Muscle, joint & pain

The SPADI: Grading Shoulder Pain and Function Separately

Save

Thirteen questions, two subscales, one number that is supposed to move. The SPADI was built in 1991 by discarding every candidate question that people answered inconsistently, or that had nothing to do with how far the shoulder could actually move. Knowing what survived that cut explains both what the form is good at and what it was never designed to tell you.

Last updated: July 2026

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

What the Shoulder Pain and Disability Index measures

The SPADI keeps two things about a painful shoulder separate: how much it hurts, and how much it stops you doing what you do. It is self-administered — the person with the shoulder fills it in, and no one measures anything — and it holds thirteen items, five forming a pain subscale and eight forming a disability subscale 1.

That split is the design, not an accident of layout. Pain and function come apart in shoulders more often than people expect. A stiff shoulder can be fairly quiet at rest and still turn getting dressed into a fifteen-minute negotiation; another can hurt viciously at three in the morning and barely interfere with the day. A single blended number would hide which of those two situations someone is in.

Self-administered means the score comes from the patient's own report rather than from an examiner's measurements. That distinction matters later, because several shoulder scores in common use are partly or entirely the examiner's.

The five pain items ask how severe the shoulder pain has been in particular circumstances. The eight disability items ask how much difficulty the shoulder causes with everyday tasks — the overhead reaching, the behind-the-back reaching, the washing and dressing and lifting that a shoulder mostly exists for 1. The questionnaire is its developers' to distribute, so the items are described here rather than reprinted.

Where the SPADI came from

It was published in 1991 by Roach, Budiman-Mak, Songsiridej and Lertratanakul, and the revealing part is what they discarded. The thirteen surviving items were reduced from a larger starting pool of candidates. A question was dropped if people answered it inconsistently when asked again a short time later, or if answers to it correlated poorly with how far the shoulder could actually move 1.

Both filters explain the shape of the finished form. The first is a stability filter: a question answered differently on Tuesday than on Monday, in someone whose shoulder has not changed, is measuring mood or wording. The second is a tethering filter: the index was deliberately kept tied to physical shoulder movement, so items floating free of range of motion did not survive.

The original derivation sample was 37 male outpatients with shoulder pain of mixed causes 1. That is a small and narrow group, and the most useful thing to know about the instrument's pedigree. Mixed causes was a deliberate choice — the SPADI was never built for one diagnosis — but a derivation cohort of that size and composition is a starting point rather than a settled account, which is why the instrument has been studied so heavily since.

How well does it hold up as a measurement?

In the development study the SPADI's internal consistency ran from 0.86 to 0.95, while its test-retest reliability ran from 0.64 to 0.66 1. Those two numbers measure different virtues, and the gap between them is the most honest thing on this page. Internal consistency asks whether the items hang together — whether the eight disability questions really are all circling the same underlying quantity. On that count the SPADI did well.

Test-retest reliability asks something harder: give the same form to the same unchanged person twice, and do you get the same answer? Internal consistency of 0.86 to 0.95, against test-retest reliability of 0.64 to 0.66 1. The second figure is modest. In practice it means a single score on a single day carries real noise, and that small differences between two administrations should not be over-read.

The developers also reported criterion validity, from high negative correlations between change in the SPADI score and change in shoulder range of motion — as measured movement improved, the reported score fell — and construct validity from a principal components factor analysis of the item structure 1. These are the original 1991 figures; a clinician quoting different reliability numbers is almost certainly drawing on the three decades of research since.

How the score is put together

Each subscale is scored on its own and then expressed as a proportion of the worst possible answer, so the pain subscale and the disability subscale both land on a 0-to-100 scale despite having different numbers of questions. The conventional total is the average of the two, reported the same way. On all three numbers, higher means worse: more pain, more difficulty.

On the SPADI, a bigger number is a worse shoulder. Several other shoulder questionnaires run the opposite way. It is worth confirming the direction on any form before reading a result off it, because a 75 is an excellent outcome on some shoulder scores and a poor one here.

The pain subscale tends to respond first to whatever calms a shoulder down; the disability subscale tends to lag, because regaining the movement and the confidence to reach overhead takes longer than losing the ache. The combined total erases that difference, which is the argument for reading all three numbers.

One limit belongs here rather than in the fine print. The development paper reports no minimal clinically important difference and no minimal detectable change for the SPADI 1. What counts as a meaningful shift in the number is a separate question, answered by separate literature.

What a SPADI score cannot tell you

It cannot tell you what is wrong. The SPADI records the consequences of a shoulder problem — how much it hurts, how much it costs you in daily function — and contains nothing that could distinguish one cause from another. Two people with an identical total can have entirely different structural problems, and the same person's total can be produced by several different combinations of answers.

The scale of what it is not distinguishing is easy to underestimate. Rotator cuff tears alone account for nearly two million visits a year in the United States 2. Many of those tears are managed without an operation, using anti-inflammatory medicines, injections and physical therapy, and most tears do not heal on their own 2. None of that shows up in a questionnaire score. A cuff tear, a stiff capsule, an irritable tendon and an unstable joint can all produce a SPADI of 60.

This is why the form is used alongside a clinical assessment rather than in place of one. Making sense of shoulder pain is a job for someone who can take a history and examine the joint. The questionnaire's contribution is the part an examination is worst at: a record of what the shoulder is doing to the person's actual week, rather than what they can reconstruct from memory at the next appointment.

Why there are so many shoulder questionnaires

Because they answer different questions and were built by different communities. The ases shoulder score, adopted by the American Shoulder and Elbow Surgeons, combines a visual analog scale for pain with a ten-item activities-of-daily-living questionnaire, each contributing half of a 0-to-100 score in which higher indicates better function and less pain 3. That is the same range as the SPADI running in the opposite direction.

The Constant-Murley score is a different animal again: a composite scored out of 100 points across pain, activities of daily living, range of motion and strength, so part of it is the patient's report and part of it is measured by a clinician, with higher scores meaning a better shoulder 4. It was designed to apply regardless of diagnosis or what the imaging shows 4. The DASH takes the widest view of all, covering symptoms and physical function across the whole upper limb rather than the shoulder alone 5. The oxford shoulder score is a fourth option in common use.

A score is unreadable without the name of the questionnaire attached to it. Ranges differ, directions differ, and some of these instruments require a clinician while others do not.

The same idea, elsewhere in the body

The SPADI belongs to a family of region-specific indices built on one shared idea: ask the patient about pain and about function, in the vocabulary of the body part in question, and score the two separately. The foot function index is the SPADI's closest relative, and not by coincidence — Budiman-Mak and Roach appear on both papers, published in the same year 16.

The FFI is a self-administered index of 23 items across three sub-scales — pain, disability and activity restriction — originally validated in 87 patients with rheumatoid arthritis, with test-retest reliability from 0.87 down to 0.69 and internal consistency from 0.96 down to 0.73 6. Read alongside the shoulder version, the resemblance is obvious: same architecture, different anatomy, and the same honest spread between the strong internal-consistency figures and the more modest stability figures.

The rest of the family names its own territory. The oswestry disability index covers the low back, the neck disability index covers the neck, and the roland-morris questionnaire offers a different route to the same back-pain question. Each was developed separately, each has its own scoring conventions, and none of them converts into any of the others.

What the family shares is a premise: what matters in a musculoskeletal problem is not only the tissue, but what the tissue is costing the person. A questionnaire is a crude instrument for that, and better than not asking.

Common questions

You do. The SPADI is self-administered, which means the entire score comes from your own answers about your own shoulder. Nobody measures your range of motion to produce it. That is a deliberate feature: it captures how the shoulder behaves across your whole week, not only how it performed during a few minutes in a clinic room.

No. It measures consequences, not causes. A tendon problem, a stiff capsule, a rotator cuff tear and an unstable joint can all produce the same total, because the questions ask about pain and difficulty rather than about anatomy. Working out the cause takes a history, an examination, and sometimes imaging — none of which a questionnaire can substitute for.

Not remotely. The ASES score, the Constant-Murley score, the Oxford Shoulder Score and the DASH are all in regular use, and they differ in range, direction and whether a clinician has to measure anything. Which one you meet usually depends on the clinic and the setting rather than on your shoulder.

Because they move independently. Pain often settles before function returns, and someone can regain most of their daily activities while a shoulder still aches at night. Blending both into one number hides which of those is happening. Keeping the two subscales visible tells a clinician which half of the problem is actually shifting.

It is one input among several, never the decision. Surgical decisions rest on the diagnosis, the structural findings, how the shoulder has responded to time and rehabilitation, and what matters to the person. A questionnaire score contributes a consistent record of the burden over time, which is genuinely useful in that conversation and not sufficient for it.

The development paper establishes no threshold for meaningful change, and its test-retest figures suggest a single score carries real noise. In practice clinicians look at direction and consistency over several administrations rather than at one gap between two numbers, and they read the pain and disability subscales separately before reading the total.

Related

Say it back

How would you explain this to someone you love?

Two or three sentences, just as you’d say it. Gale reflects back what you focused on — a mirror, not a quiz.

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

Shoulder symptoms that outrank any questionnaire

  • —Shoulder or arm pain arriving with chest pressure, shortness of breath, sweating, nausea or lightheadedness
  • —A hot, swollen, exquisitely tender shoulder joint with fever or shaking chills
  • —Inability to lift or rotate the arm at all after a fall, a dislocation or a direct blow, or a shoulder that looks visibly out of shape
  • —Numbness, progressive weakness, or a cold or pale hand on the same side as the painful shoulder

Shoulder or arm pain together with chest pressure, breathlessness, sweating or nausea can be a heart attack. Call 911 rather than driving yourself or waiting to see whether it passes.

This page explains what a shoulder questionnaire measures and how it was built. It is general education, not medical advice, and no score can identify the cause of shoulder pain or replace an assessment by a clinician who can examine you.

References

  1. 1.Roach KE, Budiman-Mak E, Songsiridej N, Lertratanakul Y (1991). Development of a shoulder pain and disability index. Arthritis Care Res. 1991;4(4):143-9. doi:10.1002/art.1790040403 ✓The SPADI's origin, authorship and construction: a self-administered index of thirteen items across two subscales, pain (five items) and disability (eight items), arrived at by item reduction from a larger pool that dropped items with poor test-retest reliability or low correlation with shoulder range of motion; the derivation cohort of 37 male outpatients with shoulder pain of mixed etiology; internal consistency 0.8604-0.9507 and test-retest reliability 0.6377-0.6552; criterion validity from high negative correlations between change in SPADI score and change in shoulder range of motion; construct validity from principal components factor analysis; and the absence of any MCID or MDC in this paper.
  2. 2.American Academy of Orthopaedic Surgeons (OrthoInfo) (2024). Rotator Cuff Tears. OrthoInfo — AAOS. link ✓That rotator cuff tears are a common cause of shoulder pain, accounting for nearly two million visits a year in the United States, that many are managed nonsurgically with anti-inflammatory medication, injections and physical therapy, and that most tears do not heal on their own — used here to show the range of causes a single questionnaire score cannot distinguish between.
  3. 3.Richards RR, An KN, Bigliani LU, et al. (1994). A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg. doi:10.1016/S1058-2746(09)80019-0 ✓The ASES Shoulder Score's structure and scoring: a patient self-evaluation combining a visual analog scale for pain with a ten-item activities-of-daily-living questionnaire, with the pain scale and the cumulative activities score each contributing half of a 0-100 total in which higher scores indicate better function and less pain.
  4. 4.Constant CR, Murley AH (1987). A clinical method of functional assessment of the shoulder. Clin Orthop Relat Res. 1987;(214):160-4. PMID 3791738 ✓That the Constant-Murley score is a composite shoulder-function measure combining subjective and objective components, scored out of 100 points across pain, activities of daily living, range of motion and strength, with higher scores indicating better shoulder function, and that the method was designed to apply irrespective of diagnosis or radiologic abnormality.
  5. 5.Hudak PL, Amadio PC, Bombardier C (Upper Extremity Collaborative Group) (1996). Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder, and hand). American Journal of Industrial Medicine. doi:10.1002/(SICI)1097-0274(199606)29:6<602::AID-AJIM4>3.0.CO;2-LThat the DASH is a self-reported measure of symptoms and physical function developed to span upper-extremity musculoskeletal disorders as a whole rather than a single joint.
  6. 6.Budiman-Mak E, Conrad KJ, Roach KE. (1991). The Foot Function Index: a measure of foot pain and disability. J Clin Epidemiol. 1991;44(6):561-70. doi:10.1016/0895-4356(91)90220-4 ✓That the Foot Function Index is a self-administered index of 23 items across three sub-scales (pain, disability and activity restriction), validated in 87 patients with rheumatoid arthritis, with test-retest reliability of total and sub-scale scores ranging 0.87 to 0.69 and internal consistency ranging 0.96 to 0.73 — and, with the SPADI paper, that Budiman-Mak and Roach co-authored both instruments in 1991.

6 sources, numbered by first appearance. General health information, not medical advice. AI-assisted editorial content — citations link their sources. Editorial policy

← Back to the health library