WHODAS 2.0 measures functioning — how much difficulty a person has actually had doing things over the past thirty days — rather than symptoms, diagnosis or distress. It covers six domains: cognition (understanding and communicating), mobility (moving and getting around), self-care (hygiene, dressing, eating, staying alone), getting along (interacting with other people), life activities (domestic responsibilities, work and school), and participation (joining in community activities, and the impact of the health condition on the respondent and their family). Each item is rated on a five-point scale from none to extreme, and the domains combine into a single summary score from 0, meaning no disability, to 100, meaning full disability.
Its distinguishing feature is that it is condition-neutral. The questions ask about difficulty in daily life without reference to the cause, so the same instrument can be used across mental, physical, neurological and substance-use conditions and the resulting scores are directly comparable. It is anchored to the International Classification of Functioning, Disability and Health, and it is a measure of disability rather than of illness severity — a patient can have florid symptoms and modest functional impairment, or the reverse, and WHODAS 2.0 will register only the second axis.
02 - Origin & purpose
Where it comes from.
WHO began developing disability assessment instruments in the late 1980s, and WHODAS 2.0 was published in 2010 in the manual edited by Üstün, Kostanjsek, Chatterji and Rehm. It was built from the ground up as a cross-cultural instrument: item development drew on linguistic and cognitive testing in nineteen countries, and field trials were conducted across a wide range of settings, so that the same construct would be recognisable in very different health systems and cultures. It replaced the older WHODAS II and was designed to sit directly on the ICF framework that WHO had adopted in 2001.
Its prominence in psychiatry comes from DSM-5. When the American Psychiatric Association retired the Global Assessment of Functioning in 2013 — on the grounds that the GAF conflated symptom severity with functional impairment and had poor inter-rater reliability — it named WHODAS 2.0 as the recommended replacement, and included the 36-item self-administered version in Section III as an emerging measure. That decision left many services with a gap: the GAF was withdrawn without WHODAS 2.0 being widely adopted in its place, and a great deal of routine practice simply continued to use the GAF informally. WHO issues the instrument in 36-item and 12-item versions, each available in self-report, proxy and interviewer-administered forms, plus a 12+24-item hybrid.
03 - Scoring & cutoffs
How scoring works.
Every item is rated none, mild, moderate, severe, or extreme/cannot do, scored 1 to 5 on the form. Two methods then convert responses into a total, and they are not interchangeable. Simple sum adds the item scores directly; it requires no software and is adequate for most clinical purposes, though WHO describes it as the less refined option. The IRT-based method, computed with the syntax supplied in the WHO manual, weights each item by its difficulty and by the level of severity endorsed, which handles the fact that reporting extreme difficulty with self-care means something different from reporting extreme difficulty with community participation. Both are then rescaled to a 0–100 metric where 0 is no disability and 100 is full disability.
There is no single validated cutoff for caseness, and WHODAS 2.0 was not designed to have one. The score is interpreted as a continuous measure of functional difficulty — against population norms, against the patient's own earlier score, or against the domain profile, which is often more useful clinically than the total. Domain scores can be reported separately and frequently should be: a total of 30 tells you much less than knowing whether it is driven by mobility or by getting along with others. If you intend to compare scores over time or across services, fix the version, the respondent (self, proxy or interviewer) and the scoring method first, because changing any of the three changes the number. Full scoring instructions and the SPSS/Stata syntax are in the WHO manual, which is free to download.
04 - Validation evidence
How well it performs.
WHODAS 2.0 has one of the broader evidence bases of any functioning measure, precisely because it is condition-neutral and was designed multinationally. Global internal consistency is reported at Cronbach's alpha 0.86, with subscale alphas ranging from about 0.82 to 0.98, and one-week test-retest reliability of 0.98 for the full instrument. Concurrent validity with other established measures of disability and health-related quality of life is consistently good.
The 2017 international systematic review by Federici and colleagues, covering the published literature to that point, concluded that WHODAS 2.0 is a valid and reliable self-report instrument for assessing disability, with a factor structure that broadly supports the six-domain model, and that it performs across a wide range of clinical populations. Subsequent reviews have extended this to specific populations including older adults and WHODAS-Child. Two limits on the evidence are worth stating plainly. First, the six-domain factor structure is supported but not uniformly replicated, and some studies find the participation domain behaves less cleanly than the others. Second, minimal important change values have been derived in particular populations — low back pain, for instance — but there is no single widely accepted MCID that transfers across conditions, so change scores should be interpreted with care.
α = 0.86
Internal consistency
Global internal consistency; subscale alphas approximately 0.82–0.98. WHO 2010 manual.
0.98
Test–retest reliability
One-week retest for the full 36-item instrument. WHO 2010 manual.
Valid and reliable across conditions
Evidence base
Conclusion of the 2017 international systematic review (Federici et al.); condition-neutral design supports comparison across diagnoses.
None
Cutoff for caseness
Interpreted continuously against norms, domain profile or the patient's own baseline. No single MCID transfers across conditions.
Retired from DSM-5 but still in wide informal use. A single clinician rating that mixes symptoms and functioning; fast, familiar, and unreliable between raters.
WHODAS 2.0
36 or 12 items, 0–100
5–20 min
When you want functioning measured separately from symptoms, comparably across diagnoses, with a self-report or proxy option. The DSM-5 successor to the GAF.
When you need a health-utility index for economic evaluation rather than a clinical picture of functioning. Licensed.
Sheehan Disability Scale
3 items, 0–30
1–2 min
A very brief measure of impairment in work, social life and family life, widely used in psychiatric trials. Far less granular than WHODAS 2.0.
06 - When to use it
Right tool, wrong tool.
Reach for it when
-You need to measure functioning separately from symptom severity — for example where a patient reports few symptoms but is not working, or the reverse.
-You want a measure that is comparable across diagnoses and across physical and mental health.
-You are looking for the DSM-5-endorsed replacement for the GAF.
-You want a domain profile, not just a total, to guide where rehabilitation effort should go.
-You need a proxy or interviewer-administered option because the patient cannot self-report.
-You are working internationally and need an instrument with validated translations and cross-cultural development.
Reach for something else when
-You need a quick clinician impression of illness severity — use the CGI-S.
-You want to screen for a disorder. WHODAS 2.0 is not diagnostic and has no caseness cutoff.
-You need a health-utility value for a cost-effectiveness analysis — use the EQ-5D.
-You want to embed the instrument in software or an electronic record without first obtaining a WHO licence.
-You need a single number to compare against historical GAF scores; the two are not convertible.
-The patient is a child — use WHODAS-Child rather than the adult version.
07 - Confidence & precision
Reading the score with care.
Reliability is high: an alpha of 0.86 and a one-week retest coefficient of 0.98 are strong figures for a self-report functioning measure, and the 0–100 metric is fine-grained enough to register clinically meaningful movement. The caution is not about precision but about comparability. A WHODAS 2.0 score is only interpretable alongside three pieces of context — which version produced it (36-item or 12-item), who answered (patient, proxy or interviewer), and which scoring method was applied (simple sum or IRT) — and changing any one of them shifts the number without any change in the patient. Fix all three at baseline and hold them constant. Beyond that, no minimal important change value transfers reliably across conditions, so where a threshold matters, use one derived in a comparable population, and treat small movements in the total as less informative than a consistent shift in a domain profile.
08 - Limitations
What it cannot tell you.
- Licensed. Clinicians may reproduce it for their own patients, but electronic use requires written permission from WHO, which puts it out of reach of most digital tools without a licensing agreement.
- No cutoff for caseness and no universally applicable minimal important change, so it supports description and monitoring better than it supports thresholds.
- Self-report, so it is influenced by mood, insight and expectation; a depressed patient may under-rate their functioning and a patient with impaired insight may over-rate it. The proxy version helps but introduces a different bias.
- Version proliferation. The 36-item, 12-item, 12+24 hybrid, self, proxy and interviewer forms produce scores that are not directly interchangeable.
- Two official scoring methods that give different numbers, with the more refined one requiring the WHO syntax.
- The six-domain factor structure is broadly but not uniformly supported, with the participation domain the least consistent.
- Thirty-day recall period, which is long enough for recall error and too long to capture rapid change on an inpatient unit.
- Adoption in routine psychiatric practice has been patchy despite the DSM-5 endorsement, so scores may be unfamiliar to colleagues who still think in GAF terms.
09 - Licensing, explained
How licensing works.
Copyright World Health Organization. WHO permits clinicians to reproduce the instrument for use with their own patients without seeking permission, but states that any other use — explicitly including electronic use — requires written permission from WHO, obtained through WHO Classifications licensing. Embedding WHODAS 2.0 in a website, app or electronic record therefore requires a licence. For that reason we publish scoring and interpretation guidance here but do not reproduce the items and do not offer a calculator. The official forms and the scoring manual are available free from WHO.
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.