HealthScorer

M-CHAT-R — Toddler Autism Screener (16-30 months)

Free M-CHAT-R toddler autism screener — Robins 2014. 20 yes/no questions, 2 minutes, AAP-recommended at the 18 and 24-month visits.

Last updated: Sources verified:

What you are about to take

The M-CHAT-R is the Modified Checklist for Autism in Toddlers, Revised — the most widely used early autism screener in the world. Twenty short yes/no questions about your child’s usual behaviour. Designed by Diana Robins, Deborah Fein and Marianne Barton, and validated in the version used today by Robins et al. in Pediatrics (2014, PMID 24366990). About two minutes. It is calibrated for children aged 16 to 30 months — outside that window, the result is not reliable. Your answers stay in your browser — we never see them. Start the test below.

  • Validated by Robins, Casagrande, Barton, Chen, Dumont-Mathieu and Fein (2014, Pediatrics)
  • Recommended by the American Academy of Pediatrics at 18 and 24-month visits (AAP 2020)
  • Used in US Preventive Services Task Force and CDC guidance
  • 2 minutes, 20 yes/no questions
  • Private — answers never leave your device

How the M-CHAT-R is scored

You answer 20 yes/no questions about your child’s typical behaviour. For most items, “no” indicates a concern — for example, if your child does not respond to their name (item 10) or does not point to share interest (item 7). Three items are scored in reverse: items 2, 5 and 12 are concerning if you answer “yes” (parental worry that the child may be deaf; unusual finger movements near eyes; distress at everyday noises). The number of items in the concerning direction sums to a total between 0 and 20.

Total scoreBandWhat it typically means
0-2Low riskTypical screening result — no further action recommended unless concern is high
3-7Medium riskAdminister the M-CHAT-R/F follow-up interview to clarify each failed item
8-20High riskBypass the follow-up — refer directly for comprehensive evaluation

The high-risk cutoff at 8 is specifically chosen so that referring directly, without the follow-up interview, does not produce excessive over-referral. The medium-risk band is where the follow-up interview adds most clinical value — it cuts roughly half of the apparent positives without missing real cases.

Why the M-CHAT-R/F follow-up matters in the medium-risk band

Many parents interpret M-CHAT-R items differently from how the developers intended. A question that read as “no” on the form often becomes “yes” when a clinician asks for an example. The follow-up interview (M-CHAT-R/F) is a structured 10-15 minute conversation, administered by your paediatrician, that walks through each failed item. It does not require special training — any paediatrician can administer it, and the form is freely available at mchatscreen.com.

The follow-up roughly halves the false-positive rate. After the interview, the indication is to refer for evaluation when 2 or more failed items remain. This is the validated next step in the 3-7 band — not a “see how it goes” approach.

What this test is — and what it isn’t

This is a screening tool, not a diagnostic instrument. A positive screen always requires a comprehensive developmental evaluation by a qualified clinician before any conclusion is drawn. The diagnostic gold standard for toddlers combines a structured parent interview (ADI-R), a structured child observation (ADOS-2 Toddler Module or Module 1), cognitive and language assessment, and a medical examination including hearing.

The age window matters. The M-CHAT-R was developed and validated specifically for children aged 16-30 months. For older children, your paediatrician would use a different screener — the Social Communication Questionnaire (SCQ) for 4+ years, or instruments like the SRS-2 across childhood. For younger children, the M-CHAT-R is not yet calibrated; surveillance of milestones is the standard approach.

Re-screening at the 24-month visit is recommended even after a negative screen at 18 months. Some signs of autism become more apparent between 24 and 36 months, and a small proportion of children who screen negative at 18 months go on to be diagnosed later. The AAP 2020 surveillance guidance is explicit on this.

Early signs across the second year of life

The M-CHAT-R is built around the developmental skills typically present in neurotypical toddlers between 16 and 30 months:

  • Joint attention — looking where you point, pointing to share interest, bringing objects to show you. Usually visible between 12-15 months. Items 1, 7, 9, 16, 19.
  • Response to name — looking up, vocalising, or stopping an activity when called. Item 10.
  • Social referencing — checking your face for emotional information when something new or unfamiliar happens. Item 19.
  • Imitation — copying simple gestures like waving, clapping, or pretending to drink from a cup. Items 3, 15.
  • Pretend play — feeding a doll, pretending to talk on a phone, using a block as a car. Item 3.
  • Eye contact in interaction — looking at you during talking, dressing, play. Item 14.

A pattern of failed items across these skill clusters carries more weight than any single item. The validation data shows that failures in joint attention and response to name are particularly specific to autism — more so than failures in motor or sensory items.

What to do at each band

  • 0-2 (low risk): Continue routine well-child visits. Re-screen at the 24-month visit if the current screen was done earlier. If you have a specific persistent concern that the screener does not capture — speech delay, regression of skills, social withdrawal — share it with your paediatrician. Parental concern is itself a clinical signal.
  • 3-7 (medium risk): Book a paediatric visit within 2-4 weeks for the M-CHAT-R/F follow-up interview. Bring the printed result. In the meantime, prioritise joint-attention play — peekaboo, naming what your child is looking at, narrating routines, picture books with feelings on faces. These are gentle, evidence-based activities regardless of where the evaluation lands.
  • 8-20 (high risk): Book a paediatric visit within 1-2 weeks. Ask for referral to a developmental paediatrician, paediatric neurologist, or multidisciplinary autism evaluation service. Contact local early-intervention services in parallel — most jurisdictions do not require a formal autism diagnosis to begin speech-language and occupational therapy.

Treatments that have evidence — and ones that do not

The two universally helpful modalities in the toddler age window are speech-language therapy and occupational therapy. They start before a formal diagnosis is in hand, in most jurisdictions. Naturalistic developmental behavioural interventions (NDBIs) — including the Early Start Denver Model and JASPER — have the strongest trial evidence for autistic toddlers specifically.

Approaches without evidence, sometimes with documented harm: chelation, hyperbaric oxygen, restrictive diets without a medically identified intolerance, megadose vitamins, stem-cell therapies marketed online. Programmes that promise to make a child “indistinguishable from peers” should be approached with the same scepticism — the evidence is for skill-building and adaptation, not transformation. Neurodiversity-affirming early intervention focuses on supporting the child’s development on their own terms rather than masking their autism.

When concern outweighs the screener

The M-CHAT-R is one input. Parental concern is itself a meaningful clinical signal — the autism literature shows that early parental worry, even when the screen is borderline, has independent predictive value. If your gut says something is off and the screener returned low-risk, raise it at the next visit anyway. Specifics help: not “I’m worried about Maya” but “Maya doesn’t look at me when I call her, doesn’t point at things, and used to wave bye-bye but stopped two months ago.”

If you have a family history — older sibling with autism, autistic parent or close relatives — surveillance is more important even with a negative screen. Sibling-recurrence rates in autism are roughly 18-20% in cohort studies.

Sources, verified 2026-05-18

  • Robins DL, Casagrande K, Barton M, Chen CA, Dumont-Mathieu T, Fein D. Validation of the Modified Checklist for Autism in Toddlers, Revised With Follow-Up (M-CHAT-R/F). Pediatrics 2014;133(1):37-45. (PMID 24366990)
  • American Academy of Pediatrics. Identification, Evaluation, and Management of Children With Autism Spectrum Disorder. Pediatrics 2020;145(1):e20193447.
  • US Preventive Services Task Force. Screening for Autism Spectrum Disorder in Young Children. JAMA 2016;315(7):691-696.
  • Centers for Disease Control and Prevention. Autism Spectrum Disorder. cdc.gov/ncbddd/autism.
  • Robins DL, Fein D, Barton ML, Green JA. The Modified Checklist for Autism in Toddlers — an initial study investigating the early detection of autism and pervasive developmental disorders. J Autism Dev Disord 2001;31(2):131-144. (PMID 11450812)

Privacy

The M-CHAT-R calculator runs entirely in your browser. Your individual answers and the calculated band never leave your device. We send one anonymous event to a privacy-respecting analytics service: your locale code and the band string (for example mchatr_low, mchatr_medium, mchatr_high). No raw answers, no per-item data, no identifier of any kind.

Frequently asked questions

What does an M-CHAT-R score of 5 mean?
A total of 5 sits in the medium-risk band (3-7). Robins 2014 (Pediatrics, PMID 24366990) calls this the interview band — the next validated step is the M-CHAT-R/F follow-up interview with your paediatrician, a structured 10-15 minute conversation that walks through each failed item. About half of children in this band turn out, after the interview, not to need further evaluation; the other half do. A score of 5 is a signal to book a paediatric visit within 2-4 weeks, not a diagnosis.
When should I start to worry?
The American Academy of Pediatrics recommends autism-specific screening at the 18- and 24-month well-child visits regardless of parental concern. If you have specific concerns earlier — no babbling by 12 months, no pointing or showing by 14-16 months, no two-word phrases by 24 months, or any loss of previously acquired skills at any age — raise them with your paediatrician without waiting for the routine screen. Parental concern is itself a meaningful clinical signal in the autism literature.
My 18-month-old is not talking — is it autism?
Not necessarily. Language delay is common at 18 months and has many causes — hearing problems, late talker pattern (the child catches up), specific language impairment, or developmental delay including autism. The M-CHAT-R looks at the broader pattern: social engagement, joint attention, response to name, pretend play. A speech-only delay with strong joint attention and social engagement is less likely to be autism than a speech delay paired with low joint attention. Either way, an 18-month visit is the moment to raise it with your paediatrician — and a hearing test is worth requesting.
M-CHAT-R score 8 or higher — what next?
Robins 2014 recommends bypassing the follow-up interview at scores of 8 or above and going directly to a comprehensive developmental evaluation. Book a paediatric visit within 1-2 weeks. Ask specifically for a referral to a developmental paediatrician, paediatric neurologist, or a multidisciplinary autism evaluation service. In parallel, contact your local early-intervention services — in the US, the Part C programme accepts referrals without a formal autism diagnosis and can begin speech-language and occupational therapy while you wait for the full evaluation.
How often does M-CHAT-R give a false positive?
Roughly half of the children flagged at the initial screen are not diagnosed with autism on full evaluation, especially in the medium-risk band (3-7). That is why Robins 2014 added the M-CHAT-R/F follow-up interview — it cuts the false-positive rate roughly in half without missing real cases. In the high-risk band (8-20), false positives are much less common; the positive predictive value rises substantially. A positive screen is a signal to investigate, never a diagnosis.
Can autism be cured?
No, and the question itself is misleading. Autism is a lifelong neurodevelopmental difference, not an illness to be cured. The goal of early intervention is not to make a child non-autistic but to support communication, social engagement, and adaptive skills so the child can thrive as themselves. Outcomes are highly variable and the strongest predictor of long-term functioning is timely, individualised support — not eliminating autism, but building the supports that help an autistic person live a full life.
What does early intervention actually involve?
Structured play and skill-building, delivered by trained therapists, with the family as a core partner. The two most universally helpful modalities in this age window are speech-language therapy (supports communication, even before words) and occupational therapy (supports sensory processing, motor skills, daily activities). Naturalistic developmental behavioural interventions (NDBIs) — like the Early Start Denver Model — have the strongest trial evidence for toddlers. Programmes that promise to make a child indistinguishable from peers should be approached with caution; the evidence is for skill-building, not transformation.
What does developmental regression mean?
Regression is the loss of previously acquired skills — words, eye contact, social interest, or play skills that the child had and then stopped using. Roughly 20-30% of autistic children show some form of regression, typically between 15 and 24 months. Regression is a separate red flag worth raising with your paediatrician even outside the regular screening flow, because it can also signal other neurological conditions and warrants evaluation. Document specifically what was lost and when.
Are girls under-diagnosed with autism?
Yes, and the literature is consistent on this. Girls are diagnosed later and at lower rates than boys, partly because diagnostic instruments were validated primarily on boys and partly because autistic girls often present differently — more socially motivated, better at mimicking peers, with special interests that look more typical (animals, fictional characters). The M-CHAT-R works in girls but may flag fewer of them at the same level of underlying difference. If your daughter has parental concern that the screener does not capture, raise it anyway.
Is autism genetic or environmental?
Predominantly genetic. Twin studies estimate 70-90% of the variance in autism risk is genetic. The remaining variance includes events that affect brain development in utero — advanced parental age, certain prenatal infections, extreme prematurity. Vaccines do not cause autism; this has been studied in cohorts of millions of children and the original 1998 Wakefield paper proposing the link was retracted by The Lancet and Wakefield lost his medical licence. Parenting style, screen time, and diet do not cause autism either.
Do vaccines cause autism?
No. The 1998 paper by Andrew Wakefield that started this concern was retracted by The Lancet in 2010 for research misconduct and Wakefield was struck off the UK medical register. Since then, cohort studies in millions of children — Danish cohort of 657,461 children (Hviid 2019, Annals of Internal Medicine), Madsen 2002, Jain 2015, Taylor 2014 meta-analysis — have all found no association between MMR or any other vaccine and autism. The scientific question is settled. Skipping vaccines exposes your child to measles, whooping cough and other preventable diseases without reducing autism risk by any measurable amount.
What happens after a positive screen?
The validated pathway is: positive M-CHAT-R screen → M-CHAT-R/F follow-up interview with your paediatrician (medium-risk band) or direct referral (high-risk band) → comprehensive developmental evaluation by a developmental paediatrician, paediatric neurologist, or multidisciplinary autism team. The full evaluation typically involves parent interview (ADI-R), structured child observation (ADOS-2), cognitive and language assessment, and a medical examination. The process can take weeks to months — start early-intervention services in parallel while you wait.
How do I talk to family about this?
Lead with what you have observed and what step you are taking, not with a label. 'I have been concerned about how Maya responds to her name, so I am going to talk to her paediatrician' lands better than 'I think Maya has autism' — and is also more accurate at the screening stage. Many families benefit from a brief written summary they can share with grandparents and other caregivers explaining what the screen found, what the next step is, and what role family members can play (talking, reading, joint play, not pressuring eye contact). Parent-led organisations have templates.
Can the test be wrong?
Yes, in both directions. False positives are common in the medium-risk band — many normally developing toddlers who are temperamentally shy, internationally adopted, dual-language, or simply having a hard week score in this range. The M-CHAT-R/F follow-up interview is specifically designed to catch these. False negatives also happen, especially in girls and in children whose autistic features are subtle. A low score with persistent parental concern is still worth raising at the next well-child visit.
Is the data private?
Yes. The M-CHAT-R calculator runs entirely in your browser. Your individual answers and the calculated band never leave your device. We send one anonymous event to a privacy-respecting analytics service: your locale code and the band string (for example `mchatr_medium`). No raw answers, no per-item data, no identifier of any kind.

Sources

  1. Validation of the Modified Checklist for Autism in Toddlers, Revised With Follow-Up (M-CHAT-R/F) — Robins DL, Casagrande K, Barton M, Chen CA, Dumont-Mathieu T, Fein D — Pediatrics (2014) (peer reviewed, retrieved 2026-05-18)
  2. Identification, Evaluation, and Management of Children With Autism Spectrum Disorder — American Academy of Pediatrics — Pediatrics (2020) (guideline, retrieved 2026-05-18)
  3. Screening for Autism Spectrum Disorder in Young Children — US Preventive Services Task Force Recommendation Statement — US Preventive Services Task Force — JAMA (2016) (guideline, retrieved 2026-05-18)
  4. Autism Spectrum Disorder — basics, signs and screening — Centers for Disease Control and Prevention (CDC) (gov health, retrieved 2026-05-18)
  5. The Modified Checklist for Autism in Toddlers — original M-CHAT (2001) — Robins DL, Fein D, Barton ML, Green JA — J Autism Dev Disord (2001) (peer reviewed, retrieved 2026-05-18)