Now offering special education grants for members.
To assess pain, stiffness, and physical function in patients with hip and / or knee osteoarthritis (OA)1
The WOMAC consists of 24 items divided into 3 subscales:
The WOMAC was developed for use among patients with knee and / or hip OA, but it has been used among patients with different conditions, including: low back pain30, rheumatoid arthritis31, juvenile rheumatoid arthritis32, systemic lupus erythematosus33, and fibromyalgia31. The WOMAC has been extensively used in both observational / epidemiological studies and to examine changes following treatments including pharmacotherapy, arthroplasty, exercise, physical therapy, knee bracing, and acupuncture.
WHO ICF Components34
b28016 (pain in joints), b289 (sensation of pain, other specified and unspecified), b134 (sleep functions), b7603 (supportive functions of arm or leg), b7800 (sensation of muscle stiffness),b7808 (sensations related to muscles and movement functions, other specified), d4100 (lying down), d4101 (squatting), d4103 (sitting), d4105 (bending), d4150 (maintaining a lying position, d4153 (maintaining a sitting position), d4154 (maintaining a standing position), d4400 (picking up), d4102 (transferring oneself while lying), d450 (walking), d4500 (walking short distances),d4501 (walking long distances), d4502 (walking on different surfaces), d4551 (climbing), d4559 (moving around unspecified), d498 (mobility, other specified), d5101 (washing whole body), d530 (toileting), d5402 (putting on footwear), d5403 (taking off footwear), d6200 (shopping), d699 (domestic life, unspecified), d9208 (recreation and leisure, unspecified), d6408 (doing housework)
Patient report questionnaire can be completed in person, over the telephone, or by computer.
The use of the WOMAC has also been validated for delivery via mobile phone (m-WOMAC)>. Osteoarthritis Index delivered by mobile phone (m-WOMAC) is valid, reliable, and responsive. Bellamy N, Wilson C, Hendrikz J, Whitehouse SL, Patel B, Dennison S, Davis T; EDC Study Group.J Clin Epidemiol. 2011 Feb;64(2):182-90. Epub 2010 Jul 6.
Training: Minimal instruction needed. User guide available1.
Time to administer/complete: Approximately 12 minutes.
Equipment needed: Copy of instrument. Ruler, if using Visual Analog format.
The WOMAC is a proprietary health status questionnaire protected by copyright and trademark. To obtain permission to use the WOMAC for research purposes, a request can be submitted at www.womac .org. Licensing processes and costs are determined on the basis of information specific to each research project.
Scale. The Likert Scale version uses the following descriptors for all items: none, mild moderate, severe, and extreme. These correspond to an ordinal scale of 0-4. The 100mm Visual Analog version uses anchors of no pain/stiffness/difficulty and extreme pain/stiffness/difficulty.
Score Range. On the Likert Scale version, the scores are summed for items in each subscale, with possible ranges as follows: pain=0-20, stiffness=0-8, physical function=0-68. On the Visual Analog version, a ruler is used to measure the distance (in mm) from the left end marker to the patient's mark. For each item, the possible range of scores is therefore 0-100. Items are summed for each subscale, resulting in possible ranges as follows: pain=0-500, stiffness=0-200, physical function=0-1700. Most commonly, a total WOMAC score is created by summing the items for all three subscales. However, other methods of aggregating scores have been used1.
Higher scores on the WOMAC indicate worse pain, stiffness, and functional limitations.
The WOMAC is typically scored by hand, using the conventions described above.
Minimal. User Guide provides instructions1.
There are no clear norms available.
Internal consistency. In a small randomized controlled trial of 2 nonsteroidal anti-inflammatory drugs among patients with knee and hip OA (N = 57), Cronbach's alphas for the Likert Scale format of the WOMAC were 0.86-0.89, 0.90-0.91, and 0.95 for the pain, stiffness, and function subscales, respectively35. In the same study, Cronbach's alphas for the Visual Analog format were 0.7.-0.81 for pain and 0.89-0.91 for function (not calculated for the stiffness subscale). Among patients in the Swedish Knee Arthroplasty Registry (n=1,014), Cronbach's alphas were 0.91 for pain and stiffness and 0.98 for physical function36. Overall, results support the internal consistency of the WOMAC subscales37.
Test-retest. In the clinical trial of nonsteroidal anti-inflammatory drugs for patients with hip or knee OA (n=57)35, with a 1-week interval, Kendall's TAU-C for Likert Scale formats were 0.68, 0.48, and 0.68 for pain, stiffness, and function subscales, respectively, and 0.68 for the total score. Corresponding values for the Visual Analog format were 0.64, 0.61, 0.72, and 0.64. Among patients with arthroscopically assessed knee OA (n=52), the test-retest reliability (intraclass correlation coefficients; ICC's) were 0.74, 0.58, and 0.92, for pain, stiffness, and physical function subscales, respectively26. In the Swedish Knee Arthroplasty Registry, Intraclass Correlations (ICC's; n=1,014) over a 3-week period were 0.95, 0.90, and 0.92 for pain, stiffness, and function, respectively36. Adequate test-retest reliability has also been confirmed for the German and Swedish versions of the WOMAC 16, 25. Overall, test-retest reliability of the WOMAC pain subscale has been variable across studies but generally meets the minimum standard; test-retest reliability has been more consistent and stronger for the physical function subscale, but the stiffness subscale has shown low test-retest reliability37.
Rater. In a study of patients undergoing hip arthroplasty (n=78), in which the WOMAC was administered by an interviewer, ICCs for intra-rater reliability ranged from 0.53-0.78 and for inter-rater reliability, 0.62-0.9738.
Content/face. Development of the WOMAC involved expert opinion (including rheumatologists and epidemiologists), reviews of existing instruments, and surveys of patients with hip and / or knee OA39.
Factorial / domain. The factor structure of the WOMAC has been confirmed in some studies of patients with hip and knee OA31, 40 (n = 317 & 2,205). However, one study of patients with hip and knee OA awaiting arthroplasty (n= 474) failed to support a single-item solution for the pain subscale41. Other studies have shown that items on the physical function subscale did not load unequivocally on that factor, and some items from the pain and physical function subscales tended to load together on a factor13, 42-45 (n =66-474). High correlations and overlapping items between the pain and physical function subscales45-47 may pose measurement problems. Specifically, it has been argued that the ability of the WOMAC physical function scale to detect change may be particularly limited in situations where there is a low correlation or “mismatch” between physical function and pain severity45, 47.
Criterion. Two studies found statistically significant Spearman correlations between patient satisfaction with knee arthroplasty and WOMAC pain (r =0.67, 0.55), stiffness (r =0.63, 0.56), and function (r =0.64, 0.48) subscales48, 49 (n =108 & 1,104). In another study, patients who were satisfied with knee arthroplasty had better WOMAC total, pain, and physical function scores than those who were not satisfied50 (n=1,193).
Construct. The WOMAC has been shown to distinguish well between patients with better vs. worse outcomes from knee arthroplasty50, patients with post-traumatic knee OA vs. normal controls51, and patients with mild / moderate vs. severe knee OA52. Several studies have shown that WOMAC and SF-36 subscales with similar constructs have moderate to high correlations, while lower correlations were observed between subscales with less similar constructs38, 52, 53. The WOMAC has shown significant Spearman correlations with performance scores for walking, stair climbing, rising from a chair, and joint range of motion among older adults with knee and hip OA54.
Sensitivity / Responsiveness. The WOMAC has been extensively used in the context of clinical trials. Prior reviews have summarized the performance of the WOMAC with respect to responsiveness in these trials37, 55.
Overall, studies have shown that the WOMAC pain and function subscales exhibit comparable or greater responsiveness to change than corresponding SF-36 subscales38, 52, 53, 56, 57. Responsiveness varies according to subscales and type of intervention37.
In a study of patients with hip and knee OA undergoing comprehensive inpatient rehabilitation, the minimal clinically important differences (MCIDs) for WOMAC global and subscale scores ranged from 0.51-1.33 for worsening and 0.67-0.75 for improvement58. In a study of outpatients with knee or hip OA, the MCIDs on the WOMAC VAS format ranged from -7.9mm to -32.6mm59. Based on these two studies, investigators used three definitions of MCID to calculate the frequency of clinically important improvement in function over 30 months in Multicenter Osteoarthritis Study (MOST) participants60. The three definitions were: MCID26% (26% improvement from baseline), MCID17% (17% improvement from baseline), and MCID tertile (low, medium, and high, based on pre-specified criteria). In that study, 24-39% of participants reached MCID based on these criteria60, suggesting that a clinically important improvement is frequent in individuals with or at high risk for knee OA. MCID values must be viewed cautiously because of limitations in methodology for calculating these values and should not be considered absolute thresholds.
The WOMAC is one of the most widely utilized self-report measures of lower extremity symptoms and function. It has been studied over a period of almost 30 years in many different contexts and patient populations, and there are abundant data regarding its utility and measurement properties. Overall, studies support the adequacy of the measurement properties of the WOMAC, though two potential weaknesses have been debated. First, there is little evidence regarding the measurement properties of the stiffness subscale, and its test-retest reliability has been low37. Second, some studies have found inadequate factorial validity of the WOMAC pain and physical function subscales, potentially leading to weaknesses in the ability of the physical function subscale to detect change when there is a weak association between pain and function. Pua et al. provide an overview of this argument47, for which further exploration is warranted.