Home › Outcome measures › Neck
Outcome measures for neck pain
Neck questionnaires that meet our evidence standard, with their reliability and the smallest change that matters. How these were chosen.
Figures describe groups, not one patient. Each figure comes from a particular group of patients and a particular version and language of the questionnaire. A change in one patient’s score that is smaller than the measurement error may be noise rather than real change.
At a glance
- Two questionnaires meet the standard: the Neck Disability Index for disability, and a 0 to 10 pain rating scale for pain intensity.
- Neck Disability Index: the most studied neck questionnaire.2 A 2024 meta-analysis pooled its reliability (intraclass correlation coefficient, ICC) at 0.91,1 but a 2012 review found its only methodologically sound reliability study gave reliability (ICC) 0.50.2
- Smallest change that matters on the Neck Disability Index: estimates range widely, from 5% to 33%; the 2024 review puts it at around 15% (7.5 of 50 points), about the same as its minimal detectable change, so a change of that size may still be measurement error.1
- Pain 0 to 10 (figures reported together with the visual analogue scale): an important change of 1.5 to 2.5 points, but a minimum detectable change of 2.6 to 4.1 points, so a change of 2 or 3 points may be measurement error.4
- The other neck-specific questionnaires in a 2012 review had no methodologically sound studies of their total-score reliability, so they are listed under ‘Reviewed but not included’.2
How these were chosenThe rule for which questionnaires are listed, and how licensed ones are handled.
To be listed here, a questionnaire must be supported by a systematic review of its measurement properties in people with neck pain, or be recommended in an international core outcome set for neck pain (a full recommendation, not interim use), and its reliability and smallest important change must be published. Where a review grades the quality of the reliability evidence, it must be at least low quality. A questionnaire built for one condition (for example, rotator cuff disease) is judged within that condition, and its entry says so. Questionnaires that need a licence are named and described, with a link to the official copy where the owner allows it; their questions are not reproduced. Open any heading for the figures and sources.
The same rule is used on every region page; see the outcome measures home page.
How to read the figuresWhat reliability, the smallest change that matters and measurement error mean, and how the evidence is rated.
Reliability (intraclass correlation coefficient, ICC): how closely the scores agree when a patient whose condition has not changed fills in the questionnaire twice. The closer to 1, the closer the agreement; the quality criteria used in COSMIN reviews (COSMIN: the consensus-based standards for the selection of health measurement instruments) count 0.70 or more as sufficient.2
Smallest change that matters (minimal important change, MIC): the smallest change in one patient’s score that patients see as a real improvement or worsening. Some studies call a similar figure the minimal clinically important difference (MCID) or clinically important difference; this page uses the term each source uses. Estimates vary between studies, so treat them as a guide.1,3
Measurement error (smallest or minimal detectable change, SDC or MDC): the smallest change that is larger than the questionnaire’s own error. The same quality criteria count measurement error as sufficient only when the smallest important change is larger than the smallest detectable change.2
Evidence rating on this page: Strong means a systematic review found high-quality (or strong) evidence that reliability is sufficient, no review contradicts it, and the published smallest detectable change is smaller than the smallest important change. Moderate means a systematic review, or a national evidence review, rated reliability as sufficient on moderate-quality evidence, or several reviews consistently report sufficient reliability without formally grading the evidence. Where reviews or studies disagree (on reliability or on measurement error), the rating is Moderate and the disagreement is stated. Low means, with no review or study disagreeing, that reliability or measurement error in the main patient group on this page (here, adults with neck pain) was low quality or insufficient (for example, the smallest detectable change is as large as, or larger than, the smallest important change), or that the support comes only from a narrower group of patients. If the Low test is met, the rating is Low even where reliability alone would be Moderate. A questionnaire built for one condition (for example, rotator cuff disease) is rated within that condition, and its entry says so. Each questionnaire gives the reason for its rating.
Each figure names its source by number (see Sources) and says whether it was checked against the full paper or only the abstract.
Questionnaires that meet the standard (2)One for disability and one for pain intensity.
Neck Disability Index (NDI)The most studied neck questionnaire; reviews disagree on its reliability and measurement error, and estimates of the smallest change that matters vary widely. Licensed; generally free for clinicians.
Evidence: Moderate (the reviews disagree on reliability and on measurement error). A 2024 meta-analysis of 79 studies pooled reliability (ICC) at 0.91, but its minimal detectable change (about 15%) was about the same as its smallest important change (about 15%);1 a 2012 COSMIN review found limited evidence that reliability is inadequate, from its only methodologically sound study, and rated measurement error indeterminate.2
| What it measures | Activities of daily living in people with neck pain (self-rated disability). It was derived from the Oswestry Disability Index.2 (source 2, full paper) |
|---|---|
| Tested in | People with neck pain: 79 studies in the 2024 meta-analysis.1 The 2012 review covered non-specific neck pain, whiplash-associated disorders and cervical radiculopathy.2 (source 1, abstract source only; source 2, full paper) |
| Items and scoring | 10 items, each with 6 response options scored 0 to 5; total 0 to 50.2 The total is doubled to give a percentage score from 0 to 100.5 (source 2, full paper; source 5, full paper) |
| Free or licensed | Copyright; distributed by Mapi Research Trust. Access is generally free for students, clinicians in practice and unfunded researchers; fees and a licence agreement may apply to health institutions, funded projects and commercial users.5,8 Official copy and licence (Mapi Research Trust ePROVIDE) (source 5, full paper; source 8, full page) |
| Reliability | Reliability (ICC) 0.91 (95% confidence interval 0.90 to 0.93), pooled across studies in a 2024 meta-analysis.1 (source 1, abstract source only) Retest reliability 0.48 to 0.99 across studies in a 2009 review, which mixed different types of correlation coefficient (the type is often not stated); the two large high-quality studies gave lower ICCs (0.50 and 0.68), and the review noted that longer retest intervals and how ‘stable’ is defined can affect the estimate.3 (source 3, full paper) Reliability (ICC) 0.50 in the only methodologically sound study in a 2012 COSMIN review, rated limited evidence of inadequate reliability.2 (source 2, full paper) |
| Smallest change that matters (MIC) | Minimal clinically important difference 5% to 33% across studies, around 15% (7.5 of 50 points) overall, in the 2024 meta-analysis.1 (source 1, abstract source only) Clinically important difference 5 points (neck pain of musculoskeletal origin) to 7 points (cervical radiculopathy) out of 50, and 19 out of 100 (about 9.5 out of 50) in a third study, in the 2009 review, which called these estimates sparse and inconsistent.3 (source 3, full paper) The 2012 review judged the estimates (3.5, 7.5 and 9.5 out of 50) too diverse to give one value.2 (source 2, full paper) |
| Measurement error | Minimal detectable change 3% to 27% across studies, around 15% overall, in the 2024 meta-analysis.1 (source 1, abstract source only) Minimal detectable change commonly taken as about 5 out of 50, but 1.7 to 10.5 across studies (10.2 in cervical radiculopathy), in the 2009 review.3 (source 3, full paper) The 2012 review rated measurement error indeterminate, because no single smallest important change could be set.2 (source 2, full paper) |
The 2012 review, which covered the original English versions, found strong evidence for internal consistency, moderate evidence for structure and responsiveness, and limited positive evidence for content validity.2 (source 2, full paper)
Numerical pain rating scale (NPRS, 0 to 10)Reliability (ICC) 0.58 to 0.93 in neck pain, but its minimum detectable change is larger than the change that matters, so measurement error is a problem.
Evidence: Low. A systematic review of 46 papers found high-to-moderate quality evidence on test-retest reliability (ICC 0.58 to 0.93), but its minimum detectable change (2.6 to 4.1 points) is larger than the important change (1.5 to 2.5 points);4 by the COSMIN quality criteria, this means measurement error is insufficient.2 The same scale is rated Low on the low back page.
| What it measures | Pain intensity.4 (source 4, abstract source only) |
|---|---|
| Tested in | Studies in which more than half the participants had neck pain (46 papers).4 (source 4, abstract source only) |
| Items and scoring | One question, scored 0 to 10.6 (source 6, full document) |
| Free or licensed | Free: the Australian Commission on Safety and Quality in Health Care lists the 0 to 10 numerical pain rating scale as free and in the public domain.6 (source 6, full document) |
| Reliability | Reliability (ICC) 0.58 to 0.93, retested 7 hours to 4 weeks apart (both scales reported together).4 (source 4, abstract source only) |
| Smallest change that matters (MIC) | Clinically important difference 1.5 to 2.5 points, moderate evidence (both scales reported together).4 (source 4, abstract source only) |
| Measurement error | Minimum detectable change 2.6 to 4.1 points, larger than the important change (both scales reported together).4 (source 4, abstract source only) |
The review’s abstract reports these figures for the 0 to 10 scale and the visual analogue scale together; it does not separate them. The authors say responsiveness and the interpretation of change scores need further work.4 (source 4, abstract source only)
Not listed on the shoulder, elbow, wrist and hand, hip, knee, ankle and foot or whole-body pages; see why on the shoulder, elbow, wrist and hand, hip, knee, ankle and foot and whole-body pages.
Reviewed but not includedOther questionnaires we checked, and why they are not listed above.
| Neck Pain and Disability Scale (NPDS) | Reason not included: no methodologically sound studies of its reliability or measurement error, and no information on its smallest important change.2 (source 2, full paper) |
|---|---|
| Neck Bournemouth Questionnaire (NBQ) | Reason not included: no methodologically sound studies of its reliability or measurement error, and its smallest important change has not been studied.2 (source 2, full paper) |
| Northwick Park Neck Pain Questionnaire (NPQ) | Reason not included: no methodologically sound studies of its reliability or measurement error.2 (source 2, full paper) |
| Whiplash Disability Questionnaire (WDQ) | Reason not included: no methodologically sound studies of its reliability or measurement error.2 (source 2, full paper) |
| Copenhagen Neck Functional Disability Scale (CNFDS) | Reason not included: no methodologically sound studies of its reliability.2 (source 2, full paper) |
| Core Neck Questionnaire (CNQ) | Reason not included: the reliability of its total score has not been studied.2 (source 2, full paper) |
| Core Whiplash Outcome Measure (CWOM) | Reason not included: no methodologically sound studies of its reliability.2 (source 2, full paper) |
| Visual analogue scale for pain (VAS) | Reason not included: it meets the same standard, from the same review, as the 0 to 10 scale, which reported both scales together. Leaving it out is our editorial choice, so the same figures are not shown twice; the figures above apply to both, with the visual analogue scale read on a 0 to 10 scale (0 to 10 cm); on a 0 to 100 mm scale, multiply by 10.4 (source 4, abstract source only) Left out of the low back page because of weak evidence there; see why there. |
| DASH and QuickDASH (Disabilities of the Arm, Shoulder and Hand, full and short forms) | Reason not included: in neck pain, a 2017 review found only limited, low-quality positive evidence of validity, reported no positive reliability evidence and unknown responsiveness; it called them promising but in need of further evaluation.7 (source 7, abstract source only) Listed on the shoulder, elbow and wrist and hand pages, where the evidence in those upper-limb conditions supports it. |
| Neck and Upper Limb Index (NULI) | Reason not included: a 2017 review found limited, low-quality positive evidence of its reliability and validity in neck pain, but the review’s abstract reports no smallest important change for it; the authors called it promising but in need of further evaluation.7 (source 7, abstract source only) |
All outcome measures · Other regions: Low back · Shoulder · Elbow · Wrist and hand · Hip · Knee · Ankle and foot · Whole body
These are findings from the cited studies, not instructions. The page is not a complete review of every outcome measure and is not a clinical protocol. Questionnaires that need a licence are described here but not reproduced; use the official copy and follow its conditions of use. The page is written for registered health practitioners and students, who remain responsible for their own clinical decisions; it is not advice for patients. See the terms of use.
Last checked against the cited sources: 13 November 2026.
Get the book for the examination itself
Rule It In, Rule It Out works through the examination of each body region: the red flags to screen first, then the special tests, with the published figures for each test and how to read them.
Look inside the book: free sample pages
Coming November 2026: 330 pages, paperback and PDF. Launch offer: subscribers get the PDF for A$39 instead of A$49, for the first two weeks after publication. Be first to know at launch
Sources (8)Each figure above is cited to a published source; figures checked only against an abstract are labelled abstract source only.
- Saltychev M, Pylkäs K, Karklins A, Juhola J. Psychometric properties of neck disability index – a systematic review and meta-analysis. Disabil Rehabil. 2024;46(23):5415-5431. Abstract source only. Link
- Schellingerhout JM, Verhagen AP, Heymans MW, Koes BW, de Vet HC, Terwee CB. Measurement properties of disease-specific questionnaires in patients with neck pain: a systematic review. Qual Life Res. 2012;21(4):659-670. Full paper. Link
- MacDermid JC, Walton DM, Avery S, et al. Measurement properties of the neck disability index: a systematic review. J Orthop Sports Phys Ther. 2009;39(5):400-417. Full paper. Link
- Modarresi S, Lukacs MJ, Ghodrati M, et al. A systematic review and synthesis of psychometric properties of the Numeric Pain Rating Scale and the Visual Analog Scale for use in people with neck pain. Clin J Pain. 2022;38(2):132-148 (online 2021). Abstract source only. Link
- Tharin F, Reinmann A, Bruyneel AV, Demoulin C. Neck Disability Index: mesure de l’impact fonctionnel des douleurs cervicales. Kinesither Rev. 2025;25(281):39-43. Full paper. Link
- Australian Commission on Safety and Quality in Health Care. Patient-Reported Outcome Measure Recommendations for Low Back Pain. Sydney: ACSQHC; September 2024. Tables 1 and 2. Full document. Link
- Alreni ASE, Harrop D, Lowe A, et al. Measures of upper limb function for people with neck pain. A systematic review of measurement and practical properties. Musculoskelet Sci Pract. 2017;29:155-163. Abstract source only. Link
- Mapi Research Trust. Neck Disability Index (NDI): ePROVIDE instrument page, conditions of use (read 13 November 2026). Full page. Link
One study, one test, once a month
Rule It In is a free evidence letter for musculoskeletal clinicians: one recent study explained, one examination test with its real numbers. From Trokos, publisher of Rule It In, Rule It Out.
Free when you confirm your email: a one-page PDF of the strongest special tests to rule in and rule out, with tests that carry a caution flagged separately.
Launch offer for subscribers: the PDF edition of Rule It In, Rule It Out for A$39 instead of A$49, for the first two weeks after publication. Sign up free to get the code on launch day.
By subscribing you agree to receive Rule It In by email, about once a month. You will get a confirmation email first, and nothing is sent until you click it. Your address is stored by Kit (Kit.com, United States) and used only for this newsletter and the subscriber launch offer; Kit records which issues are opened and which links are clicked. Unsubscribe with one click in any issue. Privacy.
