What is the ADI-R? A guide to the Autism Diagnostic Interview-Revised

The ADI-R, formally the "Autism Diagnostic Interview-Revised," is a comprehensive, standardized interview that a trained clinician conducts with a parent or caregiver as part of an autism evaluation. Its 93 items gather a detailed picture of a person's early development, language and communication, reciprocal social interaction, and restricted and repetitive behaviors. The ADI-R is one component of a comprehensive evaluation carried out by qualified clinicians; it is neither a standalone diagnostic decision nor an IQ test. Authored by Michael Rutter, Ann Le Couteur, and Catherine Lord and published in 2003 by WPS (Western Psychological Services), it remains one of the most widely used instruments in autism research and clinical practice. This article explains where the ADI-R came from, what the interview covers, how its two types of scoring algorithms work, how it is paired with direct observation, and what its strengths and limitations are.

The ADI-R is a revision of the original Autism Diagnostic Interview. In a 1994 paper in the Journal of Autism and Developmental Disorders, Lord, Rutter, and Le Couteur described the revised instrument as a "semi-structured," "investigator-based" interview for caregivers of children and adults for whom autism is a possible diagnosis. "Semi-structured" means the interviewer works from a fixed set of questions and topics but asks them conversationally, probing for detail rather than reading a rigid script. "Investigator-based" means the interviewer, and never the caregiver, decides how each answer is coded, using standardized definitions of the behaviors in question. According to the authors, the revision reorganized and shortened the original interview and linked it to the diagnostic criteria in the ICD-10 and DSM-IV classification systems. The published edition has been distributed by WPS since 2003, with a Spanish-language edition also available, and the publisher notes that the interview had been used in research for decades before that.

According to WPS, the ADI-R's 93 items focus on three functional domains: language and communication; reciprocal social interactions; and restricted, repetitive, and stereotyped behaviors and interests. These map closely onto the behavioral criteria clinicians use when considering an autism diagnosis. To get there, the interviewer works through eight content areas in a set order, recording and coding the caregiver's responses under highly standardized procedures:

• Background: family circumstances, education, previous diagnoses, and medications.

• Overview of behavior: an orienting picture of how the person behaves day to day.

• Early development: developmental milestones in the first years of life.

• Language acquisition: when and how language emerged, including any loss of language or other skills.

• Current language and communication: how the person communicates now, verbally and nonverbally.

• Social development and play: interest in other people, shared enjoyment, friendships, and pretend play.

• Interests and behaviors: unusually intense interests, routines, rituals, and repetitive movements.

• Clinically relevant behaviors: issues such as aggression, self-injury, and possible epileptic features that matter for care planning.

The early-development questions are a defining feature. Because autism is a developmental condition, evidence about how a person behaved as a young child carries real diagnostic weight, and the ADI-R is built to collect that history methodically.

The ADI-R reverses the usual testing arrangement: the person being evaluated does not answer any questions. Instead, WPS explains, an experienced clinical interviewer questions a parent or caretaker who knows the individual's developmental history and current behavior well. The person under evaluation can be a child or an adult, as long as their "mental age," an estimate of the developmental level at which they are functioning, is above 2 years, 0 months. That floor exists because many of the behaviors the interview asks about, such as pretend play and conversational back-and-forth, are meaningless to probe below that developmental level.

Administration and scoring typically require 90 to 150 minutes, or roughly 1.5 to 2.5 hours, according to the publisher. The length reflects the instrument's depth: the interviewer needs enough detail on each behavior to code it against standardized definitions rather than simply accepting a yes or no.

Valid use also depends on training. WPS states plainly that administration and coding are highly standardized and that valid assessment requires training; its clinical training program is built around 16 hours of video showing expert interviewers administering the ADI-R. Clinicians who plan to use the instrument in formal research may need additional training provided by the test authors and their colleagues.

The ADI-R does not produce an IQ-style standard score or a percentile. WPS notes that because it is an interview rather than a test, and because it focuses on behaviors that are rare in people without autism, it yields categorical results instead of norm-referenced scales. Coded items feed into five age-specific scoring algorithms of two distinct types:

• Diagnostic algorithms (two): based on the full developmental history, these are used to support formal diagnosis. They weight early development heavily because the diagnostic criteria are developmental in nature.

• Current behavior algorithms (three): focused on how the person functions right now, these are used for treatment and educational planning, where present-day needs matter more than history.

This split is one of the ADI-R's most practical features. A clinician can ask two different questions of the same interview: does the lifetime pattern support a diagnosis, and what does this person need today? A single comprehensive algorithm form lets the clinician calculate any of the five algorithms from one administration.

No responsible clinician diagnoses autism from an interview alone, and the ADI-R was never designed to be used that way. In a 2021 paper in Frontiers in Psychiatry, Kamp-Becker and colleagues describe the current diagnostic gold standard as having two essential components: direct observation of behavior by an experienced clinician and a caregiver interview, with the ADI-R serving as that interview. The two methods cover each other's blind spots. A caregiver can report on decades of history the clinician will never see, while direct observation captures behavior a caregiver may have normalized or never noticed. Experienced clinical judgment then integrates both with everything else in the file.

A comprehensive evaluation usually reaches beyond autism-specific instruments too. Teams commonly add cognitive testing, an adaptive-behavior measure such as the Vineland-3, and quantitative rating scales such as the SRS-2. The ADI-R also sits at a different stage of the process than screening tools. A brief parent questionnaire like the M-CHAT flags toddlers who should be evaluated further; the ADI-R is part of that further evaluation, in the hands of qualified professionals. No online test or questionnaire can screen for or detect autism on its own, and a diagnosis is always a clinical judgment made by qualified clinicians.

The ADI-R's reputation rests on decades of research use. WPS reports that extensive international research has provided strong evidence for the reliability and validity of its categorical results, and that the instrument has proven effective at differentiating autism from other developmental disorders. Its structured developmental history is something no observation session can replicate, and its dual algorithms serve both diagnostic and planning purposes.

The limitations are equally well documented, and honest evaluators keep them in view:

• Training burden: valid administration requires substantial specialized training, with research use often requiring more. Kamp-Becker and colleagues note that the length and in-depth training required for the gold-standard instruments confine them largely to specialty clinics, which contributes to long waiting lists for evaluations.

• Length: at 1.5 to 2.5 hours for administration and scoring, the ADI-R consumes significant clinician and family time before any observation or testing takes place.

• Reliance on caregiver recall: the same 2021 paper cautions that the ADI-R may be subject to retrospective recall biases or inaccurate caregiver memory, particularly when a caregiver was not concerned about the child's behavior early on.

• Imperfect accuracy: a 2021 meta-analysis by Lebersfeld and colleagues in the Journal of Autism and Developmental Disorders estimated the ADI-R's "sensitivity," the proportion of autistic individuals it correctly identifies, at about .75, and its "specificity," the proportion of non-autistic individuals it correctly rules out, at about .82, with weaker specificity in everyday clinical samples than in research samples. Kamp-Becker's team likewise found that for adolescents and adults, a classifier built on observation alone outperformed one combining observation with the ADI-R. Findings like these are exactly why the instrument informs, and does not replace, clinical judgment.

No. The ADI-R is one component of a comprehensive evaluation. Qualified clinicians combine it with direct observation, other testing, and clinical judgment before making any diagnosis.

According to the publisher, administration and scoring typically require 90 to 150 minutes, or about 1.5 to 2.5 hours.

A parent or caregiver who knows the individual's developmental history and current behavior well. The person being evaluated does not answer the interview questions.

An experienced clinical interviewer with specific training in the instrument. Administration and coding are highly standardized, and formal research use may require additional training from the test authors and their colleagues.

It can be used with children and adults as long as their mental age is above 2 years, 0 months.

No. It measures autism-related behavior through caregiver report and produces categorical results rather than cognitive scores. Intelligence is assessed separately with dedicated cognitive tests.

The ADI-R measures nothing about intelligence. It produces no cognitive score of any kind, and an autism evaluation and an IQ test answer different questions, which is why thorough assessment teams often administer both. Cognitive testing in that context is individually administered by a professional, and the ADI-R itself belongs strictly in trained clinical hands.

For adults who are simply curious about their own cognitive abilities, a different kind of instrument applies. The RIOT IQ test (Reasoning and Intelligence Online Test), developed by RIOT IQ with psychometrician Dr. Russell T. Warne, measures adult cognitive ability through 15 subtests across six indices: verbal reasoning, fluid reasoning, spatial ability, working memory, processing speed, and reaction time. It takes about 52 minutes and reports scores on the familiar mean-100, standard-deviation-15 IQ scale. It is an IQ test for adults 18 and older; it is not an autism assessment, and it does not replace an individually administered diagnostic evaluation of any kind. You can learn more or take the test at riotiq.com.

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. Journal of Autism and Developmental Disorders, 24(5), 659-685. doi.org

2. Rutter, M., Le Couteur, A., & Lord, C. (2003). Autism Diagnostic Interview-Revised (ADI-R) [Product page]. WPS. wpspublish.com

3. Pearson Clinical Assessment. (n.d.). Autism Diagnostic Interview-Revised (ADI-R) [Product page]. pearsonclinical.com.au

4. Kamp-Becker, I., Tauscher, J., Wolff, N., Kuepper, C., Poustka, L., Roepke, S., Roessner, V., Heider, D., & Stroth, S. (2021). Is the combination of ADOS and ADI-R necessary to classify ASD? Rethinking the "gold standard" in diagnosing ASD. Frontiers in Psychiatry, 12, 727308. pmc.ncbi.nlm.nih.gov

5. Lebersfeld, J. B., Swanson, M., Clesi, C. D., & O'Kelley, S. E. (2021). Systematic review and meta-analysis of the clinical utility of the ADOS-2 and the ADI-R in diagnosing autism spectrum disorders in children. Journal of Autism and Developmental Disorders, 51(11), 4101-4114. doi.org

Photo: Virginia State Parks staff, via Wikimedia Commons (CC BY 2.0), cropped