RAADS-14 Screen

The RAADS-14 Screen is a 14-item self-report screening instrument for autism spectrum disorder in adults, developed by Joakim Eriksson, Louise Andersen, and Susanne Bejerot at Karolinska Institutet and published in Molecular Autism in 2013 under Creative Commons Attribution 2.0. It was derived from the 80-item Ritvo Autism and Asperger Diagnostic Scale-Revised (RAADS-R, Ritvo 2011) using factor analysis to retain the items with the strongest discriminatory power for autism spectrum disorder versus other psychiatric conditions. The cutoff for a positive screen is ≥14 of a possible 42 points, with reported sensitivity of 97% and specificity that varies by comparison group.

The RAADS-14 is one of the most widely used brief adult autism screens in clinical and research settings. Its three-factor structure — Mentalizing Deficits, Social Anxiety, and Sensory Reactivity — provides sub-dimensional information that single-factor screens like the AQ-10 do not produce. The CC BY 2.0 license also makes it one of the few brief adult autism screens that can be deployed freely in commercial and public-facing applications, which has accelerated its adoption.

Autism in adults is systematically underdiagnosed. Mean age at autism diagnosis in adults seeking evaluation is often 30 to 40 years, with women, gender-diverse adults, and racially minoritized adults disproportionately represented in late-diagnosis populations. The clinical bottleneck is not desire for evaluation but capacity: comprehensive autism evaluation using the diagnostic gold standards (ADOS-2, ADI-R) takes hours of clinician time and is in chronic short supply for adult populations. Brief screening tools fill the triage gap.

The RAADS-14 specifically targets adult psychiatric outpatients, the population in which differential diagnosis is hardest. Clinicians evaluating an adult presenting with social difficulty, attention problems, or anxiety face the question of whether autism, ADHD, social anxiety disorder, or some combination best accounts for the presentation. The RAADS-14 was validated specifically in this context, with samples comparing autism to ADHD, autism to other psychiatric conditions, and autism to non-psychiatric controls. Its discriminatory power varies by comparison — lower against ADHD because of genuine symptom overlap, higher against non-psychiatric controls — which is exactly the pattern an honest screening tool should produce.

The instrument has practical advantages beyond its validation: 14 items take 3 to 5 minutes to administer, sub-dimensional scoring provides interpretable feedback to clinicians and patients, and the open license enables free public deployment. These advantages have made the RAADS-14 the de-facto standard brief adult autism screen in many psychiatric outpatient settings, despite the existence of older and equally well-validated alternatives.

The RAADS-14 derives from the Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R), developed by Riva Ariella Ritvo, Edward Ritvo, and colleagues and published in the Journal of Autism and Developmental Disorders in 2011 (international validation study). The RAADS-R is an 80-item self-report instrument designed to assist with adult autism diagnosis across four domains: language, social relatedness, sensory-motor, and circumscribed interests. RAADS-R itself derived from the original RAADS, which was published earlier and intended primarily for adults with high cognitive capacity (verbal IQ ≥ 80).

Joakim Eriksson, Louise Andersen, and Susanne Bejerot at Karolinska Institutet developed the RAADS-14 as a screening shortform for use in adult psychiatric outpatient settings. They tested all 80 RAADS-R items in a Swedish psychiatric outpatient sample and used Receiver Operating Characteristic (ROC) curve analysis to identify the items with the strongest discriminatory power for autism spectrum disorder versus other psychiatric conditions and non-psychiatric controls. The 14 items with the highest area-under-the-curve values were retained; four items with the lowest discriminatory power were specifically removed.

The validation paper was published in Molecular Autism in December 2013 (DOI: 10.1186/2040-2392-4-49) under Creative Commons Attribution 2.0. The choice of license was deliberate: Eriksson and colleagues explicitly noted that they wanted the RAADS-14 to be freely usable, including in clinical, research, and public-facing applications. This contrasts with the AQ family (Cambridge ARC restricts commercial use) and the SRS family (proprietary, licensed by Western Psychological Services), and is one of the main reasons the RAADS-14 has been adopted as the standard instrument in many free public screening tools.

The 14 items below are reproduced verbatim from Eriksson et al. (2013), Table 2, under Creative Commons Attribution 2.0. Subscale assignment derives from the factor analysis reported in Eriksson 2013, Table 4. Item 6 is the only reverse-scored item — it is worded in the neurotypical direction, so its raw response is subtracted from 3 before summing.

The three subscales

• Mentalizing Deficits — items 1, 4, 9, 11, 12, 13, 14 (7 items, max 21). Theory-of-mind, social cognition, detail-focused processing, and change rigidity.
• Social Anxiety — items 3, 5, 6, 8 (4 items, max 12; item 6 reversed). Group functioning, social uncertainty, friendship-building difficulty, and small-talk capacity.
• Sensory Reactivity — items 2, 7, 10 (3 items, max 9). Tactile sensitivity, sensory overwhelm, and auditory hyper-reactivity.

Eriksson 2013 also identified a rapid-screen sub-form using only items 1 through 5 with a cutoff of ≥4 (sensitivity 93%, specificity 45-49%). This sub-form was reported as a possible coarse triage when even 14 items is impractical. The 5-item rapid-screen has not been independently revalidated and is not the recommended primary screen.

Each of the 14 items is rated on a 4-point Likert scale. The response anchors, reproduced verbatim from Eriksson 2013, are:

• 3 — True now and when I was young
• 2 — True only now (refers to skills acquired since age 16)
• 1 — True only when I was younger than 16
• 0 — Never true (and never described me)

Total scores are computed by summing the 14 item values, with item 6 reversed. The maximum possible score is 42 (all 14 items at 3, accounting for item 6 reversal), and the minimum is 0. The published cutoff for a positive screen is ≥14.

The item-6 reversal

Item 6 (“I can chat and make small talk with people”) is the only reverse-scored item. The other 13 items are worded so that endorsement is consistent with autism (for example, “It is difficult for me to understand how other people are feeling”). Item 6 is worded so that endorsement is consistent with neurotypical social fluency, which is the opposite direction. The reversal rule is:

• Response of 0 (“Never true”) → contributes 3 to total
• Response of 1 → contributes 2
• Response of 2 → contributes 1
• Response of 3 (“True now and when young”) → contributes 0

This is mathematically equivalent to subtracting the raw response from 3 before summing. Most computerized administrations of the RAADS-14 handle this automatically; pencil-and-paper administration requires the scorer to apply the reversal manually.

Sub-dimensional scoring

Sum each subscale’s items separately to produce three sub-dimensional scores. Eriksson 2013 reports that subscale-level scoring provides interpretive value beyond the binary cutoff, particularly for distinguishing between predominant phenotypes (mentalizing-dominant vs sensory-dominant vs social-anxiety-dominant presentations). The original validation study did not establish subscale-level cutoffs, so subscale scores are best interpreted descriptively rather than diagnostically.

The RAADS-14 was validated in three Swedish samples reported in Eriksson 2013: an autism sample (Phase II, n = 135), a psychiatric outpatient sample (OPD, n = 213), and a non-psychiatric general-population sample (n = 590). Within the OPD sample, an ADHD sub-sample was further analyzed separately to assess discriminant validity in a population with high feature overlap.

Internal consistency

Cronbach’s alpha for the full 14-item scale was 0.92 in the combined ASD plus OPD sample, indicating high internal consistency. Subscale-level alphas ranged from 0.81 (Sensory, 3 items) to 0.87 (Social Anxiety, 4 items) — all in the acceptable-to-strong range despite the brevity of individual subscales.

Sensitivity, specificity, and AUC

At the published cutoff of ≥14:

The pattern of high sensitivity (97% across all comparisons) and variable specificity is characteristic of a screening instrument designed to err toward false positives rather than false negatives. The 46% specificity against ADHD reflects genuine symptom overlap between the two conditions, and is the main reason the RAADS-14 cannot reliably distinguish autism from ADHD on its own. Co-occurring autism and ADHD (AuDHD) is increasingly recognized.

Three-factor structure

Confirmatory factor analysis in Eriksson 2013 supported the three-factor structure (Mentalizing, Social Anxiety, Sensory) with adequate fit indices. Factor loadings and inter-factor correlations are reported in Eriksson 2013, Table 4. The factor structure has not been independently replicated outside Swedish populations; cross-cultural invariance work would strengthen confidence in the sub-dimensional interpretation.

What is missing

Eriksson 2013 did not report test-retest reliability for the RAADS-14. The parent RAADS-R (Ritvo 2011) reported test-retest r = 0.987, and the brief 14-item form is presumed to retain similar stability, but this has not been formally confirmed. The validation samples were also drawn entirely from Swedish adult psychiatric outpatient populations; performance in non-psychiatric general-population samples or in non-Swedish cultural contexts is less well characterized.

The validation samples in Eriksson 2013 produced the following mean total scores. These norms are descriptive of the Swedish adult validation samples and may not generalize precisely to other cultural contexts or to non-psychiatric populations.

The Eriksson 2013 paper reports the median score of the 135 ASD adults at 32 of 42 points, which has become a frequently-cited reference point in clinical interpretation. The substantial gap between the OPD mean (12.4) and the ASD mean (27.9 to 30.8) is the empirical basis for the cutoff at 14 — a score below 14 is unusual in autism and a score in the 14 to 25 range is consistent with either autism or another psychiatric condition with social-cognitive features, while scores above 26 strongly suggest autism.

The RAADS-14 has documented limitations that should be held clearly in view when interpreting results.

1. It is a screen, not a diagnostic instrument

Autism diagnosis requires structured clinical interview (typically the ADOS-2 and ADI-R), developmental history, and assessment of functional impairment. None of these can be performed by a self-report instrument. A positive RAADS-14 means a clinical conversation is reasonable; it does not mean autism.

2. Specificity against ADHD is low

At the published cutoff, specificity against ADHD is only 46%. Adults with diagnosed or suspected ADHD should interpret a positive RAADS-14 cautiously and consider the possibility of co-occurring autism and ADHD (AuDHD) rather than autism alone.

3. May under-detect in women and gender-diverse adults

Autism is systematically underdiagnosed in women and gender-diverse populations, in part due to camouflaging — the conscious masking of autistic traits to fit neurotypical expectations. The RAADS-14 was validated in samples that included both men and women, but the items themselves emphasize externally-visible social difficulty more than internalized accommodation. A negative RAADS-14 in a woman who suspects autism does not rule it out.

4. Does not distinguish autism from social anxiety in some cases

Eriksson 2013 specifically noted that items 5 and 6 do not reliably distinguish adults with autism from adults with social anxiety disorder. Users whose elevation is concentrated in the Social Anxiety subscale should consider that social anxiety disorder is the more parsimonious explanation in many cases, and that the two conditions can also co-occur.

5. Does not assess childhood developmental history

Autism is, by DSM-5 definition, a neurodevelopmental condition with onset in early developmental period. The RAADS-14 response anchors implicitly capture some childhood information, but this is much more limited than the structured developmental history clinical diagnosis requires.

6. Does not capture stereotypies

DSM-5 Criterion B includes stereotyped or repetitive motor movements (hand-flapping, rocking, etc.). The RAADS-14 captures rigidity (item 14) but does not directly assess motor stereotypies. A clinical evaluation will probe these directly; the screen does not.

7. Validation context: Swedish adult psychiatric outpatients

The instrument’s psychometric performance in non-Swedish populations and non-psychiatric general-population samples is less well characterized. The reported 95% specificity against non-psychiatric controls is encouraging but comes from a single national context.

Several validated adult autism screens exist. Each makes different trade-offs between brevity, sensitivity, specificity, and licensing.

The RAADS-14 is most often chosen for free public-facing deployment because it is the only validated, brief adult autism screen released under a license that permits commercial use with attribution. For research or clinical contexts where licensing is not a constraint, the AQ-10 is shorter (10 items vs 14) but produces a single composite score without sub-dimensional information; the SRS-A is more comprehensive but requires a clinical license.

• Autism Spectrum Disorder — the clinical condition the RAADS-14 screens for. DSM-5 criteria, prevalence, comorbidity, and lifespan considerations.
• AuDHD — co-occurring autism and ADHD. The RAADS-14’s lower specificity against ADHD reflects the genuine overlap between these conditions.
• Autism Camouflaging (Masking) — the conscious masking of autistic traits, measured separately by the CAT-Q (Hull 2019). A complementary rather than overlapping construct.
• Late-Diagnosed Autism — the population most likely to be self-administering screening tools like the RAADS-14.
• ASRS Screener — the validated adult ADHD screening instrument; commonly used alongside the RAADS-14 to assess for AuDHD.
• RAADS-R — Ritvo Autism Asperger Diagnostic Scale-Revised; the 80-item parent instrument from which RAADS-14 was derived.
• CAT-Q — Camouflaging Autistic Traits Questionnaire (Hull 2019); 25-item instrument measuring autism-specific masking behaviors.

The full methodology page documents the implementation choices in detail: scoring algorithm pseudocode, severity-band derivation, archetype thresholds, validation evidence, population norms, and limitations.