Module 17 — OCI-R — OCD Symptoms: Checking, Ordering and Intrusive Thoughts

 

What the eighteen questions of the OCI-R actually ask, how an autistic routine and an OCD compulsion can look identical from outside while running on opposite fuel, how often the two conditions travel together, and why this page draws its line at 12 rather than 21.

 

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

 

The big idea

 

If you have ever gone back to check the door a fourth time and still not felt that it was checked, or needed the tins facing the same way before you could sit down — this module is about a questionnaire that asks about both of those, and about the one thing it cannot ask. Two people can do the identical thing for opposite reasons. The reason is what matters, and it is not on the form.

 

The reasoning behind 12 is sound for what this actually is: a free web questionnaire where the cost of a false alarm is one uncomfortable half hour and a conversation, and the cost of a miss is somebody with real OCD deciding they are fine. Given those two costs, catching more and flagging more is the right trade. But you should know what you bought. Screen-positive here means "worth looking at". It does not mean "probably OCD", and at 12 it means that considerably less than it would at 21.

 

Step 1 — The lesson

 

A. Eighteen questions, six areas

 

The Obsessive-Compulsive Inventory — Revised is one of the shortest instruments in this course. Eighteen questions, three to five minutes, six areas with three questions each. It asks how much a particular experience has distressed or bothered you in the past month, and it adds your answers up.

 

It came out of a longer questionnaire. Edna Foa and colleagues cut a 42-item original down to 18 items and validated the short version on 215 people with a diagnosis of OCD, 243 people with other anxiety disorders, and 677 people with no clinical diagnosis at all.¹ That third group matters. An instrument that has only ever been given to patients cannot tell you where ordinary sits.

 

The six areas are Washing, Obsessing, Hoarding, Ordering, Checking and Neutralizing. In plainer terms: cleaning past the point of need; thoughts that arrive uninvited and will not leave; not being able to throw things away; needing objects arranged a particular way; going back to make sure; and the private mental acts — counting, repeating a word, undoing — that people perform to cancel a thought out.

 

Diagram — A · Eighteen questions, six areas. Six subscales of three items each, every one scoring 0 to 12, adding to a total of 0 to 60. Two people with the same total can have completely different shapes underneath it. The shape is the part worth reading.

 

One honest note about the version on this site. The live screener page does not state the response scale for individual items, so this module does not state one either. What it does give you is a total between 0 and 60 and six subscale scores between 0 and 12, and those are what the workbook records.

 

The six subscales have been examined on their own in a clinical sample rather than only as a total,² and the whole instrument has been put through an independent psychometric evaluation outside the laboratory that built it.³ That is more than most of the questionnaires in this course can say, and it is the reason this one sits in the top validity tier.

 

B. Same act, different engine

 

Here is the difficulty at the center of this module, and it is a real clinical difficulty rather than a rhetorical one. A questionnaire can ask what you do and how much it bothers you. It cannot ask why you do it. For an autistic adult, the why is the entire distinction.

 

Picture a shelf of spice jars, labels out, in a fixed order. Two people keep that shelf. One of them finds it steadying: the kitchen is loud and bright and unpredictable, and this one surface is not. If somebody rearranges it, the cost is real — the afternoon gets harder — but the routine itself is wanted. They would not give it up if you offered.

 

Diagram — B · Same act, different engine. The behavior is not the diagnostic information. The relationship to the behavior is. Ask whether you would keep it if the anxiety attached to it vanished tomorrow — a routine you would keep and a compulsion you would drop are different things wearing the same clothes.

 

The other person is not steadied by the shelf at all. A thought arrived — something will go wrong, something is contaminated, something is their fault — and the thought is not one they agree with or want. Straightening the jars quietens it for a few minutes. Then it comes back, and the price goes up, and the jars have to be straightened again.

 

That first description is not a symptom of anything. Thirty-two autistic adults interviewed about stimming described it as a self-regulatory mechanism that helps soothe or communicate intense emotion, and objected specifically to treatment aimed at eliminating it.⁴ A comparative review of restricted and repetitive behaviors across autism and OCD reached a related conclusion from the research side: the behaviors develop along broadly similar lines in both, but the role anxiety plays in them is where the two part company.⁵

 

None of which means autistic people do not get OCD. When 40 autistic adults were compared with 45 adults whose primary diagnosis was OCD, using a clinician-administered symptom checklist with obsessions carefully distinguished from autistic interests and stereotyped behavior, the two groups had similar frequencies of obsessive-compulsive symptoms. The OCD group was more severe. But up to half of the autistic group reported at least moderate interference from theirs.⁶ Being autistic does not protect you. It just makes the question harder to answer from the outside.

 

C. How often they travel together

 

Part Six of this course is about the conditions that turn up alongside being autistic often enough to be worth naming. OCD is one of the better-documented ones, and the documentation is unusually good because Denmark keeps registers.

 

A cohort of 3,380,170 people was followed. Among them, 18,184 had a diagnosis of an autism spectrum disorder and 11,209 had a diagnosis of OCD; 739 had both. People diagnosed as autistic were 2.18 times more likely to be diagnosed with OCD later on, and people diagnosed with OCD were 3.91 times more likely to be diagnosed as autistic later on. Put the other way round: 6.6 per cent of the people with OCD also had an autism diagnosis, against 0.5 per cent of everybody else.⁷

 

Diagram — C · How often they travel together. Two large groups with a small, dense overlap. The register also found raised odds in the other direction across generations: a parent with OCD had a higher chance of an autistic child. Related, and still not the same thing.

 

The most directly useful study for this page gave the OCI-R to 171 autistic adults, 108 adults with OCD, 54 adults with both, and 92 comparison participants. The group with both scored significantly higher than the autistic group alone, and the instrument corresponded well with clinician diagnosis. The finding that matters most, though, is this one: OCD symptoms were not significantly correlated with core autistic repetitive behavior as measured on the ADI-R or the ADOS-G. The authors concluded that OCD shows up separately from autism, with a different profile of repetitive thoughts and behavior.⁸

 

That is genuinely good news. It means the two are separable in principle, and that a high score is not just your autistic traits being counted twice.

 

One of the six subscales, though, is now in an awkward position. In 2010 a review recommended that compulsive hoarding be pulled out of OCD and given its own diagnosis,⁹ and DSM-5 did exactly that. So a sixth of this questionnaire measures something that is no longer part of the condition the questionnaire screens for. The instrument's own authors have since published a 12-item version that drops Hoarding altogether — along with Neutralizing, on the grounds that it performs poorly¹⁰ — and a German validation of that shorter version, in 102 people with OCD, 69 with an anxiety-related disorder and 248 non-clinical participants, found it held up well.¹¹ The OCI-R you take here is still the 18-item original. Read the Hoarding score as information about hoarding, not as evidence about OCD.

 

D. What the number is worth

 

The published cut-off for the OCI-R is 21. The screener on this site uses 12. That is a deliberate choice, it is unusual, and the page is the only screener on this site that explains its own reasoning. It deserves an honest account rather than a quiet one.

 

A cut-off is a line drawn across a continuum. Move the line down and you catch more of the people who really do have the condition, and you also flag more people who do not. Move it up and the reverse happens. There is no setting that avoids both errors, only a choice about which error you would rather make.

 

Look at what the line does in real samples. A Sri Lankan validation compared 89 people, patients with OCD against controls, with the diagnosis made independently by two consultant psychiatrists. Its own ROC analysis landed on 21 as well, and at that threshold the questionnaire had 84.4 per cent sensitivity and 85.6 per cent specificity.¹² That looks excellent, and it is worth knowing why it flatters the instrument: a study that compares diagnosed patients with healthy controls has removed all the difficult middle. Real screening does not get to do that.

 

Other language versions have been through the same exercise and drawn their own lines. There is a German validation in patients with OCD, anxiety disorders and depressive disorders,¹³ and a European Portuguese one in 90 patients with OCD and 246 people without a psychiatric illness, which reported internal consistency of 0.913 in the patient group and 0.888 in the control group and a correlation of 0.651 with the clinician-rated Y-BOCS.¹⁴ The instrument travels well. The threshold travels less well, because a threshold is partly a fact about the sample it was drawn in.

 

Diagram — D · What a cut-off catches. The sixteen missed are the argument for lowering the line, and this page lowers it to 12. The cost sits on the side of the picture you cannot see here: OCD has a lifetime prevalence of roughly 1.9 to 3.0 per cent, so most of the people any screener flags will not turn out to have it.

 

Now the site's own reasoning, which you should have in front of you. The page notes that the original developers proposed 21 for case finding, and that 21 catches only about 66 per cent of true cases; that a German sample proposed 18, with roughly 84 per cent sensitivity and 82 per cent specificity; and that New Path has set its line at 12 in order to favor sensitivity. Those figures come from the page's own note. This module has verified that the German validation exists and what it covers, but has not been able to read its full text, so treat the 18, the 84 and the 82 as the page's numbers rather than as numbers this module has checked.

 

Two more things the number cannot do. It cannot tell an autistic routine from a compulsion, because nothing about the item wording distinguishes wanting from resisting — that is section B's problem and no cut-off solves it. And it cannot diagnose anything. In the study of autistic adults above, the OCI-R separated autistic adults from autistic adults with OCD reasonably well, and separated autistic adults with OCD from adults with OCD alone only fairly well,⁸ and in every one of those comparisons the thing doing the actual diagnosing was a clinician.

 

INSIDE THE INSTRUMENT

 

OCI-R — Obsessive-Compulsive Inventory — Revised

 

Where it came from

 

Foa and colleagues published the OCI-R in Psychological Assessment in 2002 as a short form of the 42-item Obsessive-Compulsive Inventory, removing the original's duplicated frequency scale, simplifying scoring and reducing overlap between subscales. Validation drew on 215 participants with OCD, 243 with other anxiety disorders and 677 non-clinical participants, with receiver operating characteristic analyses used to derive screening thresholds.¹ The subscales were subsequently validated in their own right in a clinical sample² and the instrument was independently evaluated and extended in a clinical sample by a separate group.³

 

What it is made of

 

Eighteen items, six subscales of three items each — Washing, Obsessing, Hoarding, Ordering, Checking and Neutralizing — each scoring 0 to 12, summing to a total of 0 to 60. The live New Path page does not state the per-item response scale, and none has been assumed here. Band descriptions on that page are reproduced verbatim in the workbook below, including the six subscale bands.

 

How well it performs

 

Independent replications are the instrument's strongest feature. A Sri Lankan Sinhala validation of 89 participants, with diagnosis set independently by two consultant psychiatrists, derived a ROC cut-off of 21 with sensitivity 84.4%, specificity 85.6%, positive predictive value 85.4% and negative predictive value 84.6%, and Cronbach's alpha above 0.7 for all six subscales and the total.¹² A European Portuguese validation of 90 patients with OCD and 246 controls reported alpha of 0.913 and 0.888 respectively and convergent validity with the Y-BOCS at r = 0.651.¹⁴ German¹³ and German 12-item¹¹ versions exist.

 

Where it was validated

 

Original validation was on treatment-seeking clinic samples and non-clinical volunteers in the United States,¹ with later independent work in Sinhala,¹² European Portuguese,¹⁴ and German¹³ populations. For autistic adults specifically there is one directly relevant study: 171 adults with a diagnosis of autism spectrum disorder, 108 with OCD, 54 with both and 92 controls, in which the OCI-R showed good psychometric properties and corresponded well with clinician diagnosis of OCD.⁸

 

What it cannot do

 

It cannot distinguish an autistic routine from a compulsion, because no item asks about function. The items ask what is done and how distressing it is. An insistence on sameness that is regulating and wanted, and a compulsion that is resisted and driven by an unwanted intrusive thought, can produce identical answers. A comparative review of restricted and repetitive behaviors across both conditions found similar developmental trajectories with the role of anxiety as the distinguishing dimension.⁵ The differentiating question is clinical, not psychometric.

 

The Hoarding subscale measures a condition that is no longer part of OCD. Hoarding disorder was recommended as a separate diagnosis in 2010⁹ and created in DSM-5. The instrument's own authors have since published a 12-item revision dropping Hoarding and Neutralizing, the latter for weak psychometric performance,¹⁰ and a German validation of that revision in 102 patients with OCD, 69 with anxiety-related disorders and 248 non-clinical controls replicated its four-factor structure.¹¹ A raised Hoarding score on the OCI-R is a pointer to a different assessment pathway.

 

The published cut-off of 21 is not the cut-off this site uses. The New Path page sets its screen-positive band at 12 and above, explicitly to favor sensitivity, and states its reasoning on the page. Clinicians reading a client's printout should note that a total of 12 to 20 is screen-positive here and would be sub-threshold against the published literature.¹

 

Reported accuracy comes from case-control designs and will not hold in community screening. Sensitivity and specificity estimated by comparing diagnosed patients with healthy controls omit the ambiguous middle where screening actually operates. With a lifetime population prevalence of 1.9% to 3.0%,¹² positive predictive value in an unselected web population is low at any threshold and lower still at 12.

 

Elevated scores in autistic adults require a functional interview before they mean anything. OCI-R scores in autistic adults did not correlate significantly with core repetitive behavior on the ADI-R or ADOS-G, which supports separability,⁸ but obsessive-compulsive symptoms are common in this population regardless: 40 autistic adults and 45 adults with OCD showed similar symptom frequencies on a clinician-administered checklist, with up to 50% of the autistic group reporting at least moderate interference.⁶

 

What a clinician does with it

 

Use the subscale profile, not the total, as an interview schedule. For each elevated subscale establish function: is the behavior ego-syntonic and regulating, or ego-dystonic, resisted and preceded by an unwanted intrusive thought? Ask what happens when it is prevented — escalating anxiety with an urge to complete points toward compulsion, while distress at a lost regulatory strategy or an unpredictable environment points elsewhere. Note co-occurrence base rates when weighing this: in a Danish cohort of 3,380,170, an autism diagnosis carried an incidence rate ratio of 2.18 for later OCD.⁷ Treat a raised Hoarding score as a separate line of enquiry.

 

Validity tier: 1 — validated and published, calibrated by independent labs. The OCI-R has a peer-reviewed development paper, independent psychometric replication, multiple language validations with their own ROC analyses, and one study of its performance in autistic adults; the tier reflects the instrument, not the locally chosen cut-off of 12, which is a site decision rather than a published threshold.

 

Diagram — E · What a screen-positive means. The value of this questionnaire is that it turns a vague unease into four specific things you can say out loud to somebody. That is the whole job. Everything after the second box needs a person, not a form.

 

STRENGTHS LENS

 

You have been running a system that works, and paying for it in explanations.

 

Routines, order, checking and a very high standard of care about detail are not deficits with a nice framing put on them. A shelf that is always the same is a piece of infrastructure: it removes one source of decisions from a day that already contains too many. People who keep systems like this are the ones who notice the discrepancy in the invoice, remember which cable goes where, and arrive with the document everybody else assumed somebody had brought.

 

What this questionnaire can do for you is separate the part that is working from the part that is costing. If a subscale comes back high and the honest answer is that you would keep the behavior even if the anxiety went away, that is a system, and the right response is to protect it. If the honest answer is that you would drop it tomorrow, that is something else, and something else has treatments.

 

F. What helps

 

Everything below turns on one distinction, so the first item is the one to do properly. The rest follow from it.

 

1. Run the why test on each high subscale, one at a time.

 

Take each subscale that came back elevated and ask two questions about it. Would I keep doing this if the anxiety attached to it disappeared tomorrow? And did a thought I disagree with come first, or did the behavior come first on its own? A routine you would keep, with no unwanted thought behind it, is a routine. A behavior you would drop, arriving after a thought that feels alien, is closer to a compulsion. Write the answer down for each one rather than deciding in general.

 

2. Test one of them by not doing it, briefly, and watch what happens.

 

Leave the jars crooked for an hour, or the door unchecked once, and pay attention to the shape of what follows. An autistic routine interrupted usually produces a flat, grinding cost: the day is harder, capacity drops, everything after it takes more. A compulsion prevented usually produces rising anxiety with a specific urge to go and complete the act, and it peaks and then falls. Those two feel different from the inside. Do not do this with anything that frightens you badly — that version belongs with a therapist.

 

3. Give the Obsessing subscale more weight than the others.

 

Washing, Ordering and Checking are the subscales most likely to be counting autistic routines. Obsessing is different: it asks about thoughts that arrive uninvited, that you do not want and cannot dismiss. Intrusive thoughts of that kind are far less easily explained by insistence on sameness, so a high Obsessing score with moderate scores elsewhere is more informative than a high total. If Obsessing is your peak, take that to somebody.

 

4. Read the Hoarding score as its own question.

 

Hoarding became a separate diagnosis in DSM-5, and the shorter successor to this questionnaire drops the subscale entirely. If Hoarding is the reason your total crossed 12, your total has told you very little about obsessive-compulsive symptoms and something quite useful about clutter, difficulty discarding, and how much of your living space is currently unusable. That is a real thing to raise, on its own terms.

 

5. Take four sentences, not a number, to whoever you talk to.

 

Say which subscale was highest, what the behavior actually is on an ordinary Tuesday, whether you would keep it if the anxiety went, and what happens when you cannot do it. A clinician can work with that in five minutes. A total of 19 out of 60, handed over on its own, tells them almost nothing and invites them to guess. The workbook below is built to produce exactly those four sentences.

 

Step 2 — Take the screener

 

Eighteen questions, three to five minutes, free and confidential. You get a total between 0 and 60 and six subscale scores between 0 and 12 — Washing, Checking, Neutralizing, Obsessing, Ordering and Hoarding. One honest steer before you start: this page flags you as screen-positive at 12, which is well below the published clinical cut-off of 21, so read a positive result as a reason to look further rather than as a finding.

 

Take the OCI-R

 

Before you start

 

The OCI-R is a published, peer-reviewed screening questionnaire with independent psychometric replication and validations in several languages, which places it among the better-evidenced instruments in this course. It is still a screen and not a diagnosis: it counts behaviors and distress, and it cannot tell an autistic routine that is wanted and regulating apart from a compulsion that is resisted and driven by an unwanted intrusive thought. The screen-positive threshold used on this page is 12, chosen to favor sensitivity, rather than the published cut-off of 21 — a lower line catches more genuine cases and also flags more people who do not have OCD, and since OCD has a lifetime prevalence of only about 1.9% to 3.0%, most people who score above any threshold will not turn out to have it. The Hoarding subscale measures something DSM-5 now treats as a separate diagnosis, and the instrument's own authors have published a shorter version that drops it. Only a clinician can diagnose.

 

Step 3 — Your workbook

 

Your answers save to this device only — we cannot see a word of what you write. This module lets you record your total and all six subscale scores, run the why test on each raised subscale, read the Obsessing and Hoarding scores on their own terms, and turn the whole thing into four sentences you could say out loud to a clinician.

 

Your OCI-R results

 

Took the screener? Put your numbers in below. The six subscales are the useful part — a total of 19 tells you very little, and where the 19 came from tells you a lot. Entirely optional — skip it if you would rather just read.

 

Score bands: 0–7 = Typical range; 8–11 = Elevated (near threshold); 12–60 = Screen-positive

 

Fields: OCI-R · Obsessive-Compulsive Inventory — Revised; Total score (0–60); My total (enter 0–60); Washing (0–12); Washing (enter 0–12); Checking (0–12); Checking (enter 0–12); Neutralizing (0–12); Neutralizing (enter 0–12); Obsessing (0–12); Obsessing (enter 0–12); Ordering (0–12); Ordering (enter 0–12); Hoarding (0–12); Hoarding (enter 0–12)

 

The why test, subscale by subscale

 

Section F, item 1. Do this one at a time rather than in general — the answer is often different for different subscales.

 

Fields: My highest subscale; What that behavior actually is, on an ordinary Tuesday; Would I keep doing this if the anxiety attached to it vanished tomorrow?: Yes, I would keep it / No, I would drop it immediately / Some of it I would keep and some I would drop / I cannot tell yet; Did an unwanted thought come first, or did the behavior come first on its own?: A thought I do not agree with came first / The behavior came first, no thought attached / It varies / I cannot tell yet

 

What happens when you cannot do it

 

Section F, item 2. The shape of the distress is more informative than its size. Skip this one entirely if the behavior is something that frightens you.

 

Fields: What I noticed when I left it undone; Which was it closer to?: A flat, grinding cost — the whole day got harder / Rising anxiety with an urge to go and complete it / Both, at different times / Neither, nothing much happened / I have not tried this; Tick: This is something I would rather try with a therapist than alone

 

The Obsessing subscale, on its own

 

Section F, item 3. Intrusive thoughts are the part of this questionnaire least well explained by autistic routine, so this subscale carries more weight than the others.

 

Fields: Thoughts that arrive uninvited and will not leave; Do these thoughts feel like mine?: No — they feel alien and I disagree with them / Yes — they are worries I recognize as my own / A mixture / Not sure; Roughly how much of a day they take

 

The Hoarding score, as its own question

 

Section F, item 4. Hoarding became a separate diagnosis in DSM-5 and the shorter successor to this questionnaire drops the subscale, so read this score on its own terms.

 

Fields: Tick: My Hoarding score is a large part of why my total crossed 12; What discarding things is actually like for me; How much of my living space is currently unusable

 

The four sentences

 

Section F, item 5. This is what you take to an appointment instead of a number. Write it now while the screener is fresh.

 

Fields: Who I would say this to; The four sentences; Tick: I want to look into this further rather than leave it here

 

Appendix — Research companion

 

Peer-reviewed research

 

10. Abramovitch A, Abramowitz JS, McKay D (2021). The OCI-12: a syndromally valid modification of the obsessive-compulsive inventory-revised. Psychiatry Research, 298, 113808. DOI 10.1016/j.psychres.2021.113808. View the paper The paper by the OCI-R's own authorship lineage introducing a 12-item revision that removes two of the six original subscales: Hoarding, because hoarding became a separate DSM-5 diagnosis, and Neutralizing, on psychometric grounds. It is the clearest signal from within the field that a third of the 18-item questionnaire no longer measures what the instrument is for. Limitation: the full text was not accessible for this module, so the sample and accuracy figures reported in it are not quoted here; the rationale summarised is taken from the subsequent German validation of the same 12-item instrument.

 

3. Abramowitz JS, Deacon BJ (2006). Psychometric properties and construct validity of the Obsessive-Compulsive Inventory-Revised: replication and extension with a clinical sample. Journal of Anxiety Disorders, 20(8), 1016-1035. DOI 10.1016/j.janxdis.2006.03.001. View the paper The standard independent psychometric evaluation of the OCI-R, carried out outside the group that developed it, replicating and extending the original psychometric work in a clinical sample. Its existence is part of why the OCI-R sits above most self-report obsessive-compulsive measures on evidence quality: the instrument has been tested by people with no stake in the result. Limitation: a single clinical sample from one American anxiety disorders service, and its full text was not accessible for this module, so no figures from it are quoted here.

 

8. Cadman T, Spain D, Johnston P, Russell A, Mataix-Cols D, Craig M, Deeley Q, Robertson D, Murphy C, Gillan N, Wilson CE, Mendez MA, Ecker C, Daly E, Findon J, Glaser K, Happe F, Murphy D (2015). Obsessive-compulsive disorder in adults with high-functioning autism spectrum disorder: what does self-report with the OCI-R tell us?. Autism Research, 8(5), 477-485. DOI 10.1002/aur.1461. View the paper The only study of the OCI-R specifically in autistic adults. Participants were 171 adults with an autism spectrum disorder, 108 with OCD, 54 with both and 92 controls. Those with both reported significantly higher obsessive-compulsive symptoms than autistic adults alone; OCD symptoms were not significantly correlated with core autistic repetitive behaviour on the ADI-R or ADOS-G; the OCI-R showed good psychometric properties, corresponded well with clinician diagnosis and produced ROC cut-offs for total and Checking scores that discriminated well between autism and autism plus OCD and fairly well between autism plus OCD and OCD alone. Limitation: a cognitively able clinic sample recruited through a specialist service, so it does not establish performance in autistic adults with intellectual disability or in unselected self-referring web users.

 

14. Cunha GV, Moreira PS, Sousa MM, Castanho T, Pico-Perez M, Ferreira S, Morgado P (2023). The Obsessive-Compulsive Inventory-Revised (OCI-R): translation and validation of the European Portuguese version. Acta Medica Portuguesa, 36(3), 174-182. DOI 10.20344/amp.16452. View the paper A European Portuguese translation and validation in 90 individuals diagnosed with OCD and 246 control participants without psychiatric illness. Confirmatory factor analysis supported the model in both groups, internal consistency was 0.913 in the patient group and 0.888 in the control group, and convergent validity against the clinician-rated Y-BOCS was r = 0.651. It is a useful demonstration that the six-factor structure survives translation. Limitation: a two-group clinical-versus-healthy design with no other psychiatric comparison group, so it establishes structure and reliability rather than the instrument's ability to discriminate OCD from other conditions.

 

1. Foa EB, Huppert JD, Leiberg S, Langner R, Kichic R, Hajcak G, Salkovskis PM (2002). The Obsessive-Compulsive Inventory: development and validation of a short version. Psychological Assessment, 14(4), 485-496. DOI 10.1037/1040-3590.14.4.485. View the paper The development paper for the OCI-R, which reduced the 42-item Obsessive-Compulsive Inventory to 18 items by removing a duplicated frequency scale, simplifying subscale scoring and reducing cross-subscale overlap. Validation used 215 participants with OCD, 243 with other anxiety disorders and 677 non-clinical participants, retaining six three-item subscales and using receiver operating characteristic analyses to derive screening thresholds. It is the source of the 21 cut-off in general clinical use. Limitation: the samples were treatment-seeking clinic attenders and volunteers in the United States, so the derived threshold is a property of those samples as much as of the instrument, and no autistic sample was involved in setting it.

 

13. Gonner S, Leonhart R, Ecker W (2008). The Obsessive-Compulsive Inventory-Revised (OCI-R): validation of the German version in a sample of patients with OCD, anxiety disorders, and depressive disorders. Journal of Anxiety Disorders, 22(4), 734-749. DOI 10.1016/j.janxdis.2007.07.007. View the paper The German-language validation of the OCI-R, conducted in patients with OCD alongside comparison groups with anxiety disorders and with depressive disorders rather than healthy controls alone, which is a harder and more clinically realistic test than a patient-versus-control design. It is the study behind the alternative cut-off of 18 that the New Path screener page cites in its own reasoning. Limitation: the full text was not accessible for this module, so the sensitivity and specificity figures attributed to it on the screener page are not independently confirmed here, and a German clinical sample cannot set a threshold for English-language web screening in any case.

 

2. Huppert JD, Walther MR, Hajcak G, Yadin E, Foa EB, Simpson HB, Liebowitz MR (2007). The OCI-R: validation of the subscales in a clinical sample. Journal of Anxiety Disorders, 21(3), 394-406. DOI 10.1016/j.janxdis.2006.05.006. View the paper A validation of the six OCI-R subscales in their own right, rather than only as contributors to a total, in a clinical sample. It is the study most often cited for the position that the subscales carry usable information individually, which is the basis for reading a subscale profile rather than a single number. Limitation: a clinical sample recruited through specialist anxiety services, which does not establish how the subscales behave in unselected community or web-screening populations, and its full text was not accessible for this module, so no figures from it are quoted here.

 

5. Jiujias M, Kelley E, Hall L (2017). Restricted, repetitive behaviors in autism spectrum disorder and obsessive-compulsive disorder: a comparative review. Child Psychiatry and Human Development, 48(6), 944-959. DOI 10.1007/s10578-017-0717-0. View the paper A critical review comparing restricted and repetitive behaviours across autism and OCD, on the basis that the similar behavioural profiles of the two create genuine potential for diagnostic confusion. It traces developmental trajectories in both and examines three constructs proposed to underlie the behaviours - anxiety, executive functioning and sensory phenomena - concluding that while the behaviours develop with some similarity, the differing role of anxiety marks an important distinction between the conditions. Limitation: a narrative comparative review rather than a meta-analysis, with no pooled effect sizes and no primary data, so it frames the differential question rather than quantifying it.

 

7. Meier SM, Petersen L, Schendel DE, Mattheisen M, Mortensen PB, Mors O (2015). Obsessive-compulsive disorder and autism spectrum disorders: longitudinal and offspring risk. PLOS ONE, 10(11), e0141703. DOI 10.1371/journal.pone.0141703. View the paper A prospective Danish population register study of 3,380,170 cohort members, of whom 18,184 had a diagnosis of an autism spectrum disorder, 11,209 had OCD and 739 had both. People diagnosed as autistic had an incidence rate ratio of 2.18 (95% CI 1.91-2.48) for a later OCD diagnosis, people with OCD had an incidence rate ratio of 3.91 (95% CI 3.46-4.40) for a later autism diagnosis, and parental OCD carried an incidence rate ratio of 1.83 (95% CI 1.45-2.28) for autism in offspring; 6.6% of people with OCD also had an autism diagnosis against 0.5% of those without. Limitation: register data captures only diagnoses that reached specialist services, so both conditions are undercounted in adults and particularly in women, and diagnostic practice changed substantially across the observation period.

 

11. Muller CL, Fink-Lamotte J, Jelinek L, Lohse L, Ehring T, Noll-Hussong M, Berberich G, Wahl-Kordon A, Borgelt J, McKay D, Abramowitz JS, Abramovitch A, Cludius B (2025). Translation and validation of the German 12-item Obsessive-Compulsive Inventory (OCI-12) in clinical and non-clinical samples. Clinical Psychology in Europe, 7(4), e16165. DOI 10.32872/cpe.16165. View the paper A German validation of the OCI-12 in 102 participants with OCD, 69 with an anxiety-related disorder and 248 non-clinical controls, which replicated the four-factor structure of the English version with a higher-order general symptom factor and reported good internal consistency, good test-retest reliability, moderate-to-good construct validity and good-to-excellent diagnostic accuracy. Its rationale states plainly that hoarding symptoms are now better captured by a separate hoarding diagnosis and that the Neutralizing scale has relatively poor psychometric properties. Limitation: a German-speaking sample recruited largely through specialist clinics, and a validation of the shortened instrument rather than of the 18-item OCI-R that this screener actually administers.

 

6. Russell AJ, Mataix-Cols D, Anson M, Murphy DG (2005). Obsessions and compulsions in Asperger syndrome and high-functioning autism. The British Journal of Psychiatry, 186(6), 525-528. DOI 10.1192/bjp.186.6.525. View the paper Forty adults with high-functioning autism spectrum disorder were given the Yale-Brown Obsessive-Compulsive Scale Symptom Checklist and compared with 45 gender-matched adults whose primary diagnosis was OCD, with symptoms deliberately distinguished from the stereotyped behaviours and interests usually seen in autistic people. The two groups had similar frequencies of obsessive-compulsive symptoms, only somatic obsessions and repeating rituals being more common in the OCD group; the OCD group had higher severity ratings, but up to 50% of the autistic group reported at least moderate interference from their symptoms. Limitation: 85 participants in total, all cognitively able, with symptom-versus-stereotypy distinctions made by clinical judgement, which is the very judgement the field has not yet operationalised.

 

12. Senanayake B, Rajasuriya M, Suraweera C, Arambepola C (2018). How valid is obsessive-compulsive inventory-revised scale among Sri Lankan adults?. Indian Journal of Psychiatry, 60(3), 318-323. DOI 10.4103/psychiatry.indianjpsychiatry_171_18. View the paper A case-control validation of a Sinhala translation of the OCI-R in 89 OCD patients and controls recruited at the National Hospital of Sri Lanka, with the criterion diagnosis made independently by two consultant psychiatrists. Receiver operating characteristic analysis gave a cut-off of 21, with sensitivity 84.4%, specificity 85.6%, positive predictive value 85.4% and negative predictive value 84.6%; Cronbach's alpha exceeded 0.7 for all six subscales and for the total. The paper also states the general-population lifetime prevalence of OCD as 1.9% to 3.0%, which is the base rate that governs how any cut-off performs in real screening. Limitation: a case-control design comparing diagnosed patients with controls removes the ambiguous middle of the distribution, so these sensitivity and specificity figures are an optimistic ceiling rather than an estimate of web-screening performance.

 

Lived experience

 

4. Kapp SK, Steward R, Crane L, Elliott D, Elphick C, Pellicano E, Russell G (2019). People should be allowed to do what they like: autistic adults' views and experiences of stimming. Autism, 23(7), 1782-1792. DOI 10.1177/1362361319829628. View the source Interviews and focus groups with 32 autistic adults about repetitive movement, analysed thematically, with two themes: stimming as a self-regulatory mechanism, and stimming as socially unaccepted but capable of becoming accepted through understanding. Participants described it as an adaptive mechanism that soothes or communicates intense emotion and objected to treatment aimed at eliminating the behaviour, which is the clearest published statement of why an autistic repetitive behaviour is not equivalent to a compulsion. Limitation: a self-selected UK sample of 32 adults able to take part in interviews and focus groups, so it under-represents autistic people with higher support needs and cannot speak to prevalence.

 

Emerging or contested

 

9. Mataix-Cols D, Frost RO, Pertusa A, Clark LA, Saxena S, Leckman JF, Stein DJ, Matsunaga H, Wilhelm S (2010). Hoarding disorder: a new diagnosis for DSM-V?. Depression and Anxiety, 27(6), 556-572. DOI 10.1002/da.20693. View the paper The review that argued compulsive hoarding should be separated from OCD and obsessive-compulsive personality disorder, on the evidence that hoarding is frequently independent of both. It weighed whether hoarding should remain a criterion of OCD, concluded there was sufficient evidence to create a new disorder provisionally called hoarding disorder, and predicted that doing so would improve case identification and stimulate specific treatment development; DSM-5 subsequently created the diagnosis. Limitation: a focused narrative review written as a DSM working-group recommendation rather than an empirical study, with no new data and an explicit acknowledgement that the proposal was preliminary.

 

Peer-reviewed = checked by independent experts before publication. Clinical model = an established professional framework, not a single study.

 

Up next

 

Module 18 - CPQ — Clinical Perfectionism, When Your Standards Run You

 

All modules in Autistic Self-Discovery

 

The behavior is not the question. The reason is. This course was built by clinicians who are part of the New Path Family. Telling a routine that steadies you apart from a compulsion that is running you is difficult to do alone and straightforward to do with someone, and the two need opposite responses — one protecting, one treating. Therapy for clients in California and coaching worldwide, all by telehealth, are offered by our sister company New Path Family of Therapy Centers, Inc. A conversation costs nothing and there is no pressure. Saving this for later counts too. Talk with the New Path team