To score the OCI-R, add the 18 item ratings (each 0 to 4) for a total of 0 to 72, then sum each three-item subscale (0 to 12). The original cut-off is 21. Current practice scores the 15 OCD items (0 to 60, cut-off 12) separately from the three hoarding items (cut-off 6).
The Obsessive-Compulsive Inventory-Revised (OCI-R) is one of the most widely used self-report measures of OCD symptoms, and one of the quickest: NovoPsych lists its administration time at about three minutes.
How do you score the OCI-R?
The OCI-R asks the client to rate how much each of 18 experiences has distressed or bothered them during the past month, from 0 (not at all) to 4 (extremely). There is no reverse scoring and no weighting.
- Total score: add all 18 items. The range is 0 to 72.
- Subscale scores: add the three items in each subscale. Each ranges from 0 to 12.
- OCI-OCD score (recommended): add the 15 non-hoarding items. The range is 0 to 60.
- OCI-HD score: add the three hoarding items (items 1, 7 and 13). The range is 0 to 12.
Steps 3 and 4 reflect how the measure is now scored on Australian platforms such as NovoPsych, because the DSM-5 moved hoarding out of OCD and into its own diagnosis.
What are the six OCI-R subscales?
Foa and colleagues (2002) built the OCI-R as a short form of the 42-item Obsessive-Compulsive Inventory, keeping 18 items across six symptom dimensions of three items each. The item allocation, as set out in the NovoPsych technical report, is:
- Washing (items 5, 11, 17): contamination fears and excessive washing or cleaning.
- Obsessing (items 6, 12, 18): intrusive, unwanted thoughts and attempts to control them.
- Ordering (items 3, 9, 15): the need for symmetry, exactness and arranging objects.
- Checking (items 2, 8, 14): repetitive checking driven by doubt.
- Neutralising (items 4, 10, 16): mental rituals such as counting and repeating.
- Hoarding (items 1, 7, 13): difficulty discarding and problems with clutter.
The six-factor structure was confirmed by Abramowitz and Deacon (2006) and again in 1,339 adults with OCD by Abramovitch and colleagues (2020). One caution from the NovoPsych report: Neutralising and Obsessing can lose reliability in some subgroups, so when they look out of step with the clinical picture, trust the interview.
What is the OCI-R cut-off score?
There is no single answer, which is why the question gets asked so often.
21 on the 18-item total. Foa and colleagues (2002) recommended 21 to separate people with OCD from non-anxious adults, based on ROC analysis across 215 patients with OCD, 243 with other anxiety disorders and 677 non-anxious adults.
A spread of 14 to 36 in later studies. Wootton and colleagues (2015) note that later work proposed cut-offs from 14 to 36, and attribute the spread to differences between samples.
12 on the 15-item OCI-OCD score, and 6 on the hoarding score. Wootton and colleagues compared 118 people with OCD, 201 with hoarding disorder and 155 controls. A cut-off of 12 on the 15-item OCD score gave sensitivity of 0.82 and specificity of 0.83. A cut-off of 6 on the three hoarding items gave sensitivity of 0.92 and specificity of 0.93. Mean OCI-OCD scores were 23.9 in the OCD group and 2.4 in controls.
For routine Australian practice, the 12 and 6 pair is the most useful because it fits the DSM-5 split and is the basis used by NovoPsych and other local platforms. If you report against the original 21, say so, because a client with a hoarding presentation can clear 21 without meeting the OCD threshold at all.
Why should hoarding be scored separately?
When the OCI-R was published, hoarding was treated as a symptom of OCD. The DSM-5 reclassified it as hoarding disorder, a separate diagnosis. Wootton and colleagues (2015) showed the OCI-R still works well when split into a 15-item OCD component and a 3-item hoarding component, with the two components separating the diagnostic groups cleanly.
Abramovitch, Abramowitz and McKay (2021) went further and proposed the OCI-12, which drops the hoarding and neutralising items. In large OCD, anxiety-disorder and community samples it matched the full OCI-R on psychometrics and beat it on sensitivity and specificity, with its own norms and cut-off.
What do the OCI-R severity bands mean?
Abramovitch and colleagues (2020) used the clinician-rated Y-BOCS to derive severity cut-offs for the OCI-R total. Their abstract is candid: the mild-to-moderate boundary performed well, the higher cut-offs were less sensitive and specific, and total scores should be interpreted with caution. Secondary summaries such as Tacklit's assessment page report the bands as 0 to 15 mild, 16 to 27 moderate and 28 or above severe. Check the paper before quoting those figures in a report.
The NovoPsych technical report takes a percentile approach instead, using the 15-item OCI-OCD score:
- 0 to 4: within normal limits (below the 75th percentile of non-clinical adults).
- 5 to 11: subclinical elevation (75th percentile or above in non-clinical adults, below the clinical threshold).
- 12 to 21: a little distressing (roughly the 16th to 40th percentile among people with OCD).
- 22 to 36: moderately distressing (41st to 84th percentile).
- 37 to 51: a lot distressing (85th to 98th percentile).
- 52 to 60: extremely distressing (99th percentile or above).
Whichever framework you use, name it in the note. "OCI-OCD 29, moderate range (NovoPsych percentile bands)" is auditable; "OCI-R moderate" is not.
What does an OCI-R score of 24 or 47 mean?
Total 24, with Hoarding 9. The total clears Foa's cut-off of 21, but the OCI-OCD score is only 15, in the "a little distressing" band, while the hoarding score of 9 is well above the threshold of 6. The picture points towards hoarding as the primary concern, and an OCD-only formulation would miss it.
Total 47, with Hoarding 2. The OCI-OCD score is 45, in the "a lot distressing" band. If Washing and Checking are both at 11 or 12, exposure and response prevention planning can start there. Re-administer after a block of sessions and compare against the change thresholds below. In both cases the OCI-R is informing a formulation, not making a diagnosis.
How does the OCI-R compare with the Y-BOCS?
The Yale-Brown Obsessive Compulsive Scale is a clinician-rated interview of 10 items scored 0 to 4, with a total of 0 to 40, and the usual gold standard in treatment trials. The OCI-R is self-report and gives a symptom profile rather than a single severity rating. The two agree only moderately, with correlations of about 0.41 to 0.53 reported in the NovoPsych summary. Use the OCI-R for screening, profiling and session-to-session tracking, and the Y-BOCS when you need a clinician-rated severity figure.
How much change on the OCI-R is meaningful?
Foa and colleagues (2002) reported two-week test-retest reliability of 0.84 for the total score in 41 people with OCD, so the measure is stable enough to track treatment.
For deciding whether a change is real, the NovoPsych report applies a minimally important difference of half a standard deviation, rounded up: 7 points on the OCI-OCD score for people with OCD, 5 points for people with hoarding disorder, 2 points for non-clinical adults, and 2 points on the hoarding score. A drop from 38 to 30 is a meaningful improvement; a drop from 38 to 33 is not yet one.
How does the OCI-R fit Australian practice?
The OCI-R is not mandated by any Australian scheme, but it fits where outcome measurement is expected. The Better Access initiative funds psychological treatment under a GP mental health treatment plan, and the plan item descriptors (for example MBS item 2715) refer to outcome measurement tools. A baseline OCI-R at intake, a repeat at review, and both scores in your letter back to the referring GP meets that expectation cleanly for an OCD referral.
Our guide to OCD treatment plan goals and objectives shows how to turn a subscale profile into measurable goals, and our overview of mental health treatment plans and referrals covers the review cycle. Because OCD so often co-occurs with depression and anxiety, many clinicians add the PHQ-9 or the DASS-21 to separate OCD-specific change from a general lift in mood.
Is the OCI-R free to use?
The OCI-R is copyright to its authors, with Edna Foa named as the rights holder on Tacklit's assessment page. It is reproduced free of charge in many public health and clinical resources, but the same page notes that written permission should be confirmed before any commercial deployment. Hosted platforms that score it for you operate under their own terms, which do not transfer rights to the instrument.
How should you record OCI-R results in clinical notes?
Record enough that a colleague or an auditor can read it without the form:
- The version used (OCI-R, 18 items) and the date of administration.
- The OCI-OCD score out of 60 and the OCI-HD score out of 12, plus the 18-item total if your service still reports it.
- The subscale scores, the cut-off and the severity framework you applied, by name.
- Change since the previous administration, whether it exceeds the minimally important difference, and a one-line interpretation tied to the formulation.
Our guide to writing a psychological assessment report covers presenting standardised scores formally. PractaLuma is AI-native practice management software for Australian mental-health practices. Its standardised assessments feature scores instruments such as the OCI-R automatically and drops the result into the clinical note, with the subscale profile and change over time alongside it. See all features and pricing.
Frequently asked questions
Can the OCI-R diagnose OCD?
No. It is a screening and severity measure. A score above the cut-off tells you a diagnostic interview is warranted, and Abramovitch and colleagues (2020) caution against leaning on the total score alone.
What is a normal OCI-R score?
On the 15-item OCI-OCD score, 0 to 4 is within normal limits and 5 to 11 is a subclinical elevation (NovoPsych percentile bands). Controls in Wootton and colleagues (2015) averaged 2.4. On the 18-item total, anything under 21 was non-clinical in the original study.
Is there an OCI-R for children?
Yes. The Obsessive-Compulsive Inventory-Child Version (OCI-CV) is a 21-item self-report validated in young people aged 7 to 17. The adult OCI-R is intended for people aged 18 and over.
What is the OCI-12?
A 12-item revision proposed by Abramovitch, Abramowitz and McKay (2021). It removes the hoarding and neutralising items, leaving washing, obsessing, ordering and checking, with its own norms and cut-off.
This article is general information for health practitioners and does not replace clinical judgement, supervision or the published scoring manuals for each instrument.