Validation of the short-form WOMAC function scale for the evaluation of osteoarthritis of the knee K. G. Auw Yang, N. J. H. Raijmakers, A. J. Verbout, W. J. A. Dhert, D. B. F. Saris From the University Medical Center, Utrecht, The Netherlands
K. G. Auw Yang, MD, Orthopaedic Resident, PhD Student N. J. H. Raijmakers, MSc, Orthopaedic Clinical Data Manager A. J. Verbout, MD, PhD, Orthopaedic Surgeon, Professor W. J. A. Dhert, MD, PhD, FBSE, Head of Orthopaedic Research D. B. F. Saris, MD, PhD, Orthopaedic Surgeon Department of Orthopaedics University Medical Center Utrecht, P. O. Box 85500, 3508 GA Utrecht, The Netherlands. Correspondence should be sent to Dr D. B. F. Saris; e-mail:
[email protected] ©2007 British Editorial Society of Bone and Joint Surgery doi:10.1302/0301-620X.89B1. 17790 $2.00 J Bone Joint Surg [Br] 2007;89-B:50-6. Received 17 February 2006; Accepted after revision 31 August 2006
50
This study validates the short-form WOMAC function scale for assessment of conservative treatment of osteoarthritis of the knee. Data were collected before treatment and six and nine months later, from 100 patients with osteoarthritis of the knee to determine the validity, internal consistency, test-retest reliability, floor and ceiling effects, and responsiveness of the short-form WOMAC function scale. The scale showed high correlation with the traditional WOMAC and other measures. The internal consistency was good (Cronbach α: 0.88 to 0.95) and an excellent test-retest reliability was found (Lin’s ρc): 0.85 to 0.94). The responsiveness was adequate concordance correlation coefficient (ρ and comparable to that of the traditional WOMAC (standardised response mean 0.56 to 0.44 and effect size 0.64 to 0.57) and appeared not to be significantly affected by floor or ceiling effects (0% and 7%, respectively). The short-form WOMAC function scale is a valid, reliable and responsive alternative to the traditional WOMAC in the evaluation of patients with osteoarthritis of the knee managed conservatively. It is simple to use in daily practice and is therefore less of a burden for patients in clinical trials.
In order to evaluate the treatment of osteoarthritis (OA) of the knee, various rating scales have been developed. Traditionally these were based on physical findings, such as the range of movement (ROM), knee laxity, deformation of the joint axis, and radiological variables1 which are usually evaluated according to the Kellgren and Lawrence grading system.2 During the last decade, outcome assessment has focused increasingly on analysis from the perspective of the patient. Because the treatment of OA primarily involves relief of symptoms, this has been a considerable improvement.3 The critical evaluation of treatment is an increasingly important topic owing to the rising prevalence of OA and its socioeconomic consequences.4 Many questionnaires and rating scales are available for the assessment of OA of the knee and its treatment, among which the Knee Society Clinical Rating Scale,5,6 the Knee Injury and Osteoarthritis Outcome Score (KOOS)7 and the Western Ontario and McMaster’s Universities Osteoarthritis Index (WOMAC)8 are most frequently used. The Knee Society Clinical Rating Scale is widely accepted as a reliable and valid measure in assessing patients undergoing total knee replacement.5 The Knee Society questionnaire consists of items on function and physical examination. This dual rating sys-
tem appears to have an advantage as it reflects the perspective of both the physician and the patient. However, it is more labour intensive, more sensitive to bias and less responsive to changes than the WOMAC.6 The KOOS is a 42-item self-administered self-explantory questionnaire which is an extension of the WOMAC and covers five dimensions: pain, symptoms, activities of daily living, sport/recreation, and knee-related quality of life (QOL). The KOOS has been shown to be a reliable and valid instrument7 and more responsive in younger patients than the WOMAC.9 Because of its greater length, the KOOS provides a detailed insight of different dimensions. However, this might also be a disadvantage, because of reduced patient compliance and incomplete response. The WOMAC is a self-assessed diseasespecific measure for patients with OA of the knee and hip, comprising 24 items in three dimensions: pain, function and stiffness.8 Short questionnaires are known to result in improved patient compliance and response rates, and are thought to improve the quality of the response.10-12 Therefore, Whitehouse et al13 developed a reduced version of the WOMAC, the short-form WOMAC function scale, which consists of seven items of funcTHE JOURNAL OF BONE AND JOINT SURGERY
VALIDATION OF THE SHORT-FORM WOMAC FUNCTION SCALE FOR THE EVALUATION OF OSTEOARTHRITIS OF THE KNEE
51
Table I. Summary of the items scored in the three different forms of the Western Ontario and McMaster Universities osteoarthritis index (WOMAC) scale tested in this paper
Traditional WOMAC
Stiffness 1. Stiffness in the morning 2. Stiffness later in the day
• •
• •
Pain 1. How often pain 2. Twisting/pivoting 3. Straightening fully 4. Bending fully 5. Walking on flat 6. Stair climbing 7. At night while in bed 8. Sitting or lying 9. Standing upright
• • • • •
• • • • •
Daily activity 1. Descending stairs 2. Ascending stairs 3. Rising from sitting 4. Standing 5. Bending on floor 6. Walking on flat 7. Getting in/out car 8. Going shopping 9. Putting on socks 10. Rising from bed 11. Taking off socks 12. Lying in bed 13. Getting in/out bath 14. Sitting 15. Getting on/off toilet 16. Heavy domestic 17. Light domestic
• • • • • • • • • • • • • • • • •
tion, which were selected by conducting interviews with orthopaedic and rheumatology staff. It has been shown to be a practical, valid, reliable and responsive alternative to the full WOMAC scale for the evaluation of total joint replacement.13 However, it has not been validated in a broader range of patients with OA undergoing conservative treatment. The aim of this study was to evaluate the short-form WOMAC function scale for patients with OA of the knee treated conservatively. As the short-form scale lacks assessment of pain and stiffness, which were originally incorporated in the traditional WOMAC, we also tested whether inclusion of these might result in improved validity, reliability and responsiveness in a modified short form version.
Patients and Methods We studied 100 patients with varying complaints from OA of the knee treated as outpatients between February and November 2004. They had a mean age of 54.9 years (34 to 85). There were 37 women and 63 men with a mean body mass index (BMI) of 27.5 (21.7 to 40). All patients completed the study and none was lost to follow-up. The main criteria for inclusion were typical clinical symptoms of OA, radiological grades of I to III on the Kellgren and Lawrence VOL. 89-B, No. 1, JANUARY 2007
Short-form WOMAC function scale (Whitehouse et al13)
Modified shortform WOMAC
Knee injury and osteoarthritis outcome score
• •
• •
• •
• •
• •
• •
•
•
scale;3 and aged over 18 years. All patients gave signed informed consent to the study. Patients who had participated in other trials within three months of inclusion, had received surgical or intra-articular pharmacological treatment within six months, had alcohol and drug abuse, concomitant painful or disabling disease of the spine, hips or lower limbs which would interfere with evaluation of the affected knee, a suspicion of ipsilateral coxarthrosis or loosening of a hip prosthesis, clinically significant or symptomatic vascular or neurological disease of the lower limbs and crystalline, inflammatory or infectious arthropathy were excluded from the study. All the patients completed the traditional WOMAC index once, and the short-form twice. The scores of the modified short-form version were deducted from the traditional WOMAC. This dataset was used to calculate the internal consistency and the test-retest reliability of these scoring systems. The agreement of the short-form WOMAC function scale with the traditional WOMAC and the modified short-form WOMAC function scale was used for the latter assessment. The items included in the respective questionnaires are shown in Table I. At the start of the study and at three and six months after treatment, orthopaedic surgeons (including DBFS) carried out physical examinations and completed the clinical part
52
K. G. AUW YANG, N. J. H. RAIJMAKERS, A. J. VERBOUT, W. J. A. DHERT, D. B. F. SARIS
Table II. Traditional Western Ontario and McMaster Universities osteoarthritis index (WOMAC), short-form WOMAC function scale and modified short-form WOMAC means and SD (0, worst, 100 best) and Cronbach’s α values for non-operatively treated osteoarthritis patients Osteoarthritis patients
Traditional WOMAC Short-form WOMAC Modified short-form WOMAC Cronbach’s α Traditional WOMAC Short-form WOMAC Modified short-form WOMAC
Pre-intervention
At 3 months
At 6 months
53.6 (15.2) 55.3 (16.5) 53.7 (15.0)
63.9 (20.4) 65.9 (21.1) 64.7 (20.2)
63.8 (22.0) 64.7 (22.6) 64.1 (21.8)
0.95 0.88 0.92
of the Knee Society Clinical Rating Scale. The patients were asked to fill out questionnaires, which included a Visual Analogue Scale (VAS) for pain, the patient part of the Knee Society Clinical Rating Scale and the KOOS. The scores of the traditional WOMAC, the short-form WOMAC and the modified short-form WOMAC were deducted from the KOOS items and were used to study the responsiveness and the floor and ceiling effects of these systems. The scores in the systems all ranged from 0 to 100, with low levels indicating severe symptoms. The study was approved by the local ethical committee. Statistical design and evaluations were monitored by the local centre for biostatistics and epidemiology, the Julius Center for Health Science and Primary Care, Utrecht, The Netherlands. Reliability. This term encompasses the internal consistency of a scale, usually measured as Cronbach’s α, which assesses the degree of correlation among items.14 A Cronbach’s α of 0.7 is widely accepted as acceptable; values between 0.7 and 0.9 are good, and above 0.9 are excellent, but may also indicate redundancies in the scale.15,16 The test-retest reliability of the short-form WOMAC was assessed by calculating Lin’s concordance correlation coefficient (ρc)17 for both the complete questionnaire and its items separately. This indicates how well a pair of measurements conforms to, or deviates from, a straight line and determines the degree of agreement, i.e. the extent to which the observations conform to a 45˚ line in a scatter plot when the traditional WOMAC scores are plotted against those of either the short-form WOMAC function scale or the modified short-form WOMAC.17 Validity. To examine the validity of a tool of assessment, the results should be tested by calculating the correlation with an accepted method of measurement. To determine the convergent construct validity of the short-form WOMAC function scale and the modified short-form WOMAC, Lin’s concordance correlation coefficient with the traditional WOMAC scale was determined. To visualise whether the accuracy of the short-form WOMAC would depend on the degree of symptoms, a Bland-Altman plot18 was used. This plots the difference between the two scores (the traditional
0.98 0.93 0.96
0.98 0.95 0.97
WOMAC and the short-form WOMAC) against the mean of the two scores.18,19 The Bland-Altman plot was also used to determine whether the accuracy of the scale would depend on the severity of symptoms by assessing whether a funnel or cone effect was observed when looking at the scatter of the points in the plot, as such an effect would imply a dependency. The upper and lower limits of agreement in the Bland-Altman plot were indicated. These were calculated by adding and subtracting twice the standard deviation (SD) of the differences to and from the mean of the difference. A paired t-test was used to assess whether there was evidence of bias, in which case there would be a systematic difference between the two sets of measurements, with one of the scoring systems tending to give larger values than the other. The floor and ceiling ranges of the short-form WOMAC function scale, the modified WOMAC and the traditional WOMAC were determined. These were defined as the percentage of respondents who scored the minimum and maximum levels on that scale respectively. Ideally, no more than 10% should be at the bottom or the top of the scale.20 Higher floor/ceiling effects indicate an insensitivity for detecting a change of symptoms. Responsiveness. Responsiveness was determined by examining the effect size and the standardised response means. The effect size is equal to the mean change in score from the baseline to three or six months, divided by the standard deviation of the baseline score.21 The standardised response mean is calculated as the mean change in score from the baseline to three or six months, divided by the standard deviation of the change in the score.22 A large value for the effect size or standardised response mean indicates that the technique has high sensitivity to change, and small values indicate either lack of responsiveness in the technique or no change over the time in the population studied. For effect size and standardised response mean, a value of 0.2 to 0.5 was regarded as small, 0.5 to 0.8 as medium and those above 0.8 as large.21 Statistical calculations were performed using the JMP version 5.0 software package (JMP Statistical Discovery Software, Cary, North Carolina). THE JOURNAL OF BONE AND JOINT SURGERY
VALIDATION OF THE SHORT-FORM WOMAC FUNCTION SCALE FOR THE EVALUATION OF OSTEOARTHRITIS OF THE KNEE
40 Difference (test – retest)
Modified short-form WOMAC function scale (re-test)
100 80 60 40 20 0
53
30 20
Upper limit of agreement
10
Mean difference
0 -10
Lower limit of agreement
-20 -30 -40
0
20
40
60
80
100
0
20
Modified short-form WOMAC function scale (test)
40
60
80
100
Mean value [(test + retest)/2] Fig. 1b
Fig. 1a
a) Lin’s concordance correlation graph of the test-retest data for the modified short-form WOMAC function scale and b) Bland-Altman plot showing the difference in the test-retest data of the short-form WOMAC function scale plotted against the mean value of these two measurements.
100 Modified short-form WOMAC
Short-form WOMAC function scale
100
80
60
40
20
80
60
40
20
ρc = 0.96 0
0
20
40
60
ρc = 0.98 80
100
Traditional WOMAC Fig. 2a
0
0
20
40
60
80
100
Traditional WOMAC Fig. 2b
Lin’s concordance correlation between a) the short-form WOMAC function scale and the traditional WOMAC and b) the modified short-form WOMAC and the traditional WOMAC.
Results Reliability. Cronbach α for the short-form WOMAC func-
tion scale was 0.95. Similarly, the values of Cronbach’s α for the traditional WOMAC and the modified short-form WOMAC were 0.98 and 0.97, respectively. Subanalysis at the baseline, three-months and six-months showed similar values (Table II). Test-retest reliability questionnaires were recorded at a mean interval of 9.1 hours (4 to 48). The Lin’s concordance coefficient (ρc) of each question is shown in Table III and varied from 0.85 to 0.91. Lin’s concordance coefficient (ρc) of the total score of the short-form WOMAC was higher at 0.94 (Fig. 1a). The Bland-Altman plot demonstrated that the mean differences between the test and re-test values of the shortform was 0.64 (SD 7.9) (Fig. 1b). No funnel effect or skewVOL. 89-B, No. 1, JANUARY 2007
ing of the data was observed, and therefore the accuracy appeared to be unaffected by repeated completion of the questionnaire. This plot demonstrates that the upper and lower limbs of agreement were 16.4 and -15.1, but showed test-retest differences of up to 35 points. Comparing the outcome of both measurements by performing a paired t-test did not show a significant difference (p = 0.416). Validity. Lin’s concordance correlation coefficient (ρc) of the short-form WOMAC function scale with the traditional WOMAC was 0.96 (Fig. 2a). Lin’s concordance correlation coefficient (ρc) of the modified short-form WOMAC with the traditional WOMAC was 0.98 (Fig. 2b). The Bland-Altman plot demonstrated that the mean differences between the values of the short-form WOMAC function scale and the traditional WOMAC was 0.23 (SD 5.8) (Fig. 3a). No funnel effect or skewing
54
K. G. AUW YANG, N. J. H. RAIJMAKERS, A. J. VERBOUT, W. J. A. DHERT, D. B. F. SARIS
Table III. Test-retest reliability: Lin’s concordance correlation coefficients (ρc), means and mean of differences with 95% confidence interval (CI) of the short-form Western Ontario and McMaster Universities osteoarthritis index (WOMAC) function scale Items
Test mean (SD)
Ascending stairs Rising from sitting Walking on flat Getting in/out car Putting on socks Rising from bed Sitting
1.7 1.6 1.2 1.5 1.2 1.4 1.1
Total score short-form WOMAC function scale
(1.1) (1.1) (1.0) (1.0) (0.9) (1.1) (1.0)
Re-test mean (SD) 1.7 1.6 1.3 1.5 1.1 1.3 1.2
65.5 (23.0)
(1.1) (1.1) (1.0) (1.1) (1.0) (1.1) (1.0)
65.0 (23.7)
Differences mean (95% CI)
Concordance correlation
-0.04 -0.02 -0.09 -0.02 0.01 0.08 -0.10
(-0.13 to 0.05) (-0.14 to 0.10) (-0.18 to 0.00) (-0.11 to 0.07) (-0.08 to 0.10) (-0.04 to 0.20) (-0.02 to 0.00)
0.91 0.85 0.90 0.90 0.90 0.85 0.86
0.64 (-0.93 to 2.22)
0.94
Table IV. Lin’s concordance correlation coefficients (ρc) between traditional, short-form and modified Western Ontario and McMaster Universities osteoarthritis index (WOMAC) scales and various measures of symptomatology of osteoarthritis patients, pre-treatment
*
KOOS pain KOOS stiffness KOOS activities of daily living KOOS sport KOOS quality of life Visual analogue scale for pain Knee Society score, function score
Traditional WOMAC
Short-form WOMAC
Modified short-form WOMAC
0.75 0.60 0.98 0.18 0.20 0.37 0.40
0.66 0.58 0.97 0.17 0.18 0.33 0.41
0.79 0.61 0.94 0.15 0.20 0.39 0.37
* KOOS, knee injury and osteoarthritis, outcome and score
Table V. Standard response means (SRM) and effect sizes (ES) of the traditional Western Ontario and McMaster Universities osteoarthritis index (WOMAC), short-form WOMAC function scale and modified short-form WOMAC at different time points compared with baseline in patients with osteoarthritis (n = 100) Short-form Traditional WOMAC WOMAC
3 months 6 months
Modified shortform WOMAC
SRM
ES
SRM
ES
SRM
ES
0.58 0.51
0.67 0.66
0.56 0.44
0.64 0.57
0.62 0.52
0.74 0.69
of the data was observed, and therefore the accuracy appeared to be independent of the symptoms experienced by the patient. The plot also demonstrates that the upper and lower limits of agreement were 11.8 and 11.4, i.e. the maximum likely difference between a pair of observations in the two scales is 11.8, on a scale that ranges from 0 to 100. The addition of pain and stiffness to the short-form WOMAC, the modified short-form WOMAC, resulted in similar values with a mean difference of 0.43 (SD) 2.4). The lower and upper limits of agreement were -4.37 and 5.23 (Fig. 3b). Comparing the outcome of both scoring systems with the traditional WOMAC by performing a paired t-test did not show a significant difference (t = 0.405, p = 0.686, and t = 1.771, p = 0.080, for the short-form WOMAC and the modified short-form WOMAC, respectively). At the baseline and at three and six months after treatment, comparison of the traditional WOMAC and the
short-form WOMAC gave similar mean values and standard deviations (Table II). Highly variable concordance correlation coefficients were found when the traditional WOMAC, the shortform WOMAC and the modified short-form WOMAC were related to the separate KOOS items, the VAS for pain and the Knee Society Score (Table IV). The floor and ceiling effects for all tools ranged from 0% to 7%, at baseline as well as at the assessments at three and six months. Responsiveness. When comparing the baseline with the evaluation at three-months, the standardised response means assessed by the short-form WOMAC function scale and the traditional WOMAC were 0.56 and 0.58, respectively. The effect sizes were 0.64 and 0.67. The modified short-form WOMAC resulted in a standardised response mean of 0.62, and an effect size of 0.74 (Table V). All indices in Table V are ‘medium’, apart from the standardised response means for the short-form WOMAC at six months, which has a value of 0.44. THE JOURNAL OF BONE AND JOINT SURGERY
Difference [Traditional WOMAC – SF WOMAC FS]
VALIDATION OF THE SHORT-FORM WOMAC FUNCTION SCALE FOR THE EVALUATION OF OSTEOARTHRITIS OF THE KNEE
15
Upper limit of agreement
10 5
Mean difference
0 -5 -10 -15
0
20
40
60
80
Lower limit of agreement 100
Mean value [(Traditional WOMAC + SF WOMAC FS)/2]
Difference [Traditional WOMAC – Modified SF WOMAC]
Fig. 3a
15 10
Upper limit of agreement
5
Mean difference Lower limit of agreement
0 -5 -10 -15
0
20
40
60
80
100
Mean value [(Traditional WOMAC + SF WOMAC FS)/2] Fig. 3b
Bland-Altman plot showing a) the difference in the traditional WOMAC and the short-form WOMAC function scale (SF WOMAC FS) plotted against the mean value of these two scales and b) the difference in the traditional WOMAC and the modified short-form WOMAC (modified SF WOMAC) plotted against the mean value of these two scales. The dotted line marked by mean difference represents the mean difference between the two scoring systems.
Discussion This study clearly demonstrates the validity of the shortformWOMAC function scale for the evaluation of the conservative treat ment of OA of the knee, as shown by a good internal consistency, the high concordance correlation coefficients with the traditional WOMAC, excellent testretest reliability and adequate responsiveness. This finding supports our view that the short-form WOMAC captures functional status as well as the traditional WOMAC in those circumstances. Our results agree with those published by Whitehouse et al13 for total hip and knee replacement, although his patients had more severe symptoms. The mean difference in scores measured by the traditional and the short-form WOMAC function scale is small (0.23 on a scale that ranges from 0 to 100) and appears independent of the severity of the symptoms. Excellent test-retest reliability was found for the short-form WOMAC for both the VOL. 89-B, No. 1, JANUARY 2007
55
completed questionnaire and its individual items. However, when individuals are considered the mean difference in score is 11.8 (SD 2) as compared with the traditional WOMAC and 0.64 (SD 7.9) upon repeated completion of the questionnaire. These findings are similar to those in previous studies and seem to be unaffected by the shortening of the score.23-30 Although the responsiveness three months after treatment was adequate it appears to decrease considerably at six months. However, the traditional WOMAC, and the modified short-form WOMAC showed similar trends and values, and we suggest that this decrease in responsiveness can be attributed to the efficacy of the treatment rather than to shortcomings of the questionnaire. As the responsiveness of the traditional WOMAC has previously been confirmed for a similar population, the responsiveness seen in the present study was not considered to be a limitation of the short-form WOMAC scale.8,31 Despite their proven reliability, validity and apparently adequate responsiveness, investigators should consider the limitations of both the traditional and the short-form WOMAC function scales, as demonstrated by the low concordance correlation coefficients of these questionnaires with the separate KOOS items, the VAS for pain and the Knee Society Score, except for the KOOS activities of daily living items. The addition of pain and stiffness parameters from the traditional WOMAC to the short-form WOMAC function scale did not result in improved reliability and validity, but did result in a slight improvement of the concordance correlation coefficient with the KOOS and, after six months of treatment. However, the modified short-form WOMAC has twice the number of items of the short-form WOMAC (14 vs 7), and the values of Cronbach’s α suggest that the additional items are redundant. This was previously demonstrated by Ryser et al,32 who showed redundancy of these parameters in the traditional WOMAC. Therefore, despite the slightly improved responsiveness, we discourage the use of the modified short-form WOMAC and recommend the use of the short-form WOMAC function scale. Recently, Tubach et al33 published another reduced WOMAC scale. Compared with the short-form WOMAC, they suggested that their scale would be more applicable for the evaluation of patients with less severe symptoms than those described by Whitehouse et al.13 However, in the present study the short-form WOMAC scale demonstrates similar validity and reliability as that of Tubach et al.33 In our study the symptoms of the patients (Table II) were less severe than in those studied by Tubach et al.33 Uniformity of the scoring systems used in orthopaedic research should be pursued. As the short-form WOMAC function scale is validated for a greater variety of joint disorders and methods of treatment than the reduced scale described by Tubach et al,33 we feel that the use of the short-form WOMAC is preferable. This study shows that the short-form WOMAC function scale developed by Whitehouse et al13 is not only reliable
56
K. G. AUW YANG, N. J. H. RAIJMAKERS, A. J. VERBOUT, W. J. A. DHERT, D. B. F. SARIS
and valid for the evaluation of total knee and hip replacement patients, but can also be used in the evaluation of patients with OA treated conservatively. Dr Saris is supported by The Netherlands Organisation for Health Research and Development (NWO); Dr Auw Yang is supported by the Anna Foundation. The authors thank Dr H. F. van Stel from the Julius Center (University Medical Center Utrecht) and Dr M. Schipper from the Centre of Biostatistics, University of Utrecht, for their statistical assistance with this study. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
References 1. Felson DT, McAlindon TE, Anderson JJ, et al. Defining radiographic osteoarthritis for the whole knee. Osteoarthritis Cartilage 1997;5:241-50. 2. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957;16:494-502. 3. Auw Yang KG, Saris DBF, Dhert WJA, Verbout AJ. Osteoarthritis of the knee: current treatment options and future directions. Curr Orthop 2004;18:311-20. 4. Buckwalter JA, Saltzman C, Brown T. The impact of osteoarthritis: implications for research. Clin Orthop 2004;427(Suppl):6-15. 5. Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop 1989;248:13-14. 6. Lingard EA, Katz JN, Wright RJ, et al. Validity and responsiveness of the Knee Society Clinical Rating System in comparison with the SF-36 and WOMAC. J Bone Joint Surg [Am] 2001;83-A:1856-64. 7. Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee injury and osteoarthritis outcome score (KOOS): development of a self-administered outcome measure. J Orthop Sports Phys Ther 1998;28:88-96. 8. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833-40. 9. Roos EM, Toksvig-Larsen S. Knee injury and Osteoarthritis Outcome Score (KOOS): validation and comparison to the WOMAC in total knee replacement. Health Qual Life Outcomes 2003;1:17. 10. Burchell B, Marsh C. The effect of questionnaire length on survey response. Quality and Quantity 1992;26:233-44. 11. Herzog AR, Bachman JG. Effects of questionnaire length on response quality. The public opinion quarterly 1981;45:549-59. 12. Kalantar JS, Talley NJ. The effects of lottery incentive and length of questionnaire on health survey response rates: a randomised study. J Clin Epidemiol 1999;52:111722. 13. Whitehouse SL, Lingard EA, Katz JN, Learmonth ID. Development and testing of a reduced WOMAC function scale. J Bone Joint Surg [Br] 2003;85-B:706-11. 14. Cronbach LJ, Warrington WG. Time-limit tests: estimating their reliability and degree of speeding. Psychometrika 1951;16:167-88. 15. Nunnaly JC. Assessment of reliability. In: Psychometric theory. 2nd ed. New York: MacGraw-Hill Series, 1978:225-55.
16. Streiner DL, Norman GR. Health measurement scales: a practical guide to their development and use. 2nd ed. Oxford: Oxford University Press, 1995:231. 17. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989;45:255-68. 18. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. 19. Moojen DJ, Saris DB, Auw Yang KG, Dhert WJ, Verbout AJ. The correlation and reproducibility of histological scoring systems in cartilage repair. Tissue Eng 2002;8:627-34. 20. Bridwell KH, Cats-Baril W, Harrast J, et al. The validity of the SRS-22 instrument in an adult spinal deformity population compared with the Oswestry and SF-12: a study of response distribution, concurrent validity, internal consistency, and reliability. Spine 2005;30:455-61. 21. Cohen J. Statistical power analysis for the behavioral sciences. Second ed. New Jersey: Lawrence Erlbaum Associates, 1988. 22. Liang MH, Fossel AH, Larson MG. Comparisons of five health status instruments for orthopaedic evaluation. Med Care 1990;28:632-42. 23. Faucher M, Poiraudeau S, Lefevre-Colau MM, et al. Assessment of the testretest reliability and construct validity of a modified WOMAC index in knee osteoarthritis. J Bone Joint Spine 2004;71:121-7. 24. Roorda LD, Jones CA, Waltz M, et al. Satisfactory cross cultural equivalence of the Dutch WOMAC in patients with hip osteoarthritis waiting for arthroplasty. Ann Rheum Dis 2004;63:36-42. 25. Wigler I, Neumann L, Yaron M. Validation study of a Hebrew version of WOMAC in patients with osteoarthritis of the knee. Clin Rheumatol 1999;18:402-5. 26. Bae SC, Lee HS, Yun HR, et al. Cross-cultural adaptation and validation of Korean Western Ontario and McMaster Universities (WOMAC) and Lequesne osteoarthritis indices for clinical research. Osteoarthritis Cartilage 2001;9:746-50. 27. Escobar A, Quintana JM, Bilbao A, Azkarate J, Guenaga JI. Validation of the spanish version of the WOMAC questionnaire for patients with hip or knee osteoarthritis: Western Ontario and McMaster Universities Osteoarthritis Index. Clin Rheumatol 2002;21:466-71. 28. Guermazi M, Poiraudeau S, Yahia M, et al. Translation, adaptation and validation of the Western Ontario and McMaster Universities osteoarthritis index (WOMAC) for an Arab population: the Sfax modified WOMAC. Osteoarthritis Cartilage 2004;12:459-68. 29. Roos EM, Klassbo M, Lohmander LS. WOMAC osteoarthritis index: reliability, validity, and responsiveness in patients with arthroscopically assessed osteoarthritis. Scand J Rheumatol 1999;28:210-15. 30. Salaffi F, Leardini G, Canesi B, et al. Reliability and validity of the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index in Italian patients with osteoarthritis of the knee. Osteoarthritis Cartilage 2003;11:551-60. 31. Angst F, Aeschlimann A, Steiner W, Stucki G. Responsiveness of the WOMAC osteoarthritis index as compared with the SF-36 in patients with osteoarthritis of the legs undergoing a comprehensive rehabilitation intervention. Ann Rheum Dis 2001;60:834-40. 32. Ryser L, Wright BD, Aeschlimann A, Mariacher-Gehler S, Stucki G. A new look at the Western Ontario and McMaster Universities Osteoarthritis Index using Rasch analysis. Arthritis Care Res 1999;12:331-5. 33. Tubach F, Baron G, Falissard B, et al. Using patients’ and rheumatologists’ opinions to specify a short form of the WOMAC function subscale. Ann Rheum Dis 2005;64:75-9.
THE JOURNAL OF BONE AND JOINT SURGERY
