The PSFS, short for Patient-Specific Functional Scale, asks the patient to name three to five activities they can no longer perform because of their current problem, and to rate each one from 0 to 10. The score is their mean. The questionnaire is right below, ready to fill in, with the calculation and the PDF export.

0 means unable to perform the activity, 10 the ability the patient had before their problem. At follow-up, the same activities are rated again. A change of 2.3 points corresponds to a medium change as perceived by the patient: the full set of thresholds is below.

Figures taken from the 2022 COSMIN systematic review (57 studies on measurement properties) and from the validation studies listed at the end of this page. Every value carries its source where it is written.

What the PSFS measures

The principle fits in one sentence: the patient names the activities they can no longer perform, or perform with difficulty, because of their current problem, then rates each one from 0 to 10. The score is the mean of those ratings.

The consequence is that no two PSFS measure the same thing. What the tool captures has nevertheless been described precisely: Fairbairn et al. classified 2,911 activities generated by 1,050 musculoskeletal patients against the International Classification of Functioning. All of them could be classified, and the distribution is clear: 80.0% belong to the "activity" domain, 7.7% to "participation" and 7.4% to impairments. In other words, left free to choose, patients talk about concrete activities, very rarely about social roles and almost never about symptoms.

The reference systematic review, conducted to COSMIN standards on 985 screened articles of which 57 were retained for measurement properties, returns a two-part verdict worth knowing before relying on the tool. Responsiveness is sufficient, and it is the best-supported conclusion of the set: 32 studies, 13,770 participants, moderate to high certainty. Test-retest reliability is sufficient in musculoskeletal conditions, across 22 studies and 845 participants, but with only low to moderate certainty. And construct validity is rated insufficient as a measure of physical function, across 21 studies and 2,945 participants.

The practical reading fits in one sentence: the PSFS is a good instrument for tracking change, and a much more debatable instrument for measuring a construct. The same review records its use across 87 distinct conditions and a standard error of measurement ranging from 0.35 to 1.5 points depending on the study, which is enough to say that no single threshold can be applied everywhere.

Letting the patient choose the activities

This is the step that determines the quality of everything else, and it is the one that gets rushed. The instruction is to ask the patient to name three to five activities that matter to them and that they can no longer perform, or perform less well, since their problem began. Neither the therapist nor a pre-set list chooses for them.

A usable activity is specific and observable. "Feeling better" is not one; "climbing the two flights up to my flat without stopping" is. In a cohort of 513 knee osteoarthritis patients in multidisciplinary rehabilitation, the activities spontaneously named clustered around walking, climbing and descending stairs, and standing, with effect sizes of 0.60 to 0.97 depending on the activity, against 0.41 for the WOMAC physical function subscale in the same patients.

Changing the list of activities during follow-up makes comparison impossible: they are recorded in the notes at the first administration, with their exact wording.

The change that counts

The practical question is not the score, it is the gap between two administrations. The reference is a cohort of 1,708 consecutive patients from five physiotherapy practices, where the minimum important difference was derived from ROC curves anchored on change as perceived by the patient.

These values stayed relatively stable from one body region to another, which is not the case for the minimal detectable change, the one that separates real change from measurement noise. That one varies a great deal, and it is the one to look at before drawing a conclusion about a given patient.

A gap of 2 points therefore does not mean the same thing in a COPD patient and in a patient with cervical radiculopathy. In the first case it comfortably exceeds measurement noise; in the second it stays inside it.

Reliability is not a given

The last two rows of the table cover the same condition and contradict each other. In 2006, in 38 patients with cervical radiculopathy, the PSFS outperformed the Neck Disability Index: ICC 0.82 against 0.68, and a minimal detectable change of 2.1 points against 10.2. Four years later, the same team replicated the study in 165 patients and concluded that the PSFS had poor test-retest reliability, with minimal detectable change raised to 3.3 points.

This is not an anomaly to be set aside, it is the rule in this kind of literature: confidence intervals on small samples are wide, and replication at a larger scale often pulls estimates down. Where the pooling has been done, the figures are firmer: the COSMIN meta-analysis in low back pathology, across 10 studies, gives a pooled reliability of ICC = 0.89 (95% CI: 0.75 to 0.95).

Comparing two patients

This is the classic objection, and on paper it holds: two patients rate different activities, so their scores do not measure the same thing. It has been tested directly. In 1,181 consecutive patients referred to physiotherapy for a musculoskeletal disorder, assessed at the first and the last session, the authors conclude that the PSFS is valid for group-level change comparisons and for between-group discrimination.

The nuance to keep is exactly that one: comparing how two cohorts evolve, yes; reading one patient's 7 out of 10 against another patient's 7 out of 10, no. The PSFS is a measure of an individual trajectory before it is a measure of a level.

Four administration pitfalls

The patient chooses, never the therapist

Suggesting the activities yourself turns the tool into a poorly calibrated region-specific questionnaire and strips it of what makes it valuable. If the patient cannot think of anything, help them describe their day rather than feed them items.

The same activities at follow-up

Rating a different list produces a number that looks like a score and is not one. The activities are recorded in the notes at the first administration, with their exact wording.

The questionnaire is administered, not handed out

It is completed with the patient, during the session. That is what makes it possible to reword an activity that is too vague, and to check that 0 and 10 are understood as lost ability and previous ability, not as a pain scale.

The same score does not mean the same thing in two patients

The score only makes sense in relation to the activities that make it up and to its own baseline value. It is the gap that is interpreted, against the minimal detectable change of the relevant population, and not against the single 2-point threshold so often quoted from memory.

Frequently asked questions

What does PSFS stand for?

PSFS stands for Patient-Specific Functional Scale. It was published by Stratford, Gill, Westaway and Binkley in Physiotherapy Canada in 1995.

How do you score the PSFS?

The score is the arithmetic mean of the ratings given to the activities the patient named, each rated from 0, unable to perform the activity, to 10, the ability they had before their problem. Three activities rated 2, 5 and 8 give a score of 5.0. The calculator on this page averages them as you go.

How much change counts as meaningful?

Across 1,708 musculoskeletal patients, the minimum important difference is 1.3 points for a small change, 2.3 points for a medium change and 2.7 points for a large change. The minimal detectable change, however, depends on the population: from 0.44 points in COPD to 3.3 points in cervical radiculopathy.

Can the PSFS be exported as a PDF?

Yes. The questionnaire's download button produces a one-page document with the patient name, the date, the named activities, their ratings and the mean score, ready for the notes. Nothing entered leaves the browser.

Is the PSFS a validated outcome measure?

Partly, and it matters which part. The 2022 COSMIN review rates responsiveness sufficient across 32 studies and 13,770 participants, and test-retest reliability sufficient in musculoskeletal conditions, but construct validity insufficient as a measure of physical function. It is a good instrument for tracking change, a debatable one for measuring a construct.

References

28 sources, PMIDs included

1. Stratford P, Gill C, Westaway M, Binkley J. Assessing disability and change on individual patients: a report of a patient specific measure. Physiotherapy Canada 1995;47(4):258-63. The original publication of the scale. The journal was not indexed in MEDLINE at that date: this reference has no PMID, and assigning one to it would be an error.
2. Chatman AB, Hyams SP, Neel JM, Binkley JM, Stratford PW, Schomberg A, et al. The Patient-Specific Functional Scale: measurement properties in patients with knee dysfunction. Phys Ther 1997;77(8):820-9. PMID 9256870. Validation in the knee in 38 patients: test-retest reliability ICC(2,1) = 0.84 at 72 hours, sensitivity to change r = 0.78.
3. Westaway MD, Stratford PW, Binkley JM. The patient-specific functional scale: validation of its use in persons with neck dysfunction. J Orthop Sports Phys Ther 1998;27(5):331-8. PMID 9580892. Validation in the cervical spine in 31 patients: reliability R = 0.92, validity r = 0.73 to 0.83 against the Neck Disability Index.
4. Cleland JA, Fritz JM, Whitman JM, Palmer JA. The reliability and construct validity of the Neck Disability Index and patient specific functional scale in patients with cervical radiculopathy. Spine (Phila Pa 1976) 2006;31(5):598-602. PMID 16508559. 38 patients: PSFS ICC 0.82 against 0.68 for the NDI, minimal detectable change 2.1 against 10.2.
5. McMillan CR, Binhammer PA. Which outcome measure is the best? Evaluating responsiveness of the DASH, the Michigan Hand Questionnaire and the Patient-Specific Functional Scale following hand and wrist surgery. Hand (N Y) 2009;4(3):311-8. PMID 19259747. 81 subjects followed at 3 and 6 months: responsiveness depends on the diagnosis, not only on the instrument.
6. Young IA, Cleland JA, Michener LA, Brown C. Reliability, construct validity, and responsiveness of the neck disability index, patient-specific functional scale, and numeric pain rating scale in patients with cervical radiculopathy. Am J Phys Med Rehabil 2010;89(10):831-9. PMID 20657263. Replication in 165 patients that contradicts the previous one: poor test-retest reliability for the PSFS, minimal detectable change 3.3.
7. Fairbairn K, May K, Yang Y, Balasundar S, Hefford C, Abbott JH. Mapping Patient-Specific Functional Scale (PSFS) items to the International Classification of Functioning, Disability and Health (ICF). Phys Ther 2012;92(2):310-7. PMID 22074939. 2,911 items generated by 1,050 patients: 100% classifiable, 80.0% in "activity", 7.7% in "participation".
8. Hefford C, Abbott JH, Arnold R, Baxter GD. The patient-specific functional scale: validity, reliability, and responsiveness in patients with upper extremity musculoskeletal problems. J Orthop Sports Phys Ther 2012;42(2):56-65. PMID 22333510. 180 patients: ICC(2,1) = 0.713, area under the ROC curve 0.887, minimum important difference 1.2 points.
9. Cleland JA, Whitman JM, Houser JL, Wainner RS, Childs JD. Psychometric properties of selected tests in patients with lumbar spinal stenosis. Spine J 2012;12(10):921-31. PMID 22749295. 55 patients: in lumbar spinal stenosis, only the Oswestry reaches excellent test-retest reliability.
10. Abbott JH, Schmitt JS. The Patient-Specific Functional Scale was valid for group-level change comparisons and between-group discrimination. J Clin Epidemiol 2014;67(6):681-8. PMID 24556219. 1,181 consecutive patients: the direct answer to the objection that scores cannot be compared between patients.
11. Abbott JH, Schmitt J. Minimum important differences for the patient-specific functional scale, 4 region-specific outcome measures, and the numeric pain rating scale. J Orthop Sports Phys Ther 2014;44(8):560-4. PMID 24828475. 1,708 patients from 5 practices: minimum important differences of 1.3, then 2.3, then 2.7 points.
12. Koehorst ML, van Trijffel E, Lindeboom R. Evaluative measurement properties of the patient-specific functional scale for primary shoulder complaints in physical therapy practice. J Orthop Sports Phys Ther 2014;44(8):595-603. PMID 25029915. 50 shoulder patients: ICC 0.87 (95% CI 0.72 to 0.94).
13. Wright HH, O'Brien V, Valdes K, Koczan B, MacDermid J, Moore E, Finley MA. Relationship of the Patient-Specific Functional Scale to commonly used clinical measures in hand osteoarthritis. J Hand Ther 2017;30(4):538-45. PMID 28807599. 35 participants: ICC(2,1) = 0.80, standard error of measurement 0.56, MDC₉₀ 1.30 and MDC₉₅ 1.56.
14. Mathis RA, Taylor JD, Odom BH, Lairamore C. Reliability and Validity of the Patient-Specific Functional Scale in Community-Dwelling Older Adults. J Geriatr Phys Ther 2019;42(3):E67-E72. PMID 29630006. 31 community-dwelling older adults: ICC 0.82 (95% CI 0.67 to 0.91), minimal detectable change 2.8 points.
15. Thoomes-de Graaf M, Fernández-de-las-Peñas C, Cleland JA. The content and construct validity of the modified patient specific functional scale (PSFS 2.0) in individuals with neck pain. J Man Manip Ther 2020;28(1):49-59. PMID 31130088. Starting observation: response options differ across the versions of the PSFS in circulation.
16. Rysstad T, Grotle M, Klokk LP, Tveter AT. Responsiveness and minimal important change of the QuickDASH and PSFS when used among patients with shoulder pain. BMC Musculoskelet Disord 2020;21(1):328. PMID 32460743. 134 shoulder patients: area under the ROC curve 0.75 for both instruments, 8 of 9 a priori hypotheses met for the PSFS against 7 of 9 for the QuickDASH.
17. Evensen J, Soberg HL, Sveen U, Hestad KA, Bronken BA. The Applicability of the Patient-Specific Functional Scale (PSFS) in Rehabilitation for Patients with Acquired Brain Injury (ABI): A Cohort Study. J Multidiscip Healthc 2020;13:1121-32. PMID 33116558. 54 patients out of 59, or 92%, manage to complete the scale after acquired brain injury.
18. Moore HE, Corning WL, van der Esch M, Roorda LD, Dekker J, de Groot J, et al. Evaluation of treatment outcome using the Patient Specific Functional Scale in knee osteoarthritis patients undergoing multidisciplinary rehabilitation. Osteoarthr Cartil Open 2020;2(4):100098. PMID 36474879. 513 knee osteoarthritis patients: effect sizes of 0.60 to 0.97 depending on the activity, against 0.41 for WOMAC physical function.
19. Alnahdi AH, Murtada BA, Zawawi AT, Omar MT, Alsobayel HI. Cross-cultural adaptation and measurement properties of the Arabic version of the Patient-Specific Functional Scale in patients with lower extremity musculoskeletal disorders. Disabil Rehabil 2022;44(15):4104-10. PMID 33587649. 116 lower limb patients: ICC 0.86, minimal detectable change 1.49 points, no floor and no ceiling effect.
20. Heldmann P, Hummel S, Bauknecht L, Bauer JM, Werner C. Construct Validity, Test-Retest Reliability, Sensitivity to Change, and Feasibility of the Patient-Specific Functional Scale in Acutely Hospitalized Older Patients With and Without Cognitive Impairment. J Geriatr Phys Ther 2022;45(3):134-44. PMID 33734156. 120 older patients: ICC 0.76 at 24 hours, with and without cognitive impairment alike.
21. KC S, Sharma S, Ginn K, Reed D. A comparison between measurement properties of four shoulder-related outcome measures in Nepalese patients with shoulder pain. Qual Life Res 2022;31(6):1897-906. PMID 35072905. Direct comparison of PSFS, DASH, QuickDASH and SPADI in the same patients: PSFS ICC 0.82, area under the ROC curve 0.83.
22. Pathak A, Wilson R, Sharma S, Pryymachenko Y, Ribeiro DC, Chua J, Abbott JH. Measurement Properties of the Patient-Specific Functional Scale and Its Current Uses: An Updated Systematic Review of 57 Studies Using COSMIN Guidelines. J Orthop Sports Phys Ther 2022;52(5):262-75. PMID 35128944. The reference review: reliability and responsiveness sufficient in musculoskeletal conditions, construct validity insufficient.
23. Nazari G, Bobos P, Lu S, Reischl S, Almeida PH, MacDermid JC. Psychometric Properties of the Patient-Specific Functional Scale in Patients with Low Back Pathology: A Systematic Review and Meta-Analysis. Physiother Can 2022;74(1):6-14. PMID 35185242. 10 studies: pooled reliability ICC 0.89 (95% CI 0.75 to 0.95).
24. Mañago MM, Cohen ET, Cameron MH, Christiansen CL, Bade M. Reliability, Validity, and Responsiveness of the Patient-Specific Functional Scale for Measuring Mobility-Related Goals in People With Multiple Sclerosis. J Neurol Phys Ther 2023;47(3):139-45. PMID 36897202. 32 participants, EDSS 1.0 to 7.0: ICC(2,1) = 0.70, MDC₉₅ 2.1 points.
25. Evensen J, Soberg HL, Sveen U, Hestad KA, Moore JL, Bronken BA. Measurement Properties of the Patient-Specific Functional Scale in Rehabilitation for Patients With Stroke: A Prospective Observational Study. Phys Ther 2023;103(5):pzad014. PMID 37140476. 71 patients in the subacute phase of stroke: ICC(3,1) = 0.81, 80% of items classified under "activities and participation".
26. Thoomes E, Cleland JA, Falla D, Bier J, de Graaf M. Reliability, Measurement Error, Responsiveness, and Minimal Important Change of the Patient-Specific Functional Scale 2.0 for Patients With Nonspecific Neck Pain. Phys Ther 2024;104(1):pzad113. PMID 37606246. 100 patients with neck pain: no floor and no ceiling effect, ICC 0.95 (95% CI 0.92 to 0.97).
27. Cioeta M, Youssef S, Brindisino F, Venturin D, Pichero R, Giovannico G, et al. Cross-cultural adaptation and psychometric properties of the Italian version of the Patient-Specific Functional Scale (PSFS) in subjects with shoulder pain. Disabil Rehabil 2025;47(2):512-8. PMID 38700257. 109 shoulder subjects: ICC 0.866, standard error of measurement 0.7 points, minimal detectable change 1.9 points.
28. Alonazi A, Aldhahi MI, Nazer R, Albarrati A. Personalizing Functional Assessment in Chronic Obstructive Pulmonary Disease: A Validation Study of the Patient-Specific Functional Scale. J Clin Med 2025;15(1):37. PMID 41517287. 70 adults with COPD: ICC 0.94, minimal detectable change 0.44 points, no floor and no ceiling effect.