Clinical library · Outcome measures
DASH and QuickDASH
A single measure for the whole upper limb, treating arm, shoulder and hand as one functional unit. The pooled minimal clinically important difference is about 11 points — and the measurement error is about 9, which leaves less room than it appears.
In one line. Thirty items (DASH) or eleven (QuickDASH) on symptoms and ability to perform upper-limb activities, scored 0 (no disability) to 100. Optional work and sports/performing arts modules exist and are scored separately.
Its defining design choice is that it does not care which joint is affected. That is useful when a patient has more than one problem in the same limb, and it is the reason the instrument is less sensitive than a joint-specific measure when only one joint is involved.
Use a range, not a point. The 2024 meta-analysis pooled 17 MCID estimates from 12 studies and 1,677 patients. Individual estimates ranged from 8.3 to 18.0 DASH points and 8.0 to 18.1 QuickDASH points; pooled MCIDs were 11.00 (95% CI 8.59 to 13.41) and 11.97 (9.60 to 14.33). Because the pooled MDC90 was about 9 points for both, the authors propose working ranges of 12 to 14 DASH points and 12 to 15 QuickDASH points — the lower bound being the first value above measurement error. [1]
The numbers you actually need
| Property | Value | Source and caveat |
|---|---|---|
| Pooled MCID | DASH 11.00 (95% CI 8.59 to 13.41); QuickDASH 11.97 (9.60 to 14.33) | 12 studies, 1,677 patients, I-squared 0%. Great methodological heterogeneity in how the primary studies calculated MCID [1] |
| Pooled MDC90 | DASH 9.04; QuickDASH 9.03 | Same meta-analysis [1] |
| Recommended working range | DASH 12 to 14; QuickDASH 12 to 15 | Lower bound is the first measure above the pooled MDC90 [1] |
| Test-retest reliability | ICC 0.93 (DASH); 0.91 (QuickDASH) | Only 30 patients in that reliability sample [2] |
| Single-cohort MCID | DASH 10.83 (sensitivity 82%, specificity 74%); QuickDASH 15.91 (79%, 75%) | Proposed as the lower boundary of a range, with 15 and 20 points from the DASH website as the upper [2] |
| Minimal important change, shoulder disorders | QuickDASH 9.1 at 3 months; 13.6 at 6 months | Danish version, primary care physiotherapy; ROC AUC above 0.70 at both points [4] |
| QuickDASH equivalence | Alpha 0.92 or above; ICC 0.94 or above | Eleven items; the concept-retention version was closest to the full DASH [5] |
| Responsiveness (QuickDASH) | Strong negative evidence | Due to lower correlations with global estimates of change, in a best-evidence synthesis of 14 studies [3] |
What it measures
Symptoms and functional limitation across the whole upper extremity, on the explicit premise that the limb functions as a single unit. A patient scores the same items whether the problem is at the shoulder, elbow or hand.
The QuickDASH was derived by three different item-reduction approaches, all of which produced comparable measurement properties. The version adopted was the one built by retaining clinically sensible content rather than by purely statistical selection, and it ranked highest for similarity to the original. [5]
Where it misleads
1. The MCID and the measurement error are uncomfortably close
A pooled MCID of about 11 points sits only about 2 points above a pooled MDC90 of about 9. [1] That is why the meta-analysts propose a range starting at 12 rather than quoting the pooled estimate as a threshold: below about 9 points you cannot distinguish change from error at all, and between 9 and 12 you are in a zone where "important" and "detectable" are barely separable.
2. The QuickDASH has strong negative evidence for responsiveness
This is the finding least likely to be quoted by people recommending the short form. A best-evidence synthesis of 14 studies found strong positive evidence for reliability and hypothesis-testing validity, moderate positive evidence for structural validity, and strong negative evidence for responsiveness, because of lower correlations with global estimates of change. [3] It also found that information on cross-cultural adaptations is lacking or of poor methodological quality.
That does not mean the QuickDASH cannot detect change — a shoulder cohort found adequate ability to measure change over 3 and 6 months, with areas under the curve above 0.70. [4] It means the synthesised evidence across studies is unfavourable, and the two findings should be held together rather than one quoted alone.
3. A single-joint problem may be better served by a single-joint measure
The whole-limb design that makes this instrument useful in multi-site problems dilutes it when only one joint is involved: items about the other joints contribute little signal. Where the question is specifically shoulder function, the shoulder-specific alternative in the same responsiveness study had a much larger minimal important change relative to its scale, which reflects a different sensitivity profile. [4]
4. The reliability figures rest on small samples
The frequently quoted ICCs of 0.93 and 0.91 come from a reliability sample of 30 patients. [2] They are consistent with the wider literature, [3][5] but they are not a large-sample estimate.
What the evidence supports — and what it does not
Supported
- A working MCID range of 12 to 14 DASH points and 12 to 15 QuickDASH points. [1]
- High test-retest reliability for both versions. [2][5]
- The QuickDASH as an acceptable substitute for the full DASH on scores and properties. [5]
- Strong evidence for reliability and hypothesis-testing validity of the QuickDASH. [3]
- Measuring change over 3 to 6 months in shoulder disorders in primary care physiotherapy. [4]
Not supported
- Treating any change below about 9 points as real. That is the pooled MDC90. [1]
- Quoting a single MCID. Primary estimates ranged 8.3 to 18.0. [1]
- Assuming the QuickDASH is as responsive as the DASH. Strong negative evidence. [3]
- Relying on cross-cultural adaptations without checking their own validation quality. [3]
How certain is this?
Evidence grade: Moderate.
The MCID meta-analysis is the strongest item: 12 studies, 1,677 patients, and I-squared of 0% for both pooled estimates, meaning the studies agreed closely once pooled. Its own caveat is great methodological heterogeneity in how the primary studies derived their MCIDs — consistency of result despite inconsistency of method. [1]
The responsiveness picture is genuinely mixed and this page reports both sides. The best-evidence synthesis rates responsiveness negatively across 14 studies; [3] a subsequent single-cohort study in shoulder disorders found adequate responsiveness. [4] The synthesis carries more weight; the cohort is more recent and closer to physiotherapy practice.
The reliability estimates come from a 30-patient sample in the most-cited MCID study [2] and from the development paper. [5]
What would change the grade: responsiveness data in physiotherapy-managed cohorts, and validated cross-cultural versions, including for Indian languages, where the review found information lacking or of poor quality. [3]
Common questions
What change should I call clinically important?
Use the range: 12 to 14 DASH points, or 12 to 15 QuickDASH points. [1] The lower bound is deliberately set just above the pooled MDC90 of about 9 points, so anything smaller cannot be distinguished from measurement error. Pooled MCIDs were 11.00 and 11.97 respectively.
DASH or QuickDASH?
The QuickDASH gives comparable scores and properties with 11 items instead of 30, with alpha of 0.92 or above and ICC of 0.94 or above. [5] The caution is responsiveness: a best-evidence synthesis found strong negative evidence for it, driven by lower correlations with global change ratings. [3] If your primary purpose is detecting change rather than describing status, that is an argument for the full version.
Can I use it when only the shoulder is affected?
You can, and it has been validated for that: in shoulder disorders managed in primary care physiotherapy, the minimal important change was 9.1 points at 3 months and 13.6 at 6 months, with areas under the curve above 0.70. [4] Just be aware that items about the elbow and hand add noise rather than signal in a single-joint problem.
Why do published MCIDs vary so much?
Method, mostly. The meta-analysis identified great methodological heterogeneity in how primary studies calculated the MCID, with individual estimates from 8.3 to 18.0 points — yet the pooled estimates had an I-squared of 0%. [1] The scatter is in the derivation, not in the instrument.
Is there a validated version in Indian languages?
Not one this page can point to. The systematic review found that information about the measurement properties of cross-cultural adaptations is still lacking, or that the available information is of poor overall methodological quality. [3] If you use a translated version, check its own validation paper rather than inheriting the English psychometrics.
References
- Galardini L, Coppari A, Pellicciari L, et al. Minimal Clinically Important Difference of the Disabilities of the Arm, Shoulder and Hand (DASH) and the Shortened Version of the DASH (QuickDASH) in People With Musculoskeletal Disorders: A Systematic Review and Meta-Analysis. Physical Therapy. 2024 May 1;104(5):. doi:10.1093/ptj/pzae033 PMID 38438144 Meta-analysis
- Franchignoni F, Vercelli S, Giordano A, et al. Minimal clinically important difference of the disabilities of the arm, shoulder and hand outcome measure (DASH) and its shortened version (QuickDASH). Journal of Orthopaedic & Sports Physical Therapy. 2014 Jan;44(1):30–9. doi:10.2519/jospt.2014.4893 PMID 24175606 Cohort study
- Kennedy CA, Beaton DE, Smith P, et al. Measurement properties of the QuickDASH (disabilities of the arm, shoulder and hand) outcome measure and cross-cultural adaptations of the QuickDASH: a systematic review. Quality of Life Research. 2013 Nov;22(9):2509–47. doi:10.1007/s11136-013-0362-4 PMID 23479209 Systematic review
- Budtz CR, Andersen JH, de Vos Andersen NB, et al. Responsiveness and minimal important change for the quick-DASH in patients with shoulder disorders. Health and Quality of Life Outcomes. 2018 Dec 10;16(1):226. doi:10.1186/s12955-018-1052-2 PMID 30526622 Responsiveness study
- Beaton DE, Wright JG, Katz JN. Development of the QuickDASH: comparison of three item-reduction approaches. Journal of Bone and Joint Surgery. 2005 May;87(5):1038–46. doi:10.2106/JBJS.D.02060 PMID 15866967 Instrument development study
About this resource
Using this in clinic
Every figure here is traceable to its source.
Every threshold on this page is given alongside the measurement error it has to clear, because for this instrument the two are only a couple of points apart. Where a value could not be verified against the paper it came from, it is not on this page, and the omission is stated rather than filled with a number from a secondary source.