Adult ADHD Test (ASRS)

The Adult ADHD Self-Report Scale (ASRS) was developed in conjunction with the World Health Organization (WHO), while the WHO was revising its Composite International Diagnostic Interview. Its 18 questions ask how often you've had the symptoms in the DSM-IV criteria for adult ADHD. The v1.1 checklist came from a workgroup that included Lenard Adler, Ronald Kessler and Thomas Spencer.

This page uses the six-question screener, which is Part A of that checklist. In a 2005 paper in Psychological Medicine, Kessler and colleagues picked these six questions out of the 18 by stepwise logistic regression, so that answers would match clinicians' blind ratings as closely as possible. They tested it on 154 people from the US National Comorbidity Survey Replication, a sample that over-represented people who'd reported childhood ADHD and symptoms that lasted into adulthood.

In that 2005 study, the unweighted six-question screener did better than all 18 questions: sensitivity 68.7% versus 56.3%, specificity 99.5% versus 98.3%, and kappa 0.76 versus 0.58. A sensitivity of 68.7% means the screener missed some people the clinicians rated as having ADHD.

A 2007 study of 668 health plan members found internal consistency of 0.63 to 0.72, test-retest correlations of 0.58 to 0.77, and an area under the curve (AUC) of 0.90 against clinician diagnoses for a four-category version of the screener (abstract only). In a later study of 646 adults with a clinical ADHD diagnosis and 908 population controls, the short screener's AUC was 0.903.

Results depend on who's answering. When two general-population samples (UK, 642 people; US, 579) took the ASRS v1.1, it flagged probable ADHD in 26.0% and 17.3%, against an expected prevalence of 2.5%. The authors estimated a positive predictive value of only about 11.5%, which means most people who screen positive in a general sample wouldn't have ADHD.

An answer counts as flagged when it reaches the shaded range: Sometimes or higher on questions 1 to 3, Often or higher on questions 4 to 6. Four flagged answers is the published rule. The two labels are this page's own.

The screener asks about symptoms. It can't tell ADHD from other causes of the same problems, and anxiety is a common one. In 618 adults with clinical anxiety, some hyperactivity items, including feeling driven by a motor (question 6 here), loaded more strongly on anxiety than on ADHD. A survey of 72,627 people in 42 countries flagged 21.4% of women and 18.1% of men as at risk, which the authors said matches earlier evidence that adult ADHD screeners have low specificity. Scores were highest in the US, Canada and other English-speaking Western countries.

The v1.1 questions follow DSM-IV criteria. The same group later published a DSM-5 version, the ASRS-5, with different questions. The checklist's own instructions add that a clinician should also review how symptoms affect work, school and home, and whether similar problems began in childhood.

Treat a positive result as a reason to ask for a full evaluation from a psychiatrist, psychologist or primary care clinician. A negative result doesn't rule ADHD out, so if trouble with focus or organization is affecting your work, school or relationships, it's worth raising with a clinician whatever your count. The test covers the past 6 months. If you retake it, wait a few months rather than a few days.

In a US survey of 22,397 adults, those who reported an ADHD diagnosis also reported depression (58.1%), anxiety (53.1%) and sleep difficulties (37.0%) far more often than other adults. Our anxiety test, depression test and sleep test can help you check whether those are part of the picture.

More on this topic: Procrastination test, Anxiety test (GAD-7), Stress test (PSS-10), Sleep test.

Your answers are scored in your browser and aren't sent or saved anywhere. Close the page and they're gone.

• Kessler et al., Psychological Medicine, 2005 (PubMed): Development of the six-question screener, the 154-person NCS-R comparison with clinician ratings, sensitivity, specificity and kappa. Abstract only.
• Kessler et al., International Journal of Methods in Psychiatric Research, 2007 (PMC): 668 health plan members: reliability and AUC of 0.90. Abstract only; the full text PDF could not be downloaded.
• NYU Langone Technology Opportunities & Ventures, ASRS v1.1 6-Question Screener: Licensing terms and the required citation; says the screener isn't a standalone diagnostic tool.
• NYU Langone, ASRS-V1.1 Screener (patient version, PDF): The six questions, response options, shaded boxes, the four-or-more rule, and the 18-and-older note.
• ADDA copy of the ASRS-v1.1 Symptom Checklist and instructions (PDF): The 18 questions, Part A and Part B, the scoring instructions, the workgroup members, and the advice to assess impairment and childhood history. A third-party copy of the official form.
• Brevik et al., Brain and Behavior, 2020 (PubMed): 646 clinically diagnosed adults and 908 population controls; AUC of the short screener. Abstract only.
• Chamberlain, Cortese and Grant, Compr Psychiatry, 2021 (PubMed): UK and US general-population samples, positive rates and positive predictive value. Abstract only.
• Alarachi et al., Assessment, 2024 (PubMed): 618 adults with clinical anxiety; overlap between some hyperactivity items and anxiety. Abstract only.
• Lewczuk et al., Journal of Attention Disorders, 2024 (PubMed): 42 countries, 72,627 people; score differences by region. Abstract only.
• Ustun et al., JAMA Psychiatry, 2017 (PubMed): The DSM-5 version of the ASRS. Abstract only.
• Adler et al., International Journal of Clinical Practice, 2019 (PubMed): US norms and co-occurring conditions in 22,397 respondents. Abstract only.