Show the full module text

Module 10 — CAS — Clinical Anger You Can Actually Look At

What the Clinical Anger Scale asks about, why its format and its cut-offs were borrowed from a depression inventory, how emotional dysregulation sits at the centre of adult ADHD, and the honest list of everything else that raises an anger score.

ADHD Self-Discovery · Part Four — Traveling Companions · 14 min read, about 28 min with the workbook

The big idea

If you have ever heard your own voice go up and known, while it was still happening, that you were going to regret this — this module is about that, and about a twenty-one question form that tries to put a number on it. Anger is information. It is also the thing people are least willing to say out loud, which is exactly why it goes unexamined for years.

The mechanism most often described is not that the feeling is wrong. It is that it arrives at full volume with no run-up. In focus groups with 43 young adults with ADHD diagnoses, participants described emotions going from zero to one hundred without the gradations other people seem to get.⁹ If you have ever been asked why you did not just say you were annoyed earlier, that is the answer: there was no earlier.

Step 1 — The lesson

A. Twenty-one questions in a borrowed coat

The Clinical Anger Scale was published in 1995 by William Snell and four colleagues at Southeast Missouri State University, in the Journal of Clinical Psychology. Twenty-one clusters of statements, four options in each, arranged from mild to severe. You pick the one that fits. The options score 0, 1, 2 and 3, and the total runs from 0 to 63.¹

If that shape sounds familiar, it should. It is the Beck Depression Inventory, item for item, structurally. Beck and colleagues published that in 1961: twenty-one questions, each a short set of graded statements, scored 0 to 3, total 0 to 63.² The Clinical Anger Scale keeps the architecture and swaps the content.

Diagram — A · Twenty-one questions in a borrowed coat. One development paper, published in 1995. The fuller psychometric account appeared seven years later as a chapter in a book the scale's own author edited and published. Everything after that is people using the instrument, not people testing it.

Look at what the twenty-one items actually ask about. Anger about the future. Anger about past failure. Anger turned on yourself. Yelling at people. Urges to harm someone. And then, from about halfway down: whether anger has changed your interest in other people, your ability to decide things, your relationships, your work, your sleep, your energy, your appetite, your interest in sex.³ That second list is the Beck inventory's functional and bodily items with the word anger put in front of them.

This borrowing is not a scandal. It produces a form that asks about consequences rather than about temperament, which is more useful than most anger questionnaires manage. But it has two effects that run through the rest of this module. The CAS will register anything that wrecks your sleep, your appetite and your interest in people, whether or not the cause is anger. And its severity cut-offs came from somewhere other than anger research, which section D gets to.

One more thing worth having early. The people who host the scale for research use state plainly that it is not meant as a formal diagnosis of clinical anger, and note that in student samples the average score sits around ten points.⁴ Ten. That is your baseline for reading anything that comes back.

B. What the word clinical is doing there

Everyone gets angry. That is not what this measures, and the distinction is worth spelling out because most people arriving at an anger questionnaire have already decided the answer is going to be bad.

Ordinary anger has a cause you can name, a peak, and a descent. Something happened, you were furious about it, and by the evening it had turned into a story you could tell. It cost you an hour, not the week. Being angry is not a disorder, is not in any diagnostic manual as such, and has a long history of being the correct response to what was going on.

Diagram — B · What the word clinical is doing there. The right-hand column is what raises a CAS total. Note that none of it is about volume. A person who never raises their voice can score high, and a person who shouts twice a year can score low.

What the CAS is trying to catch is anger that has stopped being an event and become a climate. Present most days rather than on Tuesday. Turning up in places that have nothing to do with the original grievance. Colouring how you read a neutral email, a slow driver, a partner asking a fair question. Following you to bed. That is the syndrome the 1995 paper set out to measure — the authors described it as an objective self-report instrument for the syndrome of clinical anger.¹

It is also worth separating anger from irritability, because the two get used as though they were the same word and they are not. Anger is a state: it has an object, an intensity and an arc. Irritability is closer to a threshold — a lowered bar for getting angry at all, a proneness rather than an episode. A 2017 review in Clinical Psychology Review made the case that the field had been conflating irritability with anger, with aggression and with several neighbouring constructs, and that measures had inherited the confusion.⁵ That matters here because a person can be highly irritable and rarely angry, or rarely irritable and carrying an enormous amount of stored anger, and the CAS total does not distinguish them.

So the honest reading of a high score is not "you have an anger problem". It is "anger is currently touching a lot of your life, and something is keeping it there". The second half of that sentence is where the useful work is.

C. Why anger is in an ADHD course at all

Because emotion is not a side effect of ADHD. It is increasingly argued to be part of the thing itself.

A 2023 systematic review in PLOS ONE gathered 22 studies and made the case for emotion dysregulation as a core symptom of adult ADHD rather than an incidental extra. Adults with ADHD consistently scored lower on emotion regulation than controls and leaned on less adaptive strategies; across the studies reviewed, emotion dysregulation was estimated to affect somewhere between 34 and 70 per cent of adults with the diagnosis, and it was associated with symptom severity, executive function, psychiatric comorbidity and, at the far end, criminal conviction.⁶

Diagram — C · Why anger is in an ADHD course at all. The spread from 34 to 70 is not sloppiness. It is what happens when a dozen research groups measure the same thing with a dozen different instruments and no agreed definition. Every estimate in that range says the same thing: this is common, not rare.

The effect sizes are large by the standards of adult ADHD research. A meta-analysis of 13 studies covering 2,535 adults found a Hedges' g of 1.17 for emotion dysregulation overall and 1.20 for emotional lability, with symptom severity correlating with dysregulation at r = 0.54.⁷ A 2014 review in the American Journal of Psychiatry set out three competing accounts — that dysregulation and ADHD are separate but travel together, that dysregulation is a core feature of ADHD, or that the combination is its own category — and proposed a neural route running between striatum, amygdala and medial prefrontal cortex.⁸ The field has moved steadily towards the second.

Module 8 covered the version of this that points inwards — rejection sensitivity, the flood that follows a short reply or a piece of feedback. Anger is frequently the same wave pointing outwards. The episode that leaves one person silently rehearsing an apology leaves another snapping at whoever is nearest, and plenty of adults do both within the same hour. If your rejection sensitivity score was high, do not be surprised by this one.

There is also the plain fact of a life. Adults with ADHD accumulate a great deal of genuine provocation: the lost decade before diagnosis, the jobs that went badly for reasons nobody could name, being called lazy by people who were wrong. A questionnaire cannot tell the difference between anger that is being generated and anger that is being earned.

D. What your number does and does not mean

Now the honest part, which is mostly about where the lines on this instrument came from.

Diagram — D · What your number does and does not mean. Four bands, reproduced exactly as the live screener states them. The lowest boundary agrees with the version of the scale that circulates elsewhere. The other three do not, and no source we could find explains why either set sits where it sits.

The live screener returns four bands: 0–13 Minimal or No Clinical Anger, 14–23 Mild Clinical Anger, 24–35 Moderate Clinical Anger, 36–63 Severe Clinical Anger. The version of the scale that circulates in clinics and template libraries uses different boundaries: 0–13 minimal, 14–19 mild, 20–28 moderate, 29–63 severe.³ Those two keys disagree about a great deal of the range. A score of 22 is mild on one and moderate on the other. A score of 31 is moderate on one and severe on the other.

Here is the part worth sitting with. That second key — 0–13, 14–19, 20–28, 29–63 — is character for character the severity key of the Beck Depression Inventory-II. Minimal, mild, moderate, severe, at exactly those numbers.² The thresholds that travel with the Clinical Anger Scale appear to have been carried over from a depression inventory along with the format, rather than derived from anger data. The 1995 paper's published abstract describes a confirmed unidimensional structure, adequate internal consistency and test-retest stability, no contamination by social desirability, and sensible correlations with trait and state anger.¹ It announces no cut-off, and the research hosts of the scale say no formal cut-off scores are provided.⁴

What does exist is a small amount of real-world context. In 2023 a study of 400 undergraduate medical and sociology students in Pakistan gave the CAS alongside measures of self-esteem and adverse childhood experiences. The mean CAS score was 19.65 with a standard deviation of 13.23, and 60.2 per cent scored in the mild-to-severe range — 64 per cent of women against 54 per cent of men.¹⁰ Read that next to the average of around ten reported in student samples generally.⁴ Scores in the teens and twenties are extremely ordinary, and the proportion of people a band labels as having clinical anger depends enormously on which key you use and who you ask.

None of which means a high score is nothing. In a 2021 study of 150 adults attending an artificial limb centre in Rawalpindi, clinical anger had a significant positive relationship with suicidal ideation, and the highest mean anger score in the sample belonged to the group with spinal cord injuries, at 26.10.¹¹ Anger that has gone this far is often standing next to something else, and the something else is what needs attention.

Which brings us to the list this section exists for. A high anger score does not tell you why. Depression is the obvious candidate and the most missed: in a long-term study of 536 people with unipolar depression, 292 of them — 54.5 per cent — showed overt irritability or anger during the current episode, and that group had a more chronic course and more comorbidity than the rest.¹² Sleep debt is causal, not merely correlated: when 142 adults were randomly assigned either to keep or to restrict their sleep across two days, restricted sleep intensified anger in everybody and reversed the normal pattern of getting used to an irritating noise.¹³ Trauma is well established: a meta-analysis of anger, hostility and PTSD in trauma-exposed adults found the association strengthened with time since the trauma and was stronger after military combat.¹⁴ Chronic pain, alcohol, an overactive thyroid, grief, some medications, and a situation that is genuinely unjust all belong on the same list.

So: your number is a reading of something real. It is not a diagnosis, it is not a measurement of your character, and it does not identify its own cause. The instrument was published once, thirty-one years ago, and has never been independently re-examined. Treat the number as the start of a differential, not the end of one.

INSIDE THE INSTRUMENT

CAS — the Clinical Anger Scale

Where it came from

Snell, Gum, Shuck, Mosley and Hite, Journal of Clinical Psychology, 1995, 51(2), 215–226, from Southeast Missouri State University. The paper is titled as preliminary and reports the development and initial validation of a self-report instrument for what the authors call the syndrome of clinical anger.¹ The fuller psychometric account is a 2002 book chapter published through the first author's own imprint rather than through a peer-reviewed journal.⁴ There is no commercial manual and no test publisher.

What it is made of

Twenty-one item clusters, four graded statements each, arranged from least to most severe; the circulating version scores the options A to D as 0 to 3, giving a total of 0 to 63.³ The live page states the range but not the item scale, so nothing about the anchors is asserted here beyond what the arithmetic requires. The 1995 factor analysis supported a unidimensional structure, so there are no subscales and no profile — one number, and that is all.¹ The item content maps closely onto the Beck Depression Inventory: pessimism, past failure, self-directed anger, social interest, indecision, work, sleep, fatigue, appetite, health preoccupation and sexual interest, each reframed around anger.²

How well it performs

The published abstract reports a confirmed unidimensional item structure, adequate internal consistency and test-retest stability, independence from social desirability, and positive associations with trait anger, state anger, anger-in, anger-out and anger-control, plus predictable relationships with psychopathological symptoms, personality traits and early family environment.¹ The coefficients themselves are not in the abstract, and the full text could not be obtained from any open source at the time of writing, so no reliability or validity figure for this instrument is quoted anywhere in this module. That absence is the finding, not an oversight.

Where it was validated

In one paper, by the group that wrote it, in a United States university sample, in 1995. A search of PubMed, Crossref, Europe PMC and DOAJ in August 2026 located no independent psychometric evaluation of the CAS in any language, and no normative sample. The studies that use it are applications rather than evaluations: 400 undergraduates in Pakistan, mean 19.65 (SD 13.23), 60.2% at mild or above,¹⁰ and 150 adults with physical disabilities in Rawalpindi, where anger correlated with suicidal ideation.¹¹ The only population reference point in circulation is an average of around ten points in student samples.⁴

What it cannot do

Its severity thresholds have no visible empirical derivation. The key that circulates with the instrument — 0–13 minimal, 14–19 mild, 20–28 moderate, 29–63 severe³ — is identical to the Beck Depression Inventory-II severity key.² The live screener uses 0–13, 14–23, 24–35 and 36–63 instead. Neither set has a published derivation we could locate, and the 1995 abstract announces no cut-off at all. Both should be read as descriptive headings, not thresholds.

It cannot separate anger from depression, and its own architecture is the reason. Roughly half the items ask about sleep, appetite, fatigue, indecision, libido, health preoccupation and loss of interest in people — the somatic and functional core of a depression inventory.² A depressed patient who attributes those changes to anger will score in the same range as an angry one. Given that 54.5% of 536 patients with unipolar depression showed overt irritability or anger during the current episode,¹² this overlap is not hypothetical.

An elevated score carries no diagnostic specificity whatever. Anger rises with depression,¹² with experimentally restricted sleep,¹³ with post-traumatic stress,¹⁴ and with chronic pain, alcohol use, thyroid disease, grief and circumstances that warrant it. Nothing about a high CAS total points at ADHD rather than at anything else, and it should never be entered into a diagnostic formulation as evidence for or against a referral.

It does not assess risk, and it is not a substitute for asking. Two items touch on urges to harm others and on shouting at people, but a summed self-report score cannot estimate the probability of harm to anyone. The finding that clinical anger tracked suicidal ideation in a disability sample¹¹ argues for direct enquiry about self-harm and about safety at home, not for reading either off a total.

The name is not unique, which is a live source of confusion. An unrelated instrument, the Cluain Mhuire Clinical Anger Scale, was developed in Ireland and published in 2011: twelve items, tested on 112 psychiatric day hospital clients and 76 non-clinical comparisons, with an alpha of 0.82, test-retest r = 0.93 in a subsample of 12, and convergence with the STAXI-2.¹⁵ A search for "clinical anger scale" returns both. Check which instrument a paper or a colleague means.

What a clinician does with it

Use the total to open a conversation about function rather than to grade a patient. The useful material is the item pattern — whether the cost sits in relationships, in work, in sleep, or in anger turned inwards — together with frequency, duration, trigger specificity and what has happened to other people as a result. Before attributing a raised score to ADHD, work the differential: mood, sleep, substances, pain, thyroid function, medication timing and trauma history. Where ADHD is established, assess emotional dysregulation with instruments that have psychometrics behind them, since the construct is well evidenced even where this measure is not.⁶⁷

Validity tier: 3 — peer-reviewed once, in 1995, and never independently re-examined. The brief for this module proposed tier 1; the evidence does not support it, because no independent laboratory has calibrated the instrument, no normative sample exists, the published coefficients are not retrievable from any open source, and the severity key that travels with the scale is a depression inventory's key rather than an anger one.

Diagram — E · One reading, many roads. This is the most important caveat in the module and the reason section D is as long as it is. A raised anger score is a true statement about your life right now. It is not an explanation, and the explanation is what changes anything.

STRENGTHS LENS

Anger is the emotion that keeps count.

Anger is a detector for cost. It fires when something has been taken, when effort has gone unrecognised, when a rule was applied to you and not to the person beside you. Adults with ADHD have usually had a great deal to detect: the years of being told to try harder while trying harder than anyone knew, the appraisals that described a personality instead of a difficulty, the friendships that quietly cooled. A nervous system that registered all of that and filed it under this was not fair was working correctly. The problem was never the signal.

It is also the emotion most likely to move something. People who feel only sadness about an unfair situation tend to stay in it; anger is what gets a case reopened, a boundary set, a job left, a diagnosis pursued after the fourth dismissal. This module is honest that anger can do real damage to people who did not deserve it, and that damage is not cancelled by having had cause. But the aim is not to feel less. It is to get the gap back — the few seconds between the signal and what you do about it.

F. What helps

Anger responds well to practical work, and unusually quickly compared with most of what this course has covered. Almost none of what follows is about trying to be a calmer person. It is about the conditions that set the threshold, and the few seconds after it is crossed.

1. Rule out the body before you rule on the character.

Sleep restriction alone raised anger in a randomised experiment and reversed people's ability to adapt to an irritating noise. Before you conclude anything about who you are, look hard at sleep, alcohol, pain, blood sugar, thyroid function, when your medication wears off, and what time of day the incidents happen. A surprising number of anger problems are a four-thirty in the afternoon problem.

2. Find the number where you stop being able to choose.

Rate the intensity 0 to 10 as it climbs, for a fortnight, and mark the point at which you lose the ability to decide what happens next. Most people find it is lower and earlier than they assumed, and that it is a consistent number. Everything useful happens below it. Once you know it, you have something to act on rather than a vague instruction to calm down.

3. Buy the gap with a rule you set in advance.

The gap between signal and action is the whole game, and it cannot be created in the moment. Decide now what you will do at your number: leave the room, put the phone face down, say "I need ten minutes" and go. Tell the people you live with what the sentence is going to be, so that walking out reads as a plan rather than as an escalation.

4. Repair specifically, and without a diagnosis attached.

Where it has landed on somebody, name the thing you did, not the mechanism behind it. "I shouted at you in front of the children and that was frightening" repairs something. "My ADHD makes my emotions go from nought to a hundred" explains something and repairs nothing, because the other person was not asking for a mechanism. Explanation is for you. Repair is for them.

5. Take the differential to an appointment, and take the whole course with you.

Bring the pattern, not the number: what sets it off, how fast it peaks, how long recovery takes, what it has cost, and what else is true right now — the sleep, the mood, the drinking, the pain, the situation you cannot leave. Ask directly about depression and about emotional dysregulation, which is well evidenced in adult ADHD and can be properly assessed. That is the end of this course, and it is the same request every module has been building towards: not a score to hand over, but an accurate account of your own life, written in your own words, that a clinician can actually work from. You have been the only person with access to that evidence all along. Now it is written down.

Step 2 — Take the screener

Twenty-one questions, five to seven minutes, free and confidential. You get one total between 0 and 63 with a band — there are no subscales, so the number is all the instrument produces. Read section D before you read the band: student samples average around ten points, the boundaries between bands differ from the key used elsewhere for this same scale, and neither set of boundaries has a published derivation.

Take the CAS

Before you start

The Clinical Anger Scale was published in 1995 and has, as far as any index shows, never been independently re-examined: there is no normative sample, no published cut-off in the original paper, and no second psychometric study in PubMed, Crossref or DOAJ. Its 21-item format and its widely circulated severity thresholds were taken from the Beck Depression Inventory, which means it will register anything that disturbs sleep, appetite, concentration and interest in other people, whether or not anger is the cause. Anger scores also rise with depression, sleep loss, chronic pain, alcohol, trauma, thyroid disease and circumstances that genuinely warrant anger, so nothing here indicates that you do or do not have ADHD. This is a structured way to describe what anger is currently costing you, not a test and not a diagnosis. Only a clinician working from a full history can diagnose.

Step 3 — Your workbook

Your answers save to this device only — we cannot see a word of what you write. This module records your total, then does the work the instrument cannot: where the anger actually shows up, what else could be causing it, the intensity at which you stop being able to choose, and the rule and the repair sentence that make the practical difference.

Your CAS results

Took the screener? Put the number in below. Read it against the context in section D before you read it against the band label — averages of around ten in student samples, and a mean of 19.65 in a study of 400 undergraduates, are the reference points that make a total mean anything. Entirely optional — skip it if you would rather just read.

Score bands: 0–13 = Minimal or No Clinical Anger; 14–23 = Mild Clinical Anger; 24–35 = Moderate Clinical Anger; 36–63 = Severe Clinical Anger

Fields: CAS · Clinical Anger Scale, 21 items; Total score (0–63); My total (enter 0–63)

Where it actually shows up

Section B. The scale is one number with no subscales, so this is the part it cannot give you. Roughly half its items ask about ordinary life rather than about rage — which of those has been affected?

Fields: Tick: Sleep; Tick: Work or study; Tick: A specific relationship; Tick: How I read other people; Tick: Anger turned on myself; What that looks like on an ordinary week

The differential, honestly

Section D and section F, item 1. A high score does not name its own cause. Go through the list before you conclude anything about yourself.

Fields: How my sleep has been for the last month: Fine / Short but steady / Broken / Badly short / I have no idea; Mood, separately from anger: Roughly normal / Flat / Low most days / I have wondered about depression; Pain, alcohol, medication timing, thyroid, grief, or anything physical going on; The situation itself: is there something here that would make anyone angry

Your number, the one where choice stops

Section F, item 2. Rate the intensity 0 to 10 as it climbs, for a fortnight, then come back and fill this in.

Fields: The point at which I can no longer choose what happens next (0–10); What I can notice in my body one or two points below it; How long from first signal to that number: Seconds / Under a minute / A few minutes / Builds over hours / It varies enormously

The rule you set in advance

Section F, item 3. The gap cannot be invented in the moment. Write the sentence now, while nothing is happening.

Fields: What I will do at my number; Who I need to tell about this, so that leaving reads as a plan; Tick: I have actually told them

Repair, without the mechanism

Section F, item 4. Name the thing you did, not the reason it happened. Explanation is for you; repair is for them.

Fields: Who it landed on; The sentence, with no diagnosis in it; Tick: Said

What I am taking to an appointment

Section F, item 5, and the end of the course. Not a score to hand over — an account of your own life that a clinician can work from.

Fields: The pattern in five lines: trigger, speed, peak, recovery, cost; The one question I want to ask about depression or emotional dysregulation; Looking back across the whole course: the three things I most want understood

Appendix — Research companion

Peer-reviewed research

11. Andleeb N, Ahsan S, Zaheer S, Kiani FS (2021). Relationship between clinical anger and suicidal ideation among people with physical disabilities. Journal of the Pakistan Medical Association, 71(10), 2442-2444. DOI 10.47391/JPMA.12-1470. View the paper Purposive convenience study of 150 physically disabled adults, 87 men and 63 women, attending the Artificial Limbs Centre at Fauji Foundation Hospital, Rawalpindi, between February and November 2016, using the Clinical Anger Scale and a modified scale for suicidal ideation. Suicidal ideation had a significant positive relationship with clinical anger, women scored significantly higher than men on both, and the highest mean anger score belonged to participants with spinal cord problems at 26.10 (SD 9.25), while suicidal ideation was highest in those with polio. Limitation: a small single-site convenience sample in one country with no comparison group and no clinical interview, and it is cross-sectional, so it establishes that anger and suicidal ideation travel together in this setting rather than that either causes the other.

10. Ansar F, Naveed H, Khattak A, Khan SA (2023). Frequency of anger and its potential relationship with self-esteem and adverse childhood experiences among medical and sociology undergraduate students in Pakistan. Pakistan Journal of Medical Sciences, 39(2), 524-528. DOI 10.12669/pjms.39.2.6113. View the paper Cross-sectional study of 400 undergraduate medical and sociology students in Pakistan using the Clinical Anger Scale alongside the Adverse Childhood Experience scale and the Rosenberg Self-Esteem Scale. The mean CAS score was 19.65 (SD 13.23) and 60.2 per cent of the sample scored in the mild-to-severe range, 64 per cent of women against 54 per cent of men; 72.8 per cent showed low self-esteem and 51.5 per cent reported at least one adverse childhood experience, with anger correlating positively with both. It is one of the few studies to publish a CAS mean and distribution in any sample. Limitation: a single-country convenience sample of undergraduates, cross-sectional and self-report throughout, using the CAS as an application rather than evaluating it, so the mean is a reference point for reading a score and not a norm for any other population.

2. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J (1961). An inventory for measuring depression. Archives of General Psychiatry, 4(6), 561-571. DOI 10.1001/archpsyc.1961.01710120031004. View the paper The originating paper for the Beck Depression Inventory, the instrument whose architecture the Clinical Anger Scale reproduces: twenty-one items, each offering a graded set of at least four statements, each scored 0 to 3, producing a total from 0 to 63. Included here because the structural parallel is exact and because the severity key that circulates with the Clinical Anger Scale - 0-13 minimal, 14-19 mild, 20-28 moderate, 29-63 severe - is character for character the published severity key of the BDI-II rather than anything derived from anger data. Limitation: this is a 1961 depression measure with no connection to anger research, so it establishes the format and the origin of the thresholds but says nothing about whether either transfers validly to a different construct.

7. Beheshti A, Chavanon ML, Christiansen H (2020). Emotion dysregulation in adults with attention deficit hyperactivity disorder: a meta-analysis. BMC Psychiatry, 20(1), 120. DOI 10.1186/s12888-020-2442-7. View the paper Meta-analysis of 13 studies covering 2,535 participants comparing emotion dysregulation in adults with ADHD against healthy controls. Between-group effects were large: Hedges' g = 1.17 for emotion dysregulation overall, 1.20 for emotional lability, 1.12 for negative emotional responses and 0.63 for emotion recognition, all p < 0.001, with ADHD symptom severity correlating with general emotion dysregulation at r = 0.54 within groups. These are among the larger effects reported anywhere in adult ADHD research. Limitation: the authors report that insufficient statistical reporting in the included studies prevented meta-regression, so age, sex, medication status and comorbidity could not be tested as moderators, and several confidence intervals are wide.

12. Judd LL, Schettler PJ, Coryell W, Akiskal HS, Fiedorowicz JG (2013). Overt irritability/anger in unipolar major depressive episodes: past and current characteristics and implications for long-term course. JAMA Psychiatry, 70(11), 1171-1180. DOI 10.1001/jamapsychiatry.2013.1957. View the paper Prospective naturalistic follow-up of 536 participants with unipolar major depression, observed for up to 31 years. Overt irritability or anger was present during the current major depressive episode in 292 of the 536, or 54.5 per cent, and that group had longer index episodes, a significantly more chronic prospective course, greater comorbid anxiety and substance use, poorer impulse control, more personality disorder, lower life satisfaction and more bipolar II disorder among relatives, despite having no greater number of episodes. Limitation: an observational cohort recruited decades ago at academic centres, so the sample is not representative of primary care depression, and the paywalled full text could not be opened, with the 54.5 per cent figure confirmed from published reporting of the paper rather than read directly from the article.

13. Krizan Z, Hisler G (2019). Sleepy anger: restricted sleep amplifies angry feelings. Journal of Experimental Psychology: General, 148(7), 1239-1250. DOI 10.1037/xge0000522. View the paper Randomised experiment in which 142 community residents were assigned either to maintain or to restrict their sleep over two days, then rated anger and affect during a product-rating task carried out alongside aversive noise. Sleep restriction universally intensified anger and reversed the normal adaptation pattern in which anger diminishes with repeated exposure to the noise, with subjective sleepiness mediating most of the experimental effects. This is the strongest available evidence that sleep loss plays a causal role in anger rather than merely accompanying it. Limitation: a two-day manipulation in community volunteers using a laboratory provocation, so it demonstrates causation over a short horizon in a controlled setting and does not establish the size of the effect in chronic real-world sleep debt.

15. Merriman B, Guerin S, Horgan F, Maguire B (2011). Developing a brief tool for anger assessment in clinical settings. Irish Journal of Psychological Medicine, 28(1), 13-20. DOI 10.1017/S079096670001185X. View the paper Development study for the Cluain Mhuire Clinical Anger Scale, a separate Irish instrument that shares three words of its name with Snell's scale. An initial 16-item version was completed by 112 psychiatric day hospital clients and 76 non-clinical comparison participants; the final 12-item scale showed internal reliability of alpha = 0.82 and test-retest reliability of r = 0.93 in a subsample of 12, converged with the STAXI-2, and distinguished anger management referrals from both clinical and non-clinical groups. Included because a literature search for the phrase clinical anger scale returns both instruments and they are not interchangeable. Limitation: a single-service Irish development sample with a test-retest subsample of only 12 people, and it has not itself been independently replicated.

14. Orth U, Wieland E (2006). Anger, hostility, and posttraumatic stress disorder in trauma-exposed adults: a meta-analysis. Journal of Consulting and Clinical Psychology, 74(4), 698-706. DOI 10.1037/0022-006X.74.4.698. View the paper The standard meta-analytic synthesis of the relationship between post-traumatic stress disorder and anger or hostility in trauma-exposed adults. The association was found to be stronger the longer the interval since the trauma, and stronger where the traumatic event was military combat than where it was not, which is the basis for treating trauma as a serious candidate explanation for a raised anger score long after the events themselves. Limitation: the abstract and full text are paywalled and could not be retrieved from any open source, so the number of studies and the pooled effect sizes are not quoted anywhere in this module, and as a synthesis of cross-sectional research it describes association rather than direction of effect.

8. Shaw P, Stringaris A, Nigg J, Leibenluft E (2014). Emotion dysregulation in attention deficit hyperactivity disorder. American Journal of Psychiatry, 171(3), 276-293. DOI 10.1176/appi.ajp.2013.13070966. View the paper A review of emotion dysregulation in ADHD across the lifespan and of the impairment it produces. It proposes a neurobiological account implicating a striato-amygdalo-medial prefrontal cortical network in orienting towards, recognising and allocating attention to emotional material, and sets out three competing conceptual models: that emotion dysregulation and ADHD are separate but correlated, that dysregulation is a core feature of ADHD, or that the combination constitutes a distinct category. It also notes that established ADHD treatments frequently improve emotion dysregulation. Limitation: a narrative review rather than a meta-analysis, pooling studies that define and measure dysregulation differently, and it explicitly declines to adjudicate between the three models it describes.

1. Snell WE, Gum S, Shuck RL, Mosley JA, Hite TL (1995). The Clinical Anger Scale: preliminary reliability and validity. Journal of Clinical Psychology, 51(2), 215-226. DOI 10.1002/1097-4679(199503)51:2<215::AID-JCLP2270510211>3.0.CO;2-Z. View the paper The originating and, as far as any index shows, only development paper for the Clinical Anger Scale, written by five researchers at Southeast Missouri State University and titled by its authors as preliminary. The published abstract reports that factor analysis confirmed a unidimensional item structure, that reliability analyses demonstrated adequate internal consistency and test-retest stability, that the scale was unrelated to social desirability, and that clinical anger correlated positively with trait anger, state anger, anger-in, anger-out and anger-control as well as relating in predictable ways to psychopathological symptoms, personality traits and early family environments. Limitation: the abstract states these findings qualitatively and reports no coefficients, no sample sizes, no norms and no cut-off scores, and the full text could not be retrieved from any open source in August 2026, so not one numeric psychometric value for this instrument is quoted anywhere in this module.

6. Soler-Gutierrez AM, Perez-Gonzalez JC, Mayas J (2023). Evidence of emotion dysregulation as a core symptom of adult ADHD: a systematic review. PLOS ONE, 18(1), e0280131. DOI 10.1371/journal.pone.0280131. View the paper Systematic review of 22 studies examining emotion regulation in adults with ADHD against comparison groups. Adults with ADHD consistently showed lower emotion regulation scores and greater use of maladaptive strategies, and the review reports emotion dysregulation as affecting approximately 34 to 70 per cent of adults with ADHD, with associations to symptom severity, executive functioning, psychiatric comorbidity and criminal conviction; the authors argue for treating emotion dysregulation as a core rather than incidental feature of the adult presentation. Limitation: the included studies were heterogeneous in sample size, ADHD subtype, comorbidity and medication status, used no standardised definition of emotion regulation, and were too inconsistent to pool meta-analytically, which is precisely why the prevalence estimate spans a range as wide as 34 to 70 per cent.

5. Toohey MJ, DiGiuseppe R (2017). Defining and measuring irritability: construct clarification and differentiation. Clinical Psychology Review, 53, 93-108. DOI 10.1016/j.cpr.2017.01.009. View the paper A conceptual review in Clinical Psychology Review addressing the confusion between irritability and its neighbouring constructs, principally anger, aggression and hostility, and the way that confusion has been inherited by the instruments built to measure them. It is the standard reference for the distinction this module draws between anger as a state with an object and an arc, and irritability as a lowered threshold for entering that state. Limitation: a narrative conceptual review with no new sample and no meta-analytic pooling, so it clarifies definitions and criticises existing measures rather than establishing any empirical rate or effect, and its full abstract could not be retrieved from an open source, so no numeric claim from it is used here.

Clinical frameworks and position statements

3. This Is The Day Therapy (2023). Clinical Anger Scale (CAS) - scale and scoring key. Clinician handout reproducing the 21-item scale, retrieved 13 August 2026. View the source A clinical practice's reproduction of the full Clinical Anger Scale, used here to establish the scoring and the interpretation key that circulate with the instrument outside our own screener. It states that each cluster of statements is scored A = 0, B = 1, C = 2, D = 3, that the total range is 0 to 63, and that scores are read as 0-13 minimal clinical anger, 14-19 mild, 20-28 moderate and 29-63 severe. The same four ranges appear on independent template and clinical-software sites, so the key is genuinely the one in general circulation. Limitation: this is a redistributed handout rather than a publisher's or author's document, it carries no page-level provenance beyond a 1995 copyright line, and it gives no source for the four severity boundaries it prints.

Lived experience

9. Ginapp CM, Greenberg NR, MacDonald-Gagnon G, Angarita GA, Bold KW, Potenza MN (2023). Dysregulated not deficit: a qualitative study on symptomatology of ADHD in young adults. PLOS ONE, 18(10), e0292721. DOI 10.1371/journal.pone.0292721. View the source Nine focus groups with 43 young adults aged 18 to 35 with ADHD diagnoses, 84 per cent female and 86 per cent from the US and Canada, analysed with interpretive phenomenological analysis. Most participants said the diagnostic criteria failed to capture their experience and described emotional dysregulation as central, with emotions moving from zero to one hundred without gradation, alongside rumination and self-blame after perceived rejection. This is the source for the description of the missing run-up that appears in section C. Limitation: a self-selected, predominantly female and predominantly North American online sample of 43 people, so it describes the texture of the experience richly and says nothing about how common or how severe it is in the wider ADHD population.

Emerging or contested

4. Stoet G (2026). Clinical Anger Scale (CAS) - PsyToolkit survey library entry. PsyToolkit online survey library, retrieved 13 August 2026. View the paper The research-facing documentation for the Clinical Anger Scale in a widely used academic survey library. It records 21 items with four-choice responses scored 0 to 3, a total range of 0 to 63, that the CAS is not meant as a formal diagnosis of clinical anger, that no formal cut-off scores are provided, and that in student samples the average score lies around ten points. It also gives the second source for the instrument's psychometrics as a 2002 chapter by Snell published through Snell Publications, the author's own imprint, rather than in a peer-reviewed journal. Limitation: a library entry rather than a study, with no sample described behind the figure of around ten points, so that number is a useful orientation for reading a score and not a normative statistic.

Up next

All modules in ADHD Self-Discovery

Anger is information, not a verdict. This course was built by clinicians who are part of the New Path family of therapy centers. Anger is the subject people put off raising longest, and it is one of the most workable things we see — usually because the differential underneath it has never been done properly. 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