The RAADS-R: A Clinician Guide to Adult Autism Screening

The Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R) is an 80-item self-report scale designed to assist the diagnosis of Autism Spectrum Disorder (ASD) in adults. It is one of the longer instruments in routine clinical use - roughly 25 minutes to complete - reflecting its goal of capturing lifelong patterns rather than a recent symptom snapshot. This guide covers what the RAADS-R measures, how its reverse-keyed items are scored, its published cut-off, and an important limitation in how LetPsyc reports results.

What the RAADS-R measures

The RAADS-R is a self-report screen completed by the adult patient, designed to assist the diagnosis of Autism Spectrum Disorder. What sets it apart from a typical symptom screener is its response format: rather than asking only "is this true now," each of the 80 items asks the respondent to place the statement in time, distinguishing symptoms present since childhood from symptoms that emerged only in adulthood. This is a deliberate design choice - ASD is by definition a lifelong neurodevelopmental condition, so a scale intended to assist its diagnosis needs to separate "this has always been true of me" from "this only became true recently."

Response scale

Each item is rated on a four-point scale:

• 3 - True now and when I was young
• 2 - Only true now
• 1 - True only when I was younger than 16
• 0 - Never true

Administration takes about 25 minutes across all 80 items, making it substantially longer than most other screens in routine use - a cost that reflects the depth of the lifelong symptom history it is trying to capture.

How to score the RAADS-R

Most items are scored directly from the 0-3 scale above, but 17 of the 80 items are reverse-keyed: items 1, 6, 11, 18, 23, 26, 33, 37, 43, 47, 48, 53, 58, 62, 68, 72, and 77. For these items, the response values are reversed before summing, so that a higher raw response contributes a lower point value to the total (and vice versa) - this corrects for items that are worded in the opposite direction from the rest of the scale, so that a higher total always means more autism-consistent responses throughout, not a mix of directions. Once every item is scored in the correct direction, all 80 item scores are summed for a total ranging from 0 to 240.

Because a full RAADS-R scoring pass means checking 80 individual responses against a 17-item reversal list before summing, hand-scoring is a realistic place for a single flipped item to go unnoticed - exactly the kind of consistent, rule-based task suited to automated scoring.

RAADS-R cut-off and interpretation

The total score is interpreted against a single published cut-off:

• 0-64: Below the RAADS-R screening cut-off
• 65-240: At or above the cut-off of 65 - consistent with ASD, warranting full diagnostic evaluation

This cut-off of 65 is worth a note on provenance: it comes from the RAADS-R's published validation literature (Ritvo et al., 2011), not from a scoring page inside the item-sheet source document itself, which carries no scoring section at all. Clinicians should treat 65 as the field-standard published cut-off, while remembering - as with every screening threshold - that it separates "warrants full diagnostic evaluation" from "does not," not "has ASD" from "does not."

An important limitation: subscales are not reported

The published RAADS-R organizes its 80 items into four subscales - Language, Social Relatedness, Sensory-Motor, and Circumscribed Interests - which in principle let a clinician see which domains are driving an elevated total. LetPsyc's implementation reports the total score only; the four subscales are not implemented. This is a deliberate choice, not an oversight: the exact item-by-item composition of each subscale is not present in the source material held here, and rather than reconstruct that mapping from memory and risk misattributing individual items to the wrong subscale, only the total score and its published cut-off are reported. Clinicians who need the subscale breakdown should consult the original RAADS-R publication or a source that reproduces the full subscale key.

Who completes it, age range, and time required

The RAADS-R is self-completed by adult patients and takes approximately 25 minutes for all 80 items.

RAADS-R compared with the M-CHAT-R

The RAADS-R and the M-CHAT-R both screen for autism spectrum risk, but at opposite ends of the lifespan and with opposite reporting structures. The M-CHAT-R is a brief 20-item parent-report checklist for toddlers aged 16-30 months, taking about 5 minutes, because a toddler cannot self-report. The RAADS-R is an 80-item adult self-report scale taking about 25 minutes, built to reconstruct a lifelong symptom history the respondent reports on themselves. Where the M-CHAT-R asks a caregiver about a small child's present-day behavior, the RAADS-R asks an adult to distinguish childhood-onset from adult-onset traits across their own lifetime - reflecting the very different clinical question each instrument is built to answer.

Limitations

As a self-report instrument relying on retrospective recall back to childhood, the RAADS-R depends on the respondent's memory and self-insight, which can be imperfect regardless of diagnosis. Its length (80 items, roughly 25 minutes) is both a strength - allowing broader symptom coverage - and a burden in a busy clinical setting. As noted above, LetPsyc's implementation reports the total score only, without the four published subscales, so clinicians needing that domain-level breakdown will need to supplement with the original instrument or clinical interview. Most importantly, the RAADS-R is a screening instrument only: a score at or above 65 is consistent with ASD and warrants full diagnostic evaluation, but it does not itself constitute a diagnosis.

Using the RAADS-R digitally

A digital RAADS-R applies the 17-item reversal key automatically before summing, so the total is accurate without a manual check against the reversal list, and flags whether the total is at or above the published cut-off of 65 in the clinician's report. This removes a realistic source of scoring error given the instrument's length, while keeping the reporting scope honest about what is - and is not - included (total only, no subscales).

Key takeaways

• The RAADS-R is an 80-item self-report scale assisting adult ASD diagnosis, taking about 25 minutes.
• Items are scored 0-3, distinguishing childhood-onset from adult-onset symptoms; 17 items are reverse-keyed before summing.
• The total score ranges from 0 to 240; a score of 65 or above is the published cut-off consistent with ASD, warranting full diagnostic evaluation.
• The 65-point cut-off comes from the published validation literature (Ritvo et al., 2011), not from a scoring page in the item sheet itself.
• LetPsyc reports the total score only - the four published RAADS-R subscales (Language, Social Relatedness, Sensory-Motor, Circumscribed Interests) are not implemented.