The DOCS measures the severity of obsessive-compulsive symptoms across four dimensions rather than producing a single undifferentiated score. The dimensions are contamination; responsibility for harm, injury or bad luck; unacceptable thoughts; and symmetry, completeness and the need for things to be "just right". Hoarding is deliberately excluded, reflecting the evidence that emerged during the scale's development that hoarding is a distinct condition rather than an OCD symptom dimension.
Its structural innovation is that the same five severity parameters are rated within each dimension: time occupied, avoidance, distress, functional interference, and difficulty disregarding obsessions or refraining from compulsions. Each of the twenty items is rated 0 to 4 over the past month, giving four subscale scores of 0 to 20 and a total of 0 to 80. This means the DOCS separates what a person's symptoms are about from how severe they are - a distinction that matters clinically, because two patients with the same total may have entirely different treatment targets.
02 - Origin & purpose
Where it comes from.
Abramowitz and a large collaborating group published the DOCS in Psychological Assessment in 2010, developing and evaluating it across clinical and non-clinical samples. The intent was to correct two limitations in the existing self-report measures: instruments such as the OCI-R weight symptom categories by the number of items assigned to them rather than by severity, and most measures confound symptom content with symptom severity. By holding the five severity parameters constant across all four dimensions, the DOCS allows content and severity to be read separately.
The scale is written at roughly a ninth-grade reading level and takes about five minutes to complete. It has been translated into around sixteen languages, and a Norwegian short form (DOCS-SF) was published in 2017.
03 - Scoring & cutoffs
How scoring works.
Each item is rated 0 (no symptoms) to 4 (extreme symptoms) with reference to the past month. Subscale scores are the unweighted sum of the five items within a dimension (0-20); the total is the sum of all twenty (0-80). There is no reverse scoring and no weighting.
Two total-score cutoffs were derived in the original validation. A cutoff of 21 distinguished patients with OCD from patients with other anxiety disorders at approximately 70% sensitivity and 70% specificity. A cutoff of 18 distinguished OCD from non-clinical adults at approximately 78% sensitivity and 78% specificity. The area under the curve for the total score was 0.86 (95% CI 0.84-0.89) against non-clinical adults and 0.77 (0.73-0.82) against other anxiety disorders - in both cases significantly better than the OCI-R total. A Spanish validation proposed a lower cutoff of 15 against a non-clinical comparison group.
There are no validated mild, moderate or severe severity bands. The scale's authors state directly that cutoff scores for those categories have not been established, and clinicians should not treat any published banding as authoritative. Reference means from the original study give more useful context: 30.06 (SD 15.49) in the OCD sample, 16.75 (13.14) in patients with other anxiety disorders, and 11.93 (9.87) in students.
04 - Validation evidence
How well it performs.
Internal consistency is high and consistent across samples. In the original validation, Cronbach's alpha for the total score was 0.90 in the OCD sample, 0.92 in the other-anxiety-disorder sample and 0.93 in students, with subscale alphas between 0.83 and 0.96. Later cross-language studies report total-score alphas from 0.87 to 0.95.
Test-retest reliability is the weaker property. Over a twelve-week interval in 210 students the total score correlated at r = 0.66, with subscale coefficients between 0.55 and 0.66. The authors note that retest was assessed only in non-clinical participants, so stability in patients under treatment is not established from this source.
Convergent validity is good. The total score correlated at r = 0.54 with the clinician-rated Y-BOCS in the OCD sample and between 0.65 and 0.71 with the OCI-R across the three samples. Correlations between each DOCS subscale and its conceptual OCI-R counterpart were generally strong, with contamination and washing correlating at 0.88 in the OCD sample. Discriminant validity held: correlations with depression, anxiety and social anxiety measures ranged from 0.08 to 0.52 and were significantly weaker than the obsessive-compulsive symptom correlations within every group. The four-factor structure was supported by both exploratory and confirmatory analysis across all three samples, with a higher-order general factor, and the scale has been shown to be sensitive to the effects of exposure and response prevention.
α = 0.90-0.93
INTERNAL CONSISTENCY OF THE TOTAL SCORE ACROSS THREE SAMPLES (ABRAMOWITZ 2010)
AUC 0.86
TOTAL SCORE DISCRIMINATING OCD FROM NON-CLINICAL ADULTS (ABRAMOWITZ 2010)
21 / 18
CUTOFFS VS OTHER ANXIETY DISORDERS (~70%) AND VS NON-CLINICAL ADULTS (~78%)
r = 0.54
CORRELATION WITH THE CLINICIAN-RATED Y-BOCS, OCD SAMPLE (ABRAMOWITZ 2010)
05 - How it compares
How it compares to the alternatives.
Instrument
Items
Time
When to reach for it
DOCS
20
~5 min
Range 0-80. Copyright retained; free for clinical and research use, fee for commercial use. Reach for it when you need symptom dimensions and severity separated, and dimension-specific treatment targets.
Range 0-40. Copyright retained; widely used with permission. The reference standard for clinician-rated OCD severity and the usual primary outcome in trials.
Range 0-40. Copyright retained. Children and adolescents; the paediatric counterpart of the Y-BOCS.
DOCS-SF
5 + checklist
~2 min
Range 0-40. Same copyright as the DOCS. Brief screening where the full DOCS is too long; validated in Norwegian with a cutoff of 16 (sensitivity 96%, specificity 94%).
06 - When to use it
Right tool, wrong tool.
If you need a free self-report OCD measure you can administer and reproduce without permission, the OCI-R is the appropriate choice. For clinician-rated severity, the Y-BOCS remains the reference standard, and the CY-BOCS is its paediatric counterpart.
Reach for it when
-You want dimension-specific severity to guide exposure planning rather than a single total.
-You are tracking response to exposure and response prevention, where the scale has demonstrated sensitivity to treatment effects.
-Research where the confounding of symptom content with severity in older instruments is a methodological problem.
-You need a self-report measure that correlates well with the Y-BOCS without requiring a clinician interview.
Reach for something else when
-You need an instrument you can reproduce on a commercial platform or charge for without seeking permission. Use the OCI-R.
-Hoarding is a presenting concern. The DOCS excludes it by design; the OCI-R includes a hoarding subscale, and dedicated hoarding measures exist.
-The patient is a child or adolescent. The DOCS is validated in adults from age 18; a child version is described as in development and has no verified peer-reviewed psychometrics.
-You need a formal severity grade. No mild, moderate or severe bands have been established.
-You need a diagnosis. The cutoffs discriminate at around 70-78% accuracy and are screening aids, not diagnostic thresholds.
07 - Confidence & precision
Reading the score with care.
Reliability of the total score is strong (alpha 0.90 to 0.93), but twelve-week test-retest reliability is only moderate (r = 0.66) and was assessed in students rather than patients. Discrimination from other anxiety disorders is modest, with sensitivity and specificity near 70% at a total of 21. No mild, moderate or severe severity bands have been established, and no minimal clinically important difference has been published, so single scores should be read alongside the dimension profile and the clinical picture rather than as a precise severity grade.
08 - Limitations
What it cannot tell you.
Modest discrimination from other anxiety disorders. At the cutoff of 21, both sensitivity and specificity sit near 70%, so roughly three in ten classifications will be wrong in a mixed anxiety population.
No established severity bands, which limits its usefulness for communicating severity to patients or in correspondence.
Twelve-week test-retest reliability of 0.66 for the total score is moderate, and was assessed only in students.
Hoarding is not covered, so a hoarding presentation will be missed entirely.
Licensing restricts what platforms can do with it, which is why it appears less often in digital tools than the OCI-R despite arguably better psychometrics.
Adults only; validated from age 18.
Self-report, and obsessive-compulsive symptoms involving taboo thoughts are among the most under-disclosed in psychiatry - a low unacceptable-thoughts score should not be taken at face value early in an assessment.
See how Aisel removes friction where it costs most. A 20-minute walkthrough tailored to your clinic.
We value your privacy
We use cookies to analyse site usage and improve your experience. Analytics and embedded media (e.g. YouTube) only load if you accept. Read our cookie policy.