Toddler · Development

M-CHAT-R Autism Screening Test

The AAP-recommended autism screen for 16-30 month-olds. 20 yes/no questions, 5 minutes. Tells you whether your toddler needs a developmental conversation — and exactly what to do next if so.

Last reviewed 28 May 2026

M-CHAT-R — autism screen, 16-30 months

Modified Checklist for Autism in Toddlers (Revised)

Answer based on your child’s usual behaviour. If your child does the behaviour rarely (you’ve only seen it once or twice), answer “no”. Recommended ages: 16-30 months. For ages 18 and 24 months, this is the AAP-recommended autism screen.
1. If you point at something across the room, does your child look at it?
2. Have you ever wondered if your child might be deaf?
3. Does your child play pretend or make-believe? (E.g. pretend to drink from an empty cup)
4. Does your child like climbing on things? (E.g. furniture, playground equipment, stairs)
5. Does your child make unusual finger movements near his or her eyes?
6. Does your child point with one finger to ask for something or to get help?
7. Does your child point with one finger to show you something interesting?
8. Is your child interested in other children? (E.g. watches them, smiles, or goes to them)
9. Does your child show you things by bringing them or holding them up for you to see — not to get help, just to share?
10. Does your child respond when you call their name?
11. When you smile at your child, does he or she smile back at you?
12. Does your child get upset by everyday noises? (E.g. does your child scream or cry over noise such as a vacuum or loud music)
13. Does your child walk?
14. Does your child look you in the eye when you are talking to him or her, playing, or dressing them?
15. Does your child try to copy what you do? (E.g. wave bye-bye, clap, or make a funny noise when you do)
16. If you turn your head to look at something, does your child look around to see what you are looking at?
17. Does your child try to get you to watch them? (E.g. saying "Look" or "Watch me")
18. Does your child understand when you tell them to do something? (E.g. if you don't point, can your child understand "put the book on the chair")
19. If something new happens, does your child look at your face to see how you feel about it?
20. Does your child like movement activities? (E.g. being swung, bounced on your knee)
Answer all 20 questions to see your total.

How do I screen my toddler for autism at home?

Take the 20-question M-CHAT-R above. Answer based on what your child usually does — not what they did once or twice. The screen takes 5 minutes; the result gives you a risk band and clear next steps. It’s the same screen the AAP recommends at the 18- and 24-month well-child visits in the US, and what NHS / RCPCH developmental services use too.

What are the earliest signs of autism in toddlers?

The AAP Bright Futures “Act Early” red flags — any ONE of these is a reason to talk to your paediatrician / health visitor:

  • By 12 months: no babbling, no pointing, no waving, no other gestures.
  • By 16 months: no single words.
  • By 24 months: no spontaneous (not just repeating) two-word phrases.
  • At any age: any LOSS of language or social skills.

Other features that should prompt a conversation:

  • Limited or fleeting eye contact.
  • Doesn’t consistently respond to their name by 12 months.
  • Prefers being alone, even when invited to play.
  • Intense focus on specific objects, parts of objects, or topics.
  • Repetitive movements — hand-flapping, rocking, spinning, lining up toys.
  • Big distress at small changes in routine.
  • Strong sensory reactions to sounds, textures, lights, or food types.
  • Limited or no pretend play (e.g. feeding a doll, talking on a toy phone).

Why screen at 18 and 24 months specifically?

The behavioural signs of autism reliably differentiate from typical development between 12 and 24 months. Before 12 months, autistic and typical infants look similar. The window of greatest brain plasticity is the first 3 years; identification by 24 months means intervention starts in that window. The median US age at autism diagnosis is still 4-5 years — years after parents’ first concerns. Universal screening narrows that gap significantly. The AAP recommends screening at BOTH 18 and 24 months because some children miss the 18-month screen and only differentiate by 24 months.

How is the M-CHAT-R scored?

Each item is 0 (typical answer) or 1 (atypical answer). Items 2, 5, and 12 are reverse-scored (“Yes” counts as 1 for those). Total 0-20. The widget above does the scoring for you.

  • Score 0-2 — Low risk. No further action. Re-screen at the next well-child visit.
  • Score 3-7 — Medium risk. Your paediatrician administers the M-CHAT-R/F Follow-Up — a structured interview asking for more detail on each failed item to clarify the picture.
  • Score 8-20 — High risk. Direct referral for diagnostic assessment. The Follow-Up step is skipped at this level.

What does a positive screen actually mean?

A positive M-CHAT-R is a signal, not a diagnosis. About 80% of children scoring as HIGH risk eventually receive an autism diagnosis on full assessment, and around 50% of MEDIUM-risk children who also fail the Follow-Up interview. Children who don’t end up with autism usually have other developmental needs — language delay, social communication disorder, global developmental delay, or sometimes undiagnosed hearing loss — all of which benefit from the same early-support pathway. Either way, a positive screen leads somewhere useful.

What to do next if your child screens positive

  1. Talk to your paediatrician / GP / health visitor. Bring the M-CHAT-R result. They may administer the Follow-Up interview, examine your child, and make referrals.
  2. Refer for diagnostic evaluation. Developmental paediatrician, paediatric neurologist, or developmental psychologist. Gold-standard tools: ADOS-2 (Autism Diagnostic Observation Schedule) and ADI-R (Autism Diagnostic Interview-Revised).
  3. Refer to Early Intervention in parallel. US: Part C of IDEA — free, every state, doesn’t require diagnosis. UK: health visitor / community paediatrics / Portage. Don’t wait for diagnosis to start support.
  4. Get a hearing test. 1-2% of children referred for autism evaluation have undiagnosed hearing loss accounting for the social-communication concerns.
  5. Take care of yourself. A positive screen is a lot for parents. Look up parent peer support groups (Autistica, NAS in UK; Autism Speaks, ASAN in US). Talk to your partner; book a check-in for yourself.

Practical scenarios — what positive screens often look like

Scenario 1: 18 months, no words, doesn’t point, doesn’t respond to name

Very likely a high-risk screen. Three classic flags. Direct referral for assessment + EI / SaLT immediately. Hearing test first to rule out hearing loss.

Scenario 2: 24 months, has 30 words, doesn’t pretend play, lines up cars, big distress at small routine changes

Medium risk screen likely. Language is present but social use of language and play are different. Follow-Up interview will clarify; assessment referral likely.

Scenario 3: 20 months girl, lots of language, social with adults, loves animals/books, no other concerns

Likely low-risk screen. Reassuring — but remember girls are under-identified. If you still have concerns, raise them with the paediatrician.

Scenario 4: 18 months, formerly babbling and waving, then went quiet over the last 3 months

Regression — loss of language / social skills is a strong red flag regardless of M-CHAT-R score. Urgent paediatric referral and consider EEG to rule out Landau-Kleffner / acquired epileptic aphasia (rare).

Scenario 5: 24 months bilingual household, fewer words than typical English-only peers

Bilingual children often have slightly fewer words in either language at 24 months but the COMBINED vocabulary is typical. M-CHAT-R is still reliable. If social communication (pointing, joint attention, pretend play) is intact, language delay alone — consider SaLT referral; not necessarily autism workup.

How accurate is the M-CHAT-R?

Original validation (Robins 2014, 16,071 children): sensitivity 0.85, specificity 0.99, positive predictive value 47.5% overall (higher with the Follow-Up step). A 2023 meta-analysis (Yuen, JAMA Pediatrics) of diverse populations found pooled sensitivity 0.83 and specificity 0.94. PPV varies with the base rate of autism in the population — higher in clinic populations, lower in general-population screening. The bottom line: M-CHAT-R is the best-evidenced free screen for this age, but a screen never replaces a full diagnostic assessment.

Common myths about autism — debunked

  • “MMR vaccine causes autism” — No. The 1998 Wakefield paper was retracted; Wakefield was struck off. Subsequent studies of >1 million children show no link. The rise in autism prevalence is from broader criteria + better recognition, not vaccines.
  • “Bad parenting causes autism” — No. Autism has strong genetic basis (heritability ~80%). Parenting style doesn’t cause it (the “refrigerator mother” theory was completely discredited decades ago).
  • “Autistic kids don’t feel emotions / love their parents” — Completely false. Autistic children feel emotions deeply; expression may look different from neurotypical expectations.
  • “If they make eye contact, they can’t be autistic” — Many autistic children make eye contact, especially with familiar people. The pattern is more about how eye contact is used in social back-and-forth than presence/absence.
  • “They’ll grow out of it” — Autism is a lifelong neurodevelopmental difference. Children grow and develop, sometimes dramatically — but they don’t “become non-autistic”.
  • “Wait and see” — Outdated advice. AAP 2020 is explicit: refer for evaluation AND support in parallel when there’s concern.

Limitations of this screen

  • Screen, not a diagnosis — diagnosis requires full clinical evaluation.
  • Follow-Up interview is recommended for medium-risk results — your paediatrician administers it. The widget gives the raw score; the clinical conversation interprets it.
  • PPV varies with population — higher in clinic referral contexts, lower in general low-risk screening.
  • May under-identify autism in girls, who often camouflage social differences more.
  • Cultural variation in eye-contact norms, climbing access, and pretend-play exposure can affect specific items.
  • Doesn’t screen for ADHD, anxiety, intellectual disability, or speech-only delays — those have their own pathways.
  • Children < 16 months or > 30 months should use age-appropriate tools (ASQ-3, PEDS, or full developmental evaluation).

Sources

  • Robins DL, Casagrande K, Barton M, et al. Validation of the Modified Checklist for Autism in Toddlers, Revised With Follow-up (M-CHAT-R/F). Pediatrics 2014;133:37-45.
  • Hyman SL, Levy SE, Myers SM; AAP. Identification, Evaluation, and Management of Children With Autism Spectrum Disorder. Pediatrics 2020;145:e20193447.
  • AAP. Bright Futures Periodicity Schedule. 4th ed.
  • Yuen T, et al. Performance of the M-CHAT Revised With Follow-up in Population Screening: A Meta-analysis. JAMA Pediatr 2023.
  • CDC. Autism and Developmental Disabilities Monitoring (ADDM) Network: Prevalence and Characteristics — 2023.
  • Dawson G, et al. Randomized, controlled trial of the Early Start Denver Model. Pediatrics 2010;125:e17-e23.
  • Hviid A, et al. Measles, Mumps, Rubella Vaccination and Autism: A Nationwide Cohort Study. Ann Intern Med 2019;170:513-520.
  • NICE CG128. Autism spectrum disorder in under 19s: recognition, referral and diagnosis.

Frequently asked questions

What is the M-CHAT-R autism screening test?
The Modified Checklist for Autism in Toddlers, Revised — a free 20-question yes/no questionnaire that parents fill in for their 16-30 month old. Takes about 5 minutes. Developed by Diana Robins and Deborah Fein, formally validated in Pediatrics 2014. The American Academy of Pediatrics recommends it (or a similar validated tool) at both the 18-month AND 24-month well-child visits. It's a SCREEN — it flags children who might benefit from a fuller assessment, not a diagnosis.
What age is the M-CHAT-R for?
16-30 months (about 1 year 4 months to 2 years 6 months). Under 16 months: too early — typical and autistic toddlers haven't differentiated enough behaviourally yet. Over 30 months: too late — different screens are validated for older toddlers and preschoolers (SCQ, SRS-2, ASQ-SE). M-CHAT-R is taken twice: once at the 18-month check, again at 24 months. About 80% of children who go on to autism diagnosis are picked up at one of these two screens.
What are the earliest signs of autism in toddlers?
AAP 'Act Early' red flags: no babbling, pointing, or other gestures by 12 months; no single words by 16 months; no spontaneous (non-echolalia) 2-word phrases by 24 months; ANY loss of language or social skills at any age. Other features that should prompt a developmental conversation: limited eye contact, not consistently responding to name, preference for being alone, intense focus on specific objects/topics, repetitive movements (hand-flapping, rocking, spinning), big distress at small routine changes, strong sensory reactions (sound, texture, light). Any of these = talk to your paediatrician / health visitor, don't wait.
How are M-CHAT-R results scored?
Each item scores 0 (typical response) or 1 (atypical response). Items 2, 5 and 12 are reverse-scored — 'Yes' counts as 1 (atypical) for those three. Total: 0-20. RISK BANDS: 0-2 = LOW risk (no action; re-screen at the next well-child visit). 3-7 = MEDIUM risk (administer the M-CHAT-R/F Follow-Up structured interview to clarify; about 50% of medium-risk children who fail Follow-Up have autism). 8-20 = HIGH risk (immediate referral for diagnostic evaluation; the Follow-Up is skipped at this level; about 80% have autism on full assessment).
My child failed the M-CHAT-R — does this mean they have autism?
Not necessarily — it means they need a fuller assessment. The screen is sensitive (catches most autistic children) but not specific (some non-autistic children also fail). At HIGH risk (8+), about 80% of children go on to receive an autism diagnosis. At MEDIUM risk (3-7) that fails the Follow-Up, about 50%. Children who DON'T receive an autism diagnosis after failing the screen usually have other developmental needs — language delay, social communication disorder, global developmental delay, sometimes hearing loss — that ALSO benefit from early support. A positive screen leads somewhere useful either way.
What happens next if my child screens positive?
Three things in PARALLEL — not sequentially: (1) Referral for diagnostic assessment — developmental paediatrician, paediatric neurologist, or developmental psychologist. NHS pathway varies by area; private assessment is available. (2) Referral to Early Intervention — in the US, every state has free Part C of IDEA services for under-3s with developmental concerns; doesn't require a diagnosis. In the UK, your health visitor / GP refers to community paediatric services and Portage / SaLT. (3) Hearing test — 1-2% of children referred for autism actually have undiagnosed hearing loss. The AAP 2020 statement (Hyman) is explicit: don't wait for diagnosis to start support.
How long are the waiting lists for autism assessment?
Long. UK NHS: ranges from 6 months to 3+ years depending on area (Children's Commissioner 2024 report). US: median 1-2 years. Private UK assessment by an MDT can be 6-12 weeks but expensive (£2,500-£5,000+). Because of these waits, the AAP / NICE message is: don't wait for the diagnosis to start interventions. Referrals to Early Intervention / SaLT / Portage / health visitor support all happen on the basis of CONCERN, not on diagnosis.
Is the M-CHAT-R culturally fair across different backgrounds?
Reasonably. It's translated into 100+ languages and validated in many populations. Some items are imperfect across cultures — eye contact norms differ; climbing access depends on housing (item 4 makes less sense for a small flat with no playground); pretend play exposure varies. The pooled performance holds up across most studies (Yuen 2023 meta-analysis), but a positive screen always merits a culturally-informed clinical conversation rather than algorithm-only action. If you're unsure about an item, answer based on what you see most often, not your idealised expectation.
Should I wait and see instead of getting tested?
No. The 'wait and see' approach is outdated and harmful. AAP 2020 (Hyman, Pediatrics) is explicit: when there's concern, refer for evaluation AND Early Intervention in parallel — both can happen without a diagnosis. Children who turn out NOT to have autism still benefit from developmental support; children who DO have autism benefit measurably from earlier intervention. The 'wait and see' approach has caused enormous delays historically — the median US age at autism diagnosis is still 4-5 years, despite signs being noticeable from 12-24 months.
What does early intervention for autism actually involve?
Naturalistic developmental behavioural interventions (NDBIs) are the evidence-based first line. Most studied: Early Start Denver Model (ESDM), JASPER (Joint Attention, Symbolic Play, Engagement, Regulation), Pivotal Response Treatment. They focus on building social-communication skills through play and natural routines, often parent-coached so it can happen at home. Speech and language therapy (SaLT). Occupational therapy for sensory and motor needs. In the UK: Early Bird (NAS programme), Portage. Intensive ABA (Applied Behaviour Analysis) is more controversial — many autistic adults and clinicians have concerns about older intensive ABA approaches; modern naturalistic versions are gentler. Talk to the team about evidence and ethics.
Can autism be diagnosed before age 2?
Yes — by an experienced specialist team using ADOS-2 toddler module, ADI-R or DISCO interview, developmental history, and observation. Stability of diagnosis from 24 months is high (~90%). Diagnosis at 18 months is possible but less stable. The point of the screen at 18 months isn't necessarily to diagnose then — it's to flag children for the assessment pathway and start early intervention referrals.
How does autism in girls present differently?
Autistic girls and women are diagnosed later on average — often missed in childhood and identified in adolescence or adulthood. Possible reasons: girls camouflage / mask social differences more (mimicking peers consciously); special interests may be more socially-typical-appearing (animals, dolls, books) rather than the stereotypically 'autistic' interests; lower rates of disruptive behaviour leads to less referral; diagnostic criteria were originally developed from male samples. The M-CHAT-R was validated with mixed-sex samples but probably under-identifies autism in girls. If you have a girl with developmental concerns and a borderline screen, push for the assessment anyway.
What's the difference between autism and a language delay?
Significant overlap, and they often coexist. Pure language delay: language production / understanding is delayed, but social communication (joint attention, eye contact, pointing to share, pretend play, social smile) is intact. Autism: language delay is common but accompanied by qualitative differences in social communication AND restricted/repetitive behaviours and interests. The diagnostic assessment teases these apart with structured observation and history. If language is the only concern, an isolated speech-and-language therapy referral may be sufficient.
Does the MMR vaccine cause autism?
No. The 1998 paper by Andrew Wakefield that suggested this was retracted by The Lancet in 2010 for serious methodological flaws and research misconduct; Wakefield was struck off the UK medical register. Since then, large studies have looked at over a million children combined (Madsen 2002, Jain 2015, Hviid 2019 Danish cohort 657,000 children) — NO link between MMR and autism. Autism rates are similar in vaccinated and unvaccinated children. The increased awareness and broader diagnostic criteria explain the rise in autism prevalence, not vaccines. Withholding MMR exposes the child to measles, mumps, and rubella — all with real risks.
What is the prevalence of autism today?
About 1 in 36 8-year-olds in the US (CDC ADDM 2023 — the 2020 surveillance cohort). UK estimates are similar (~1-2%). Up from ~1 in 150 reported in 2000. The increase is mostly explained by: (1) broader diagnostic criteria (DSM-IV → DSM-5 added Asperger / PDD-NOS into ASD), (2) better recognition in females, minorities, and milder presentations, (3) better awareness and screening. Probably a small genuine increase too, but the dominant driver is recognition catching up with prevalence.
My child seems to develop fine — should I still take the screen?
Yes — universal screening at 18 and 24 months is the AAP recommendation. The point of universal screening is to catch the children whose parents haven't yet noticed differences, or who haven't realised that what they're seeing isn't typical. Even confident parents miss subtle social-communication differences. Filling in the screen takes 5 minutes and the LOW-risk result is itself reassuring.
Is autism a disorder or a difference?
Both perspectives are valid. The medical / DSM-5 framework calls it Autism Spectrum Disorder because diagnosis depends on the differences causing functional difficulty. The neurodiversity perspective frames autism as a different way of processing the world — strengths and challenges, not just deficits. Many autistic adults strongly prefer the latter view. Modern best practice: identify autism early to access support that suits the individual, but avoid framing the child as broken. The M-CHAT-R doesn't predict severity or outcome — it just flags for further evaluation.
How does this relate to other calculators on BumpBites?
Companion: /calculators/milestone-tracker for the CDC Learn-The-Signs Act-Early developmental milestone checklist; /calculators/asq-3 descriptive page for the broader Ages & Stages Questionnaire; /calculators/peds-tool descriptive page for the PEDS developmental screen; /calculators/baby-percentile for growth monitoring; /calculators/vaccine-scheduler for the routine immunisation schedule.

Continue with BumpBites

CDC Milestone Tracker

Learn-The-Signs Act-Early checklist.

Open

ASQ-3 Description

Ages & Stages broader screen.

Open

PEDS Tool Description

Parents' Evaluation of Developmental Status.

Open
Browse all pregnancy calculators