Is there a blood test for autism? What a lab can and cannot tell you

The short answer

No blood test can diagnose autism. Autism is diagnosed by a clinician who talks with you about your child’s development and watches how your child communicates, plays, and interacts — not by anything a lab can find in a vial of blood. The Centers for Disease Control and Prevention (CDC) says this plainly: there is no medical test, such as a blood test, to diagnose autism spectrum disorder (Centers for Disease Control and Prevention, 2025). The American Academy of Pediatrics (AAP) clinical report on identifying and managing autism in children describes the same clinical, observation-based process (Hyman et al., Pediatrics, 2020).

That does not mean blood is never drawn. After a diagnosis, a clinician may order genetic testing and, occasionally, other blood work — for reasons this page explains below. None of it is what made the diagnosis.

If your child has not yet been diagnosed, start with getting your child assessed and what autism testing actually involves; this page picks up from there, for the blood-test question specifically.

What blood is actually drawn for after a diagnosis

Once your child has a clinical diagnosis, a blood or saliva sample is often collected for one main reason: genetic testing. This looks for a change in your child’s DNA that might explain the autism, alongside other features. It is worth understanding clearly, because it is easy to mistake for the diagnosis itself.

Genetic testing does not diagnose autism — the diagnosis already happened. What it can do is find an underlying genetic explanation in a meaningful share of children. Combining chromosomal microarray and exome sequencing, one study of 95 children who received both tests found an identifiable genetic cause in about 1 in 6 of them — 15.8 percent (Tammimies et al., JAMA, 2015). That leaves most children without a single identified genetic cause on current tests, which is common and does not mean nothing is there — testing technology keeps improving, and some families revisit it years later.

The full range of what genetic testing can offer — chromosomal microarray, fragile X testing, and exome sequencing — plus how to prepare for the blood draw and what a result can mean for your family, is on genetic testing, explained. If your clinician specifically offers whole exome sequencing, here is what it finds and does not find.

A clinical genetics evaluation is built around your child’s specific history and exam findings, not a fixed panel run on everyone (Schaefer and Mendelsohn, Genetics in Medicine, 2013). That is why your clinician might, case by case, add something beyond the standard genetic test — a lead level if your child eats non-food items or has had exposure to old paint or pipes, for example, or another specific blood test if something in the history or exam points that way. None of this is unique to autism; it is the same individualized reasoning a genetics team uses for any child they evaluate.

The blood tests you may see advertised

Alongside genetic testing, you may come across labs or clinics advertising a blood test that claims to detect or screen for autism directly — most often built around folate receptor autoantibodies, sometimes shortened to FRAT or FRAA.

Here is what the actual research shows. One study measured folate receptor autoantibodies in the blood of 93 autistic children and found them in about three-quarters of the group — 75.3 percent. Sixteen of those children also had cerebrospinal fluid folate levels checked, and all came back below the typical range. Children with the antibodies who were then given oral leucovorin, a form of folate, showed more improvement over about four months — on measures of verbal communication, language, attention, and repetitive behavior — than a comparison group who did not receive it, though only about a third showed moderate to substantial improvement (Frye et al., Molecular Psychiatry, 2013).

That is a real finding worth taking seriously, and it is also not a diagnostic test. The study measured a difference across a research sample; it did not test whether the antibody result can reliably tell an autistic child from a non-autistic one, which is a different kind of study with a different kind of evidence. It has not been validated at the scale, or with the independent replication, that turning a research finding into a diagnostic test requires. Neither of the two leading US clinical guidelines for evaluating autism lists a blood biomarker test as part of a standard evaluation: not the American Academy of Pediatrics’ clinical report (Hyman et al., 2020), and not the American College of Medical Genetics’ evaluation guideline (Schaefer and Mendelsohn, 2013).

If a clinic offers this test or one like it, that does not automatically make it worthless — it may sit inside a legitimate research question, or be relevant if your child has separate signs of a folate-related condition. It does mean the marketing may be ahead of the evidence. Ask what the test claims to show, whether it is approved to diagnose autism specifically (none currently are), and what a positive or negative result would actually change about your child’s care.

What “a blood test for autism” headlines mean

You will keep seeing headlines like this, because biomarker research is an active, well-funded field — the search for a faster, more measurable way to identify autism than hours of a clinician’s time. A 2026 review of that research across several approaches, including brain-activity recordings and eye-tracking, concluded that despite real scientific progress, a biomarker that is actually usable in the clinic “remains elusive” (Griffin et al., Biological Psychiatry, 2026).

Here is the gap between a headline and a usable test, in plain terms. A study that finds autistic children have, on average, a different level of some blood marker than non-autistic children has found a group difference — useful for understanding biology, not yet useful for testing one individual child. To become a diagnostic test, a marker has to be checked against two numbers: sensitivity, how many autistic children it correctly flags, and specificity, how many non-autistic children it correctly clears. A marker can show a real group-level difference and still perform poorly on both, because individual children’s values overlap too much between groups to sort any one child reliably.

That is the step almost no proposed autism blood marker has cleared yet. So when a headline says scientists have found “a blood test for autism,” the honest translation is usually: researchers found a biological difference worth studying further — not a test your child’s clinician can order today to diagnose autism.

What to say to a relative who read about a blood test

A relative who read about this is trying to help, in the same way you are. A short, warm answer keeps the conversation useful instead of defensive.

Try something like: “I looked into that — there isn’t a blood test that diagnoses autism yet. [Child]’s diagnosis came from a clinician watching how [child] communicates and plays, over more than one visit, which is actually how it’s done. We did have some blood drawn after the diagnosis — that’s a genetic test, to check for an identifiable cause, not a test for autism itself. I’ll let you know if we learn anything from it.”

If they push back with a specific product or article, you can add: “That one measures something real in the blood, but it hasn’t been shown to reliably tell an autistic child from a non-autistic one — the two leading US clinical guidelines for autism don’t recommend it for diagnosis.” You do not owe anyone a longer explanation than that, and you are allowed to end the conversation there.

What to ask if a clinic offers a test

If a clinic offers you a blood test marketed for autism — beyond the genetic testing your child’s diagnosing clinician already discussed — five direct questions get you a straight answer fast:

1. What does this test actually measure, in plain language?
2. Is it approved to diagnose autism specifically?
3. What published, independent research backs it up, and has it been repeated by other labs?
4. What would a positive result change about my child’s care that we are not already doing?
5. What would a negative result change?

A clinic confident in its evidence will answer all five without hesitation. Vague answers, or an answer that leans on testimonials instead of published research, are your signal to slow down and check with your child’s regular clinician before you pay for anything.

Questions parents ask after reading this

1. If we pursue genetic testing, what specifically will the result change about our care this year?
2. If a relative or clinic mentions a specific blood test, what is the first question we ask before agreeing to anything?
3. Has fragile X, or another specific condition our clinician mentioned, already been ruled out — or is that still open?
4. Who is the right person to call if a lab result comes back and we do not understand it?

Write the answers in the same notebook you started for the assessment. Future clinicians will ask some of the same things.

If you only do three things

1. If your child has not had genetic testing since diagnosis, ask your clinician whether it is worth pursuing now — genetic testing, explained covers what to expect.
2. If a relative sends you a headline about “a blood test for autism,” use the script above rather than arguing from scratch.
3. If a clinic offers a blood test beyond standard genetic testing, ask the five questions above before you agree to it.

You are allowed to say no to a test that has not earned its evidence yet — and yes to the ones that have.