You filled in a questionnaire on a clipboard while your toddler emptied the toy box. Twenty questions, two minutes, mostly yes or no. Then the paediatrician came back in with a slightly different face and said the words "we'd like to refer you".
You are now in the car park googling M-CHAT autism screening on your phone with the engine off.
Here is what nobody said clearly in that room, because there were eleven minutes and you were only half in the conversation.
What does an M-CHAT autism screening result actually mean?
It means your child fell into a group worth looking at more closely. It does not mean your child is autistic, and it is not a diagnosis. The M-CHAT is a screening questionnaire, a deliberately blunt instrument designed to over-refer rather than under-refer, because the cost of missing a child is much higher than the cost of an unnecessary appointment.
Most children who screen positive turn out not to be autistic. That is not a flaw in the tool. That is what the tool is for.
What the M-CHAT is, in plain terms
The version most practices use is the M-CHAT-R, sometimes with a follow-up interview, in which case you will see it written as M-CHAT-R/F. Twenty yes-or-no questions about a child between roughly 16 and 30 months, which is why it turns up at the 18-month and 24-month check-ups.
The questions cover the early social-communication behaviours that tend to differ in autistic toddlers: whether your child points at things to show you, follows your point, responds to their name, brings things over to share, imitates you, and joins in pretend play. It does not test your child. It asks you.
That matters more than it sounds.
The instrument's raw material is a tired parent's recollection in a waiting room. The result is sensitive to how you were feeling that morning, whether you read a question the way it was meant, and whether you happened to think of the one time your child did point rather than the twenty times they did not.
The score bands and what each one triggers
If nobody told you the number, ring the practice and ask for it: "Can you tell me my child's total score on the M-CHAT-R, and whether the follow-up was completed?" You are entitled to it, it takes them thirty seconds, and without it the bands below are not much use to you. If the screen was done on paper and handed back, the total is the count of scored items, not the count of questions.
| Total score | Band | What should happen next |
|---|---|---|
| 0 to 2 | Low | No further action from this screen. Rescreen at the next check-up if your child is under 24 months. |
| 3 to 7 | Medium | The follow-up interview should be done, usually by the practice, to clarify the flagged items. If concerns remain after it, refer for evaluation. |
| 8 to 20 | High | Skip the follow-up. Refer straight for a diagnostic evaluation and for early intervention, at the same time. |
| Low, but you are still worried | Not a band, but common | Say so and ask for the referral anyway. Parental concern is its own indicator, and you can self-refer to Early Intervention or your school district without anyone's agreement. |
Two things go wrong in real practices. The first is that the medium band gets treated as a fail, and families are sent off frightened without the follow-up interview that exists precisely to sort this group out. The second is the opposite: the medium band gets treated as a pass, nothing is done, and nobody rescreens.
If your child scored in the medium band and you were not asked any follow-up questions, that is a reasonable thing to ring the practice about. The specific ask is: "She scored in the medium range. Was the M-CHAT follow-up interview completed? If not, can we do it?"
If they decline, or say they do not do the follow-up, you do not have to negotiate. Ask for a referral to a diagnostic evaluation instead, and call Early Intervention or your school district yourself the same day. Neither of them needs the practice's agreement, or the score.
Why the follow-up interview exists
Because the yes-or-no format loses everything interesting. The follow-up takes each flagged item and asks for an example, then asks whether that example is typical or a one-off.
Two examples, both real patterns.
A parent who ticked "no" to pointing may be thinking of last Tuesday. Under questioning, it turns out their child points at aeroplanes constantly and then checks their face. That item clears. Another parent ticked "yes" to pointing, meaning their child points at the biscuit cupboard, which is requesting rather than sharing. The questionnaire is not fine-grained enough to catch that distinction. The follow-up is.
The interview reclassifies a substantial share of medium-band children as low concern. It is the most useful part of the process and the part most often skipped for time.
Screening is not diagnosis, and here is the practical difference
A screen sorts a population into "look further" and "carry on". A diagnostic evaluation looks at one child in depth, usually over several hours, often with more than one professional, using structured observation, developmental history and direct interaction.
Nothing you do on a two-minute questionnaire, including doing it online at 1am, which most parents in this position have done, can substitute for that. Online versions are real and free and it is fine to look at one. But a score you generated yourself is a prompt to make an appointment, not information about your child, and the urge to answer with the single best example of each behaviour makes it useless.
The other practical difference: a screen does not open doors. A diagnostic evaluation, and in the US an Early Intervention or school district evaluation, does. Which is why the move after a positive screen is to start two processes, not one.
What to do this week, in order
1. Ask for the referral in writing, and ask what the wait is — you want it on record and you want to know whether you are looking at six weeks or fourteen months, because the answer changes everything you do next. If it is months rather than weeks, what to do while waiting for an autism assessment is the next thing to read.
2. If your child is under three, call Early Intervention yourself, today — in the US this is your state's Part C programme, and you do not need a diagnosis, a referral, or a positive screen to self-refer. The evaluation is free. This is the highest-value thing you can do this week, and most families do not know it exists.
3. If your child is over three, write to your school district instead — one paragraph by email, asking for an evaluation for special education eligibility, including speech and language. The written date starts their clock, so email rather than ring.
4. Book a hearing test — always, and separately. Hearing loss and glue ear mimic several of the behaviours the M-CHAT flags, are common at this age, and are quick to rule out. Do not let this get folded into the autism referral and lost.
5. Ask for a speech and language evaluation in parallel — do not wait for the autism question to resolve first. Communication support is useful regardless of the eventual answer, and ASHA's autism practice portal sets out what a speech-language pathologist assesses in this exact situation.
6. Start writing things down — dated notes, short videos, a running word list with the date each word first appeared. You will be asked to remember all of this months from now, and you will not. Our free developmental assessment is one way to organise those observations, though it is a communication tool and not an autism screener.
What if the screen came back low and you still feel something is off
Trust yourself, and act on it anyway.
The M-CHAT is a screen for a specific age band and a specific presentation. It is weakest with children who have strong language, with children whose social difficulties are subtle, and with children who compensate, which is a large part of why so many autistic girls are missed entirely at this stage. It also depends on how you answered on the day, and plenty of parents underreport because the questions felt like accusations.
A low score is not a certificate. If your gut is still tugging, go back and say this: "The screening was negative, and I remain concerned. I would like a referral for a full developmental evaluation." Parental concern is itself a recognised indicator, and a good clinician will act on it. If yours will not, you can self-refer without their agreement.
If your child failed the M-CHAT screening, read this part
Sitting with a positive screen is horrible, and the standard reassurance, that most positives are not autism, often lands as a brush-off rather than as comfort. So here is the version that actually helps.
A positive screen means your child showed enough early social-communication differences that a two-minute questionnaire noticed. Some of those children are autistic. Many have a language delay, a hearing problem, another kind of developmental difference, or nothing at all. In three of those four cases the referral you were just given still leads somewhere useful, because the evaluation looks at development broadly, not just at autism.
The screen did not tell you something bad about your child. It told you to go and find out, sooner rather than later. That is a better position than the one you were in yesterday, when the same things were true and nobody had noticed.
FAQ
Can my child fail the M-CHAT?
There is no pass or fail, though everyone including some clinicians uses that language. There is a score that sorts children into three groups with three different next steps. The score has no meaning about your child beyond that.
My child scored high at 18 months. Should we rescreen at 24 months instead of referring?
No. A high score is the one result that is meant to bypass further screening and go straight to evaluation and to early intervention. Rescreening in that situation costs six months.
Is the M-CHAT used outside the US?
It is used widely, though routine universal screening at fixed ages is a US practice and is not standard everywhere. Where it is not routine, you can still raise specific concerns with your GP or health visitor and request a developmental referral.
Does a positive screen affect insurance or school records?
A screening result sits in your child's medical record like any other. It is not a diagnosis and does not by itself create an educational label. If you have concerns about how records are used, ask the practice directly rather than avoiding the screen.