Patient acceptable symptom state and OMERACTeOARSI set of responder criteria in joint replacement. Identification of cut-off values

Transcription

1 Osteoarthritis and Cartilage 20 (2012) 87e92 Patient acceptable symptom state and OMERACTeOARSI set of responder criteria in joint replacement. Identification of cut-off values A. Escobar y *, M. Gonzalez y, J.M. Quintana z, K. Vrotsou y, A. Bilbao x, C. Herrera-Espiñeira k, L. Garcia-Perez {, F. Aizpuru #, C. Sarasqueta yy y Unidad de Investigación, Hospital de Basurto e CIBER Epidemiología y Salud Pública (CIBERESP), Bilbao, Bizkaia, Spain z Unidad de Investigación, Hospital de Galdakao e CIBER Epidemiología y Salud Pública (CIBERESP), Galdakao, Bizkaia, Spain x Fundación Vasca de Innovación e Investigación Sanitarias (BIOEF) e CIBER Epidemiología y Salud Pública (CIBERESP), Sondika, Bizkaia, Spain k Hospital Universitario Virgen de las Nieves, Granada, Spain { Planning and Evaluation Service, Canary Islands Health Service e CIBER de Epidemiología y Salud Pública (CIBERESP), Santa Cruz de Tenerife, Spain # Unidad de Investigacion, Hospital de Txagorritxu e CIBER de Epidemiología y Salud Pública (CIBERESP), Vitoria-Gasteiz, Spain yy Unidad de Investigación, Hospital de Donostia e CIBER de Epidemiología y Salud Pública (CIBERESP), Donostia-San Sebastián, Spain article info summary Article history: Received 23 August 2011 Accepted 14 November 2011 Keywords: Joint replacement Response criteria Outcomes assessment Satisfaction Objective: To identify new cut-off values beyond which patients can be considered as satisfied or as responders through patient acceptable symptom state (PASS) and OMERACTeOARSI (Outcome Measures in RheumatologyeOsteoarthritis Research Society International) set of responder criteria in total joint replacement. Methods: Secondary analysis of a 1-year prospective multicenter study of 861 patients, 510 with total knee replacement (TKR) and 351 with total hip prosthesis (THR). Pain and function data were collected by the reverse scoring option of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). PASS values were identified with the 25th centile estimation using an anchoring question about satisfaction with actual symptoms. OMERACTeOARSI set of responder criteria was based on a combination of absolute and relative change of pain, function and global patient s assessment. Receiver operating characteristic (ROC) analysis was used as a complementary approach. Results: The values for PASS were about 80 and 69 for pain and function in THR, while these values were 80 and 68 when using OMERACTeOARSI criteria. Regarding TKR, PASS values were about 75 and 67 in pain and function with both criteria. ROC values were slightly lower in all cases. PASS and OMERACTeOARSI values varied moderately across tertiles of baseline severity. Conclusion: With the provided data we can establish when a patient can be considered as satisfied/ responder in joint replacement. The scores achieved at 1 year were very similar according to both criteria. Ó 2011 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved. Introduction Although total joint replacement (TJR) has proven to be an effective procedure in terms of improvements in pain and function evaluated by means of some patient s reported outcomes as healthrelated quality-of-life (HRQoL) instruments 1, we actually do not know what kind of patients can be classified as responders to the treatment and/or what would be the cut-off points in the scores which discriminate responders and non-responders patients. In * Address correspondence and reprint requests to: A. Escobar, Unidad de Investigación, Hospital de Basurto, Avenida de Montevideo, 18, Bilbao, Bizkaia, Spain. Tel: ; Fax: address: antonio.escobarmartinez@osakidetza.net (A. Escobar). general, studies measure baseline scores and post surgical scores at some point in time, 6 months or 1 year, and compare the change in scores that patients as a whole have undergone. Several questionnaires measure outcomes in TJR, but nearly all of them measure pain and function. One of the existing problems is that results are expressed as continuous data, which may not be useful for the clinicians as they do not know how to interpret changes in the follow-up period. More specifically, clinicians should know how many patients have a satisfactory response to the administered therapy. Therefore the problem is defining what would be a success in terms of HRQoL 2. There are new approaches to measure the patient s response to the medical treatments. Amongst these we have the OMER- ACTeOARSI set of responder criteria for osteoarthritis (OA) 3 and the patient acceptable symptom state (PASS75) 4. The first one classifies /$ e see front matter Ó 2011 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved. doi: /j.joca

2 88 A. Escobar et al. / Osteoarthritis and Cartilage 20 (2012) 87e92 patients as responders or non-responders based on a combination of absolute and relative changes of pain, function and global patient s assessment. The concept of PASS is based on wellbeing or satisfaction with the actual symptoms and is defined as the value beyond which patients consider themselves well 4. Although OMERACTeOARSI (Outcome Measures in Rheumatologye Osteoarthritis Research Society International) set of responder criteria was developed for patients with OA who entered into clinical trials, we believe that it could be applied to patients with more severe stages of the disease, those requiring joint replacement for example, given the fact that responder s criteria are based on both absolute and relative changes in the scores. Regarding patient reported outcomes based on HRQoL we should draw conclusions from two view points of view: firstly, score changes in each dimension from baseline to some point in time, say 1-year post surgery; secondly, the level of the patient s current symptom perception. Patients can improve from their baseline status, but their final state might not be satisfactory for them. The aim of the present study was to determine the cut-off values of the pain and function dimensions of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) beyond which patients can be considered as responders to TJR based on satisfaction with symptoms (PASS) and the OMERACTeOARSI criteria set at 1-year post surgical management. In addition, we have calculated these cut-off points through receiver operating curve characteristics (ROC) analysis. Finally, we have evaluated whether the PASS is stable over three time periods. Methods This is a secondary analysis of a 1-year prospective study that took place in 15 hospitals; three in Andalusia, three in Canary Islands and nine in the Basque Country (Spain). The Institutional Review Boards of the participant Hospitals approved the study. All patients received a letter informing them about the study and asking for voluntary participation. We considered that patients who returned their questionnaires agreed to participate in the study. Consecutive patients who were to undergo primary TJR for knee or hip OA between March 2005 and December 2006 were eligible for the study. Patients were excluded if they had psychiatric disorders that would compromise their ability to properly answer the questionnaires. We developed data-collection questionnaires to retrieve data from the medical records and directly from the patients. In the main study, we sent the questionnaires to the patients at baseline, a month before surgery, at 3, 6 and 12 months post surgery. For the purpose of this study, inclusion criteria were patients with TJR due to OA, complete baseline data, 1-year measurements of HRQoL questionnaire (WOMAC) and the items related to the derivation of the OMERACTeOARSI criteria, and finally the anchoring question about satisfaction with symptoms to calculate PASS scores. All patients had unilateral TJR. Questionnaire We used the WOMAC questionnaire that is a disease-specific, selfadministered questionnaire developed to study patients with hip or knee OA 5. It has a multidimensional scale made up of 24 items grouped into three dimensions: pain (five items), stiffness (two items), and physical function (17 items). We studied the pain and function dimensions. We used the Likert version with five response levels for each item, representing different degrees of intensity (none, mild, moderate, severe or extreme) scored from 0 to 4. The data were standardized to a range of values from 0 to 100. According to recent recommendations 6 in this study we used the reverse option, where 0 represents the worst status and 100 the best possible status. The WOMAC has been translated and validated into Spanish 7,8. Statistical analysis We used three different statistical methods to calculate cut-off values. First, following the paper of Tubach et al. 4, we used the concept of PASS that was developed taking into account the opinions from the patient. It was based on their opinion about actual satisfaction with their symptoms. This question was If you had to be the rest of your life with the symptoms you have now, how would you feel? We used a Likert version with four response levels: very satisfied, somewhat satisfied, somewhat dissatisfied and very dissatisfied. Then patients who were very satisfied or somewhat satisfied were considered as satisfied while considering the remainder as unsatisfied (Yes/No). This question has been used as anchor to calculate cut-off values. As we used the reverse option of WOMAC scores, where 0 represents the worst status and 100 the best possible status, we defined the PASS as the 25th centile of the final score at 1 year instead of the 75th centile. The same was done to calculate PASS at 3 and 6 months. Secondly, we used the OMERACTeOARSI set of responder criteria 3. This set is as follows: if the patient has an improvement in pain or function 50% and absolute change 20 then, he is a responder. If the patient does not reach these criteria but has an improvement in at least two of the three following criteria he will also be a responder: 1. Pain 20% and absolute change 10; 2. Function 20% and absolute change 10; 3. Patient s global assessment of the disease 20% and absolute change 10. This initiative used different tools to assess pain, functional disability and global patient s assessment. To measure pain and function, the visual analog scale (VAS) and WOMAC dimensions were the tools most used, while for global assessment the VAS and Likert scales were the most often used. In our study, pain and function were measured by each WOMAC dimension. Patient s global assessment was measured by a fivepoint Likert scale. To create the OMERACTeOARSI criteria a change of one point was considered as a 20% global assessment change. As with PASS, we have also used the 25th centile approach for the final score at 1 year. We stratified cut-off values of PASS and OMERACTeOARSI scores by baseline tertiles of severity in each dimension. Finally, a ROC analysis was performed, using as a gold standard patients satisfaction with symptoms (Yes/No). As optimal cut-off value for the 1-year score of each dimension was considered the one that maximized the sum of sensitivity and specificity. We draw 500 bootstrap samples, calculated their respective ROC curves and derived a cut-off value for each. The cut-off values given are the mean values of the 500 cut-off points and presented along with their derived 95% confidence interval (CI). This approach was used to study the robustness of the cut-off values across both criteria, PASS and OMERACTeOARSI. Descriptive statistics included frequencies and percentages for categorical variables and means and standard deviations (SD) for continuous variables. Chi-square and Fisher s exact tests were used for comparing categorical variables, whereas the Student s t test was used for continuous variables. Statistical analyses were performed using SPSS, v. 17 for Windows (SPSS Inc) and Confidence Interval Analysis (CIA) v.2.0. Results For the present secondary analysis there were 861 patients who met all three inclusion criteria. Of those, 510 patients (59.2%) had undergone knee replacement and 351 (40.8%) hip arthroplasty.

3 A. Escobar et al. / Osteoarthritis and Cartilage 20 (2012) 87e92 89 Regarding sample demographics, the knee sample had more females and patients were older and more obese. Table I shows baseline data and data at 1 year on the WOMAC dimensions. There was no baseline difference between the two joints in the pain dimension, but patients who had undergone hip replacement reached better scores at 1 year (P < 0.001), so they had higher improvement. Regarding the function dimension, although there was baseline difference (P ¼ 0.001) between hip and knee joints, the improvement at 1 year was also higher in patients undergoing hip replacement (P < 0.001). In Table II we can see the values of the WOMAC scores at 1 year achieved by patients who fulfilled each criterion studied, PASS, OMERACTeOARSI and the values that maximized sensitivity and specificity through ROC and their corresponding CIs. PASS In patients considering their current state as satisfactory by PASS anchoring question (84% for hip replacement and about 79% for knee replacement), the cut-off points in the hip cohort for WOMAC dimensions at 1 year based on 25th centile were 80 in the pain dimension and about 69 in function. In the patients who had undergone knee replacement these values were 75 points and 68, respectively. Responders by OMERACTeOARSI criteria In the hip cohort about 95% of the patients were responders to the OMERACTeOARSI criteria, and about 86% when considering patients with knee replacement. When using the 25th centile approach based on these responders patients, the scores at 1 year in the patients with hip OA were 80 points in pain and about 68 in the function dimension. Regarding knee-operated patients, these cutoff values were 75 in pain and 66 points in function. ROC Regarding the best cut-off points by ROC analysis in the pain dimension at 1 year, these values were 70 and 71 in the hip and knee patients, respectively. The values for WOMAC functional dimension were about 69 and 63 points in hip and knee patients. Considering these cut-off points, in all cases but pain in hip replacement patients, the area under the ROC curve (AUC) was higher than The estimates of the PASS and OMERACTeOARSI criteria stratified by baseline tertiles of severity are displayed in Table III. Table I Patient s characteristics and WOMAC dimension scores at baseline and at 1 year Hip replacement Knee replacement P-value n ¼ 351 Mean (SD) n ¼ 510 Mean (SD) Gender (females n%) 155 (44.3%) 349 (68.4%) <0.001 Age (years) 65.4 (12.5) 71.4 (6.9) <0.001 BMI * 28.0 (3.8) 30.4 (4.8) <0.001 WOMACy Baseline pain 44.8 (17.4) 44.7 (18.6) 0.9 Pain at 1 year 84.6 (16.9) 78.7 (19.6) <0.001 Change in pain 39.8 (22.3) 34.0 (24.1) <0.001 Baseline function 35.1 (16.5) 39.0 (17.5) Function at 1 year 75.9 (18.8) 71.8 (21.0) Change in function 40.8 (21.7) 32.8 (23.8) <0.001 * Body mass index. y Western Ontario and McMaster Universities Osteoarthritis Index. From 0 (worst) to 100 (best). Table II WOMAC scores at 1 year according to the different criteria and joints WOMAC* dimensions Regarding pain, although the scores were very similar, from 70e75 to 80e85 points, the better the patient s baseline score the higher the cut-off value at 1 year in pain domain in both joints. These differences between groups were around 5e10 points in both, the PASS and OMERACTeOARSI criteria and in both joints. When looking at the function dimension, this trend was clearer, from about 60 points to 75 with differences of around 15 points between those patients in the worst baseline situation and those in the better situation. In the hip cohort, the percentage of satisfied patients did not vary according to the baseline tertiles of pain (P ¼ 0.9) or function (P ¼ 0.8). Regarding knee-operated patients, there was a statistically significant difference in the pain dimension across baseline tertiles (P ¼ 0.04), but not in function (P ¼ 0.1). Finally, Table IV lists the cut-off values with their 95% CIs over time of WOMAC dimensions based on the 25th centile approach in patients considering their current state as satisfactory by PASS anchoring question. In general data of both dimensions were similar at 6 months and 1 year. Likewise the percentage of patients who considered themselves satisfied was about 82% in hip prosthesis and 77% in the knee replacement cohort. However, the estimates corresponding to the first evaluation carried out at 3 months post surgery were lower, both in the scores and in the percentage of satisfied patients. Discussion Hip replacement (n ¼ 351) Knee replacement (n ¼ 510) Pain PASSy, 25th centile (95% CIz) 80.0 (75.0e85.0) 75.0 (70.0e80.0) ROC** curve (95% CI) 70.0 (69.4e70.6) 71.4 (71.1e71.8) AUCyy (IC 95%) 0.77 (0.70e0.84) 0.83 (0.78e0.87) OMERACTeOARSI, 25th 80.0 (75.0e85.0) 75.0 (70.0e80.0) centile (95% CI) Function PASSy, 25th centile (95% CI) 69.1 (67.6e72.1) 67.7 (64.7e70.6) ROC** curve (95% CI) 68.8 (68.4e69.3) 63.3 (62.9e63.6) AUCyy (IC 95%) 0.81 (0.75e0.87) 0.83 (0.79e0.88) OMERACTeOARSI, 25th 67.6 (64.7e68.8) 66.2 (64.7e69.1) centile (95% CI) PASSy responders 84.0% (80.2e87.8) 78.8% (75.3e82.3) (satisfied), % (95% CI) OMERACTeOARSI responders, % (95% CI) 94.6% (92.2e97.0) 86.3% (83.3e89.3) * Western Ontario and McMaster Universities Osteoarthritis Index. From 0 (worst) to 100 (best). y Patient acceptable symptom state. z Confidence interval. ** Receiver operating characteristics curve. yy Area under the curve, to study the discriminatory ability of the cut-off point based on the ROC curve to predict PASS-responders. In this prospective study we applied the concepts of PASS, the OMERACTeOARSI set of responder criteria on patients who have undergone hip or knee replacement and to validate previous data, we performed ROC analysis. The scores derived could be considered as cut-off points for considering a patient as satisfied or responder when performing TJR. The concept of PASS represents an absolute value, not a change and can be considered a clinical treatment target. PASS has been defined as the value beyond which patients can consider themselves well 9 and has been widely used in rheumatology in several diseases such as hip and knee OA 4,10, rheumatoid arthritis 11, ankylosing spondylitis 12, rotator cuff syndrome 13 or Reynaud s phenomenon 14 and recently in hip replacement 15.

4 90 A. Escobar et al. / Osteoarthritis and Cartilage 20 (2012) 87e92 Table III WOMAC scores at 1 year according to the PASS and OMERACTeOARSI criteria and joint divided into baseline severity tertiles Low (worst) Hip replacement (n ¼ 351) High (best) Low (worst) Knee replacement (n ¼ 510) High (best) Baseline tertiles Baseline tertiles Medium Medium WOMAC* dimension Pain Tertiles < e54.9 >55.0 < e54.9 >55.0 PASSy, 25th (95% CIz) 75.0 (70.0e85.0) 80.0 (75.0e85.0) 80.0 (75.0e85.0) 70.0 (65.0e75.0) 75.0 (70.0e80.0) 80.0 (75.0e85.0) OMERACTeOARSI, 75.0 (70.0e85.0) 80.0 (75.0e85.0) 85.0 (75.0e85.0) 70.0 (65.0e75.0) 75.0 (70.0e80.0) 85.0 (80.0e87.5) 25th (95% CI) PASSy responders 85.0% (78.5e91.5) 83.3% (76.5e90.1) 83.7% (77.0e90.4) 75.1% (68.6e81.6) 75.9% (69.5e82.3) 85.1% (79.7e90.5) (satisfied), % (95% CI) Function Tertiles < e42.5 >42.5 < e45.5 >45.5 PASSy, 25th (95% CI) 60.1 (57.3e67.2) 72.1 (63.2e76.5) 75.0 (70.6e79.4) 60.3 (51.6e66.2) 67.7 (64.7e72.1) 72.8 (69.1e75.0) OMERACTeOARSI, 58.8 (54.4e62.5) 72.0 (63.2e75.0) 73.5 (69.1e76.5) 58.8 (52.9e63.2) 66.7 (63.2e69.1) 74.6 (70.6e76.5) 25th (95% CI) PASSy responders (satisfied), % (95% CI) 82.5% (75.6e89.4) 85.6% (79.2e92.0) 84.2% (77.6e90.8) 75.0% (68.5e81.5) 83.8% (78.3e89.3) 78.9% (72.8e85.0) * Western Ontario and McMaster Universities Osteoarthritis Index. From 0 (worst) to 100 (best). y Patient acceptable symptom state. z Confidence interval. On the other side, OMERACTeOARSI criteria are based on relative and absolute changes in scores. Although these criteria have been used extensively in the literature to compare different interventions such as drugs 16,17, dietary supplements 18, medical devices 19,20 or the impact of rehabilitation programs 21 to the best of our knowledge this is the first time they have been used after TJR to establish cut-off values. In this study we have used the reverse WOMAC scores, that is from 0 (worst situation) to 100 points (best situation), so the approach to the cut-off points corresponds to the 25th centile in both criteria PASS and OMERACTeOARSI. Regarding ROC analysis the cut-off point was that which maximized sensitivity and specificity. There were statistically significant improvements in HRQoL dimensions, pain and functionality for both joints. Although baseline scores were similar in hip and knee patients those who underwent hip replacement had higher improvements in both dimensions at 1-year follow-up. This is a fact already reported in the literature 1,22e25. These higher improvements in both pain and Table IV PASS scores estimate over time by joint WOMAC * dimension Hip Pain, 25th Function, 25th PASS-responders (satisfied), % Knee Pain, 25th Function, 25th PASS-responders (satisfied), % Follow-up time 3 months 6 months 12 months 75.0 (70.0e80.0) 80.0 (75.0e81.2) 80.0 (75.0e85.0) 63.2 (58.6e64.7) 67.6 (63.2e70.6) 69.1 (67.6e72.1) 74% (69.4e78.6) 81% (76.9e85.1) 84% (80.2e87.8) 70.0 (65.0e75.0) 75.0 (70.0e80.0) 75.0 (70.0e80.0) 63.9 (58.8e64.7) 64.7 (63.3e67.6) 67.7 (64.7e70.6) 67.7% (63.6e71.8) 77% (73.3e80.7) 78.8% (75.3e82.3) * Western Ontario and McMaster Universities Osteoarthritis Index. From 0 (worst) to 100 (best). y Confidence interval. function in patients undergoing hip replacement may explain the higher percentage of patients who meet the criteria, whether applying the PASS or considering the OMERACTeOARSI. In the hip prosthesis cohort, the cut-off points at 1 year of the pain dimension were similar in both criteria and a little lower according to ROC. In the function dimension all three values were similar. Likewise, the patients who had undergone knee replacement had values nearly equal for pain according to the three criteria. In the function dimension these values were again similar in both criteria and a little lower according to ROC. The lower cutoff values obtained by ROC analysis have been previously described in other pathologies such as ankylosing spondylitis 12 or rheumatoid arthritis 11. The higher percentages of satisfied patients measured by PASS and responder patients by OMERACTeOARSI, in those who have undergone hip replacement possibly agree with the greater improvements that experienced these patients in both dimensions; pain and function as measured by WOMAC. We should note that, although in different ways, both approaches, PASS and OMER- ACTeOARSI criteria are based on pain and function, PASS subjectively and the other objectively. Our results agree with others studies 26 which have recently showed that 81% of patients claimed that they were satisfied or very satisfied 1 year after total knee replacement or that the patients with hip replacement had faster and greater improvements than those with knee replacement 23. As in other studies 4 our PASS estimates varied moderately across the tertiles of baseline severity. In this case also, values of both criteria were quite similar in both joints within each tertile. Although measured with other patient reported outcome, the Oxford Hip Score 15, our results at 1 year were similar and show that the percentage of PASS responder patients did not vary according to the baseline tertiles of severity in the hip cohort, neither in pain nor in the function dimension. In patients with knee replacement there were slight differences according to baseline tertiles of pain with the higher percentage of PASS responder patients in the high baseline tertile. Therefore, it appears that the cut-off values and percentage of satisfied patients can be considered independent of baseline scores. Regarding stability over time, PASS values can be considered stable from the 6th month onwards. This stability could be applicable to the scores of both dimensions and to the percentage of satisfied patients measured by the PASS concept.

5 A. Escobar et al. / Osteoarthritis and Cartilage 20 (2012) 87e92 91 There are some possible limitations in the study. As in most cohort studies, missing data can bias the results. When we compared included patients with non-included (data not shown), the observed statistically significant differences were minor in terms of points, in fact there were only differences of about three and four points in baseline pain and function dimensions. No differences regarding gender, age or body mass index (BMI) were observed. Thus, although bias may be present by missing data it is likely to be minor and we believe the results can be generalized. Regarding OMERACTeOARSI criteria, there was a variable, the patient s global assessment, which can be measured on a VAS or Likert scale. In our case, this item was measured by means of a fivepoint Likert scale. To reproduce the original changes we considered the 20% and absolute change 10 as one point in our scale. The problem might be to consider this change as a measurement error instead of a real change. On the other hand, the original authors measured pain and function disability by VAS or WOMAC while we used only WOMAC dimensions. Another possible limitation is related to the PASS question. There are two main differences with the original paper about PASS 4. First, the original question about patient s opinion had different wording and was recorded by answering yes or no (Question: Taking into account all the activities you carry out during your daily life, your level of pain, and also your functional impairment, do you consider that your current state is satisfactory?). In our case the question, as cited above, was also focused on satisfaction with symptoms but not as specific as in the original. In addition the answers were rated on a four-point Likert scale which was dichotomized as satisfied (very satisfied/satisfied) and not satisfied (somewhat dissatisfied/very dissatisfied). According to the OMERACT eight suggestions 2, the issue of time spent in the state was rest of your life given that it is assumed that patients expect greater effects from surgery than from medical therapy. In conclusion, this study has provided data to know when a patient can be considered as satisfied or a responder to the treatment according to the PASS and OMERACTeOARSI criteria. The scores achieved at 1-year post surgical management by the criteria set are very close to those achieved by patients who considered themselves satisfied. Role of funding sources Regarding this study, both public sources are only funders of the main research project. The study sponsors have not had involvement in any process of this manuscript. Author contribution All authors have contributed to each of three activities: (1) the conception and design of the study, or acquisition of data, or analysis and interpretation of data, (2) drafting the article or revising it critically for important intellectual content, and (3) approval of the final version, and they will take public responsibility for the content of the paper. The content has not been published, nor is it being considered elsewhere. Conflict of interest No conflicts of interest (e.g., funding sources for consultancies or studies of products) exist in this study. Acknowledgments We thank the staff members of the different departments, research and quality units, as well as the medical record sections of the participating hospitals and the patients for their collaboration. This study was supported by grants from the Instituto de Salud Carlos III-Fondo de Investigación Sanitaria, (PI 04/0938, PI 04/0542, and PI 04/2577) and from the Basque Country Health Department ( ). References 1. Ethgen O, Bruyere O, Richy F, et al. Health-related quality of life in total hip and total knee arthroplasty. A qualitative and systematic review of the literature. J Bone Joint Surg Am 2004 May;86-A(5):963e Tubach F, Ravaud P, Beaton D, et al. Minimal clinically important improvement and patient acceptable symptom state for subjective outcome measures in rheumatic disorders. J Rheumatol 2007 May;34(5):1188e Pham T, van der HD, Altman RD, et al. OMERACT-OARSI initiative: Osteoarthritis Research Society International set of responder criteria for osteoarthritis clinical trials revisited. Osteoarthritis Cartilage 2004 May;12(5):389e Tubach F, Ravaud P, Baron G, et al. Evaluation of clinically relevant states in patient reported outcomes in knee and hip osteoarthritis: the patient acceptable symptom state. Ann Rheum Dis 2005 Jan;64(1):34e7. 5. Bellamy N, Buchanan WW, Goldsmith CH, et al. 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 Dec;15(12):1833e Singh J, Sloan JA, Johanson NA. Challenges with health-related quality of life assessment in arthroplasty patients: problems and solutions. J Am Acad Orthop Surg 2010 Feb;18(2):72e Batlle-Gualda E, Esteve-Vives J, Piera MC, et al. Tradución y adaptación al español del cuestionario WOMAC específico para artrosis de rodilla y cadera. Rev Esp Reumatol 1999;26:38e Escobar A, Quintana JM, Bilbao A, et al. 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 Nov 19;21:466e Kvien TK, Heiberg T, Hagen KB. Minimal clinically important improvement/difference (MCII/MCID) and patient acceptable symptom state (PASS): what do these concepts mean? Ann Rheum Dis 2007 Nov;66(Suppl 3):iii40e Dougados M, Moore A, Yu S, et al. Evaluation of the patient acceptable symptom state in a pooled analysis of two multicentre, randomised, double-blind, placebo-controlled studies evaluating lumiracoxib and celecoxib in patients with osteoarthritis. Arthritis Res Ther 2007;9(1):R Heiberg T, Kvien TK, Mowinckel P, et al. Identification of disease activity and health status cut-off points for the symptom state acceptable to patients with rheumatoid arthritis. Ann Rheum Dis 2008 Jul;67(7):967e Tubach F, Pham T, Skomsvoll JF, et al. Stability of the patient acceptable symptomatic state over time in outcome criteria in ankylosing spondylitis. Arthritis Rheum 2006 Dec 15;55(6): 960e Tubach F, Dougados M, Falissard B, et al. Feeling good rather than feeling better matters more to patients. Arthritis Rheum 2006 Aug 15;55(4):526e Khanna PP, Maranian P, Gregory J, et al. The minimally important difference and patient acceptable symptom state for

6 92 A. Escobar et al. / Osteoarthritis and Cartilage 20 (2012) 87e92 the Raynaud s condition score in patients with Raynaud s phenomenon in a large randomised controlled clinical trial. Ann Rheum Dis 2010 Mar;69(3):588e Arden NK, Kiran A, Judge A, et al. What is a good patient reported outcome after total hip replacement? Osteoarthritis Cartilage 2011 Feb;19(2):155e Rother M, Lavins BJ, Kneer W, et al. Efficacy and safety of epicutaneous ketoprofen in Transfersome (IDEA-033) versus oral celecoxib and placebo in osteoarthritis of the knee: multicentre randomised controlled trial. Ann Rheum Dis 2007 Sep;66(9):1178e Verkleij SP, Luijsterburg PA, Koes BW, et al. Effectiveness of diclofenac versus acetaminophen in primary care patients with knee osteoarthritis: [NTR1485], DIPA-trial: design of a randomized clinical trial. BMC Musculoskelet Disord 2010; 11: Ruff KJ, Winkler A, Jackson RW, et al. Eggshell membrane in the treatment of pain and stiffness from osteoarthritis of the knee: a randomized, multicenter, double-blind, placebo-controlled clinical study. Clin Rheumatol 2009 Aug;28(8):907e van Raaij TM, Reijman M, Brouwer RW, et al. Medial knee osteoarthritis treated by insoles or braces: a randomized trial. Clin Orthop Relat Res 2010 Jul;468(7):1926e Bar-Ziv Y, Beer Y, Ran Y, et al. A treatment applying a biomechanical device to the feet of patients with knee osteoarthritis results in reduced pain and improved function: a prospective controlled study. BMC Musculoskelet Disord 2010;11: Lin CW, March L, Crosbie J, et al. Maximum recovery after knee replacement e the MARKER study rationale and protocol. BMC Musculoskelet Disord 2009;10: Rissanen P, Aro S, Slatis P, et al. Health and quality of life before and after hip or knee arthroplasty. J Arthroplasty 1995 Apr; 10(2):169e Wylde V, Dieppe P, Hewlett S, et al. Total knee replacement: is it really an effective procedure for all? Knee 2007 Dec;14(6): 417e Chesworth BM, Mahomed NN, Bourne RB, et al. Willingness to go through surgery again validated the WOMAC clinically important difference from THR/TKR surgery. J Clin Epidemiol 2008 Sep;61(9):907e Bourne RB, Chesworth B, Davis A, et al. Comparing patient outcomes after THA and TKA: is there a difference? Clin Orthop Relat Res 2010 Feb;468(2):542e Bourne RB, Chesworth BM, Davis AM, et al. Patient satisfaction after total knee arthroplasty: who is satisfied and who is not? Clin Orthop Relat Res 2010 Jan;468(1):57e63.