Show the full module text

Module 2 — ASRS v1.1 — The WHO's Six-Question Adult ADHD Screener

How the World Health Organization's adult screener works, why the threshold is different on different questions, what a positive result is genuinely worth, and why the negative result is the more trustworthy one.

ADHD Self-Discovery · Part Two — The Broad Screeners · 13 min read, about 24 min with the workbook

The big idea

If someone has told you to just try harder — and you knew, privately, that you were already trying harder than anyone could see — this module is about the six questions the World Health Organization settled on, what they were built to do, and the one thing they genuinely cannot do.

Step 1 — The lesson

A. Big engine, small rudder

The word "attention" has done a lot of damage to how ADHD gets understood. It suggests a shortage — not enough focus, not enough drive, not enough care. Most adults who get here recognise the opposite problem.

There is usually plenty of engine. There are ideas, urgency, interest, the ability to work for nine hours straight on something that has hold of you. What is in short supply is the steering: choosing what the engine points at, starting when the thing is dull, stopping when the thing is not, and holding a course when nothing external is forcing it.

Diagram — A · Big engine, small rudder. Not a shortage of power. A shortage of steering for the power you already have. This is why the same person can lose an afternoon to a task nobody asked for and be unable to start a fifteen-minute one that matters.

Some of the most useful clinical models of ADHD describe it primarily as a difficulty with self-regulation rather than with attention itself — the capacity to hold a future goal in mind and let it govern what you do now, when something more immediate is competing for the wheel.

B. Nobody claps for the plates that stay up

The second reason adults get missed for decades is that the work is invisible until it fails.

You build systems. Alarms on top of alarms. Arriving forty minutes early because you cannot calibrate twenty. Doing three days of work in the one night before it is due, and delivering something good, so nobody ever sees the cost. From the outside all of that reads as fine. It reads as fine right up until one plate drops, and then the drop is the only data anyone has.

Diagram — B · Nobody claps for the plates that stay up. The effort that keeps everything spinning is invisible by design. Only the one that falls gets noticed — which is why the feedback people get is about carelessness, and never about the enormous work that preceded it.

This is the mechanism behind the population numbers. Adult ADHD sits at roughly 2.8% across twenty national surveys, and the same research describes it as vastly under-recognised and undertreated, with most people seeking help for something else entirely.¹ Recorded adult diagnosis roughly doubled in one large US health system between 2007 and 2016 and still sat far below the epidemiological estimate.² That gap is not a mystery. It is a room full of people whose plates mostly stayed up.

There is a well-documented sex pattern here too. Girls meet criteria at roughly half the rate of boys in childhood, and the ratio moves close to equal in adulthood — a shift that points at recognition rather than biology. Clinical referral ratios have run anywhere from 3:1 to 16:1 in favour of boys against a community ratio of around 3:1.³

C. Six keys out of eighteen

The ASRS v1.1 has 18 questions, all asking how often something has happened in the last six months, answered never, rarely, sometimes, often, or very often.

But only the first six carry the screen. Those six were not chosen because they sound the most like ADHD — they were chosen statistically, as the subset that best separated people who did meet criteria from people who did not. Four are about attention and organisation; two are about restlessness. Notably, none of them is about impulsivity.

Diagram — C · Six keys out of eighteen. Part A is the tested screen. Part B is twelve further questions the instrument itself says carry no total score and no diagnostic likelihood — they exist to give a clinician more to ask about.

There is one genuinely unusual feature worth knowing before you take it, because it surprises people who go looking at the scoring afterwards. The threshold is not the same on every question. On the first three, answering sometimes already counts as a positive. On the next three, it has to be often or more.

That is not sloppiness. It is the whole point. When the developers checked each symptom against clinical assessment, agreement varied enormously from item to item. "Sometimes I have trouble getting things in order" separates people; "sometimes I fidget" does not, because nearly everyone fidgets sometimes. Each question got the cut-point that made it useful. Four or more positives across those six is the screen.

D. A signal with only two lamps

The instrument's own instruction, in its own words, is that four or more marks in the shaded boxes means symptoms "highly consistent with ADHD in adults" and that further investigation is warranted. That is the entire claim it makes. It is worth reading twice, because it is much more modest than what people usually take away.

Diagram — D · A signal with only two lamps. Every screener in this course is a two-lamp signal: low or no indication, or possible or elevated indication. There is no third lamp that lights up to say you have it. That lamp does not exist on any of these instruments.

Here is what the ASRS does not ask about, and each of these is a required part of an actual diagnosis. It does not ask whether anything was present before you were twelve — every question is about the last six months. It does not ask whether it shows up in more than one part of your life. It does not ask whether it causes real impairment. And it does not ask whether something else explains it better — depression, anxiety, trauma, chronic sleep debt, thyroid problems and a long list of others produce overlapping answers.

That last one is not theoretical. In a study of adults being treated for depression, the ASRS Part A returned a positive predictive value of 21% — roughly four in five positives were the depression talking. The negative predictive value in the same study was 92%.⁴ Which is the pattern across this entire literature, and it is genuinely useful news: a negative ASRS is the more trustworthy result.

INSIDE THE INSTRUMENT

ASRS v1.1 — Adult ADHD Self-Report Scale (WHO)

Where it came from

Kessler, Adler, Ames, Demler, Faraone, Hiripi, Howes, Jin, Secnik, Spencer, Ustun and Walters (2005), Psychological Medicine 35(2), 245-256.⁵ Developed alongside the revision of the WHO Composite International Diagnostic Interview so that a short self-report could stand in for interview items in population surveys. The 6-item Part A was selected from the 18 by stepwise logistic regression to optimise concordance with clinical classification. Copyright is held jointly by New York University and Ronald Kessler.

What it is made of

18 items covering DSM-IV Criterion A symptoms, each rated on a 5-point frequency scale for the past six months. Part A is items 1-6 (four inattention/executive, two hyperactivity, no impulsivity item); Part B is items 7-18, which the instrument states carry no total score and no diagnostic likelihood. Scoring uses a variable threshold: items 1, 2 and 3 count as positive at sometimes or above, items 4, 5 and 6 at often or above. Four or more positives across Part A is a positive screen.

How well it performs

In the original clinical reappraisal sample (n = 154, oversampled for reported childhood ADHD), the unweighted six-question Part A outperformed the full 18 items: sensitivity 68.7% versus 56.3%, specificity 99.5% versus 98.3%, total classification accuracy 97.9% versus 96.2%, kappa 0.76 versus 0.58. Note that sensitivity was only 68.7%, and that the very high specificity reflects a low base rate. Independent validation in 668 health-plan subscribers found internal consistency 0.63-0.72, test-retest 0.58-0.77, and AUC 0.90.⁶ The largest independent known-groups study (646 diagnosed adults versus 908 controls) reported AUC 0.904 for the full scale and 0.903 for the short screener.⁷

Where it was validated

US population and health-plan samples, against DSM-IV criteria, in the mid-2000s. It has never been re-derived for DSM-5, which changed symptom wording, raised the age-of-onset threshold from 7 to 12, and lowered the adult symptom threshold from 6 to 5.

What it cannot do

It substantially over-identifies in unselected populations. Two independent general-population samples (UK n = 642, USA n = 579) found the ASRS indicated probable ADHD in 26.0% and 17.3% of participants against an expected prevalence of 2.5%, with an estimated positive predictive value around 11.5%; the authors concluded it produces seven- to ten-fold over-identification.⁸ A primary-care study across eight practices found a positive predictive value of 0.52 within sample, falling to 0.12 when adjusted to population prevalence.⁹

Specificity collapses in psychiatric populations. 21.4% positive predictive value in treatment-seeking depression;⁴ 0.26 in 1,138 treatment-seeking substance-use patients even at a 13% base rate.¹⁰

Scores are not equivalent across sex and age. In 8,580 British adults, all six screener items showed age measurement noninvariance and a subset showed gender noninvariance, with males and younger respondents endorsing items at lower thresholds.¹¹ This means identical scores do not carry identical meaning across groups. It does not demonstrate that the ASRS under-detects women — no study we could find shows that — and the documented under-recognition of ADHD in women sits upstream of the questionnaire, in referral and clinical judgement.³

It has no validity or effort scale and no way to detect inconsistent or exaggerated responding.⁹

DSM-5 rescoring of these items has been formally withdrawn. In a memo dated 8 December 2023, Kessler withdrew the 2018 guidance for rescoring the DSM-IV ASRS items to DSM-5, citing a scoring-rule error on two items and over-fitting that substantially over-estimated prevalence, and advised users to return to the original DSM-IV scoring rules.¹² Answers to v1.1 cannot be converted into an ASRS-5 result.

The ASRS-5 update

Ustun and colleagues (2017), JAMA Psychiatry 74(5), 520-527, built a DSM-5-aligned successor from a 29-item pool using a machine-learning selection method, with six differently weighted items, a 0-24 total and a cut-point of 14. Reported general-population sensitivity 91.4%, specificity 96.0%, AUC 0.94; clinical-sample sensitivity 91.9%, specificity 74.0%.¹³ It does not formally supersede v1.1 — both remain distributed — and its six items are not the same six.

What a clinician does with it

Treats a positive Part A as a trigger for a full assessment: developmental history back to before age 12, corroborative or informant history where available, evidence of symptoms across more than one setting, documented functional impairment, mental state examination, and a differential that screens for mood, anxiety, substance use, sleep and learning disorders. Where the symptom count is borderline, a clinician-administered instrument rather than a second self-report.

The UK national guideline states that a diagnosis "should not be made solely on the basis of rating scale or observational data."¹⁴ A 2023 Royal College of Psychiatrists report is blunter still, recording that in the group's clinical experience symptom screeners such as the ASRS have proved "undiscriminating for triage purposes."¹⁵ We include that here rather than omitting it.

Validity tier: 1 — validated and published. Among the best-evidenced screening instruments in adult mental health, and still wrong far more often than it is right when it says yes.

Diagram — E · Mostly bottle caps. A metal detector that never beeped for a bottle cap would miss the rings. When a screen is tuned to catch what matters in a population where the thing is rare, most of the beeps are caps. The beep is real information. It is a reason to dig, not a reason to celebrate.

STRENGTHS LENS

The engine was never the problem.

Every trait on this questionnaire has a second face. Difficulty sustaining attention on the dull thing is the same machinery as absorption so complete that eight hours vanish. Trouble waiting your turn shares wiring with speed — the answer arriving before anyone has finished the question. Restlessness is drive with nowhere to go yet.

Screeners can only ask about the cost side, because that is what they were built to detect. Reading one and concluding that you are a list of deficits is a mistake the instrument invites and does not intend. Whatever your result, the aim is a rudder proportionate to the engine — not a smaller engine.

F. What helps

Four things worth doing with the result, whichever way it lands.

1. Answer for an ordinary six months, not your worst week.

Every item asks about the last six months on purpose. A fortnight of crisis is not the window, and neither is the calmest stretch you have had in years. If you cannot find an ordinary six months, that is itself worth writing down.

2. Take a negative result seriously.

This is the one place these instruments are genuinely strong. Negative predictive values across the adult ADHD screening literature run above 92% and often above 96%. If Part A comes back low, it is reasonably good evidence — and if something is still clearly wrong, that is a reason to look somewhere else rather than to keep retaking this.

3. Collect the childhood evidence before you book anything.

The ASRS cannot ask about age twelve, and an assessment will. School reports, old reports cards, a parent or older sibling who remembers, your own memories of homework and mornings and being told you had potential. Gathering this in advance is the single most useful thing you can do between a positive screen and an appointment.

4. Rule the other things in or out, not just ADHD.

Sleep debt, depression, anxiety, thyroid function and long-running stress all generate a positive on this questionnaire. That is not a reason to dismiss your result — it is a reason for a proper assessment rather than a self-diagnosis, because several of those are treatable in their own right and some of them travel alongside ADHD rather than instead of it. Module 4 of this course is about exactly that overlap.

5. Write down what it costs you, in specifics.

An assessor will ask about impairment, and "I am disorganised" is much weaker evidence than "I have paid late fees on the same bill nine times this year and I have missed two of my own deadlines this quarter." Spend ten minutes on the concrete version before any appointment. The workbook below is built to collect exactly that.

Step 2 — Take the screener

Eighteen questions, five to seven minutes, free and confidential. The first six carry the screen; the other twelve give you more to talk about. Read section C first if you can — the scoring threshold is deliberately different on different questions, and knowing that in advance stops the result feeling arbitrary.

Take the ASRS v1.1

Before you start

The ASRS v1.1 (Kessler et al., 2005) is a validated screening questionnaire developed with the World Health Organization. It reliably flags symptom patterns worth investigating, but it is a screen, not a diagnosis — only a clinician can diagnose. It maps to DSM-IV criteria, asks nothing about childhood, and over-identifies substantially in general populations, so a positive result means look further rather than case closed. Treat your score as a mirror for noticing patterns and a starting point for that conversation.

Step 3 — Your workbook

Your answers save to this device only — we cannot see a word of what you write. This module records your Part A result, then does the work the questionnaire cannot: the childhood evidence, the other explanations, and where it actually costs you.

Your Part A result

The ASRS is not currently set up with a score widget in our screener library, so record it yourself here. Four or more positives across the six Part A questions is a positive screen — remember items 1 to 3 count at 'sometimes' and items 4 to 6 at 'often'.

Fields: How many of the six Part A boxes fell in the shaded range?: 0 / 1 / 2 / 3 / 4 / 5 / 6 / I have not taken it yet; Date I took it

Which of the six landed hardest?

Not the total — the specific question that made you stop and reread it.

Fields: The question, and why

The plates you keep spinning

Section B. List the systems you have built that nobody knows about — the alarms, the buffers, the all-nighters.

Fields: What I am doing that looks like nothing from outside

Childhood evidence

Section F, item 3. The ASRS cannot ask about age twelve. An assessment will. Start the list now.

Fields: Tick: School reports or report cards I could find; Tick: A parent, older sibling or relative who remembers; Tick: My own specific memories of homework, mornings, or being told I had potential; Tick: Anything a teacher wrote or said that stuck; What I remember, in whatever order it comes

What else could explain this?

Section F, item 4. Being honest here makes an assessment more useful, not less.

Fields: Tick: I have been sleeping badly for a long time; Tick: I am low or anxious most days; Tick: Something significant happened that I have not fully dealt with; Tick: There is a physical health thing that has not been looked at; Notes

Where it actually costs you

Impairment is one of the things a screener cannot ask about and an assessment must establish.

Fields: Work or study; Home, money, admin; Relationships

The engine side

Strengths Lens. The questionnaire can only ask about cost. Write down the other half yourself.

Fields: What the same wiring does for me when it is pointed at the right thing

Your one next step

One. Not a plan.

Fields: What I am going to do before this time next week

Appendix — Research companion

Peer-reviewed research

7. Brevik EJ, Lundervold AJ, Haavik J, Posserud MB (2020). Validity and accuracy of the Adult Attention-Deficit/Hyperactivity Disorder (ADHD) Self-Report Scale (ASRS) and the Wender Utah Rating Scale (WURS) symptom checklists in discriminating between adults with and without ADHD. Brain and Behavior, 10(6), e01605. DOI 10.1002/brb3.1605. View the paper 646 clinically diagnosed adults versus 908 population controls in Norway. Full 18-item ASRS AUC 0.904; 6-item screener AUC 0.903. Limitation: a known-groups design comparing diagnosed patients with healthy controls overstates real-world discrimination, since the hard clinical task is separating ADHD from other psychiatric presentations.

8. Chamberlain SR, Cortese S, Grant JE (2021). Screening for adult ADHD using brief rating tools: what can we conclude from a positive screen? Some caveats. Comprehensive Psychiatry, 106, 152224. DOI 10.1016/j.comppsych.2021.152224. View the paper Two independent general-population samples (UK n = 642, USA n = 579). The ASRS indicated probable ADHD in 26.0% and 17.3% against an expected prevalence of 2.5%; estimated positive predictive value about 11.5%, implying seven- to ten-fold over-identification. Limitation: no diagnostic interviews were conducted, so the false-positive estimate is inferred from expected prevalence.

2. Chung W, Jiang S-F, Paksarian D, Nikolaidis A, Castellanos FX, Merikangas KR, Milham MP (2019). Trends in the prevalence and incidence of attention-deficit/hyperactivity disorder among adults and children of different racial and ethnic groups. JAMA Network Open, 2(11), e1914344. DOI 10.1001/jamanetworkopen.2019.14344. View the paper Cohort of 5,282,877 adults in a large US health system. Recorded adult ADHD prevalence rose from 0.43% in 2007 to 0.96% in 2016, roughly doubling while remaining an order of magnitude below epidemiological estimates. Limitation: administrative diagnosis data from a single insured health system, so it measures recorded diagnosis rather than true prevalence.

4. Dunlop BW, Wu R, Helms K (2018). Performance of the Adult ADHD Self-Report Scale-v1.1 in adults with major depressive disorder. Behavioral Sciences, 8(4), 37. DOI 10.3390/bs8040037. View the paper 40 treatment-seeking adults with primary major depressive disorder versus 55 healthy controls. ASRS Part A gave sensitivity 60%, specificity 68.6%, positive predictive value 21.4%, negative predictive value 92.3%. Limitation: small sample, but the clearest published demonstration that depressive cognitive symptoms inflate ASRS scores.

1. Fayyad J, Sampson NA, Hwang I, Adamowski T, Aguilar-Gaxiola S, Al-Hamzawi A, Andrade LHSG, et al. (2017). The descriptive epidemiology of DSM-IV adult ADHD in the World Health Organization World Mental Health Surveys. ADHD Attention Deficit and Hyperactivity Disorders, 9(1), 47-65. DOI 10.1007/s12402-016-0208-3. View the paper Twenty nationally or regionally representative surveys, n = 26,744, interview-administered. Current adult ADHD prevalence averaged 2.8%; treatment seeking was low everywhere and directed mainly at comorbid conditions. Limitation: DSM-IV criteria and lay-administered interviews, so not directly comparable to DSM-5 figures.

9. Hines JL, King TS, Curry WJ (2012). The Adult ADHD Self-Report Scale for screening for adult attention deficit-hyperactivity disorder (ADHD). Journal of the American Board of Family Medicine, 25(6), 847-853. DOI 10.3122/jabfm.2012.06.120065. View the paper 200 adults across eight primary care practices. Sensitivity 1.0, specificity 0.71, negative predictive value 1.0, positive predictive value 0.52 within sample, falling to 0.12 adjusted to 4.4% population prevalence. The authors also note the ASRS offers no way to assess inconsistency or malingering. Limitation: modest sample and a single comparator instrument.

3. Hinshaw SP, Nguyen PT, O'Grady SM, Rosenthal EA (2022). Annual Research Review: Attention-deficit/hyperactivity disorder in girls and women: underrepresentation, longitudinal processes, and key directions. Journal of Child Psychology and Psychiatry, 63(4), 484-496. DOI 10.1111/jcpp.13480. View the paper Review of cross-sectional and longitudinal evidence. Girls meet criteria at just under half the rate of boys, a ratio approaching equality by adulthood; clinicians may overlook symptoms in females because of less overt presentations and frequent compensatory strategies. Limitation: a narrative rather than systematic review, addressing clinical recognition generally rather than the ASRS.

5. Kessler RC, Adler L, Ames M, Demler O, Faraone S, Hiripi E, Howes MJ, Jin R, Secnik K, Spencer T, Ustun TB, Walters EE (2005). The World Health Organization Adult ADHD Self-Report Scale (ASRS): a short screening scale for use in the general population. Psychological Medicine, 35(2), 245-256. DOI 10.1017/s0033291704002892. View the paper Original development of the 18-item ASRS and the 6-item Part A screener. Clinical reappraisal sample n = 154, oversampled for reported childhood ADHD. Part A sensitivity 68.7%, specificity 99.5%, accuracy 97.9%, kappa 0.76. Limitation: small oversampled sample; the very high specificity reflects a low base rate and does not transfer to unselected populations; items map to DSM-IV, not DSM-5.

6. Kessler RC, Adler LA, Gruber MJ, Sarawate CA, Spencer T, Van Brunt DL (2007). Validity of the World Health Organization Adult ADHD Self-Report Scale (ASRS) Screener in a representative sample of health plan members. International Journal of Methods in Psychiatric Research, 16(2), 52-65. DOI 10.1002/mpr.208. View the paper Independent validation in 668 health-plan subscribers, each screened three times including once alongside a blinded clinical interview. Internal consistency 0.63-0.72, test-retest 0.58-0.77, AUC 0.90. Limitation: internal consistency is modest for a clinical instrument and the sample was an insured convenience subsample.

13. Ustun B, Adler LA, Rudin C, Faraone SV, Spencer TJ, Berglund P, Gruber MJ, Kessler RC (2017). The World Health Organization Adult Attention-Deficit/Hyperactivity Disorder Self-Report Screening Scale for DSM-5. JAMA Psychiatry, 74(5), 520-527. DOI 10.1001/jamapsychiatry.2017.0298. View the paper The ASRS-5: six new DSM-5-based items selected from a 29-item pool by a machine-learning algorithm, with weighted response options, total 0-24 and cut-point 14. General population sensitivity 91.4%, specificity 96.0%, AUC 0.94; clinical sample sensitivity 91.9%, specificity 74.0%. Limitation: development samples were small and enriched for ADHD, so the reported positive predictive value will not hold in unselected primary care, and its items are not the same six as v1.1.

11. Zhang KA, Markon KE (2021). Age and gender measurement noninvariance of the Adult ADHD Self-Report Scale Screener. Journal of Attention Disorders, 25(3), 403-413. DOI 10.1177/1087054718808059. View the paper n = 8,580 British adults aged 16-95 from the 2007 Adult Psychiatric Morbidity Survey. All six screener items showed age measurement noninvariance and a subset showed gender noninvariance, with males and younger respondents endorsing at lower thresholds; a three-factor structure fit better than a single dimension. Limitation: this establishes that identical scores do not carry identical meaning across groups; it does not show the scale under-detects women.

10. van de Glind G, van den Brink W, Koeter MWJ, Carpentier PJ, van Emmerik-van Oortmerssen K, Kaye S, et al. (2013). Validity of the Adult ADHD Self-Report Scale (ASRS) as a screener for adult ADHD in treatment seeking substance use disorder patients. Drug and Alcohol Dependence, 132(3), 587-596. DOI 10.1016/j.drugalcdep.2013.04.010. View the paper International multi-country study, n = 1,138 treatment-seeking substance use disorder patients, ADHD prevalence 13.0%. Sensitivity 0.84-0.88, specificity 0.56-0.76 by primary substance, positive predictive value 0.26, negative predictive value 0.97. Limitation: even at a 13% base rate roughly three quarters of positives were false, and specificity varied substantially by drug of choice.

Clinical frameworks and position statements

12. Kessler RC (2023). DSM-5 scoring of the DSM-IV ASRS screening questions (memorandum to users, 8 December 2023). Harvard Medical School, Department of Health Care Policy. View the source Memo from the ASRS's lead author formally withdrawing the February 2018 guidance for rescoring the DSM-IV ASRS items to DSM-5, citing a scoring-rule error on two items and over-fitting that substantially over-estimated prevalence; users are advised to return to the original DSM-IV scoring rules. Limitation: an unrefereed institutional memo, but the authoritative statement from the copyright holder.

14. National Institute for Health and Care Excellence (NICE) (2019). Attention deficit hyperactivity disorder: diagnosis and management (NICE guideline NG87). NICE, London (published March 2018, last updated September 2019). View the source UK national clinical guideline. Requires diagnosis by a specialist based on a full clinical and psychosocial assessment across settings, a full developmental and psychiatric history, and observer reports; recommendation 1.3.2 states a diagnosis should not be made solely on the basis of rating scale or observational data. Limitation: UK-specific and does not name the ASRS.

15. Royal College of Psychiatrists in Scotland, Special Interest Group in Adult ADHD (2023). Attention deficit hyperactivity disorder (ADHD) in adults: good practice guidance (College Report CR235). Royal College of Psychiatrists, London. View the source Professional consensus guidance. Records that symptom screeners such as the ASRS have been found undiscriminating for triage purposes in the group's clinical experience, and that rating scales support but never substitute for careful history taking and clinical evaluation. Limitation: expert consensus rather than a systematic evidence review, and UK-specific.

Up next

Module 3 — SAAST — A Longer, Structured Look at Adult ADHD

All modules in ADHD Self-Discovery

A positive screen is a good reason to talk to someone. This course was built by clinicians who are part of the New Path family of therapy centers. Working out whether this is ADHD, something alongside it, or something else entirely is genuinely hard to do alone — and it is the kind of question a first conversation is good at. Therapy for clients in California, coaching worldwide, all by telehealth. A conversation costs nothing and there is no pressure. Saving this for later counts too. Talk with the New Path team

NEW PATH FAMILY