RehabMeasures Instrument
  1. Rehabilitation Measures Database
  2. Oxford Knee Score

Oxford Knee Score

Last Updated

Purpose

Developed as an outcome measure to be used with patients having a total knee replacement (TKR). It provides a measure of outcome for TKR that is short, practical, reliable, valid, and sensitive to clinically important changes over time.

Link to Instrument

Instrument details

Acronym OKS

Area of Assessment

Activities of Daily Living
Functional Mobility
Gait
Life Participation
Occupational Performance
Pain
Quality of Life
Seating
Sleep

Assessment Type

Patient Reported Outcomes

Administration Mode

Paper & Pencil

Cost

Free

Cost Description

Licenses given to publicly funded healthcare and non-commercially funded academic research institutes are free of charge; commercial use of the OKS is fee-based.
Licenses can be obtained by following the link below:
http://process.innovation.ox.ac.uk/

Diagnosis/Conditions

  • Arthritis + Joint Conditions
  • Pain Management

Populations

Osteoarthritis

Key Descriptions

  • 12-item self-report measure containing questions regarding the patient’s pain and level of function.
  • Originally, each item was rated on a scale of 1-5, from least to most difficulty or severity.
  • The lowest total score was a 12, indicating the patient was fully functional with no complaints, and the highest total score was a 60, indicating maximum difficulty.
  • The modified scoring system ranges from 0-4 on each question with 4 representing maximum function and 0 representing poorest function.
  • Using this scale, the lowest, worst score is a 0, and the highest, best score is 48.
  • The OKS has 2 subscales:
    1) Pain
    2) Function
  • The pain component score (OKS-PCS) consists of items 2, 3, 7, 11 and 12, and the functional component score consists of items 1, 4, 5, 6, 8, 9, and 10.

Number of Items

12

Time to Administer

5-10 minutes

Required Training

No Training

Age Ranges

Adult

18 - 64

years

Elderly Adult

65 +

years

Instrument Reviewers

Chris Ball, SPT, Mark Boles, SPT, Deidra Debnam, SPT, Heather Eustis, SPT, Caitlin Gallagher, SPT, Brittany Stapleton, SPT, Sarah van der Horst, SPT in November 2013.

Body Part

Lower Extremity

ICF Domain

Body Function
Activity
Participation

Measurement Domain

Activities of Daily Living

Considerations

Knee OA:

(Harris et al, 2013) 

  • The OKS could be used in clinical practice to monitor disease progression in individual patients undergoing non-operative management for knee OA. 

TKA:

(Collins et al, 2011) 

  • Clinicians should be aware that some patients may require explanation of individual items, which could introduce interviewer bias. 

(Dawson et al, 1998) 

  • This tool is intended specifically with knee surgery alone.

 Do you see an error or have a suggestion for this instrument summary? Please e-mail us! 

Osteoarthritis

back to Populations

Standard Error of Measurement (SEM)

Non-operative Knee Osteoarthritis (Knee OA):

(Harris et al, 2013; n = 134; mean age = 59)

  • SEM = 2.65 (OKS)
  • SEM = 6.2 (OKS-FCS)
  • SEM = 6.9 (OKS-PCS) 

 

Total Knee Replacement (TKR):

(Collins et al, 2011; pooled subject data from Dawson et al. 1998, Impellizzeri et al. 2010, Dunbar et al. 2000, Dunbar et al. 2001)

  • SEM = 2.2 

(Dawson et al, 1998; n = 117; median age = 73)

 

 

 

12 item OKS before TKR and at 6-month follow up

 

 

 

 

 

 

Standard Deviation

r value

(1-r)

Square Root of (1-r)

SEM

Q1

0.6

0.6

0.4

0.63

0.37

Q2

1.04

0.53

0.47

0.68

0.71

Q3

0.79

0.64

0.36

0.6

0.47

Q4

0.97

0.54

0.46

0.67

0.65

Q5

0.85

0.57

0.43

0.65

0.55

Q6

0.9

0.45

0.55

0.74

0.66

Q7

0.9

0.56

0.44

0.66

0.59

Q8

1.08

0.47

0.53

0.72

0.78

Q9

0.84

0.69

0.31

0.55

0.46

Q10

1.22

0.47

0.53

0.72

0.88

Q11

1.13

0.61

0.39

0.62

0.70

Q12

0.88

0.65

0.35

0.59

0.52

SUM

10.6

6.18

5.18

 

 

AVG

0.88

0.515

0.485

0.69

0.61

 

(Impellizzeri et al, 2010; n = 79; mean age = 69 (7) years; 6 months post TKA)

  • SEM = 2.2

Minimal Detectable Change (MDC)

Knee OA: 

(Harris et al, 2013)

  • MDC = ±6 (OKS)
  • MDC = ±16 (OKS-PCS) 
  • MDC = ±15 (OKS-FCS) 

 

Total Knee Replacement (TKR):

(Collins et al, 2011)

  • MDC = 6.1 

 

(Impellizzeri et al, 2010)

  • MDC = 6.1

Minimally Clinically Important Difference (MCID)

Knee OA: 

(Harris et al, 2013)

  • MCID = 6 (OKS)
  • MCID = 14 (OKS-PCS)
  • MCID = 10 (OKS-FCS)

Normative Data

Knee OA:

(Harris et al, 2013)

  • Baseline Mean OKS = 30.29 (10)
  • Baseline Mean OKS-PCS = 59.36 (22)
  • Baseline Mean OKS-FCS = 67.22 (21)
  • 3 Month Follow Up Mean OKS = 32.15 (11)
  • 3 Month Follow Up Mean OKS-PCS = 65.13 (24)
  • 3 Month Follow Up Mean OKS-FCS = 68.66 (23) 

 

Total Knee Replacement (TKR):

(Dawson et al, 1998)

  • Pre-operative Mean = 44.61
  • 6 Month Follow Up Mean = 29.33 

 

(Jenny et al, 2012; n = 200; mean age = 71 (9)) 

  • Mean OKS = 43.7 (before surgery)
  • Mean OKS = 20.5 (after surgery) 

 

(Madsen et al, 2013; n = 80)

OKS in patients in group-based rehab vs. supervised home training s/p TKA

 

 

 

 

Intervention

Control

p-value

3 months

7.9 ± 5.9 

8.4 ± 6.7

0.7

6 months

10.5 ± 7.5

9.7 ± 8.5

0.7

 

Test/Retest Reliability

Knee OA: 

(Harris et al, 2013) 

  • Excellent Test-Retest Reliability
  • ICC = 0.93 (OKS)
  • ICC = 0.91 (OKS-PCS)
  • ICC = 0.92 (OKS-FCS) 

 

Total Knee Replacement (TKR):

(Collins et al. 2011) 

  • Excellent Test-Retest Reliability
  • ICC = 0.91-0.94
  • Coefficient of Reliability = 6.45 (Bland and Altman method) 

 

(Dawson et al, 1998) 

  • Excellent Test-Retest Reliability
  • r = 0.92 

 

(Impellizzeri et al, 2010) 

  • Excellent Test-Retest Reliability
  • ICC = 0.91

Internal Consistency

Knee OA: 

(Harris et al, 2013) 

  • Excellent Internal Consistency
  • Cronbach’s α = 0.94 (OKS)
  • Cronbach’s α = 0.88 (OKS-FCS)
  • Cronbach’s α = 0.90 (OKS-PCS) 

 

(Xie et al, 2011; n = 187; mean age = 64) 

  • Excellent Internal Consistency
  • Cronbach’s α = .896 

 

Total Knee Replacement (TKR):

(Collins et al, 2011) 

  • Excellent Internal Consistency
  • Cronbach’s α = 0.87-0.93 

 

(Dawson et al, 1998) 

  • Excellent Internal Consistency
  • Cronbach’s α = 0.87 (pre-op)
  • Cronbach’s α = 0.93 (6-months post-op) 

 

(Dunbar et al, 2001; n = 1,200; mean age = 78) 

  • Excellent Internal Consistency
  • Cronbach’s α = 0.93 

 

(Jenny et al, 2013: n = 100; mean age = 72) 

  • Excellent Internal Consistency
  • Cronbach’s α = 0.91 

 

(Jenny et al, 2012) 

  • Satisfactory Internal Consistency
  • Cronbach’s α = .88 (pre-op)
  • Cronbach’s α = .66 (post-op)

Criterion Validity (Predictive/Concurrent)

Total Knee Replacement (TKR):

(Jenny et al, 2012) 

Concurrent Validity compared to AKS

  • Strong negative correlation pre operatively between OKS and both the AKS knee and function scores (r = -.33; r = -.47)
  • Weak negative correlation post operatively between OKS and AKS knee score (r = -.19)
  • Strong negative correlation post operatively between OKS and AKS function score (r = -.49)

Construct Validity

Knee OA:

(Harris et al, 2013)

Baseline Spearman's correlations between scores

 

 

 

 

OKS

OKS-PCS

OKS-FCS

ICOAP

-0.88

-0.88

-0.79

KOOS-PS

-0.85

-0.78

-0.87

PCS-12

0.65

 

 

MCS-12

0.37

 

 

 

 

 

 

All correlations were significant at the 0.01 level

 

 

 

 

3-month Spearman's correlations between scores

 

 

 

 

OKS

OKS-PCS

OKS-FCS

ICOAP

-0.67

-0.67

-0.60

KOOS-PS

-0.62

-0.55

-0.62

 

 

 

 

All correlations were significant at the 0.01 level

ICOAP = Intermittent and Constant Osteoarthritis Pain KOOS-PS = Knee Injury and Osteoarthritis Score - Physical Function Short Form                                                    PCS-12 = Physical Component Summary of SF-12         MCS-12 = Mental Component Summary of SF-12

 

 

 

 

(Xie et al, 2011) 

  • Convergent Validity
    • OKS strongly correlated with SF 6D (role limitation, social function and pain) and with EQ- SD (mobility, self care, usual activities and pain/discomfort) and Mobility-VAS (p =.51 -.82)
    • OKS moderately correlated with SF-6D physical functioning (p=.44)
    • OKS weakly correlated with SF-6D vitality (p = .35) 
  • Divergent Validity
    • OKS strongly correlated with SF-6D mental health (p =.51)
    • OKS moderately correlated with EQ-5D anxiety/depression (p= .41) 

 

Total Knee Replacement (TKR):

(Collins et al, 2011) 

  • Good correlation with knee-specific and general health questionnaires, such as the Western Ontario and McMaster Universities Osteoarthritis Index, American Knee Society Score, Knee Outcome Survey Activities of Daily Living Scale, and pain and physical function components of the Short Form 36 and Health Assessment Questionnaire. See Dawson et al. and Impellizzeri et al. 

 

(Dawson et al, 1998) 

  • Moderate correlation with AKS
  • Significant agreement with parts of the SF36 and HAQ (physical function and pain)

 

Correlation between the 12-item knee score and the AKS score, SF36 and HAQ assessments in 117 patients before operation and in 85 at the six-month follow-up

 

 

 

 

 

Correlation Coefficient

 

Test

Preop

Postop

AKS assessment

 

 

    Knee Score

-0.47**

--

    Function

-0.54**

--

SF36

 

 

    Physical

-0.69**

-0.66**

    Pain

-0.71**

-0.78**

    Role-physical

-0.52**

-0.43**

    Role-mental

-0.25**

-0.46**

    Social function

-0.56**

-0.78**

    Mental health

-0.19*

-0.41**

    Energy

-0.35**

-0.62**

    Health perceptions

-0.07

-0.41**

HAQ

 

 

    Pain VAS

0.53**

--

    Disability index

0.61**

--

 

 

 

*p<0.05, **p<0.01

 

 

 

(Dunbar et al, 2000) 

  • The Oxford-12 correlated most closely with the physical domains. 
  • The Oxford-12 correlated poorly with the eating domain of the SIP (rho = 0.14) thus demonstrating good divergent construct validity. 

 

(Impellizzeri et al, 2010)

Average Correlations between Instruments

 

 

Oxford Knee Score

WOMAC Pain

0.67*

WOMAC Stiffness

0.45*

WOMAC Function

0.67*

KOS-ADLS Symptoms

-0.48*

KOS-ADLS Function

-0.62*

KOS-ADLS Total Score

-0.62*

SF-12 PC

-0.50*

SF-12 MC

-0.17

 

 

*P<.001

 

 

  • The OKS was highly correlated with KOS-ADLS function and WOMAC pain and function subscales but less well correlated with those of the WOMAC stiffness and KOS-ADLS symptom scales. 

 

(Jenny et al, 2013) 

  • Poor Correlation - OKS with AKSS: 0.20 (P<.001)
  • Poor Correlation - OKS with HAAS: 0.19 (P<.001)

Content Validity

Knee OA:

(Xie et al, 2011) 

  • All items were shown to be relevant. 

 

Total Knee Replacement (TKR):

(Collins et al, 2011) 

  • Extensive input from patients in the development of the OKS ensures content validity. 

 

(Dawson et al, 1998) 

  • Items were selected based on interviews with patients considering a TKR to identify how they identified and reported problems with their knees. A questionnaire was drafted and tested on a different set of patients, requesting additional comments related to their knee problems that were not addressed. Modifications were made and this procedure was repeated 2 additional times to produce the final questionnaire.

Floor/Ceiling Effects

Total Knee Replacement (TKR):

(Dunbar et al, 2000)

  • Floor effect in 6.8% of patients
  • Ceiling effect in 0.1% of patients 

 

(Impellizzeri et al, 2010)

  • No floor or ceiling effects prior to TKA
  • No floor effects 6 months postoperatively
  • Ceiling effects 6 months postoperatively in 27% of patients 

 

(Jenny et al, 2012)

  • Pre-Op Group: no floor effect, small ceiling effect (7%)
  • Post-Op Group: substantial floor effect (33%), no ceiling effect
  • Calculated using original scoring scale

 

(Jenny et al, 2013)

  • Ceiling effect = 33%

Responsiveness

Knee OA: 

(Harris et al, 2013) 

  • Small change from baseline to 3 month follow-up
    • OKS: ES = 0.19
    • OKS-PCS: ES = 0.26
    • OKS-FCS: ES = 0.07 

 

Total Knee Replacement (TKR):

(Collins et al, 2011)

  • ES = 0.9 – 2.19
  • SRM = 0.7 

 

(Dawson et al, 1998) 

  • Large responsiveness from pre-operative to the 6 month follow-up (ES = 2.19) 

 

(Impellizzeri et al, 2010) 

  • Large Responsiveness
  • Effect Size = 0.9
  • SRM = 0.7 

 

(Lin et al, 2013) 

  • Excellent Responsiveness 
  • Most responsive compared to EQ-VAS and EQ-5D
  • SRM: 1.50 (large magnitude of effect)

Bibliography

Collins, Natalie J, Misra, Devyani, Felson, David T, Crossley, Kay M, & Roos, Ewa M. (2011). Measures of knee function: International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form, Knee Injury and Osteoarthritis Outcome Score (KOOS), Knee Injury and Osteoarthritis Outcome Score Physical Function Short Form (KOOS‐PS), Knee Outcome Survey Activities of Daily Living Scale (KOS‐ADL), Lysholm Knee Scoring Scale, Oxford Knee Score (OKS), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), Activity Rating Scale (ARS), and Tegner Activity Score (TAS). Arthritis care & research, 63(S11), S208-S228. 

Dawson, Jill, Fitzpatrick, Ray, Murray, David, & Carr, Andrew. (1998). Questionnaire on the perceptions of patients about total knee replacement. Journal of Bone & Joint Surgery, British Volume, 80(1), 63-69. 

Dunbar, Michael J, Robertsson, Otto, Ryd, Leif, & Lidgren, Lars. (2000). Translation and validation of the Oxford-12 item knee score for use in Sweden. Acta Orthopaedica, 71(3), 268-274. 

Dunbar, MJ, Robertsson, Otto, Ryd, Leif, & Lidgren, Lars. (2001). Appropriate questionnaires for knee arthroplasty RESULTS OF A SURVEY OF 3600 PATIENTS FROM THE SWEDISH KNEE ARTHROPLASTY REGISTRY. Journal of Bone & Joint Surgery, British Volume, 83(3), 339-344x. 

Harris, Kristina K, Dawson, Jill, Jones, Luke D, Beard, David J, & Price, Andrew J. (2013). Extending the use of PROMs in the NHS—using the Oxford Knee Score in patients undergoing non-operative management for knee osteoarthritis: a validation study. BMJ open, 3(8), e003365. 

Impellizzeri, Franco M, Mannion, Anne F, Leunig, Michael, Bizzini, Mario, & Naal, Florian D. (2011). Comparison of the reliability, responsiveness, and construct validity of 4 different questionnaires for evaluating outcomes after total knee arthroplasty. The Journal of arthroplasty, 26(6), 861-869. 

Jenny, J-Y, & Diesinger, Y. (2012). The Oxford Knee Score: compared performance before and after knee replacement. Orthopaedics & Traumatology: Surgery & Research, 98(4), 409-412. 

Jenny, Jean-Yves, Louis, Pascal, & Diesinger, Yann. (2013). High Activity Arthroplasty Score has a Lower Ceiling Effect Than Standard Scores After Knee Arthroplasty. The Journal of arthroplasty. 

Lin, Fang-Ju, Samp, Jennifer, Munoz, Alexis, Wong, Pei Shieen, & Pickard, A Simon. (2013). Evaluating change using patient-reported outcome measures in knee replacement: the complementary nature of the EQ-5D index and VAS scores. The European Journal of Health Economics, 1-8. 

Madsen, Majbritt, Larsen, Kristian, Madsen, Inger Kirkegård, Søe, Hanne, & Hansen, Torben Bæk. (2013). Late group-based rehabilitation has no advantages compared with supervised home-exercises after total knee arthroplasty. Danish medical journal, 60(4), A4607-A4607. 

Roland, Martin, & Morris, Richard. (1983). A study of the natural history of back pain: part I: development of a reliable and sensitive measure of disability in low-back pain. spine, 8(2), 141-144. 

Xie, Feng, Ye, Hua, Zhang, Yu, Liu, Xia, Lei, Ting, & LI, Shu‐Chuen. (2011). Extension from inpatients to outpatients: validity and reliability of the Oxford Knee Score in measuring health outcomes in patients with knee osteoarthritis. International journal of rheumatic diseases, 14(2), 206-210.

rehabilitation measures

More Instruments Like This

We have reviewed more than 500 instruments for use with a number of diagnoses including stroke, spinal cord injury and traumatic brain injury among several others.

updated Jan 9, 2024

Performance Assessment of ...

read more

updated Apr 10, 2023

Patient Activation Measure

read more

updated Oct 29, 2021

Brief Pain Inventory

read more