectual disability and sex; epilepsy, other medical conditions, suicide, and external causes were among the investigated contributors. These associations do not establish that autism itself directly causes excess deaths, and the average age among people who died during follow-up is not equivalent to population life expectancy. (cambridge.org)\n\nElizabeth O’Nions and colleagues’ UK matched-cohort study, published online in November 2023 and collected in a January 2024 journal issue, estimated an apparent life-expectancy deficit of 6.14 years for diagnosed autistic men without intellectual disability and 6.45 years for corresponding women. With intellectual disability, estimated deficits were 7.28 and 14.59 years, respectively. The authors emphasized that underdiagnosis—particularly among people with fewer support needs—could bias these estimates toward a less healthy diagnosed subgroup. The estimates summarize mortality rates during the studied period, rather than predicting any individual’s lifespan, and cannot automatically be transferred to other countries or generations. (pubmed.ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Education, employment, and legal recognition",
      "text": "In the United States, autism is an explicit eligibility category under the Individuals with Disabilities Education Act (IDEA). Eligible pupils receive a free appropriate public education through an individualized education program (IEP), which records current academic and functional performance, goals, and required services. IDEA also requires education in the least restrictive environment: removal from regular classrooms is appropriate only when education there cannot be achieved satisfactorily with supplementary aids and services. In Endrew F. v. Douglas County School District (2017), a case involving an autistic pupil, the Supreme Court rejected a standard requiring only minimal educational benefit and held that an educational program must be reasonably calculated to enable progress appropriate to the child’s circumstances. (sites.ed.gov)\n\nOther jurisdictions use different planning and funding mechanisms. In England, schools provide special educational needs support, while children and young people requiring provision beyond that support may receive an education, health and care (EHC) plan, potentially extending to age 25. Local authorities assess needs and specify additional educational, health, and social support; decisions about assessment, provision, or school placement can be challenged through the SEND Tribunal. These arrangements differ from the American IEP system in administrative responsibility and scope, rather than constituting a uniform international entitlement. (gov.uk)\n\nAdjustments address particular barriers rather than prescribing an identical environment for every autistic person. School support can include additional teaching assistance, small-group instruction, and help with communication or classroom participation. Workplace examples include written instructions, predictable schedules, reduced noise or altered lighting, and advance notice of changes. The Job Accommodation Network also describes modifications to interviews and supervisory communication. Such reasonable accommodations can change how a task is communicated or performed without removing its essential requirements; their suitability depends on the individual and the setting. (gov.uk)\n\nEmployment inequalities are documented, although estimates depend on population definitions. The British Buckland Review of Autism Employment, published on February 28, 2024 and led by Robert Buckland, reported that approximately three in ten working-age autistic disabled people were employed, compared with five in ten disabled people overall and eight in ten non-disabled people. This is an employment rate, not an unemployment rate: people outside the labour market are included in its denominator. The review identified ambiguous interview questions, emphasis on social presentation rather than job skills, and inaccessible sensory environments as barriers, and recommended changes to recruitment and workplace support. (gov.uk)\n\nStructured vocational programs have been tested experimentally. Paul Wehman and colleagues’ multisite randomized controlled trial, published in 2020, enrolled 156 autistic young people aged 18–21 with substantial support needs. Project SEARCH plus ASD Supports combined workplace internships, classroom instruction, employment planning, and continuing support after graduation. Within one year of graduation, 73.4% of the intervention group had obtained competitive employment at or above minimum wage, compared with 17% of controls for whom outcome data were available. Employed intervention participants averaged 21.2 working hours weekly. These results concern a defined transition-age population and program, rather than demonstrating equivalent effects for all autistic adults or autism-focused hiring schemes. (pubmed.ncbi.nlm.nih.gov)\n\nLegal recognition links access to services with protection against discrimination. In the United States, the Americans with Disabilities Act protects qualified disabled applicants and employees; its employment provisions generally cover employers with at least 15 employees and require reasonable accommodation unless this creates undue hardship. Internationally, the Convention on the Rights of Persons with Disabilities, adopted on December 13, 2006 and effective from May 3, 2008, supplies a broader human rights framework. Article 24 requires inclusive education and individualized support, while Article 27 addresses equal access to work, nondiscrimination, and workplace accommodation. These obligations apply to states parties and require implementation through national laws and institutions. (eeoc.gov)"
    },
    {
      "heading": "Neurodiversity, self-advocacy, and autistic perspectives",
      "text": "The neurodiversity movement frames autism as part of natural variation in human cognition and challenges the assumption that becoming less visibly autistic is necessarily a desirable outcome. This position does not inherently deny disability or substantial support needs. The Autistic Self Advocacy Network (ASAN), founded in 2006, explicitly describes autism as a developmental disability while advocating acceptance, communication access, and assistance with daily living rather than elimination of autistic identity. Its stated distinction is between support that expands a person’s choices and intervention intended principally to make that person resemble non-autistic peers. Speech therapy, occupational therapy, and alternative communication can therefore be compatible with its neurodiversity framework. (autisticadvocacy.org)\n\nSelf-advocacy encompasses both individual decisions—such as expressing preferences about care, relationships, or living arrangements—and collective participation in policies affecting autistic people. ASAN’s account includes communication through actions as well as speech and rejects the idea that extensive disability makes someone incapable of advocacy. This emphasis also changes whose judgments count in research. A 2014 UK study by Elizabeth Pellicano, Adam Dinsmore, and Tony Charman used interviews, focus groups, and a subsequent survey to examine community priorities. Participants emphasized research with practical effects on everyday life, including services and support, revealing a mismatch with the prevailing emphasis on biological mechanisms. This approach places quality of life, communication, and self-determination alongside clinical measures rather than treating reductions in observable autistic traits as sufficient evidence of benefit. (autisticadvocacy.org)\n\nDisagreement over intervention goals is particularly evident in discussions of stimming, including repetitive movements or sounds. In a 2019 qualitative study, Steven K. Kapp, Robyn Steward, and colleagues examined autistic adults’ accounts of these behaviors. Participants described stimming as a means of regulating overwhelming emotions or sensory experiences and reported that other people’s disapproval could pressure them to suppress it. These accounts distinguish a behavior’s visible unusualness from its function for the person performing it. ASAN’s 2021 intervention ethics paper similarly criticized goals centered on unquestioning compliance, arguing that the ability to refuse and communicate boundaries is part of self-determination. These are explicitly articulated advocacy positions and qualitative findings, not evidence that every repetitive behavior is harmless or that all interventions have identical effects. (pubmed.ncbi.nlm.nih.gov)\n\nResearch incorporating autistic perspectives has also examined social contributions to psychological distress. Monique Botha and David Frost’s 2018 study of 111 autistic respondents found that everyday discrimination, internalized stigma, and concealment were associated with poorer mental health after accounting for general stress exposure. Their minority-stress framework treats stigmatized social status as a possible additional source of distress, rather than attributing all mental-health difficulties directly to autism. The observational design establishes associations rather than causation, but identifies experiences that conventional symptom-focused assessments may overlook. (discovery.ucl.ac.uk)\n\nService-access research provides concrete examples of participatory work. Dora Raymaker and colleagues’ 2017 study used community-based participatory methods to develop a healthcare-barriers checklist and surveyed 437 adults, including 209 autistic participants. Among autistic respondents, 32% reported difficulty processing information quickly enough for real-time healthcare discussions, 30% reported sensory problems in facilities, and 29% reported difficulty communicating with providers. Related work by Christina Nicolaidis and colleagues developed and evaluated an online primary-care toolkit through an academic–community partnership. Such projects translate first-person reports into assessment instruments and service tools, making communication and environmental barriers explicit targets of healthcare improvement. (pubmed.ncbi.nlm.nih.gov)\n\nTerminology preferences remain heterogeneous. Lorcan Kenny and colleagues’ UK survey, published in 2016, collected responses from 3,470 autistic people, relatives, professionals, and other community members. “Autistic” was endorsed more frequently by autistic adults and parents than by professionals, whereas almost half of professionals endorsed “person with autism.” Identity-first wording can express autism’s place in personal identity; person-first wording separates the person linguistically from the diagnosis. The survey found no universally preferred terminology, so neither professional convention nor a majority preference represents every autistic individual. (pubmed.ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Controversies and misconceptions",
      "text": "The claim that vaccines cause autism became prominent after Andrew Wakefield and colleagues published a 1998 Lancet report describing 12 children. It relied partly on parents’ recollections of symptom onset following measles, mumps, and rubella (MMR) vaccination; this uncontrolled case series could not establish causation. On February 2, 2010, The Lancet fully retracted the paper following findings by the UK General Medical Council. The retraction specifically identified false statements about consecutive patient referral and approval by a local ethics committee. The withdrawal was therefore not simply a consequence of disagreement over interpretation, but involved defects in the reported research procedures. (pubmed.ncbi.nlm.nih.gov)\n\nLarge population studies subsequently tested the vaccine hypothesis directly. In 2019, Anders Hviid and colleagues reported a Danish cohort of 657,461 children: the adjusted autism hazard ratio comparing MMR-vaccinated with unvaccinated children was 0.93, with a 95% confidence interval of 0.85–1.02, providing no evidence of increased risk. Analyses also found no consistent increase among children with an autistic sibling or during specified periods after vaccination. In December 2025, the World Health Organization’s Global Advisory Committee on Vaccine Safety reaffirmed that vaccines do not cause autism, including vaccines containing thiomersal or aluminium. Its assessment distinguished methodologically rigorous studies from positive findings with serious bias; two studies concerning aluminium and autism were ecological studies, whose population-level comparisons could not establish individual causation. (pubmed.ncbi.nlm.nih.gov)\n\nParent-blaming explanations, particularly the “refrigerator mother” hypothesis associated with Bruno Bettelheim, attributed autism to insufficient parental affection. These explanations have been discredited: autism is not caused by emotionally cold or otherwise “bad” parenting. Contemporary research instead investigates its biological neurodevelopmental origins, including genetic influences and developmental exposures. An NIH-funded study reported in 2019 estimated that inherited genetic factors accounted for approximately 80% of variation in autism liability. Such a population estimate does not mean that 80% of an individual’s autism has a genetic cause, nor does it establish a single mechanism common to all autistic people. (leedsautism.org.uk)\n\nCommercial claims of a cure through exclusion diets, supplements, chelation, or stem-cell treatments go beyond reliable evidence. NICE’s evidence review found insufficient evidence of the safety and efficacy of exclusion diets and vitamin or mineral supplements for autistic behaviours in adults; its childhood guideline likewise excludes gluten- or casein-free diets as treatments for core features. Chelation uses substances that bind metals to facilitate their removal, but its legitimate use for certain poisonings does not establish efficacy for autism. The US Food and Drug Administration warns that chelation can cause dehydration, kidney failure, and death. It also states that regenerative medicine therapies are not approved to treat autism and has received reports of infections and tumour formation from unapproved regenerative products generally; these reports should not be misrepresented as autism-specific complication rates. (nice.org.uk)\n\nRising recorded prevalence presents a different kind of controversy: the increase is observable, but its causes cannot be inferred from diagnosis counts alone. In a 2015 study of 677,915 Danish children born during 1980–1991, Stefan Hansen and colleagues estimated that changes in diagnostic criteria in 1994 and the inclusion of outpatient diagnoses in registries in 1995 together explained 60% of the increase in reported prevalence, with a 95% confidence interval of 33%–87%. That estimate concerns a particular population and reporting system, not a universal proportion. The unexplained remainder does not by itself establish increased biological incidence or identify an environmental cause. Similarly, an exposure’s correlation with recorded diagnoses is not equivalent to a causal effect: changes in case ascertainment must be separated from changes in underlying occurrence. (pubmed.ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Open questions and research directions",
      "text": "Research on objective biomarkers seeks to distinguish three uses: supporting diagnosis, identifying biologically informative subgroups, and measuring responses to interventions. These uses require different validation: a measure associated with autism does not necessarily predict an individual’s diagnosis or treatment response. The Autism Biomarkers Consortium for Clinical Trials, established in 2015, has investigated electroencephalography and eye tracking in children aged 6–11, measuring electrical brain activity and attention to social information as possible tools for stratifying trial participants. No definitive biological test replaces developmental and behavioral assessment. Nevertheless, some digital diagnostic aids have entered clinical practice: FDA-cleared EarliPoint devices analyze looking behavior while children watch videos. These are diagnostic aids, not evidence that a universal biological marker has been established. (nimh.nih.gov)\n\nSubtyping research combines phenotypic information with genetic data to investigate whether apparently similar presentations arise through different developmental pathways. In July 2025, Aviya Litman, Natalie Sauerwald and colleagues reported an analysis of 239 features in 5,392 autistic children from the SPARK cohort. A statistical mixture model identified four classes, labeled social/behavioral, mixed autism with developmental delay, moderate challenges, and broadly affected. The researchers replicated comparable phenotypic patterns in 861 participants from the Simons Simplex Collection and found differences between classes in rare genetic variants and developmental gene-expression patterns. These are research classifications rather than established diagnostic categories; the study did not demonstrate subtype-specific treatment effectiveness. Their potential value lies in testing more precise hypotheses about developmental trajectories and biological mechanisms instead of assuming that one mechanism explains the entire spectrum. (nature.com)\n\nThe male-to-female diagnosis ratio raises a related question: how much reflects biological susceptibility, and how much reflects ascertainment? Rachel Loomes, Laura Hull and William Mandy’s 2017 meta-analysis included 54 studies and 53,712 autistic participants. Studies that screened populations regardless of existing diagnosis produced an estimated male-to-female ratio of 3.25, compared with 4.56 in studies relying on previously diagnosed cases. This difference supports under-identification of girls but does not quantify the biological contribution. Separating susceptibility from recognition therefore requires population-based recruitment, rather than examining only people who have already reached diagnostic services. (pubmed.ncbi.nlm.nih.gov)\n\nLarge cohorts also seek to distinguish genetic influences, prenatal exposures and developmental outcomes. NIEHS-supported resources include CHARGE, with more than 1,800 children enrolled, and MARBLES, with more than 400 mother–child pairs. MARBLES collects exposure information before, during and after pregnancy, while EARLI enrolled mothers who already had an autistic child and followed the subsequently born sibling through age three. These prospective designs establish exposure timing and provide stored biological specimens for investigating gene–environment relationships. However, the associations described by these programs are not themselves proof of causation; identifying causal pathways remains distinct from identifying statistical predictors. (niehs.nih.gov)\n\nDigital screening introduces questions about validation across populations. A 2023 study, Early detection of autism using digital behavioral phenotyping, used computer vision and machine learning to quantify children’s gaze, facial movements and other behaviors during an app-based assessment. It explicitly addressed the reduced accuracy of conventional screening in real-world settings, particularly for girls and children of color. Research priorities also concern which outcomes tools and interventions measure. Christina Nicolaidis and colleagues’ 2019 AASPIRE guidelines specify shared decision-making, fair compensation, accessible consent and multiple participation formats for autistic research partners and participants. A 2025 AASPIRE consensus study involving 53 experts identified outcomes including quality of life, self-determination, access to communication, community participation and satisfaction with services, while finding that most available instruments needed adaptation for autistic adults or their proxies. (pubmed.ncbi.nlm.nih.gov)"
    }
  ],
  "links": [
    "neuroscience",
    "public-health-surveillance",
    "public-health",
    "disease",
    "gene",
    "human-rights",
    "diagnosis",
    "international-classification-of-diseases",
    "schizophrenia",
    "language-acquisition",
    "language",
    "prevalence",
    "biomarker",
    "anxiety-disorders",
    "allele",
    "genome",
    "dna",
    "mutation",
    "synapse",
    "gene-expression",
    "dna-methylation",
    "confounding",
    "causation",
    "pregnancy",
    "epilepsy",
    "immune-system",
    "cytokine",
    "neuron",
    "randomized-controlled-trial",
    "quality-of-life",
    "dopamine",
    "serotonin",
    "depression",
    "cognitive-behavioral-therapy",
    "health-system",
    "life-expectancy",
    "education",
    "labor-market",
    "disability",
    "ethics",
    "genetics",
    "correlation",
    "phenotype",
    "machine-learning",
    "informed-consent"
  ],
  "sources": [
    {
      "title": "Autism Spectrum Disorder",
      "url": "https://www.psychiatry.org/File%20Library/Psychiatrists/Practice/DSM/APA_DSM-5-Autism-Spectrum-Disorder.pdf"
    },
    {
      "title": "Psychiatry.org - About DSM-5-TR",
      "url": "https://www.psychiatry.org/psychiatrists/practice/dsm/about-dsm"
    },
    {
      "title": "Clinical Testing and Diagnosis for Autism Spectrum Disorder",
      "url": "https://www.cdc.gov/autism/hcp/diagnosis/index.html?trk=article-ssr-frontend-pulse_little-text-block"
    },
    {
      "title": "Autism Spectrum Disorder",
      "url": "https://www.psychiatry.org/File%20Library/Psychiatrists/Practice/DSM/DSM-5-TR/APA-DSM5TR-AutismSpectrumDisorder.pdf"
    },
    {
      "title": "Highlights of Changes from DSM-IV-TR to DSM-5",
      "url": "https://www.psychiatry.org/File%20Library/Psychiatrists/Practice/DSM/APA_DSM_Changes_from_DSM-IV-TR_-to_DSM-5.pdf"
    },
    {
      "title": "Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders",
      "url": "https://iris.who.int/bitstream/handle/10665/375767/9789240077263-eng.pdf"
    },
    {
      "title": "Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders",
      "url": "https://iris.who.int/bitstream/handle/10665/375767/9789240077263-eng.pdf?sequence=1"
    },
    {
      "title": "Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders (CDDR)",
      "url": "https://www.who.int/publications/i/item/9789240077263"
    },
    {
      "title": "How Did Autism Become Autism?",
      "url": "https://sparkforautism.org/discover_article/autism-history/"
    },
    {
      "title": "Autistic Disturbances of Affective Contact",
      "url": "https://neurodiversity.net/library_kanner_1943.html"
    },
    {
      "title": "Asperger's syndrome: a clinical account",
      "url": "https://pubmed.ncbi.nlm.nih.gov/7208735/"
    },
    {
      "title": "Asperger's syndrome: a clinical account",
      "url": "https://media.autism.org.uk/m/284ff8b7e289a1d8/original/Asperger-Syndrome-A-Clinical-Account-by-Lorna-Wing.pdf"
    },
    {
      "title": "‘Autistic psychopathy’ in childhood",
      "url": "https://www.cambridge.org/core/books/abs/autism-and-asperger-syndrome/autistic-psychopathy-in-childhood/C8439B5AC1215CE53946943845E2FD3F"
    },
    {
      "title": "The empty fortress; infantile autism and the birth of the self",
      "url": "https://www.ncbi.nlm.nih.gov/nlmcatalog/113761"
    },
    {
      "title": "Severe impairments of social interaction and associated abnormalities in children: epidemiology and classification",
      "url": "https://pubmed.ncbi.nlm.nih.gov/155684/"
    },
    {
      "title": "Language, social, and cognitive impairments in autism and severe mental retardation",
      "url": "https://pubmed.ncbi.nlm.nih.gov/6927697/"
    },
    {
      "title": "DSM History",
      "url": "https://www.psychiatry.org/psychiatrists/practice/dsm/about-dsm/history-of-the-dsm"
    },
    {
      "title": "Which terms should be used to describe autism? Perspectives from the UK autism community",
      "url": "https://pubmed.ncbi.nlm.nih.gov/26134030/"
    },
    {
      "title": "Autism Spectrum Disorder - National Institute of Mental Health (NIMH)",
      "url": "https://www.nimh.nih.gov/health/publications/autism-spectrum-disorder"
    },
    {
      "title": "Signs and Symptoms of Autism Spectrum Disorder",
      "url": "https://www.cdc.gov/autism/signs-symptoms/index.html"
    },
    {
      "title": "Sensory features in autism: Findings from a large population-based surveillance system",
      "url": "https://stacks.cdc.gov/view/cdc/117054"
    },
    {
      "title": "Autism Spectrum Disorder: Communication Problems in Children",
      "url": "https://www.nidcd.nih.gov/health/autism-spectrum-disorder-communication-problems-children"
    },
    {
      "title": "Prevalence and Early Identification of Autism Spectrum Disorder Among Children Aged 4 and 8 Years — Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022",
      "url": "https://www.cdc.gov/mmwr/volumes/74/ss/ss7402a1.htm"
    },
    {
      "title": "Data and Statistics on Autism Spectrum Disorder",
      "url": "https://www.cdc.gov/autism/data-research/"
    },
    {
      "title": "What Is the Male-to-Female Ratio in Autism Spectrum Disorder? A Systematic Review and Meta-Analysis",
      "url": "https://pubmed.ncbi.nlm.nih.gov/28545751/"
    },
    {
      "title": "Global prevalence of autism: A systematic review update",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC9310578/"
    },
    {
      "title": "Autism",
      "url": "https://www.who.int/news-room/fact-sheets/detail/autism-spectrum-disorders"
    },
    {
      "title": "Recommendations: Autism spectrum disorder in under 19s—recognition, referral and diagnosis",
      "url": "https://www.nice.org.uk/guidance/cg128/chapter/Recommendations"
    },
    {
      "title": "Clinical Screening for Autism Spectrum Disorder",
      "url": "https://www.cdc.gov/autism/hcp/diagnosis/screening.html"
    },
    {
      "title": "M-CHAT-R Follow-Up (M-CHAT-R/F)",
      "url": "https://www.mchatscreen.com/wp-content/uploads/2025/03/M-CHAT-R_F_Rev_March2025.pdf"
    },
    {
      "title": "Autism Diagnostic Observation Schedule, Second Edition",
      "url": "https://www.wpspublish.com/ados-2-autism-diagnostic-observation-schedule-second-edition?hsa_acc=6243382947&hsa_ad=350562097364&hsa_cam=1933626877&hsa_grp=71360921235&hsa_kw=ados+2&hsa_mt=b&hsa_net=adwords&hsa_src=g&hsa_tgt=aud-549698309755%3Akwd-302490665885&hsa_ver=3"
    },
    {
      "title": "How do health care providers diagnose autism?",
      "url": "https://www.nichd.nih.gov/health/topics/autism/conditioninfo/diagnose"
    },
    {
      "title": "Autism, by Bryan King",
      "url": "https://depts.washington.edu/uwhatc/FCAP/resources/Autism_by_Bryan_King_MD.pdf"
    },
    {
      "title": "Disparities in Service Use Among Children With Autism: A Systematic Review",
      "url": "https://pubmed.ncbi.nlm.nih.gov/32238530/"
    },
    {
      "title": "Racial Differences in the Prevalence of Autism Spectrum Disorder: A Systematic Review",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC12367818/"
    },
    {
      "title": "Recommendations: Autism spectrum disorder in adults—diagnosis and management",
      "url": "https://www.nice.org.uk/guidance/CG142/chapter/Recommendations"
    },
    {
      "title": "Heritability of autism spectrum disorders: a meta-analysis of twin studies",
      "url": "https://pubmed.ncbi.nlm.nih.gov/26709141/"
    },
    {
      "title": "Association of Genetic and Environmental Factors With Autism in a 5-Country Cohort",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6646998/"
    },
    {
      "title": "Identification of common genetic risk variants for autism spectrum disorder",
      "url": "https://pubmed.ncbi.nlm.nih.gov/30804558/"
    },
    {
      "title": "A phenotypic spectrum of autism is attributable to the combined effects of rare variants, polygenic risk and sex",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC9474668/"
    },
    {
      "title": "Large-Scale Exome Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology of Autism",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/pmid/31981491/"
    },
    {
      "title": "Rare coding variation provides insight into the genetic architecture and phenotypic context of autism",
      "url": "https://escholarship.org/content/qt04m356rr/qt04m356rr_noSplash_2783e7a157788c54595e8aa100045a4d.pdf"
    },
    {
      "title": "Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC1925054/"
    },
    {
      "title": "Mutations in the gene encoding the synaptic scaffolding protein SHANK3 are associated with autism spectrum disorders",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2082049/"
    },
    {
      "title": "Autism-Associated Mutations in ProSAP2/Shank3 Impair Synaptic Transmission and Neurexin–Neuroligin-Mediated Transsynaptic Signaling",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3752148/"
    },
    {
      "title": "FMR1 Disorders",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK1384/?report=reader"
    },
    {
      "title": "Tuberous Sclerosis Complex",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK1220/?report=reader"
    },
    {
      "title": "Tuberous sclerosis complex gene products, Tuberin and Hamartin, control mTOR signaling by acting as a GTPase-activating protein complex toward Rheb",
      "url": "https://pubmed.ncbi.nlm.nih.gov/12906785/"
    },
    {
      "title": "The joint effect of air pollution exposure and copy number variation on risk for autism",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6364847/"
    },
    {
      "title": "Autism in thalidomide embryopathy: a population study",
      "url": "https://pubmed.ncbi.nlm.nih.gov/8157157/"
    },
    {
      "title": "The maternal interleukin-17a pathway in mice promotes autism-like phenotypes in offspring",
      "url": "https://pubmed.ncbi.nlm.nih.gov/26822608/"
    },
    {
      "title": "Maternal infection during pregnancy and likelihood of autism and intellectual disability in children in Sweden: a negative control and sibling comparison cohort study",
      "url": "https://pubmed.ncbi.nlm.nih.gov/36087610/"
    },
    {
      "title": "Association of maternal diabetes with autism in offspring",
      "url": "https://pubmed.ncbi.nlm.nih.gov/25871668/"
    },
    {
      "title": "Autism risk associated with parental age and with increasing difference in age between the parents",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5414073/"
    },
    {
      "title": "Preterm or Early Term Birth and Risk of Autism",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC9809198/"
    },
    {
      "title": "The CHARGE Study: Childhood Autism Risks From Genetics and the Environment #2066",
      "url": "https://nda.nih.gov/edit_collection.html?id=2066"
    },
    {
      "title": "Blood Mercury Concentrations in CHARGE Study Children with and without Autism",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2831962/"
    },
    {
      "title": "Stereological study of the neuronal number and volume of 38 brain subdivisions of subjects diagnosed with autism reveals significant alterations restricted to the striatum, amygdala and cerebellum",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4177256/"
    },
    {
      "title": "Morphological features of the medial superior olive in autism",
      "url": "https://pubmed.ncbi.nlm.nih.gov/18291353/"
    },
    {
      "title": "Early brain development in infants at high risk for autism spectrum disorder",
      "url": "https://www.nature.com/articles/nature21369"
    },
    {
      "title": "Early brain development in infants at high risk for autism spectrum disorder",
      "url": "https://pubmed.ncbi.nlm.nih.gov/28202961/"
    },
    {
      "title": "Neocortical excitation/inhibition balance in information processing and social dysfunction",
      "url": "https://www.nature.com/articles/nature10360"
    },
    {
      "title": "Loss of mTOR-dependent macroautophagy causes autistic-like synaptic pruning deficits",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4159743/"
    },
    {
      "title": "Autism subtypes identified using cross-species functional connectivity analyses",
      "url": "https://www.nature.com/articles/s41593-026-02287-z"
    },
    {
      "title": "Broad transcriptomic dysregulation occurs across the cerebral cortex in ASD",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC9668748/"
    },
    {
      "title": "The maternal interleukin-17a pathway in mice promotes autismlike phenotypes in offspring",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4782964/"
    },
    {
      "title": "mTOR-related synaptic pathology causes autism spectrum disorder-associated functional hyperconnectivity",
      "url": "https://www.nature.com/articles/s41467-021-26131-z"
    },
    {
      "title": "Treatment and Intervention for Autism Spectrum Disorder",
      "url": "https://www.cdc.gov/autism/treatment/"
    },
    {
      "title": "Autism spectrum disorder in under 19s: support and management",
      "url": "https://www.nice.org.uk/guidance/cg170/resources/autism-spectrum-disorder-in-under-19s-support-and-management-pdf-35109745515205"
    },
    {
      "title": "Autism and Autism Spectrum Disorder",
      "url": "https://www.asha.org/Practice-Portal/Clinical-Topics/Autism/"
    },
    {
      "title": "Early intensive behavioral intervention (EIBI) for increasing functional behaviors and skills in young children with autism spectrum disorders (ASD)",
      "url": "https://www.cochrane.org/evidence/CD009260_early-intensive-behavioral-intervention-eibi-increasing-functional-behaviors-and-skills-young"
    },
    {
      "title": "Randomized, controlled trial of an intervention for toddlers with autism: the Early Start Denver Model",
      "url": "https://pubmed.ncbi.nlm.nih.gov/19948568/"
    },
    {
      "title": "Augmentative and Alternative Communication (AAC)",
      "url": "https://www.asha.org/practice-portal/professional-issues/augmentative-and-alternative-communication/"
    },
    {
      "title": "What is occupational therapy?",
      "url": "https://www.aota.org/about/what-is-ot"
    },
    {
      "title": "Autistic experiences of applied behavior analysis",
      "url": "https://pubmed.ncbi.nlm.nih.gov/35999706/"
    },
    {
      "title": "Recommendations: Autism spectrum disorder in under 19s: support and management",
      "url": "https://www.nice.org.uk/guidance/cg170/chapter/Recommendations"
    },
    {
      "title": "Recommendations: Autism spectrum disorder in adults: diagnosis and management",
      "url": "https://www.nice.org.uk/guidance/cg142/chapter/recommendations"
    },
    {
      "title": "RISPERDAL prescribing information",
      "url": "https://www.accessdata.fda.gov/drugsatfda_docs/label/2025/020272s088%2C020588s074%2C021444s060lbl.pdf"
    },
    {
      "title": "Aripiprazole prescribing information",
      "url": "https://www.accessdata.fda.gov/drugsatfda_docs/label/2025/216655s001lbl.pdf"
    },
    {
      "title": "Randomized, controlled, crossover trial of methylphenidate in pervasive developmental disorders with hyperactivity",
      "url": "https://pubmed.ncbi.nlm.nih.gov/16275814/"
    },
    {
      "title": "Principles of treatment, safety, monitoring and withdrawal: Epilepsies in children, young people and adults",
      "url": "https://www.nice.org.uk/guidance/ng217/chapter/principles-of-treatment-safety-monitoring-and-withdrawal"
    },
    {
      "title": "Living with autism without knowing: receiving a diagnosis in later life",
      "url": "https://pubmed.ncbi.nlm.nih.gov/34040855/"
    },
    {
      "title": "Adult outcome for children with autism",
      "url": "https://pubmed.ncbi.nlm.nih.gov/14982237/"
    },
    {
      "title": "Life Course Health Development in Autism Spectrum Disorders",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK543697/"
    },
    {
      "title": "Post-high school service use among young adults with an autism spectrum disorder",
      "url": "https://pubmed.ncbi.nlm.nih.gov/21300654/"
    },
    {
      "title": "The health status of adults on the autism spectrum",
      "url": "https://pubmed.ncbi.nlm.nih.gov/25911091/"
    },
    {
      "title": "“I’m dealing with a health care system that doesn’t get it”: Barriers and Facilitators to Inclusive Healthcare for Autistic Adults",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11132937/"
    },
    {
      "title": "Healthcare access for autistic adults: A systematic review",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7373620/"
    },
    {
      "title": "Premature mortality in autism spectrum disorder",
      "url": "https://www.cambridge.org/core/services/aop-cambridge-core/content/view/4C9260DB64DFC29AF945D32D1C15E8F2/S0007125000279385a.pdf/div-class-title-premature-mortality-in-autism-spectrum-disorder-div.pdf"
    },
    {
      "title": "Estimating life expectancy and years of life lost for autistic people in the UK: a matched cohort study",
      "url": "https://pubmed.ncbi.nlm.nih.gov/38188276/"
    },
    {
      "title": "Estimating life expectancy and years of life lost for autistic people in the UK: a matched cohort study",
      "url": "https://discovery.ucl.ac.uk/id/eprint/10182794/1/O%27Nions2023_Lancet_Europe_Autism%20lifeExp.pdf"
    },
    {
      "title": "Children with special educational needs and disabilities (SEND): Extra help",
      "url": "https://www.gov.uk/children-with-special-educational-needs/extra-SEN-help"
    },
    {
      "title": "Children with special educational needs and disabilities (SEND): Special educational needs support",
      "url": "https://www.gov.uk/children-with-special-educational-needs/special-educational-needs-support"
    },
    {
      "title": "Autism Spectrum",
      "url": "https://askjan.org/disabilities/Autism-Spectrum.cfm"
    },
    {
      "title": "Small Employers and Reasonable Accommodation",
      "url": "https://www.eeoc.gov/laws/guidance/small-employers-and-reasonable-accommodation"
    },
    {
      "title": "The Buckland Review of Autism Employment: report and recommendations",
      "url": "https://www.gov.uk/government/publications/the-buckland-review-of-autism-employment-report-and-recommendations/the-buckland-review-of-autism-employment-report-and-recommendations"
    },
    {
      "title": "Competitive Employment for Transition-Aged Youth with Significant Impact from Autism: A Multi-site Randomized Clinical Trial",
      "url": "https://pubmed.ncbi.nlm.nih.gov/30825082/"
    },
    {
      "title": "Competitive employment for transition-aged youth with significant impact from autism: A multi-site randomized clinical trial (Wehman et al., 2020)",
      "url": "https://beta.dol.gov/research-data/clear/studies/competitive-employment-transition-aged-youth-significant-impact-autism-multi-site-randomized"
    },
    {
      "title": "Disability Discrimination and Employment Decisions",
      "url": "https://www.eeoc.gov/disability-discrimination-and-employment-decisions"
    },
    {
      "title": "UNTC: Convention on the Rights of Persons with Disabilities",
      "url": "https://treaties.un.org/pages/ViewDetails.aspx?chapter=4&mtdsg_no=IV-15&src=TREATY"
    },
    {
      "title": "Who We Are - Autistic Self Advocacy Network",
      "url": "https://autisticadvocacy.org/about-asan/who-we-are/"
    },
    {
      "title": "What We Believe - Autistic Self Advocacy Network",
      "url": "https://autisticadvocacy.org/about-asan/what-we-believe/"
    },
    {
      "title": "Self-Advocacy - Autistic Self Advocacy Network",
      "url": "https://autisticadvocacy.org/self-advocacy/"
    },
    {
      "title": "What should autism research focus upon? Community views and priorities from the United Kingdom",
      "url": "https://pubmed.ncbi.nlm.nih.gov/24789871/"
    },
    {
      "title": "'People should be allowed to do what they like': Autistic adults' views and experiences of stimming",
      "url": "https://pubmed.ncbi.nlm.nih.gov/30818970/"
    },
    {
      "title": "The Ethics (or Lack Thereof) of Autism Interventions",
      "url": "https://autisticadvocacy.org/wp-content/uploads/2021/12/ACWP-Ethics-of-Intervention.pdf"
    },
    {
      "title": "Extending the Minority Stress Model to Understand Mental Health Problems Experienced by the Autistic Population",
      "url": "https://discovery.ucl.ac.uk/10060929/1/Frost_Botha%20and%20Frost%202018.pdf"
    },
    {
      "title": "Barriers to healthcare: Instrument development and comparison between autistic adults and adults with and without other disabilities",
      "url": "https://pubmed.ncbi.nlm.nih.gov/27663266/"
    },
    {
      "title": "The Development and Evaluation of an Online Healthcare Toolkit for Autistic Adults and their Primary Care Providers",
      "url": "https://pubmed.ncbi.nlm.nih.gov/27271730/"
    },
    {
      "title": "Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children",
      "url": "https://pubmed.ncbi.nlm.nih.gov/9500320/?dopt=Abstract"
    },
    {
      "title": "Retraction—Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children",
      "url": "https://briandeer.com/solved/lancet-retraction-2010.pdf"
    },
    {
      "title": "Measles, Mumps, Rubella Vaccination and Autism: A Nationwide Cohort Study",
      "url": "https://pubmed.ncbi.nlm.nih.gov/30831578/"
    },
    {
      "title": "Statement of the WHO Global Advisory Committee on Vaccine Safety (GACVS) on vaccines and autism",
      "url": "https://www.who.int/news/item/11-12-2025-statement-gacvs-vaccines-autism"
    },
    {
      "title": "What is autism? | Leeds Autism Services",
      "url": "https://www.leedsautism.org.uk/pages/faqs/category/autism-explained"
    },
    {
      "title": "NHS 111 Wales - Health A-Z: Autism",
      "url": "https://111.wales.nhs.uk/Encyclopaedia/a/article/autism"
    },
    {
      "title": "Science Update: Risk for autism spectrum disorder mainly influenced by genetics, suggests NICHD-funded study",
      "url": "https://www.nichd.nih.gov/newsroom/news/080119-autism"
    },
    {
      "title": "CG142 Autism in adults full guideline",
      "url": "https://www.nice.org.uk/guidance/cg142/resources/full-guideline-pdf-186587677"
    },
    {
      "title": "Recommendations: Autism spectrum disorder in under 19s: support and management",
      "url": "https://www.nice.org.uk/guidance/cg170/chapter/recommendations"
    },
    {
      "title": "Questions and Answers on Unapproved Chelation Products",
      "url": "https://www.fda.gov/drugs/medication-health-fraud/questions-and-answers-unapproved-chelation-products"
    },
    {
      "title": "Important Patient and Consumer Information About Regenerative Medicine Therapies",
      "url": "https://www.fda.gov/vaccines-blood-biologics/consumers-biologics/important-patient-and-consumer-information-about-regenerative-medicine-therapies"
    },
    {
      "title": "Explaining the increase in the prevalence of autism spectrum disorders: the proportion attributable to changes in reporting practices",
      "url": "https://pubmed.ncbi.nlm.nih.gov/25365033/"
    },
    {
      "title": "Testing and Refining Biomarkers to Support Intervention Research for Children with Autism",
      "url": "https://www.nimh.nih.gov/news/science-updates/2020/testing-and-refining-biomarkers-to-support-intervention-research-for-children-with-autism"
    },
    {
      "title": "Autism Biomarkers for Clinical Trials",
      "url": "https://www.nimh.nih.gov/funding/grant-writing-and-application-process/concept-clearances/2019/autism-biomarkers-for-clinical-trials"
    },
    {
      "title": "Community testing practices for autism within the autism and developmental disabilities monitoring network",
      "url": "https://stacks.cdc.gov/view/cdc/159346/cdc_159346_DS1.pdf"
    },
    {
      "title": "EarliPoint System 510(k) summary, K230337",
      "url": "https://www.accessdata.fda.gov/cdrh_docs/pdf23/K230337.pdf"
    },
    {
      "title": "Decomposition of phenotypic heterogeneity in autism reveals underlying genetic programs",
      "url": "https://www.nature.com/articles/s41588-025-02224-z.pdf"
    },
    {
      "title": "Autism",
      "url": "https://www.niehs.nih.gov/health/topics/conditions/autism"
    },
    {
      "title": "Early detection of autism using digital behavioral phenotyping",
      "url": "https://pubmed.ncbi.nlm.nih.gov/37783967/"
    },
    {
      "title": "The AASPIRE practice-based guidelines for the inclusion of autistic adults in research as co-researchers and study participants",
      "url": "https://pubmed.ncbi.nlm.nih.gov/30939892/"
    },
    {
      "title": "Consensus on high-priority outcomes to be used in the evaluation of services for autistic adults: Results from a \"CBPR-Nested Delphi Process\"",
      "url": "https://pubmed.ncbi.nlm.nih.gov/40468818/"
    }
  ],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}