Validity and reliability of the Japanese Orthopaedic Association score for osteoarthritic knees

J Orthop Sci (2012) 17:750–756 DOI 10.1007/s00776-012-0274-0

ORIGINAL ARTICLE

Validity and reliability of the Japanese Orthopaedic Association score for osteoarthritic knees Masayoshi Okuda • Shohei Omokawa • Kohjiro Okahashi • Manabu Akahane • Yasuhito Tanaka

Received: 18 August 2011 / Accepted: 16 July 2012 / Published online: 7 August 2012 Ó The Japanese Orthopaedic Association 2012

Abstract Background A variety of outcome measures are available to evaluate physical impairment and disability in patients with knee osteoarthritis, and most physician-rated measures are not validated. The objective of this study was to assess the validity and reliability of an observer-based knee scoring system of the Japanese Orthopaedic Association (the JOA) commonly used in Japanese clinical practice, and to determine demographic variables affecting the score. Methods A consecutive series of 85 patients with primary knee osteoarthritis completed the JOA (four domains pain on walking, pain on ascending or descending stairs, range of motion, and joint effusion), two validated patient-rated measures including the generic instrument of the Medical Outcomes Study 36-Item Short-Form (the SF-36) Health Survey, and the disease-specific scale of the Japanese Knee Osteoarthritis Measure (the JKOM), and a performance based timed-up-and-go test (TUG). Concurrent validity was determined by examining correlations of the JOA with

M. Okuda (&) Department of Orthopaedic Surgery, Ishinkai-Yao General Hospital, 1-41 Numa, Yao, Osaka 581-0036, Japan e-mail: [email protected] S. Omokawa
Y. Tanaka Department of Orthopaedic Surgery, Nara Medical University, Nara, Japan K. Okahashi Department of Orthopaedic Surgery, Nara Prefectural Nara Hospital, Nara, Japan M. Akahane Department of Public Health, Health Management and Policy, Nara Medical University School of Medicine, Nara, Japan

123

the SF-36 and the JKOM. Construct validity was verified by correlating each domain of the JOA with objective measurements of TUG using Spearman’s rank correlation coefficient. Intra- and interobserver reliability and internal consistency of the JOA were evaluated with another cohort of 32 patients who had a knee disorder at baseline and again at a mean of 18 days later. Results The JOA was significantly correlated with validated patient-rated outcome measures (the JKOM, the SF-36), indicating concurrent validity of the JOA. Domains of the JOA had significant correlations with the TUG, showing adequate construct validity. Intra- and interobserver reliability for the JOA showed a moderate to almost perfect agreement, and internal consistency of Cronbach’s a indicated that the JOA score was a highly reliable instrument to assess knee osteoarthritis. As a demographic variable, age was highly correlated with the JOA. Conclusions The JOA, generally used as an observerderived knee scoring system, is a valid and reliable tool for evaluating functional status in patients with knee osteoarthritis.

Introduction Symptomatic knee osteoarthritis (OA) is present in one of eight persons older than 60 years and is associated with significant limitation of activities in daily living in England [1]. Several outcome measures have been devised to assess the functional status of patients with knee problems. Although several instruments are available to evaluate physical impairment and disability in patients with knee OA, most of the physician-rated the measures are not validated, and it is obscure how these measures demonstrate the exact function of the knee joint. The Japanese Orthopaedic

Validity and reliability of JOA score for knee OA

Association (the JOA) score for knee osteoarthritis is one of the most frequently used for knee scoring in Japanese clinical practice. In the field of clinical outcome assessment, there has been increased demand for high-quality outcome measures after medical and surgical interventions. These need to be valid, reliable, and have high responsiveness, for as low a cost as possible. Several new self-administered questionnaires have been developed [2–7]. These tests consist of three kinds of instruments. First, there are generic tests such as the Medical Outcomes Study 36-Item ShortForm Health Survey (the SF-36) with questions regarding patients’ health in general, influenced by comorbidity. Second, there are disease-specific scores such as the Japanese Knee Osteoarthritis Measure (the JKOM) and the Western Ontario and McMaster Universities Osteoarthritis Index (the WOMAC). Last, there are patient-specific tests such as the McMaster Toronto Arthritis Patient Preference Disability Questionnaire. General Health Questionnaires allow the effect of treatment to be compared in various medical conditions [8]. These questionnaires have been tested for validity, reliability, and responsiveness. However, there is a lack of information concerning the validity, reliability, and responsiveness of the JOA. The objective of this study was to assess the validity and reliability of the JOA, which is commonly used for the evaluation of patients with OA, and to determine demographic variables affecting the score.

Materials and methods Patients The study sample comprised 117 consecutive patients with clinical and radiographic diagnosis of primary knee OA according to the American Academy of Rheumatology criteria [9]. All patients had OA with a Kellgren Lawrence grade of 2 or more with narrowing of the joint space and sclerosis of the subchondral bone. Patients who had subjective symptoms of pain and/or stiffness of the unilateral knee were included in this study. Those who had previous fractures around the knee joint, acute disorders including a bruise and/or sprain of the knee, and inflammatory knee arthritis were excluded. Patients who had neurological problems such as either hemiplegia or symptomatic lumber canal stenosis were not included in the study. Significant symptoms of the ipsilateral hip joint, back or the contralateral leg, and other systemic joint diseases were also excluded. All patients were given a study packet and asked to complete an informed consent form on the initial and follow-up visits. The informed consent form and protocol were approved by the institutional review board before enrolling the patients.

751

Eighty-five of the 117 patients completed the JOA physician-rated measure, two validated patient-rated measures including the SF-36 and the JKOM. In addition, the patients underwent a performance-based 3-m timed-upand-go test (TUG) on the initial visit to the hospital. The mean age of patients was 67.3 years (range 40–83), 65 were women, and mean body mass index (BMI) was 23.8. We conducted multiple measurements of the JOA in the other 32 patients to assess intra- and interobserver reliability. The mean age of these patients was 69.8 years (range 51–79); 20 were women. Two independent observers (two board-certified orthopedic physicians) met all 32 patients, and measurements of the JOA were performed. One observer (M.O.) performed a second measurement at a mean of 18 days (range 14–28 days) after the first measurement. Measurement scales The JOA consists of four domains with a maximum of 100 points: pain on walking (30 points), pain on ascending or descending stairs (25 points), mobility (35 points), and joint effusion (10 points) (Table 1). High scores indicate good function. The questionnaire can be completed in a few minutes. The SF-36, which is one of the most frequently used health-status measures, is self-administered and can be completed in about 10 min. It consists of 36 questions divided into eight domains: physical function, social function, role-physical, role-emotional, bodily pain, general health, mental health, and vitality [10]. The SF-36 was translated into Japanese and tested by Fukuhara et al. [5] for validity and reliability. A stratified comparison with the general population in Japan was also done according to age in decades and gender. The JKOM is a self-administered, disease-specific measure consisting of 25 items, which include patient pain in level walking, standing or climbing stairs, and physical functions related to the activities of daily living and social functions including participation. It can be completed in about 10 min. It is divided into four domains: pain and stiffness in the knees, condition in daily life, general activities, and health conditions. Low scores indicate good function. The JKOM was proved to have sufficient reliability and validity by means of statistical evaluation and comparisons with two other health-related scales, the WOMAC and the SF-36 [7]. Statistical analyses We calculated the intraclass correlation coefficient (ICC) to assess intra- and interobserver reliability of the JOA. The interpretation of ICC value was based on the criteria

123

752

M. Okuda et al.

Table 1 The JOA

Table 2 ICC for intra- and interobserver reliability for the JOA

Pain on walking

Intraobserver reliability

Interobserver reliability

Pain on walking

0.54 (0.24–0.74)

0.75 (0.55–0.87)

20

Pain on ascending or descending stairs

0.69 (0.45–0.83)

0.67 (0.43–0.83)

Walking 100 m or more, but less than 500 m without regard to pain

15

Range of motions Joint effusion

0.77 (0.58–0.88) 0.82 (0.67–0.91)

0.54 (0.01–0.61) 0.58 (0.30–0.77)

Walking indoors or nearby, but less than 100 m without regard to pain

10

Total

0.90 (0.81–0.95)

0.71 (0.50–0.85)

Walking 1 km or more usually with no pain, but without regard to mild pain, rarely felt, with certain activity

30

Walking 1 km or more regardless of pain

25

Walking 500 m or more, but less than 1 km without regard to pain

Values in parentheses indicate 95 % confidence interval

Inability to walk

5

Inability to stand

0

ICC interclass correlation coefficient

Pain on ascending or descending stairs No pain

25

Pain relieved by using handrails

15

Pain with handrails, but no pain with step-by-step ambulation

20

Pain with step-by-step ambulation: pain relieved by using handrails

10

Pain even with step-by-step ambulation and handrail use

5

Inability to ascend or descend because of pain

0

Range of motion Squatting

35

Sideways or cross-legged sitting

30

Flexion or arc of motion of 110° or more

25

Flexion or arc of motion of 75° or more

20

Flexion or arc of motion of 35° or more

10

Flexion or arc of motion less than 35° including ankylosis or severe flexion contracture

0

Joint effusion No edema, no swelling

10

Occasional puncture required

5

Frequent puncture required

0

JOA the Japanese Orthopaedic Association score for osteoarthritic knees. Total 100 points

proposed by Landis and Koch [11]. An ICC value of 0.00–0.20 represents slight agreement, 0.21–0.40 fair agreement, 0.41–0.60 moderate agreement, 0.61–0.80 substantial agreement, and [0.81 almost perfect agreement. The 32 patients were used to establish internal consistency of the JOA. Overall internal consistency for all four domains was determined. Cronbach’s a [0.70 was considered acceptable [12]. Questionnaires with good content validity are expected to have fewer answers that score the lowest and highest possible (floor and ceiling effects). Construct validity was verified by examining for any significant correlation of the JOA with the objective TUG measurement. Divergent construct validity and convergent construct validity in the domains of pain and physical function and total score were tested using Spearman’s correlation coefficient. We hypothesized that when one domain in the JOA correlated with the corresponding

123

domain in the SF-36 and the JKOM, it was considered convergent construct validity. When one domain correlated with the corresponding domain but not with other domain, it was considered divergent construct validity. Concurrent validity was expected to be present when the scores correlated with a similar validated measure (the SF-36, the JKOM) of the condition being evaluated. Because the SF-36 is the gold standard of health-related QOL and the JKOM is a validated measure to assess the pain and disability of knee OA, we used these scores for verification of concurrent validity of the JOA. For divergent and convergent construct validity, and concurrent validity, a significance level for Spearman’s q was set greater than 0.40 and P \ 0.01. The correlations between demographic variables including the age and BMI and the JOA were analyzed with using Spearman’s correlation coefficient in order to determine demographic variables affecting the score. Significance level for Spearman’s q was set greater than 0.40 and P \ 0.01. We used SPSS version 13.0 for Windows for the statistical analysis (SPSS Inc., Chicago, IL, USA).

Results All the 117 patients completed evaluations. One patient did not answer some of the individual items of the SF-36. Intraobserver reliability There was almost perfect intraobserver reliability for the overall JOA (ICC 0.90) (Table 2). Intraobserver reliability for the JOA showed a moderate agreement for the domains of pain on walking (ICC 0.54), a substantial level for the domains of pain on ascending or descending stairs (ICC 0.69) and the domains of mobility (ICC 0.77), and an almost perfect level for the domains of joint effusion (ICC 0.82).

Validity and reliability of JOA score for knee OA

753

Table 3 Result of domain scores for the JOA, the JKOM, and the SF-36 Score

Domain

N

Mean

Median

JOA

Pain on walking

85

20

20

SF-36

JKOM

SD 7.5

Range

Minimum (no. of floor)

Maximum (no. of ceiling)

25

5

30 (13)

Pain on stairs

85

12

10

7.8

25

0 (7)

25 (7)

Function

85

27

25

5.4

25

10

35 (17) 58.7

Physical function

85

27

30

18.0

70

-11.8

Role physical

85

33

32

15.8

55

1.7

56.2

Bodily pain

84

38

40

10.0

44

17.2

61.4

Social function Pain

84 85

40 22

44 22

13.6 7.0

53 30

4.5 10 (1)

57.1 40 (2)

Physical function

85

23

21

9.6

36

10 (2)

46

Social function

85

13

11

5.7

20

5 (2)

25 (4)

General health

85

5

5

1.9

8

4 (2)

10 (2)

JOA the Japanese Orthopaedic Association score for osteoarthritic knees, SF36 the Medical Outcomes Study 36-Item Short-Form Health Survey, JKOM the Japanese Knee Osteoarthritis Measure

There was substantial interobserver reliability for the overall JOA (ICC 0.71). Interobserver reliability for the JOA showed substantial agreement for the domains of pain on walking (ICC 0.75) and the domains of pain on ascending or descending stairs (ICC 0.67), and a moderate level for the domains of mobility (ICC 0.54) and the domains of joint effusion (ICC 0.58).

Table 4 Correlation of demographic variables (age and BMI) and TUG with the scores determined using Spearman’s correlation Score

JOA

SF-36

JKOM

Age

-0.54*

-0.53*

0.50*

BMI

0.07

0.05

0.07

TUG

-0.66*

-0.59*

0.73*

BMI body mass index, TUG times up-and-go test

Internal consistency Cronbach’s a was 0.84, indicating acceptable internal consistency of more than 0.70 based on the Cortina’s criteria [12]. Content validity The means, medians, standard deviations, and ranges of the JOA, JKOM and SF-36 are shown in Table 3. The numbers of ceiling scores of the JOA were identified. Seven patients had the minimum disability score of zero (floor) for pain when climbing stairs. Thirteen, 7, and 17 patients had a maximum disability score (ceiling) for pain on walking, pain on ascending or descending stairs, and daily function, respectively. There were fewer ceiling values for both pain and function in the JKOM and the SF-36 than in the JOA. Construct validity The total JOA score was significantly correlated with TUG (q = -0.66; P \ 0.0001) (Table 4). TUG correlated with domains of pain in the JOA (pain and disability during walking domain, q = -0.64; P \ 0.0001 and pain and disability during stairs domain, q = -0.63; P \ 0.0001). The function domain in the JOA was correlated with the function scores in the SF-36 (q = 0.44; P \ 0.0001) and

* P \ 0.0001 (two-tailed)

Table 5 Correlations between the domains and the total scores in the JOA, the SF-36, and the JKOM determined using Spearman’s correlation Score

SF-36

JKOM

Domain

Pain

Function

Total

Pain

Function

Total

Pain

0.35

0.43*

0.54*

-0.54*

-0.58*

-0.60*

Function

0.37

0.44*

0.49*

-0.46*

-0.64*

-0.61*

Total

0.41*

0.49*

0.55*

-0.58*

-0.70*

-0.69*

* P \ 0.000 1 (2 tailed), n = 85

the JKOM (q = -0.64; P \ 0.0001) (Table 5). The pain domain of the JOA had a significant correlation with the pain score of the JKOM (q = -0.54; P \ 0.0001) but had no significant correlation with the pain score of the SF-36 (q = 0.35). The function domain of the JOA had no significant correlation with the pain scores of the SF-36 (q = 0.37) but had a significant correlation with the pain scores of the JKOM (q = -0.46; P \ 0.0001). The pain domain of the JOA had a significant correlation with the function scores of the SF-36 (q = 0.43; P \ 0.0001) and the JKOM (q = -0.58; P \ 0.0001).

123

754

M. Okuda et al.

Discussion

the JOA

y= 0.6249x + 48.204

the SF-36 Fig. 1 Relationship between the total scores of the JOA and the SF-36. There was a significant correlation between the JOA and the SF-36 (q = 0.55; P \ 0.0001)

the JOA

y= -0.6457x + 105.99

the JKOM

Fig. 2 Relationship between the total scores of the JOA and the JKOM. There was a significant correlation between the JOA and the JKOM (q = -0.69; P \ 0.0001)

Concurrent validity The total score for the JOA had a significant correlation with those of the SF-36 (q = 0.55; Fig. 1) and the JKOM (q = -0.69; Fig. 2). As a demographic variable, age demonstrated a moderate correlation with the JOA (q = -0.54; P \ 0.001), the SF-36 (q = -0.53; P \ 0.001), and the JKOM (q = 0.50; P \ 0.001) (Table 4). There were almost no correlations between the three questionnaires (the JOA, the SF-36 and the JKOM) and the BMI (q = 0.07, 0.05, and 0.07; P = 0.64, 0.75 and 0.62, respectively) (Table 4).

123

This is one of the first studies to validate the JOA, which is commonly used in Japanese clinical practice as an observer-derived knee scoring system in patients with knee OA. The number of available outcome instruments has increased substantially, and physician-rated scoring systems are frequently used in clinical practice for evaluating a patient’s functional status. Recently, there have been several reports examining the validity and reliability of these scoring systems. Briggs et al. [13] examined the validity of the Lysholm and Tegner scores using the SF-12 and the International Knee Documentation Committee score. They showed that use of these scores—a generic measure of health-related quality of life and a measure of patient satisfaction, respectively—provided a comprehensive outcome assessment for patients with anterior cruciate ligament injuries. The Harris Hip score is one of the most frequently used disease-specific but not self-administered hip scoring systems. So¨derman et al. [14] examined the validity and reliability of the Harris hip score by analyzing correlation of the score with both the SF-36 and the WOMAC. They concluded the Harris hip score to be useful for a physician or a physiotherapist in studying clinical outcomes after hip replacement surgery. Based on reliability testing in the current study, the JOA was found to have good intra- and interobserver reliability and internal consistency. An ICC value exceeded 0.9, indicating that the score was highly reliable in repeated testing and was stable over time. We observed Cronbach’s a of 0.84 for internal consistency, indicating that the reliability of the JOA was superior to that of the Lysholm score (Cronbach’s a of 0.72), which is a similar physicianrated knee scoring system [13]. The significant correlations between the same domains of the JOA and the JKOM and between those of the JOA and the SF-36 indicate that the JOA showed convergent construct validity as a physician-rated measure. No significant correlation between the function domain of the JOA and the pain domain of the SF-36 partly indicated the divergent construct validity of the JOA. However, the JOA showed a limited divergent construct validity, because there were significant correlations between the pain domain of the JOA and the function domains of the JKOM and the SF-36. We consider that this result was attributable to the uniqueness of the JOA, in which the pain domain includes functions of gait and stepping the stairs ability. The TUG is an objective performance test and is a useful tool for quantifying function of mobility because of its validity and reliability [15]. The total score and each domain of the pain on walking and on ascending or descending stairs in the JOA had significant correlations with TUG, showing a sufficient construct validity for the JOA.

Validity and reliability of JOA score for knee OA

Several reports have described the validity of the SF-36 and the WOMAC applied to patients with OA and other conditions [3–8, 10, 16]. So¨derman et al. [10] demonstrated that both the generic instruments (the NHP and the SF-36) and the disease-specific test (WOMAC) had high validity, measuring what they were supposed to measure in patients with hip OA. Akai et al. [7] verified the validity and reliability of the JKOM for patients with knee OA by comparison with both the SF-36 and the WOMAC. The JOA was significantly correlated with both the SF-36 and the JKOM, which are validated outcome measures, indicating high concurrent validity of the JOA. The JOA was correlated with both the generic instrument (the SF-36) and the disease-specific test (the JKOM). Correlations of the JOA were markedly higher with the JKOM (q = -0.69) than with the SF-36 (q = 0.55). This result showed that the JOA, which is a region-specific measure, was similar to a disease-specific measure. Disease-specific scales were more sensitive than general health status measures for evaluating the outcomes of orthopedic procedures [6, 8, 10]. The SF-36 has been validated in this population and is widely used for measuring health outcomes [3, 16–18]. The SF-36 is a generic questionnaire and has the potential limitation of being less sensitive to clinical change in patients with complaints specific to an anatomical region or disease process. The JKOM and a variety of region-specific and disease-specific questionnaires were designed to increase responsiveness in measuring the outcomes of patients with knee complaints. A complete evaluation of function following surgery such as knee arthroplasty requires the use of both a specific and a general questionnaire [2, 19, 20]. Adequate evaluations can be executed not only by an observer-based scoring system but also by a patient-based measure when we try to assess the function of patients. The SF-36 includes several questions to measure functional health of the lower extremity, and kneeling and bending activity of items 3–6 reflect the specific functions related to the Japanese lifestyle. The JOA is measured based on the activities of daily living in Japan. These similarities of questions may have contributed to the significant correlation between the function domains of the JOA and the SF-36. Understandably, the patient’s age affected all of the outcome measures in the current study, demonstrating that the patient’s functional status decreased with age. This result supports the theory that the SF-36 measure is distributed by sex, age, social class, and use of health services [3, 4]. One limitation of this study was that the sample size was small. Nonetheless, the current analysis demonstrated a significant validity and reliability of the JOA score. Relatively strict exclusion criteria of the patients might have contributed to the homogeneity of the patient group,

755

leading to significant results. Another limitation was that the responsiveness of the JOA was not investigated because we only examined the primary status of the knee before interventions. Further studies will obtain more evidence on the responsiveness of the JOA in other patient populations related to knee problems.

Conclusion The JOA, which is generally used as an observer-derived knee scoring system, is a reliable and valid tool for evaluating functional status in patients with knee osteoarthritis. Acknowledgements We wish to thank Akio Iida M.D. and Takumi Ohkawa M.D. for their assistance of the measurement. We are also grateful to all of the orthopaedic staff members in Ishinkai General Hospital. Conflict of interest The authors did not receive and will not receive any benefits or funding from any commercial party related directly or indirectly to the subject of this article.

References 1. Conagham PG, Emerton M, Tennant A. Internal construct validity of the Oxford Knee Scale: evidence from Rasch measurement. Arthr Rheum. 2007;57(8):1363–7. 2. Bellamy N, Buchanan W, 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 and the knee. J Rheumatol. 1988;15:1833–40. 3. Brasier JE, Harper R, Jones NMB, O’Cathain A, Thomas KJ, Usherwood T, Weslake L. Validating the SF-36 health survey questionnaire: new outcome measure for primary care. BMJ. 1992;305:160–4. 4. Ware JE, Sherbourne CD. 36-Item Short-Form Health Survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30:473–83. 5. Fukuhara S, Bito S, Green J, Hsiao J, Kurokawa K. Translation, adaptation, and validation of the SF-36 health survey for use in Japan. J Clin Epidemiol. 1998;51:1037–44. 6. Nelson FS, Allen PM, John GS, Gray RM. Evaluation of the construct validity of the DASH questionnaire by correlation with the SF-36. J Hand Surg [Am]. 2002;27:537–41. 7. Akai M, Doi T, Fujino K, Iwata T, Kurosawa H, Nasu T. An outcome measure for Japanese people with knee osteoarthritis. J Rheumatol. 2005;32:1524–31. 8. Bombardier C, Melfi CA, Paul J, Green R, Hawker G, Wright J, Coyte P. Comparison of a generic and a disease-specific measure of pain and physical function after knee replacement surgery. Med Care. 1995;33:131–44. 9. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, Christy W, Cooke TD, Greenwald R, Hochberg M. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Arthr Rheum. 1986;29(8):1039–49. 10. So¨derman P, Malchau H. Validity and reliability of Swedish WOMAC osteoarthritis index: a self-administered disease-

123

756

11. 12. 13.

14.

15.

16.

M. Okuda et al. specific questionnaire (WOMAC) versus generic instruments (SF-36 and NHP). Acta Orthop Scand. 2000;71:39–46. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159–74. Cortina JM. What is coefficient alpha? An examination of theory and applications. J Appl Psychol. 1993, 98–104. Briggs KK, Lysholm J, Tegner Y, Rodkey WG, Kocher MS, Steadman JR. The reliability, validity, and responsiveness of the Lysholm score and Tegner activity scale for anterior cruciate ligament injuries of the knee. Am J Sports Med. 2009;37:890–7. So¨derman P, Malchau H. Is the Harris Hip Score System useful to study the outcome of total hip replacement? Clin Orthop. 2001;384:189–97. Podsiadlo D, Richardson S. The timed ‘‘Up & Go’’: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 2000;48:104–5. Martin DP, Engelberg R, Agel J, Swiontkowski MF. Comparison of the musculoskeletal function assessment questionnaire with the

123

17.

18.

19.

20.

Short Form-36, the Western Ontario and McMaster Universities Osteoarthritis Index, and the Sickness Impact Profile health-status measures. J Bone Joint Surg Am. 1997;79:1323–35. Liang MH, Fossel AH, Larson MG. Comparisons of five health status instruments for orthopedic evaluation. Med Care. 1990;28:632–42. Bessette L, Sangha O, Kuntz KM, Keller RB, Lew RA, Fossel AH, Katz JN. Comparative responsiveness of generic versus diseasespecific and weighted versus unweighted health status measures in carpal tunnel syndrome. Med Care. 1998;36:491–502. Rorabeck CH, Bourne RB, Laupacis A, Feeny D, Wong C, Tugwell P, Leslit K, Bullas R. A double-blind study of 250 cases comparing cemented with cementless total hip arthroplasty. Clin Orthop. 1994;298:156–64. Wright JG, Young NL. The Patient-Specific Index: asking patients what they want. J Bone Joint Surg Am. 1997;79:974–83.