Free SPADI calculator. Score all 13 pain and disability items in about two minutes, apply the official N/A pro-rating rule, and see MCID/MDC context before calling a change real.
The Shoulder Pain and Disability Index (SPADI) is a 13-item patient-reported outcome measure that assesses shoulder pain and disability over the past week. It takes most patients about two to three minutes to complete, is free to use, and includes separate pain and disability subscales — one of the few shoulder outcome measures structured that way. It has kept its place in clinics since Roach and colleagues first published it in 1991.
Despite its simplicity, SPADI scoring is not always presented consistently online, particularly when calculating the overall total score or handling not-applicable responses. Score it correctly below, then see how the two total-score formulas differ, how to handle N/A responses, and what the evidence on interpretation will and will not support.
The SPADI is a self-administered questionnaire measuring shoulder pain and shoulder-related disability over the past week. Roach, Budiman-Mak, Songsiridej and Lertratanakul published the original version in Arthritis Care & Research in 1991, using visual analogue scales similar in format to a Visual Analog Scale (VAS). Williams, Holleman and Simel produced the eleven-point numeric rating scale version in 1995 — the same 0-to-10 format used on a Numeric Pain Rating Scale — so the SPADI could be administered by telephone. Agreement between the two formats is strong (ICC 0.86), and the numeric version is what essentially everyone uses today; Shirley Ryan AbilityLab states plainly that the original visual analogue version "is not commonly used."
The SPADI has been validated across adhesive capsulitis, rotator cuff pathology, shoulder arthroplasty, rheumatoid arthritis with shoulder involvement, and mixed primary-care shoulder pain, and it has been translated into 21 languages across 26 translations.
A few wording variants exist across editions and are not errors: UK and Australian forms say "jumper" rather than "pullover sweater," some forms add the metric conversion to the 10-pound carrying item, and some label the disability anchor "unable to do" rather than "so difficult it required help."
Each subscale is scored as a percentage of its own maximum:
Pain % = (sum of the 5 pain items ÷ 50) × 100
Disability % = (sum of the 8 disability items ÷ 80) × 100
The total — and the formula conflict nobody flags
Two methods are in circulation for the overall total, and they do not agree:
These are not equivalent, and they diverge as the two subscales separate. A patient with a pain raw score of 25/50 (50%) and a disability raw score of 20/80 (25%) scores 34.6 by method A and 37.5 by method B.
Standardize on method A, the formula printed on the instrument. Then do the thing that actually protects you: state your method in the documentation, and report both subscale percentages alongside the total.
For this calculator, we use Method A: Total SPADI = (sum of all answered items ÷ maximum possible score for those items) × 100. The pain and disability subscale percentages are also reported separately, so it's clear whether the overall score is being driven primarily by pain or by functional limitation.
The N/A rule most calculators ignore
The SPADI permits a "not applicable" response and pro-rates for it. The rule as originally stated: a patient may mark one item per subscale as not applicable and it is omitted from the total; if more than two items are marked non-applicable, no score is calculated. In practice, reduce the denominator by 10 for each omitted item — one skipped pain item means dividing by 40, not 50. Sources differ slightly on the cut-off ("more than two," "two or more," or Shirley Ryan AbilityLab's "two-thirds of items per subscale must be answered"). For this calculator, the pro-rating rule follows the original SPADI instrument instructions: at most one pain item and two disability items may be marked not applicable; beyond that, no score is calculated.
All SPADI scores run 0 (best) to 100 (worst).
One further caution: some clinical forms also display a US Medicare G-code severity mapping (0 = CH, 1–19 = CI, 20–39 = CJ, and so on). These administrative categories should not be interpreted as validated SPADI clinical severity bands.
Two thresholds answer two different questions. MDC (minimum detectable change) is the smallest change exceeding measurement error — below it, you cannot tell improvement from noise. MCID (minimal clinically important difference) is the smallest change the patient considers worthwhile.
The SPADI has generally demonstrated good reliability and responsiveness, although reported values vary substantially between studies and populations.
The disability subscale does not achieve interval-level measurement. The largest structural study of the SPADI — Jerosch-Herold et al., 1,030 physiotherapy referrals — used Rasch analysis and found that while the pain subscale achieved acceptable unidimensionality after iteration, the disability subscale showed persistent misfit from uniform differential item functioning. The authors concluded the subscales should be treated separately and that the disability subscale is not suitable for interval-level measurement in clinical trials. That directly undercuts the validity of the single combined total — which is the headline number every calculator reports. Report the subscales.
Clinical takeaway: the SPADI total provides a useful summary, but the pain and disability subscales should also be reviewed. They can show whether a change in the total score reflects improvement in pain, function, or both.
Differential item functioning by age and sex. The pain subscale shows DIF by age and gender, meaning identical scores may not represent identical severity across demographic groups. The item content offers an obvious explanation: "removing something from your back pocket" and "putting on an undershirt or pullover sweater" are culturally and generationally specific in ways that a 1991 sample would not have surfaced.
A ceiling effect for high-demand shoulders. Because the disability items focus primarily on basic activities of daily living, the SPADI may be less sensitive to limitations in high-demand activities such as overhead sport or heavy occupational tasks. Add a Patient-Specific Functional Scale for these patients. Note that the empirical studies finding "no ceiling effect" were conducted in impaired clinical populations, where the ceiling would not surface.
Thin foundations and uneven translations. The original reliability work used 37 patients in a single, all-male sample. Across 34 studies and 26 translations, fewer than half the measurement properties tested were of adequate quality, pooled structural validity was rated insufficient, and only the Danish, Dutch and Nepali versions confirmed their properties with sound methodology (KC et al., 2021 systematic review).
It is not diagnostic. The SPADI measures impact, not pathology. It cannot distinguish a rotator cuff tear from adhesive capsulitis from pain referred from the cervical spine — which is why a high SPADI in a patient with concurrent neck symptoms should prompt a cervical screen as well.
Roy, MacDermid and Woodhouse (2009) reviewed 71 studies across four shoulder measures; a later review of 506 trials and 36,553 patients adds condition-level usage patterns. For a broader primer on choosing between outcome measures, see Spry's Patient-Reported Outcome Measures (PROMs) guide.
A sensible default for an outpatient shoulder caseload: SPADI as the primary measure, with ASES added when finer sensitivity is needed in rotator cuff work, and QuickDASH substituted when the presentation spans the whole upper limb.
SPADI vs QuickDASH: neither is universally better. Choose SPADI when the clinical problem is primarily shoulder pain and disability; choose QuickDASH when symptoms involve the broader upper extremity.
Track SPADI without burying the score in a note. Spry lets clinics capture patient-reported outcomes as structured data, track change across visits, and review outcome trends alongside the patient's clinical record. See Spry in action.
Sum the 5 pain items and divide by 50 for the pain percentage; sum the 8 disability items and divide by 80 for the disability percentage. For the total, divide the sum of all 13 items by 130 and multiply by 100. All scores run 0 (best) to 100 (worst).
Lower is better, but there is no validated cut-off for "good." The SPADI was designed to measure change, not to classify severity. Compare against the patient's own baseline and against published means for their condition — 61.3 for adhesive capsulitis, 46.7 for primary-care shoulder pain, 40 to 43 for mixed musculoskeletal caseloads.
A SPADI score of 50 means the overall score is halfway between the best possible score (0) and worst possible score (100). It does not correspond to a validated "moderate" or "severe" category, because no universally accepted SPADI severity bands have been established. Interpret it alongside the patient's diagnosis, baseline score, subscale scores, and change over time.
Most commonly cited as 8–13 points, with population-specific values from 8 (new shoulder pain) to 20 (Dutch primary care). Because the MCID is smaller than the roughly 18-point MDC in several studies, an MDC-based threshold is a more cautious reference point for judging change in an individual patient.
Approximately 18 points, with reported values from 17.0 to 21.5 across populations. A change smaller than this cannot be reliably distinguished from measurement error.
No. No authoritative source publishes validated severity cut-offs. The mild/moderate/severe bands that appear on many calculator sites are an informal web convention with no peer-reviewed origin.
The SPADI allows a "not applicable" response and pro-rates for it: reduce the denominator by 10 for each omitted item. At most one pain item and two disability items may be omitted; beyond that, no valid score can be calculated.
They are similarly responsive (AUC 0.85 vs 0.85 at six months in 767 physiotherapy patients), so choose on content fit. Use the SPADI when the problem is confined to the shoulder and pain is a primary outcome; use the QuickDASH when the elbow, wrist, or hand may also be involved or you need cross-condition comparability.
Two to three minutes to complete and about a minute to score by hand. A digital calculator scores it instantly, including pro-rating.
.avif)