Child development

The Parent Interview Behind a Diagnosis

Save

Most of what families brace for in an autism evaluation is about the child. One long piece of it is about you: a caregiver interview that walks backward through a life, asking what your child did at two and what they do now. Here is what that interview is for, why it can feel like an interrogation, and what its result can and cannot settle.

Last updated: July 2026

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

What Is the ADI-R, and Who Answers It?

The Autism Diagnostic Interview-Revised is a semistructured, investigator-based diagnostic interview conducted with the caregivers of a person who may have a developmental disorder 1. A trained interviewer asks the questions, listens to how you describe a behavior, and decides whether what you described meets the interview's own definition of it. You supply the evidence; the interviewer applies the standard. Nobody scores your child in the room.

An investigator-based interview works differently from the questionnaires most families have already filled out. On a questionnaire you read the item and choose the answer. Here the interviewer probes for concrete examples and asks follow-ups until they understand what actually happens, then applies a coding definition you were never shown. Two parents who would tick the same box on a form can be coded differently, because the coding turns on the detail underneath the box.

It was published in 1994 by Lord, Rutter and Le Couteur as a reorganized and shortened revision of an earlier interview, extended to cover people with a mental age of roughly 18 months and upward into adulthood, and explicitly keyed to the ICD-10 and DSM-IV criteria in force at the time 1. Mental age rather than chronological age is the boundary, because many of the behaviors the interview asks about only become meaningful past a certain point in development.

Why the Questions Reach So Far Back

There is no blood test for autism. The diagnosis is built from developmental history and from behavior a clinician observes, which is why a comprehensive evaluation draws on people trained to read both — developmental pediatricians, child psychologists and psychiatrists, sometimes neurologists 2. The caregiver interview is where the history half comes from, and history here means years, not the last month.

Autism is developmental, so its shape lives in a timeline: what came when, what came late, what came and then went. An afternoon with a child shows who that child is on that afternoon. Only someone who was there can say what they were like at eighteen months.

That puts real weight on memory. In CDC surveillance of 8-year-olds at 11 US sites in 2020, the median age of earliest known autism diagnosis was 49 months 3. By the time most families sit down for this interview, the years being asked about are already behind them. Records help more than effort does, and home video most of all.

That same gap is the argument for acting early on autism rather than waiting to see whether something resolves on its own, and autism evaluation wait times widen it further, putting months between the moment a parent first worried and the moment anyone asks them about it.

How the Interview Is Scored

The interviewer codes behaviors on a defined scale as the interview goes, and a subset of those codes feeds a diagnostic algorithm written against ICD-10 and DSM-IV criteria 1. The algorithm's output is a classification: the coded history either does or does not meet the interview's threshold. That classification enters the evaluation as one piece of evidence alongside everything else. It is not the diagnosis.

The algorithm classifies. The clinician diagnoses. Those are different jobs, and the second takes in what the first cannot see — the direct observation session, cognitive and language testing, medical history, school reports, and what the clinician makes of the person in front of them.

This page does not print the ADI-R's items, its per-domain thresholds, or its algorithm cutoffs. Two reasons, and the second matters more. The instrument is a copyrighted clinical tool, so its items are not ours to publish. And a cutoff without the training that produces the codes underneath it is not information — it is a number a frightened parent can apply to their own child at two in the morning and reach a conclusion no clinician would sign. There is no home version of this interview.

One limitation is worth carrying in: the 1994 report validated the revision in a preschool sample 1, and the instrument is used well beyond preschoolers now. Its algorithm was also keyed to the diagnostic criteria of the early 1990s, so a fair question for your evaluator is which criteria their version scores against, and how much weight they give the algorithm result next to everything else they have.

What the ADI-R Cannot Settle

It cannot rule other explanations in or out on its own. A history that codes positive here is consistent with autism, and in places also consistent with a language disorder, a hearing loss nobody caught, an intellectual disability, a trauma history, or several of those together. Sorting between them is the differential diagnosis, and it belongs to the clinician, not to the algorithm.

It also cannot measure progress. The ADI-R was built as a diagnostic interview 1 — an instrument for deciding whether a pattern is present, not for tracking how much of it is present this year against last. No established threshold defines meaningful change on it, and a second ADI-R result should never be offered as proof that a therapy is working. Measuring change is a different job for different instruments.

And it carries whatever your memory and framing bring into the room. A parent who has never watched a neurotypical two-year-old closely has a different baseline than one with three older children. A parent afraid of being judged rounds the hard parts down. Neither is dishonesty; both move the coding. The interviewer's probing exists to get underneath exactly that, which is part of why the questioning can feel relentless.

Where the Interview Fits in an Adult Assessment

The interview reaches into adulthood by design, covering mental ages from roughly 18 months upward into adult life 1, and adult assessment does lean on developmental history. NICE's UK guideline for diagnosing autism in adults describes a multidisciplinary assessment that includes taking that developmental history alongside direct assessment of the person themselves 4.

The difficulty becomes obvious once named. A caregiver interview needs a caregiver who was there and remembers. Many adults seeking assessment in their thirties or fifties have parents who have died, parents they are estranged from, or parents whose version of a difficult childhood is itself part of what is being sorted out. Some have school records; many have none. Assessment in that situation leans harder on the adult's own account, on whatever documentation survived, and sometimes on a sibling or partner who can speak to the long view.

NICE is UK guidance rather than a US coverage rule or legal standard, worth saying plainly, because US adult assessment pathways vary by state, by plan, and by who nearby does adult work at all.

Can the Caregiver Interview Happen Over Video?

A caregiver interview is the part of an autism evaluation that ports most cleanly to video, because nothing in it needs the clinician to be in the same room as the child. The observation half is harder, and purpose-built tools exist for it: TELE-ASD-PEDS has the caregiver run a set of play activities while the clinician watches remotely and codes what they see 5.

Whether a full telehealth evaluation suits a particular child depends on their age and presentation and on what the evaluating clinic is set up to do. Preliminary work from the period when remote autism evaluation was scaled rapidly during COVID-19 reported that it was feasible and acceptable to the clinicians using it 6 — feasible and acceptable being a lower bar than definitively as accurate, which that work did not claim to have established.

In practice a remote evaluation is often a mix: the interview by video, the observation by video with the caregiver acting as the clinician's hands, and an in-person visit for whatever the screen cannot carry.

What Actually Helps Beforehand

Bring documents rather than a script. Early well-child notes, any prior speech, hearing or early-intervention evaluation, school and daycare comments, and phone video of ordinary moments — a meal, a playground, a transition that went badly — do more for the accuracy of this interview than a week of preparing for the evaluation ever will.

Do not rehearse, and do not summarize. The interview runs on concrete instances. "Last Tuesday at the store he did this, and it went on for about twenty minutes" is worth more than "he has trouble with transitions." If a question makes you realize you do not actually know, say so. A guess coded as fact is worse than an acknowledged gap.

Resist the pull to present your child at their best. Every parent feels it, and it is the most common way a caregiver interview quietly understates what a child needs. Nobody in that room is assessing your parenting. They are building a description of what your child finds hard, because that description is what services get written from.

Leave room afterward, too. Spending a long morning cataloguing everything difficult about your own child is its own kind of day, whatever the result turns out to be, and coping with the diagnosis — or with the wait for one — is a real part of this rather than a soft one.

Common questions

Longer than any other single conversation in most evaluations. The revised version was deliberately shortened compared with the interview it replaced, but it still moves through development in detail, and how long that takes depends on how much history there is to cover. The honest answer is to ask the clinic when you book, so you can arrange childcare and not be rushed.

Usually not both, but whoever holds the early years best should be. If one parent was the primary caregiver in toddlerhood and the other was not, the first one's memory is the more useful material. Two caregivers together can fill each other's gaps, which helps; a clinic may also have practical limits on who can attend, so ask when scheduling.

That is not what the interview is built to find. Its questions are about what the child does — how they play, communicate, respond, and react to change — not about how you handled it. The detail can feel exposing because it is detailed, not because it is aimed at you. Evaluators are assembling a description of need, not a verdict on a family.

Say so, and bring whatever remembers for you. Well-child visit notes, a baby book, daycare or preschool comments, and home video all count, and a relative who was around can fill in stretches you were too exhausted to encode. "I do not know" is a legitimate answer in this interview. A confident guess that turns out to be wrong is the outcome to avoid.

You can ask for the evaluation report, and asking is worth doing. The interview items themselves are a copyrighted clinical instrument and are not distributed to families, so the useful request is different: ask the evaluator to walk you through what the interview contributed, where it agreed with the rest of the assessment, and where it did not.

No. Meeting the algorithm's threshold means the developmental history you described fits the pattern the interview was built to detect. A diagnosis comes from a clinician weighing that alongside direct observation, testing, medical history, and school information. Interviews can meet the threshold in children who are not diagnosed, and children are diagnosed whose interviews did not.

Related

Say it back

How would you explain this to someone you love?

Two or three sentences, just as you’d say it. Gale reflects back what you focused on — a mirror, not a quiz.

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

Do not wait on an evaluation date for these

  • —Loss of speech, babble, gestures, or social skills a child previously had, at any age — a regression warrants a call to the pediatrician this week rather than waiting for a scheduled evaluation.
  • —A child who has stopped responding to their name or seems not to hear — hearing loss can look like autism, is separately treatable, and is tested for quickly.
  • —Staring spells with unresponsiveness, or convulsive movements, which need medical assessment rather than a developmental one.
  • —An older child, teenager, or adult awaiting assessment who talks about wanting to die, or about hurting themselves.

If someone waiting on an autism assessment talks about wanting to die or about hurting themselves, call or text 988, the Suicide and Crisis Lifeline. A first seizure, a convulsion, or unresponsiveness is a 911 call or an emergency-room visit, not something to raise at the next appointment.

This article explains what a caregiver interview is and how its result is used. It does not assess any particular child, and nothing here can be scored or applied at home. A diagnosis comes from a qualified clinician who has evaluated the person directly.

References

  1. 1.Lord C, Rutter M, Le Couteur A. (1994). Autism Diagnostic Interview-Revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. J Autism Dev Disord. 1994;24(5):659-85.. doi:10.1007/BF02172145 ✓That the ADI-R is a semistructured, investigator-based diagnostic interview administered to caregivers; that it was published in 1994 as a reorganized, shortened revision extended to individuals with mental ages from roughly 18 months into adulthood; that it is explicitly linked to ICD-10 and DSM-IV criteria; and that its original psychometric findings come from a preschool sample. Also that it is a diagnostic interview rather than a change-sensitive outcome measure.
  2. 2.Centers for Disease Control and Prevention (2024). Clinical Testing and Diagnosis for Autism Spectrum Disorder. CDC — Autism Spectrum Disorder (ASD), Healthcare Providers. linkThat autism diagnosis relies on developmental history and observed behavior rather than a blood test, and that a comprehensive evaluation may involve developmental pediatricians, child psychologists or psychiatrists, and neurologists.
  3. 3.Maenner MJ, Warren Z, Williams AR, et al. (CDC ADDM Network) (2023). Prevalence and Characteristics of Autism Spectrum Disorder Among Children Aged 8 Years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2020. MMWR Surveillance Summaries. PMID 36952288 ✓The median age of earliest known autism diagnosis of 49 months in CDC ADDM surveillance of 8-year-olds at 11 US sites in 2020, used here to show how far back a caregiver interview must reach.
  4. 4.National Institute for Health and Care Excellence (NICE) (2021). Autism spectrum disorder in adults: diagnosis and management (CG142). NICE Clinical Guideline (via NCBI Bookshelf). link ✓That adult autism assessment is conducted as a multidisciplinary assessment that includes taking a developmental history, cited as UK guidance rather than US rules.
  5. 5.Vanderbilt Kennedy Center, TRIAD (2024). TELE-ASD-PEDS (TAP). Vanderbilt Kennedy Center — TRIAD. link ✓That TELE-ASD-PEDS is a caregiver-administered set of play activities observed remotely by a clinician, used to structure the observation half of a telehealth autism evaluation.
  6. 6.Wagner L, Corona LL, Weitlauf AS, et al. (2020). Use of the TELE-ASD-PEDS for Autism Evaluations in Response to COVID-19: Preliminary Outcomes and Clinician Acceptability. Journal of Autism and Developmental Disorders. link ✓That telehealth-based autism evaluation was scaled during COVID-19 and reported as feasible and acceptable to clinicians in preliminary work, explicitly framed here as an implementation finding rather than a definitive accuracy result.

6 sources, numbered by first appearance. General health information, not medical advice. AI-assisted editorial content — citations link their sources. Editorial policy

← Back to the health library