SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Objective

To estimate the minimum clinically important improvement (MCII) and patient acceptable symptom state (PASS) values for 4 generic outcomes in 5 rheumatic diseases and 7 countries.

Methods

We conducted a multinational (Australia, France, Italy, Lebanon, Morocco, Spain, and The Netherlands) 4-week cohort study involving 1,532 patients who were prescribed nonsteroidal antiinflammatory drugs for ankylosing spondylitis, chronic back pain, hand osteoarthritis, hip and/or knee osteoarthritis, or rheumatoid arthritis. The MCII and PASS values were estimated with the 75th percentile approach for 4 generic outcomes: pain, patient global assessment, functional disability, and physician global assessment, all normalized to a 0–100 score.

Results

For the whole sample, the estimated MCII values for absolute change at 4 weeks were −17 (95% confidence interval [95% CI] −18, −15) for pain; −15 (95% CI −16, −14) for patient global assessment; −12 (95% CI −13, −11) for functional disability assessment; and −14 (95% CI −15, −14) for physician global assessment. For the whole sample, the estimated PASS values were 42 (95% CI 40, 44) for pain; 43 (95% CI 41, 45) for patient global assessment; 43 (95% CI 41, 44) for functional disability assessment; and 39 (95% CI 37, 40) for physician global assessment. Estimates were consistent across diseases and countries (for subgroups ≥20 patients).

Conclusion

This work allows for promoting the use of values of MCII (15 of 100 for absolute improvement, 20% for relative improvement) and PASS (40 of 100) in reporting the results of trials of any of the 5 involved rheumatic diseases with pain, patient global assessment, physical function, or physician global assessment used as outcome criteria.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

The choice of an outcome measure is a major step in the design of clinical trials. Trials of rheumatology ideally should systematically evaluate pain, functional impairment, and patient global assessment. For most musculoskeletal disorders, such as osteoarthritis (OA), an outcome criterion is recommended for each of these domains (e.g., for OA, a 0–10 numerical rating scale [NRS] for pain and the Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC] for functional impairment) (1–3). General outcome criteria, such as patient global assessment, are also recommended.

In general, the results of clinical trials are reported at the group level, e.g., with the mean change from baseline by treatment group or with the effect size. Because a statistically significant difference is mostly a matter of sample size and SD of change in the variable, the most difficult issue is determining whether an observed difference is clinically important (4). How does a mean result at the group level (e.g., a mean change in pain of 1 point on a 0–10-point NRS) translate to a proportion of patients with clinically relevant improvement? To take into account the patient's perspective and for easier interpretation of results, considering whether the change from baseline is important at the individual patient level and then reporting the percentage of patients with improved condition by treatment group is an interesting approach.

This approach requires, for each patient, that the continuous outcome measure (e.g., change in pain from baseline on a 0–10-point NRS) be dichotomized into a binary variable (important change in pain from baseline, i.e., change greater than a cutoff defining an important improvement in the patient's perspective). Several cutoffs have been proposed to define the minimum clinically important difference (or minimum important difference) (5, 6). Recently, the minimum clinically important improvement (MCII) (7) was proposed because the amount of change patients consider clinically important is not the same with worsened or improved conditions (8–10). Since only the patient can perceive whether the change in symptom is important, the definition of the MCII relies on an anchor-based method, with the patient's opinion as the external anchor (6, 11, 12).

The MCII reflects the concept of improvement (“feeling better”). The patient acceptable symptom state (PASS) has been proposed to address the concept of partial symptomatic remission (“feeling good”) (13). These concepts are complementary: a patient's condition can be markedly improved with treatment but can still be poor. For example, a score for pain that decreases from 9 to 5 on a 0–10-point NRS is considered a success in terms of improved condition, but an absolute value of 5 still reflects a painful and/or disabling condition. Therefore, reporting the results of a trial in terms of both proportion of patients with improved condition and proportion in an acceptable state at the end of the trial is useful. The PASS value is a clinically relevant cutoff from the patient's perspective, which allows for classifying patients at the end of the trial as being in “an acceptable state” (with the outcome score ≤ the PASS) or not (with the outcome score > the PASS). The PASS was found more relevant than the MCII because the MCII is the amount of change needed to reach the PASS (14). However, when dichotomizing a continuous variable, reporting the between-groups difference in the proportion of patients with improved condition and the proportion in an acceptable state at the end of the trial is less powerful than reporting the effect size (15). Reporting both is considered an interesting approach (16–19).

The concepts of MCII and PASS are supported by the Outcome Measures in Rheumatology (OMERACT) international network, which has focused 2 meeting sessions (2004 and 2006) on this issue (19, 20) and has contributed importantly to understanding in this area. To date, MCII and PASS cutoff values have been estimated for hip and knee OA in France (7, 13) and acute painful shoulder in France (14). PASS values have been estimated for ankylosing spondylitis (AS), rheumatoid arthritis (RA), and psoriatic arthritis in Norway (21, 22); for RA in The Netherlands (23); and for AS in the US and Europe (24, 25) and Canada (26).

This prospective multinational cohort study aimed to estimate the MCII and PASS for the main symptomatic criteria used in rheumatology for 5 different diseases and in 7 different countries.

Significance & Innovations

  • Defining a minimum important improvement or an acceptable state to use in reporting the results of trials or cohorts is widely considered important.

  • This article reports the minimum clinically important improvement (MCII) and the patient acceptable symptom state (PASS) values estimated for the first time for 4 generic outcomes in 5 chronic rheumatic diseases (rheumatic arthritis, ankylosing spondylitis, hand osteoarthritis, hip and/or knee osteoarthritis, and chronic back pain) and 7 countries in the same cohort study.

  • The estimates of the MCII and PASS were similar across diseases and countries, whatever the outcome measure. This consistency reinforces the robustness of these concepts and allows for promoting the use of values of MCII (15 of 100 for absolute improvement, 20% for relative improvement) and PASS (40 of 100) when reporting the results of trials of the 5 involved rheumatic diseases with pain, patient global assessment, physical function, or physician global assessment used as outcome criteria.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Study design and study population.

The Rheumatological Evaluation of Facts Leading to Excellent Treatment (REFLECT) study was a prospective, multicenter, multinational, observational 4-week cohort study.

Patients were involved from a hospital rheumatology department in Australia, France, Italy, Lebanon, Morocco, Spain, and The Netherlands. To be included in the REFLECT study, outpatients had to be age >18 years, experiencing pain from musculoskeletal disease (≥3 on a 0–10-point NRS), have a nonsteroidal antiinflammatory drug (NSAID) prescribed for the next 4 weeks, and be able to understand the objectives of the study and complete questionnaires in the national language of the country in which they live (English in Australia, French in France, Italian in Italy, Arabic in Lebanon and Morocco, Castilian in Spain, and Dutch in The Netherlands). This study involved patients with 1 of the following musculoskeletal diseases: RA as defined by the American College of Rheumatology (ACR) (27); AS as defined by the modified New York criteria (28) with a painful axial involvement at baseline; hand OA as defined by the ACR (29); hip and/or knee OA as defined by the ACR (30, 31); or mechanical back pain (BP) defined as BP for at least 3 months resulting from a mechanical disorder.

Inclusion could begin with the start of a first NSAID or a switch from one NSAID to another. Patients were excluded if they had a prosthesis on the assessed joint or if they had received an intraarticular injection in the 4 weeks before the study began. All patients initially visited the rheumatologist in charge, and an NSAID was prescribed (the drug and dose were chosen by the physician). A final visit to the same rheumatologist was scheduled for 4 weeks later.

Data collection.

Data were collected at 2 visits (baseline visit and 4-week followup). At the baseline visit, data were collected on demographics (age, sex, weight, height) and disease characteristics (date of symptom onset for each disease and specific characteristics for each evaluated disease). At both visits, patients assessed their status regarding their musculoskeletal disease by general and disease-specific patient-reported outcomes (PROs). General PROs were pain during the 48 hours before the visit, patient global assessment of disease activity, and functional disability. These 3 outcomes were measured on a 0–10-point NRS. Disease-specific PROs, according to each disease, were as follows: 1) the Health Assessment Questionnaire score for RA (32), 2) the Bath Ankylosing Spondylitis Disease Activity Index (33) and Bath Ankylosing Spondylitis Functional Index (34) for AS, 3) the Australian/Canadian Hand Osteoarthritis Index (35) for hand OA, 4) WOMAC (36) for hip or knee OA, and 5) the Roland and Morris questionnaire (37) for BP. At the baseline and final visits, the rheumatologist also rated global assessment of the disease activity (physician global assessment) on a 0–10-point NRS (a general physician-reported outcome).

In this article we report the results for the 4 general outcome measures: pain, patient global assessment of disease activity, functional disability, and physician global assessment of disease activity. Each of these generic outcomes was measured on a 0–10-point NRS (Supplementary Appendix A, available in the online version of this article at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)2151-4658), with 0 being the best score and 10 the worst. For this report, all outcome scores were normalized to a 0–100 score.

A specific case report form was available in each of the 6 languages involved. The questionnaires were translated to be as conceptually similar as possible by a single group with experience in translation and linguistic validation of patient questionnaires.

Outcome measures.

As previously described (7, 13), to determine the MCII and the PASS, we used an external anchoring method based on patient perspective. The external anchors, collected at the final visit, were as follows: for the MCII determination, the patients assessed the change from baseline on a 3-point Likert scale (improved, no change, or worse). If they reported an improvement, they were asked how important this improvement was to them (very important, moderately important, slightly important, or not at all important). The MCII was estimated for both the absolute difference (final value − baseline value) and relative difference (final value − baseline value/baseline value) in each PRO. The MCII was defined as the 75th percentile of the distribution of change in PRO scores for patients who considered they had a slightly or moderately important improvement during the study (i.e., the change in symptoms [the MCII] or a lower change was achieved by 75% of patients with a slightly or moderately important improvement) (7).

For the PASS determination, patients were asked, “If you were to remain for the rest of your life as you were during the last 48 hours, would this be acceptable or unacceptable for you?” with a dichotomous response mode as acceptable or unacceptable. The PASS was defined as the 75th percentile of the distribution of the PRO scores at the final visit for patients who considered their state acceptable at the end of the study (13). The wording of the external anchors for determining the MCII and PASS in the different languages is given in Supplementary Appendix B (available in the online version of this article at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)2151-4658).

Statistical analysis.

Analyses were conducted first by disease and then by country. Logistic regression was used to model the observations and compute MCII and PASS estimates (according to the 75th percentile of the logistic curve, as described above) and their 95% confidence intervals (95% CIs). Data for patients with missing data in one of the outcomes (general outcome measures or external anchor) were excluded from the analyses concerning this outcome.

If the number of patients for which the MCII or the PASS had to be estimated was too low (<20), the cutoffs were not estimated for that subgroup. Data for patients lost to followup were excluded from the analysis. Statistical analysis involved use of SAS, version 8.2, and Splus software, version 6.2 (TIBCO).

Ethics.

This study was conducted in compliance with the good clinical practices and Declaration of Helsinki principles. This study was an observational study and not a clinical trial. The conduct of the study did not interfere with the usual care of patients. If local law required it, the national principal investigator obtained institutional review board approval of the protocol, as well as approval of all subsequent major changes. Each patient gave their signed informed consent to be in the study if local law required it.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

A total of 1,532 patients were enrolled in the study, and 1,505 (98%) patients completed the final visit. The number of patients by disease and country is shown in Table 1. Clinical and demographic data are in Table 2. In all, 668 (44.5%) patients reported a slightly or moderately important improvement, and 1,014 (67.4%) patients reported being in an acceptable state at the end of the study.

Table 1. Number of patients included in the study by disease and country*
 DiseaseTotal
ASBPHand OAHip/ knee OARA
  • *

    AS = ankylosing spondylitis; BP = back pain; OA = osteoarthritis; RA = rheumatoid arthritis.

Countries      
 Australia5555525
 France95106102109101513
 Italy1350646776270
 Lebanon40004441125
 Morocco335005030163
 The Netherlands150002338
 Spain8278787882398
Total2832892493533581,532
Table 2. Baseline characteristics of patients and changes from baseline in patient-reported outcomes*
 AS (n = 283)BP (n = 289)Hand OA (n = 249)Knee/hip OA (n = 353)RA (n = 358)
  • *

    Values are the mean ± SD unless indicated otherwise. AS = ankylosing spondylitis; BP = back pain; OA = osteoarthritis; RA = rheumatoid arthritis; IQR = interquartile range; NRS = numerical rating scale.

Male sex, no. (%)214 (77)87 (30)30 (12)66 (19)62 (18)
Age, years43 ± 1456 ± 1564 ± 1165 ± 1254 ± 15
Body mass index, kg/m225 ± 427 ± 525 ± 429 ± 525 ± 5
Disease duration, years13 ± 106 ± 87 ± 66 ± 610 ± 9
Disease duration, median (IQR) years11 (5–19)4 (1–8)5 (2–10)5 (2–9)6 (3–14)
Pain (0–10 NRS)60 ± 2065 ± 1863 ± 1767 ± 1859 ± 22
 Week 0, median (IQR)60 (50–70)70 (50–80)60 (50–80)70 (50–80)60 (40–80)
 Absolute change (week 0–week 4)−20 ± 24−23 ± 23−21 ± 23−22 ± 21−18 ± 23
 Relative change (week 0–week 4)−30 ± 44−34 ± 37−31 ± 33−31 ± 33−28 ± 39
Patient global assessment (0–10 NRS)60 ± 1964 ± 1760 ± 1865 ± 1859 ± 19
 Week 0, median (IQR)60 (50–70)60 (50–80)60 (50–70)70 (50–80)60 (50–70)
 Absolute change (week 0–week 4)−17 ± 23−21 ± 22−17 ± 20−18 ± 20−15 ± 21
 Relative change (week 0–week 4)−25 ± 40−30 ± 43−26 ± 37−24 ± 35−22 ± 40
Physical function (0–10 NRS)58 ± 2261 ± 2059 ± 1963 ± 1957 ± 21
 Week 0, median (IQR)60 (40–70)60 (50–70)60 (50–70)60 (50–80)60 (40–70)
 Absolute change (week 0–week 4)−16 ± 24−19 ± 22−15 ± 22−16 ± 21−15 ± 21
 Relative change (week 0–week 4)−26 ± 40−30 ± 38−22 ± 44−20 ± 53−23 ± 42
Physician global assessment (0–10 NRS)56 ± 1856 ± 1557 ± 1661 ± 1652 ± 18
 Week 0, median (IQR)60 (40–70)60 (40–70)60 (50–70)60 (50–70)50 (40–70)
 Absolute change (week 0–week 4)−16 ± 20−19 ± 20−14 ± 17−18 ± 16−14 ± 17
 Relative change (week 0–week 4)−27 ± 36−33 ± 35−24 ± 30−29 ± 30−24 ± 37
Patients reporting a slightly or moderately important improvement, no. (%)121 (44)123 (43)104 (43)169 (49)151 (43)
Patients who considered their state at the end of the study acceptable, no. (%)178 (64)186 (65)182 (75)220 (64)248 (70)

Baseline scores on a 0–10-point NRS varied between diseases. Patients with hip or knee OA had the highest scores and those with RA the lowest. Figures 1 and 2 show the estimates of MCII and PASS with their 95% CIs for the 4 general outcome measures.

thumbnail image

Figure 1. Estimates of the absolute and relative minimum clinically important improvement (MCII) and 95% confidence intervals (95% CIs) for the 4 general outcome measures. Small squares represent point estimates for each disease and each country. Large squares represent the overall MCII estimate (for the whole sample) with its 95% CI. Australia and The Netherlands are not represented because of insufficient sample size for estimation. UCL = upper 95% CI; LCL = lower 95% CI.

Download figure to PowerPoint

thumbnail image

Figure 2. Estimates of the patient acceptable symptom state (PASS) with 95% confidence intervals (95% CIs) for the 4 general outcome measures. Small squares represent point estimates for each disease and each country. Large squares represent the overall PASS estimate (for the whole sample) with its 95% CI. Sample sizes for which the PASS was estimated are shown in parentheses. Australia and The Netherlands are not represented because of insufficient sample size for estimation. UCL = upper 95% CI; LCL = lower 95% CI.

Download figure to PowerPoint

MCII.

For pain in the whole sample, the MCII for absolute difference was −17 (95% CI −18, −15) and the MCII for relative difference was −21% (95% CI −24%, −19%). For patient global assessment in the whole sample, the MCII for absolute difference was −15 (95% CI −16, −14) and the MCII for relative difference was −20% (95% CI −21%, −18%). For functional disability assessment in the whole sample, the MCII for absolute difference was −12 (95% CI −13, −11) and the MCII for relative difference was −15% (95% CI −18%, −13%). For physician global assessment in the whole sample, the MCII for absolute difference was −14 (95% CI −15, −14) and the MCII for relative difference was −18% (95% CI −20%, −16%).

PASS.

In the whole sample, the PASS was 42 (95% CI 40, 44) for pain, 43 (95% CI 41, 45) for patient global assessment, 43 (95% CI 41, 44) for functional disability assessment, and 39 (95% CI 37, 40) for physician global assessment. Globally, the MCII and PASS estimates were consistent across diseases and across countries when the samples were sufficient. Australia, Lebanon, Morocco, and The Netherlands often had extreme values and wide 95% CIs because of MCII and PASS estimation for small groups of patients.

For each PRO, patients with RA who had the lowest baseline scores had the lowest PASS. Similarly, patients with hip or knee OA who had the highest baseline scores had the highest PASS. The proportion of patients with both change in score higher than the MCII and final score lower than the PASS for the outcome criteria was 44.6% for pain, 39.8% for patient global assessment, 37.9% for functional disability assessment, and 30.1% for physician global assessment.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

The importance of defining a minimum important improvement or an acceptable state to use in trials reporting is widely accepted (38, 39). This prospective multinational study allowed for estimating in the same cohort, globally by disease and nationally by the MCII and PASS values for 4 generic outcomes used in clinical trials of 5 chronic rheumatic diseases (RA, AS, hand OA, hip and/or knee OA, and chronic BP). Furthermore, this is the first determination of the MCII and PASS for hand OA and chronic BP.

Our estimates of the MCII and PASS were similar across diseases and countries, whatever the outcome measure. Results that deviated from the overall findings could be related to low enrollment (with large 95% CIs). Few patients were involved in this estimation of the MCII and PASS from Australia and The Netherlands as compared with France, Italy, and Spain. The consistency of the MCII and PASS estimates reinforces the robustness of these concepts and allows for promoting the same estimates of the MCII and PASS, whatever the disease or country, in reporting the results of trials of any of the 5 rheumatic diseases.

Indeed, we need to agree on a single value (for each criterion) that could be used for reporting in all trials. Previously, values chosen for MCII or PASS cutoffs, among a range of relevant MCII and PASS values, had a low impact on the difference in success rate between arms in a trial (i.e., the proportion of patients benefiting from treatment) (40). Thus, the slight differences observed across diseases would not result in a distortion of results. We recommend the use of values of MCII (15 of 100 for absolute improvement, 20% for relative improvement) and PASS (40 of 100) in reporting the results of trials using pain, patient global assessment, functional disability, or physician global assessment as outcome criteria for any of the 5 involved rheumatic diseases (Table 3).

Table 3. Recommended values for MCII and PASS for pain, patient global assessment, functional disability, and physician global assessment*
 Outcome measured on 0–100 VASOutcome measured on 0–10 NRS
  • *

    Important improvement in condition is defined as decrease in score higher than or equal to the minimum clinically important improvement (MCII) cutoff. Acceptable state is defined as a score strictly lower than the PASS cutoff. PASS = patient acceptable symptom state; VAS = visual analog scale; NRS = numerical rating scale.

MCII  
 For absolute improvement152
 For relative improvement, %2020
PASS404

Musculoskeletal conditions have a great impact on quality of life. Indeed, among patients with a wide range of chronic diseases, those with musculoskeletal diseases report the poorest levels of physical functioning and pain (41). The consistency of PASS estimates across diseases allows for using the proportion of patients achieving the PASS for any the 4 generic outcome measures to compare the burden of the different diseases.

Only 1 study has investigated differences in PASS estimates across countries. This study involved a multinational controlled trial database of AS and compared the US and different European countries. PASS estimates were higher for English-speaking than non–English-speaking countries (25). We did not include enough English-speaking patients to investigate this issue, which may need further research.

Our MCII estimates for global pain and patient global assessment are consistent with those from previous studies, ranging from −15 to −20 (7, 13, 25), and the PASS estimates in this study are slightly higher than that found in most studies, i.e., approximately 35 (7, 13, 25), but lower than that found in a Canadian study, i.e., approximately 50 (26).

For the 4 outcome measures, the PASS cutoff appears to be slightly higher for hip or knee OA than the other diseases. However, hip or knee OA patients had the highest scores for the 4 outcome measures at baseline. This finding agrees with previous results showing that patients with a chronic disease and with the most severe symptoms would consider satisfactory a higher level of symptoms than would patients with few symptoms (14).

The main strength of this study is that it is the first estimation of the MCII and PASS in the same cohort. The methodology used was the same for each disease (5 diseases) and each country (7 countries), with the same wording for the external anchor, treatment, followup duration, and outcome measures, which allows for comparisons across diseases and countries.

One limitation of the study is that for each of the 4 outcome measures, we used a discrete 0–10-point NRS to measure the outcome and then standardized it into a continuous 0–100 scale. This method relies on the assumption that the transformation was valid in this context. Recent work suggests that transformed WOMAC visual analog scale (VAS) values are clinically acceptable approximations of native NRS values (42). Thus, for example, for pain the overall MCII value will be −17 if pain is measured on a 0–100-point VAS, but the individual-level MCII will be −2 if pain is measured on a 0–10-point NRS, because the NRS is a discrete variable. Similarly, the PASS value for pain will be 42 with a 0–100-point VAS and 4 with a 0–10-point NRS.

Another limitation of the study is that it involved patients receiving NSAIDs. Whether similar MCII and PASS estimates would be obtained for patients receiving tumor necrosis factor blockers, for example, is an issue. Patients receiving NSAIDs and biotherapy may have different expectations for treatment efficacy, with a more ambitious goal for biologic agents, which may affect MCII and PASS estimates. PASS values have been previously estimated in a trial of AS patients comparing adalimumab to placebo (24). PASS values for patient global assessment were 29.5 (95% CI 20.0, 39.0) for the adalimumab group and 53.0 (95% CI 32.0, 81.0) for the placebo group. Similar values were observed for total back pain. However, these data were from small samples (88 patients in the adalimumab group and 24 in the placebo group), and the large 95% CIs do not allow for concluding any difference between the 2 groups and comparison to our results.

This article reports the MCII and PASS estimates for the 4 generic outcome measures recommended for use in trials of rheumatic disorders. The MCII and PASS estimates for disease-specific PROs will be reported in further articles. The consistency of the MCII and PASS estimates for the generic outcome measures allows for using the same estimate across the 5 conditions studied. Addressing the individual level by describing the proportion of patients with an important improvement (as defined by the MCII) or achieving the PASS cutoff provides additional meaningful information from clinical trials and longitudinal observation studies. This approach will aid in the interpretation of results, add useful information for daily practice (43), and provide information that is complementary to the conventional presentation of results at the group level (mean changes in scores). Reporting the proportion of patients achieving the MCII and the PASS aims at complementing, not replacing, information on the effect size, because the effect size remains a more powerful approach.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Tubach had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Tubach, Ravaud, Martin-Mola, Awada, Bellamy, Bombardier, Hajjaj-Hassouni, Hochberg, Logeart, van der Heijde, Dougados.

Acquisition of data. Martin-Mola, Awada, Bellamy, Matucci-Cerinic, van de Laar, van der Heijde, Dougados.

Analysis and interpretation of data. Tubach, Ravaud, Bellamy, Bombardier, Felson, Hajjaj-Hassouni, Hochberg, Logeart, Matucci-Cerinic, van der Heijde, Dougados.

ROLE OF THE STUDY SPONSOR

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Merck, Sharp, & Dohme sponsored the study, participated in the study design, and implemented the study, including the translation of the questionnaires. Merck, Sharp, & Dohme reviewed the manuscript, but had no role in the statistical analysis nor was publication of the article contingent on the approval of Merck, Sharp, & Dohme.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

We acknowledge Gabriel Baron, Arthur Allignol, and Blandine Pasquet for their technical help in statistical analysis and reporting, and Dr. F. Nacci and S. Guiducci for their efforts in implementing the study in Italy.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information
  • 1
    Altman R, Brandt K, Hochberg M, Moskowitz R, Bellamy N, Bloch DA, et al. Design and conduct of clinical trials in patients with osteoarthritis: recommendations from a task force of the Osteoarthritis Research Society. Results from a workshop. Osteoarthritis Cartilage 1996; 4: 21743.
  • 2
    Bellamy N, Kirwan J, Boers M, Brooks P, Strand V, Tugwell P, et al. Recommendations for a core set of outcome measures for future phase III clinical trials in knee, hip, and hand osteoarthritis: consensus development at OMERACT III. J Rheumatol 1997; 24: 799802.
  • 3
    Group for the Respect of Ethics and Excellence in Science (GREES): osteoarthritis section. Recommendations for the registration of drugs used in the treatment of osteoarthritis. Ann Rheum Dis 1996; 55: 5527.
  • 4
    Wright JG. Interpreting health-related quality of life scores: the simple rule of seven may not be so simple. Med Care 2003; 41: 5978.
  • 5
    Beaton DE, Boers M, Wells GA. Many faces of the minimal clinically important difference (MCID): a literature review and directions for future research. Curr Opin Rheumatol 2002; 14: 10914.
  • 6
    Crosby RD, Kolotkin RL, Williams GR. Defining clinically meaningful change in health-related quality of life. J Clin Epidemiol 2003; 56: 395407.
  • 7
    Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, et al. Evaluation of clinically relevant changes in patient-reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement. Ann Rheum Dis 2005; 64: 2933.
  • 8
    Cella D, Hahn EA, Dineen K. Meaningful change in cancer-specific quality of life scores: differences between improvement and worsening. Qual Life Res 2002; 11: 20721.
  • 9
    Guyatt GH, Jaeschke RJ. Reassessing quality-of-life instruments in the evaluation of new drugs. Pharmacoeconomics 1997; 12: 6216.
  • 10
    Schwartz AL, Meek PM, Nail LM, Fargo J, Lundquist M, Donofrio M, et al. Measurement of fatigue: determining minimally important clinical differences. J Clin Epidemiol 2002; 55: 23944.
  • 11
    Jaeschke R, Singer J, Guyatt GH. Measurement of health status: ascertaining the minimal clinically important difference. Control Clin Trials 1989; 10: 40715.
  • 12
    Juniper EF, Guyatt GH, Willan A, Griffith LE. Determining a minimal important change in a disease-specific quality of life questionnaire. J Clin Epidemiol 1994; 47: 817.
  • 13
    Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, 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; 64: 347.
  • 14
    Tubach F, Dougados M, Falissard B, Baron G, Logeart I, Ravaud P. Feeling good rather than feeling better matters more to patients. Arthritis Rheum 2006; 55: 52630.
  • 15
    Cohen J. The cost of dichotomization. Appl Psychol Meas 1983; 7: 24953.
  • 16
    Altman DG, Royston P. The cost of dichotomising continuous variables. BMJ 2006; 332: 1080.
  • 17
    Brozek JL, Guyatt GH, Schunemann HJ. How a well-grounded minimal important difference can enhance transparency of labelling claims and improve interpretation of a patient reported outcome measure. Health Qual Life Outcomes 2006; 4: 69.
  • 18
    Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, Jensen MP, Katz NP, et al. Core outcome measures for chronic pain clinical trials: IMMPACT recommendations. Pain 2005; 113: 919.
  • 19
    Tubach F, Ravaud P, Beaton D, Boers M, Bombardier C, Felson DT, et al. Minimal clinically important improvement and patient acceptable symptom state for subjective outcome measures in rheumatic disorders. J Rheumatol 2007; 34: 118893.
  • 20
    Tubach F, Wells GA, Ravaud P, Dougados M. Minimal clinically important difference, low disease activity state, and patient acceptable symptom state: methodological issues. J Rheumatol 2005; 32: 20259.
  • 21
    Kvamme M, Kristiansen I, Lie E, Kvien T. Identification of cutpoints for acceptable health status and important improvement in patient-reported outcomes, in rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis. J Rheumatol 2009; 37: 2631.
  • 22
    Tubach F, Pham T, Skomsvoll JF, Mikkelsen K, Bjorneboe O, Ravaud P, et al. Stability of the patient acceptable symptomatic state over time in outcome criteria in ankylosing spondylitis. Arthritis Rheum 2006; 55: 9603.
  • 23
    Ten Klooster PM, Drossaers-Bakker KW, Taal E, van de Laar MA. Patient-perceived satisfactory improvement (PPSI): interpreting meaningful change in pain from the patient's perspective. Pain 2006; 121: 1517.
  • 24
    Dougados M, Luo M, Maksymowych W, Chmiel J, Chen N, Wong R, et al. Evaluation of the patient acceptable symptom state as an outcome measure in patients with ankylosing spondylitis: data from a randomized controlled trial. Arthritis Rheum 2008; 59: 55360.
  • 25
    Maksymowych W, Gooch K, Dougados M, Wong R, Chen N, Kupper H, et al. Thresholds of patient-reported outcomes that define the patient acceptable symptom state in ankylosing spondylitis vary over time and by treatment and patient characteristics. Arthritis Care Res (Hoboken) 2010; 62: 82634.
  • 26
    Maksymowych WP, Richardson R, Mallon C, van der Heijde D, Boonen A. Evaluation and validation of the patient acceptable symptom state (PASS) in patients with ankylosing spondylitis. Arthritis Rheum 2007; 57: 1339.
  • 27
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 28
    Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis: a proposal for modification of the New York criteria. Arthritis Rheum 1984; 27: 3618.
  • 29
    Altman R, Alarcon G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hand. Arthritis Rheum 1990; 33: 160110.
  • 30
    Altman R, Alarcon G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip. Arthritis Rheum 1991; 34: 50514.
  • 31
    Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum 1986; 29: 103949.
  • 32
    Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum 1980; 23: 13745.
  • 33
    Garrett S, Jenkinson T, Kennedy LG, Whitelock H, Gaisford P, Calin A. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol 1994; 21: 228691.
  • 34
    Calin A, Garrett S, Whitelock H, Kennedy LG, O'Hea J, Mallorie P, et al. A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index. J Rheumatol 1994; 21: 22815.
  • 35
    Kelli DA, Joanne MJ, Jordan BR, Virginia BK. Validity, factor structure, and clinical relevance of the AUSCAN osteoarthritis hand index. Arthritis Rheum 2006; 54: 5516.
  • 36
    Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthitis of the hip or knee. J Rheumatol 1995; 15: 183340.
  • 37
    Roland M, Morris R. A study of the natural history of back pain. Part I. Development of a reliable and sensitive measure of disability in low-back pain. Spine 1983; 8: 1414.
  • 38
    Wells G, Anderson J, Beaton D, Bellamy N, Boers M, Bombardier C, et al. Minimal clinically important difference module: summary, recommendations, and research agenda. J Rheumatol 2001; 28: 4524.
  • 39
    Wells G, Boers M, Shea B, Anderson J, Felson D, Johnson K, et al. MCID/low disease activity state workshop: low disease activity state in rheumatoid arthritis. J Rheumatol 2003; 30: 11101.
  • 40
    Tubach F, Giraudeau B, Ravaud P. The variability in MCII and PASS estimates did not have an impact on treatment effect estimates. J Clin Epidemiol 2009; 62: 7258.
  • 41
    Sprangers MA, de Regt EB, Andries F, van Agt HM, Bijl RV, de Boer JB, et al. Which chronic conditions are associated with better or poorer quality of life? J Clin Epidemiol 2000; 53: 895907.
  • 42
    Bellamy N, Hendrikz J, Wilson C. Comparison of transformed visual analogue and native numerical rating scaled patient responses to the WOMAC Index [abstract]. Int Med J 2011; 41 Suppl: 23.
  • 43
    Wells G, Anderson J, Boers M, Felson D, Heiberg T, Hewlett S, et al. MCID/low disease activity state workshop: summary, recommendations, and research agenda. J Rheumatol 2003; 30: 11158.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
ACR_21747_sm_Appendix1.doc23KSupplementary Data
ACR_21747_sm_Appendix2.doc110KSupplementary Data

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.