The Upper Extremity Functional Index, Explained

What the Upper Extremity Functional Index measures

It measures function across the whole upper limb as a single unit. Rather than asking separate questions about a shoulder, an elbow and a wrist, a region measure asks about what an arm does — reaching, lifting, carrying, dressing, working at a keyboard — and returns one number for the limb. The DASH was developed on exactly that premise, as a self-reported measure of symptoms and physical function spanning upper-extremity musculoskeletal disorders rather than one diagnosis 1Ref 1Hudak 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).The development of the DASH as a validated self-reported measure of symptoms and physical function spanning upper-extremity musculoskeletal disorders rather than a single diagnosis.. The QuickDASH, derived from it, is explicit about the design choice: it treats physical function and symptoms of the upper limb as a single region rather than a joint-specific problem 2Ref 2Beaton DE, Wright JG, Katz JN; Upper Extremity Collaborative Group (2005).Development of the QuickDASH: comparison of three item-reduction approaches.The QuickDASH's construct — physical function and symptoms of the upper limb treated as a single region rather than joint-specific — plus its 11-item count, its derivation from the 30-item DASH by comparing concept retention, equidiscriminative item-total correlation and item response theory against field data from 407 patients, and its reported psychometrics (Cronbach alpha at or above 0.92, ICCs above 0.94, construct validity r at or above 0.64 against single-item pain and function indices). Also cited for what it does not report: no MCID, no MDC, no score range, no severity cutoff..

The reasoning is practical. Arm problems rarely stay inside one joint — a stiff shoulder changes how an elbow is used, a painful wrist changes how a shoulder is loaded — so a score anchored to a single joint can miss where the disability actually lives. A region score also means one form covers every upper-limb patient who walks into a clinic, which is why departments standardise on them.

A region measure answers 'how well does this arm work', not 'how bad is this joint'. Those are different questions, and only one of them describes a person's day.

Where the index comes from: the lower-limb original

It is the upper-limb counterpart to the lower extremity functional scale, and the resemblance is deliberate rather than coincidental. That scale was developed for outpatients with lower-extremity musculoskeletal dysfunction, using 107 patients across twelve outpatient physical therapy clinics, and was constructed as a new instrument rather than adapted from an existing questionnaire 3Ref 3Binkley JM, Stratford PW, Lott SA, Riddle DL (1999).The Lower Extremity Functional Scale (LEFS): Scale Development, Measurement Properties, and Clinical Application.The Lower Extremity Functional Scale as the lower-limb instrument this one is modelled on: its origin as a purpose-built scale, its intended population of outpatients with lower-extremity musculoskeletal dysfunction, its derivation from 107 patients across twelve outpatient physical therapy clinics, test-retest reliability of R = .94, sensitivity to change exceeding the SF-36 physical function subscale, and its published MCID and MDC of 9 scale points with point-in-time measurement error of ±5.3 scale points. No item count, score range or direction is attributed to it here..

What it established is why the format spread. Its sensitivity to change exceeded that of the physical-function subscale of the SF-36 in the same population, and its test-retest reliability was R = .94 3Ref 3Binkley JM, Stratford PW, Lott SA, Riddle DL (1999).The Lower Extremity Functional Scale (LEFS): Scale Development, Measurement Properties, and Clinical Application.The Lower Extremity Functional Scale as the lower-limb instrument this one is modelled on: its origin as a purpose-built scale, its intended population of outpatients with lower-extremity musculoskeletal dysfunction, its derivation from 107 patients across twelve outpatient physical therapy clinics, test-retest reliability of R = .94, sensitivity to change exceeding the SF-36 physical function subscale, and its published MCID and MDC of 9 scale points with point-in-time measurement error of ±5.3 scale points. No item count, score range or direction is attributed to it here.. It also published the figures that let a clinician read one patient's change rather than only a group average.

The lower-limb original reports a minimal clinically important difference and a minimal detectable change of 9 scale points, alongside point-in-time measurement error of ±5.3 points 3Ref 3Binkley JM, Stratford PW, Lott SA, Riddle DL (1999).The Lower Extremity Functional Scale (LEFS): Scale Development, Measurement Properties, and Clinical Application.The Lower Extremity Functional Scale as the lower-limb instrument this one is modelled on: its origin as a purpose-built scale, its intended population of outpatients with lower-extremity musculoskeletal dysfunction, its derivation from 107 patients across twelve outpatient physical therapy clinics, test-retest reliability of R = .94, sensitivity to change exceeding the SF-36 physical function subscale, and its published MCID and MDC of 9 scale points with point-in-time measurement error of ±5.3 scale points. No item count, score range or direction is attributed to it here..

Those numbers belong to that instrument. Calling something the upper-extremity counterpart of a scale does not transfer the scale's measurement properties across, and this page does not lend them.

What this page does not print, and why

No item count, score range or change threshold for the UEFI appears anywhere on this page. The sources behind this article do not establish them, and printing a number a reader might use to interpret their own form — sourced from somewhere else, or from nowhere — would be worse than leaving the gap visible. This is a common situation with outcome measures rather than a peculiarity of this one.

The pattern is easy to see in a well-documented sibling. The QuickDASH's defining paper reports internal consistency of at least 0.92, intraclass correlation coefficients above 0.94, and construct validity of r ≥ 0.64 against single-item indices of pain and function — and reports no minimal clinically important difference, no minimal detectable change, no score range and no severity cutoff 2Ref 2Beaton DE, Wright JG, Katz JN; Upper Extremity Collaborative Group (2005).Development of the QuickDASH: comparison of three item-reduction approaches.The QuickDASH's construct — physical function and symptoms of the upper limb treated as a single region rather than joint-specific — plus its 11-item count, its derivation from the 30-item DASH by comparing concept retention, equidiscriminative item-total correlation and item response theory against field data from 407 patients, and its reported psychometrics (Cronbach alpha at or above 0.92, ICCs above 0.94, construct validity r at or above 0.64 against single-item pain and function indices). Also cited for what it does not report: no MCID, no MDC, no score range, no severity cutoff.. Those interpretive values do exist, but they were established by separate responsiveness studies and they vary by condition and population 2Ref 2Beaton DE, Wright JG, Katz JN; Upper Extremity Collaborative Group (2005).Development of the QuickDASH: comparison of three item-reduction approaches.The QuickDASH's construct — physical function and symptoms of the upper limb treated as a single region rather than joint-specific — plus its 11-item count, its derivation from the 30-item DASH by comparing concept retention, equidiscriminative item-total correlation and item response theory against field data from 407 patients, and its reported psychometrics (Cronbach alpha at or above 0.92, ICCs above 0.94, construct validity r at or above 0.64 against single-item pain and function indices). Also cited for what it does not report: no MCID, no MDC, no score range, no severity cutoff..

The absolute number on a functional index carries less information than the difference between it and the same person's earlier score on the same form.

The route around the gap is short. The form prints its own scoring line, and the therapist who chose it knows which version is in use, which direction it runs, and what a meaningful change looks like for the condition being treated. Asking is routine rather than an imposition.

How it compares with the DASH and the QuickDASH

The dash questionnaire is the older and longer instrument in this family, and the QuickDASH is its short form: eleven items reduced from the parent thirty, derived by comparing three item-reduction methods — concept retention, equidiscriminative item-total correlation, and item response theory — against field-testing data from 407 patients with upper-limb musculoskeletal conditions 2Ref 2Beaton DE, Wright JG, Katz JN; Upper Extremity Collaborative Group (2005).Development of the QuickDASH: comparison of three item-reduction approaches.The QuickDASH's construct — physical function and symptoms of the upper limb treated as a single region rather than joint-specific — plus its 11-item count, its derivation from the 30-item DASH by comparing concept retention, equidiscriminative item-total correlation and item response theory against field data from 407 patients, and its reported psychometrics (Cronbach alpha at or above 0.92, ICCs above 0.94, construct validity r at or above 0.64 against single-item pain and function indices). Also cited for what it does not report: no MCID, no MDC, no score range, no severity cutoff.. The concept-retention version is the one that became the QuickDASH 2Ref 2Beaton DE, Wright JG, Katz JN; Upper Extremity Collaborative Group (2005).Development of the QuickDASH: comparison of three item-reduction approaches.The QuickDASH's construct — physical function and symptoms of the upper limb treated as a single region rather than joint-specific — plus its 11-item count, its derivation from the 30-item DASH by comparing concept retention, equidiscriminative item-total correlation and item response theory against field data from 407 patients, and its reported psychometrics (Cronbach alpha at or above 0.92, ICCs above 0.94, construct validity r at or above 0.64 against single-item pain and function indices). Also cited for what it does not report: no MCID, no MDC, no score range, no severity cutoff..

The honest summary is that these instruments overlap heavily in the territory they cover and are not interchangeable in what they produce. No crosswalk between them appears in the evidence behind this page, so a score on one is not a translation of a score on another — and a clinic that switches forms partway through a course of care loses the comparison that made the measurement worth collecting.

When a joint-specific or patient-specific measure fits better

A limb-wide score can average away a problem that lives in one joint. Where the diagnosis is narrow, a narrow instrument often reads it better. The patient-rated tennis elbow evaluation was validated in 78 players with chronic, unilateral, MRI-confirmed lateral elbow tendinopathy and out-performed its comparators on responsiveness, with a standardized response mean of 2.1 against 1.5 to 1.7 for the visual analogue scale, the DASH and the other measures tested 4Ref 4Rompe JD, Overend TJ, MacDermid JC (2007).Validation of the Patient-rated Tennis Elbow Evaluation Questionnaire.The PRTEE as a joint-specific alternative: validation in 78 tennis players with chronic, unilateral, MRI-confirmed lateral elbow tendinopathy, subscale internal consistency of 0.94 for pain, 0.93 for specific activities and 0.85 for usual activities, and responsiveness expressed as a standardized response mean of 2.1 against 1.5 to 1.7 for the comparator measures, supporting its use as a primary outcome measure in tennis elbow research.. Its subscales held up individually — 0.94 for pain, 0.93 for specific activities, 0.85 for usual activities — which is what supports its use as the primary outcome in elbow tendinopathy research rather than a general arm score 4Ref 4Rompe JD, Overend TJ, MacDermid JC (2007).Validation of the Patient-rated Tennis Elbow Evaluation Questionnaire.The PRTEE as a joint-specific alternative: validation in 78 tennis players with chronic, unilateral, MRI-confirmed lateral elbow tendinopathy, subscale internal consistency of 0.94 for pain, 0.93 for specific activities and 0.85 for usual activities, and responsiveness expressed as a standardized response mean of 2.1 against 1.5 to 1.7 for the comparator measures, supporting its use as a primary outcome measure in tennis elbow research..

A different kind of instrument is not purely self-report at all. The Constant-Murley shoulder score is a composite scored out of 100 across pain, activities of daily living, range of motion and strength, with a higher score indicating better shoulder function — which means part of it has to be measured by a clinician rather than filled in at reception 5Ref 5Constant CR, Murley AH (1987).A clinical method of functional assessment of the shoulder.The Constant-Murley score as a composite shoulder measure combining subjective and objective components across pain, activities of daily living, range of motion and strength, scored out of 100 with a higher score indicating better shoulder function — and therefore requiring clinician-taken measurements rather than being purely patient-completed..

A patient-reported outcome measure is completed by the patient alone; a composite score such as the Constant-Murley adds measurements a clinician takes during the examination.

A third approach abandons fixed items altogether. On the patient-specific functional scale, the patient nominates up to five activities they find difficult because of their condition and rates each from 0, unable to perform, to 10, able to perform at the prior level; the reported score is the mean across the nominated activities, so the item count is variable by design rather than fixed 6Ref 6Stratford P, Gill C, Westaway M, Binkley J. (1995).Assessing Disability and Change on Individual Patients: A Report of a Patient Specific Measure.The Patient-Specific Functional Scale as the patient-nominated alternative: the patient names up to five activities made difficult by their condition and rates each on an 11-point scale from 0 (unable to perform) to 10 (able to perform at prior level), with the reported score being the mean across nominated activities and the item count therefore variable rather than fixed; developed and assessed in 63 outpatients with mechanical low back pain against the Roland-Morris Questionnaire.. It was developed in 63 outpatients with mechanical low back pain and assessed against the Roland-Morris Questionnaire 6Ref 6Stratford P, Gill C, Westaway M, Binkley J. (1995).Assessing Disability and Change on Individual Patients: A Report of a Patient Specific Measure.The Patient-Specific Functional Scale as the patient-nominated alternative: the patient names up to five activities made difficult by their condition and rates each on an 11-point scale from 0 (unable to perform) to 10 (able to perform at prior level), with the reported score being the mean across nominated activities and the item count therefore variable rather than fixed; developed and assessed in 63 outpatients with mechanical low back pain against the Roland-Morris Questionnaire..

What repeating the score is actually for

Serially, it answers one question: is this arm working better than it was three weeks ago? That is harder to answer from memory than it sounds. Pain and function move independently, a good week distorts the recollection of a bad one, and a person partway through a course of therapy is a poor judge of their own starting point. A repeated score restores the baseline.

This is what the format was designed around. The lower-limb original was published under a title naming scale development, measurement properties and clinical application, and it reports a minimal detectable change precisely so that one patient's change can be read rather than only a difference between groups 3Ref 3Binkley JM, Stratford PW, Lott SA, Riddle DL (1999).The Lower Extremity Functional Scale (LEFS): Scale Development, Measurement Properties, and Clinical Application.The Lower Extremity Functional Scale as the lower-limb instrument this one is modelled on: its origin as a purpose-built scale, its intended population of outpatients with lower-extremity musculoskeletal dysfunction, its derivation from 107 patients across twelve outpatient physical therapy clinics, test-retest reliability of R = .94, sensitivity to change exceeding the SF-36 physical function subscale, and its published MCID and MDC of 9 scale points with point-in-time measurement error of ±5.3 scale points. No item count, score range or direction is attributed to it here..

There is a second use, discussed less often with patients: documentation. A course of care that produced a measurable functional change is easier to justify to a payer, and easier to review internally, than one described only in narrative notes.

And a third, which is the one patients tend to find valuable. A completed form is a structured conversation. When the total has not moved, the useful follow-up is which specific activities were rated lowest, and whether those are the activities the treatment plan has actually been aimed at.

The same idea in every other body region

Each region of the body has its own version of this form, and clinics choose by where the problem is rather than by preference. Ask at a low-back clinic and the questionnaire is likely to be the oswestry disability index; ask at a neck clinic and it is likely to be the neck disability index; ask about a knee or an ankle and it is the scale this one was modelled on, developed for outpatients with lower-extremity musculoskeletal dysfunction 3Ref 3Binkley JM, Stratford PW, Lott SA, Riddle DL (1999).The Lower Extremity Functional Scale (LEFS): Scale Development, Measurement Properties, and Clinical Application.The Lower Extremity Functional Scale as the lower-limb instrument this one is modelled on: its origin as a purpose-built scale, its intended population of outpatients with lower-extremity musculoskeletal dysfunction, its derivation from 107 patients across twelve outpatient physical therapy clinics, test-retest reliability of R = .94, sensitivity to change exceeding the SF-36 physical function subscale, and its published MCID and MDC of 9 scale points with point-in-time measurement error of ±5.3 scale points. No item count, score range or direction is attributed to it here..

Standardising this way has a specific payoff. When every clinician in a department uses the same form for the same region, a score recorded by one therapist means the same thing to the next one, and a patient who returns two years later has a comparable baseline already sitting in the record.

It also sets a reasonable expectation for the paperwork. Being handed a questionnaire at a first appointment is not an administrative formality, and it is not a screening test for anything. It is the first data point in a series, and it is the only one in that series that cannot be collected later.