The minimal clinically important difference of the scleroderma clinical trials consortium damage index

Abstract

Objective:

The Scleroderma Clinical Trials Consortium Damage Index is an index of global damage in systemic sclerosis. The objective of this study is to determine the minimal clinically important difference of the Scleroderma Clinical Trials Consortium Damage Index.

Methods:

Patients in the Canadian Scleroderma Research Group registry and the Australian Scleroderma Cohort Study who completed Scleroderma Clinical Trials Consortium Damage Index scores and the SF36v2 at baseline and the first full follow-up visit were studied. To calculate the minimal clinically important difference, an anchor question came from SF36v2: “Compared to one year ago, how would you rate your health in general?.” Options were: much better, somewhat better, about the same, somewhat worse and much worse. We use the “somewhat worse” or “much worse” categories to indicate those with any worsening. We used four anchor methods: receiver operating characteristic curve, change difference, regression analysis, and average change.

Results:

We studied 1672 patients. Mean disease duration was 11.4 ± 10.0 years; 62.5% had diffuse cutaneous systemic sclerosis. Baseline mean Damage Index was 5.3 ± 4.2; mean change of Damage Index over 1 year was 0.9 ± 1.8 units. The calculated minimal clinically important difference values were 1 for receiver operating characteristic method, 0.625 for change difference, 0.1879 for regression analysis, and 1.37 for average change. Omitting the regression analysis method as an outlier, the mean of the other methods was 1.

Conclusion:

The most appropriate minimal clinically important difference for the Scleroderma Clinical Trials Consortium Damage Index is a change of ⩾ 1.0 units in the Scleroderma Clinical Trials Consortium Damage Index as is already recognized by patients as a significant change after 1 year. This can be applied to group means as well as to individuals where an ordinal change is required.

Keywords

MCID scleroderma systemic sclerosis outcome measures damage

Systemic sclerosis (SSc) is a rare systemic autoimmune disease associated with high mortality. It is characterized by immune dysregulation, vasculopathy, and fibroblast dysfunction, which ultimately leads to increased deposition of extracellular matrix, fibrosis, and irreversible organ damage.¹ To measure damage in SSc, the Scleroderma Clinical Trials Consortium Damage Index (SCTC-DI) was developed and validated in 2019.² As the first validated SSc damage measurement, the SCTC-DI may allow us to better study the natural evolution of the disease and be useful in both the clinical and trial design setting.

As we have shown that the SCTC-DI worsens over time even in early SSc,³ it is important that we understand what degree of change in this measure of damage is appreciated by patients. This may become especially important if damage accrual prevention is considered to be a possible endpoint of clinical trials in SSc.

The minimal clinically important difference (MCID) is defined as “the smallest difference in score in the domain of interest which patients perceive as beneficial.”⁴ As the SCTC-DI can only get worse over time, the definition in this case is the minimal change in the score that patients perceive as worsening.

There are two major classes of methods to determine the MCID of a measure; anchor-based and distribution-based methods.⁵ Our study used several anchor-based methods to determine the MCID of the SCTC -DI over 1 year.

Methods

Patient cohort

Participants were recruited from the Canadian Scleroderma Research Group (CSRG) registry and from the Australian Scleroderma Cohort Study (ASCS). The CSRG recruits and follows patients from 15 centers in Canada and Mexico. These centers see local and regional referrals. All patients must have a diagnosis of SSc (confirmed by an experienced rheumatologist), be ⩾ 18 years of age, provide informed consent, and be fluent in English, French, or Spanish. Over 98% of the cohort meets the 2013 ACR/EULAR classification criteria for SSc.⁶ The ASCS is an Australian multicentre cohort study of risk and prognostic factors in SSc. The CSRG registry and the ASCS have been approved by all Human Research Ethics Committees of participating sites. All participants met ACR/EULAR criteria for SSc.⁶ Written informed consent was obtained from all participants at recruitment. All sites have the approval of their institutional review boards and patients provide written consent. Data are collected at yearly visits using a standardized data collection protocol, which includes clinical (both patient- and physician-reported) and laboratory data and is recorded in a customized database. Patients for this study were those who had a calculable SCTC-DI over two visits 0.75–1.25 years apart and an available SF-36 at the second visit. The first such set of visits in the database for each patient was employed for the study. Disease duration is time from first non-Raynaud’s manifestation.

This study was approved by the ethics committee at the Jewish General Hospital in Montreal, Canada (Approval number project MP-05-2017-563, 16-210, Form: F9 H-40322).

Scleroderma Clinical Trials Consortium Damage Index

The SCTC-DI measures global irreversible damage in SSc patients and was developed to be highly correlated with mortality and morbidity (measured by Short Form-36).² Validated and published in 2019, its development was an international collaboration between 22 experts with input from patient partners using a combined approach of consensus and data-driven methods. The index is composed of 23 differently weighted items in several organ systems (musculoskeletal, skin, vascular, gastrointestinal, respiratory, cardiovascular, and renal). Low, medium, and high SCTC-DI scores are defined as < 5, 6–12, and ⩾ 13, respectively, with a maximum score of 55. For the definition of the individual components of the SCTC-DI, please refer to the original paper.²

In this study, the SCTC-DI was calculated using registry data, but three items (calcinosis complicated by infection or requiring surgery, gastric antral vascular ectasia (GAVE), and right ventricular dysfunction) were removed since they were not collected in the CSRG database, and one item (small joint contractures) was removed due to missing data (> 20% of visits). The maximum SCTC-DI score possible was therefore 42. Scores are always whole integers as are changes in scores over time.

The SCTC-DI is scored such that patients cannot improve; they can only stay the same or worsen.

Anchor

The anchor question used in our analyses came from SF36v2—“Compared to one year ago, how would you rate your health in general?”. The options were: much better, somewhat better, about the same, somewhat worse, and much worse. We use the “somewhat worse” or “much worse” categories to indicate those with any worsening.

SF-36 score

The SF-36 includes one multi-item scale that measures eight health domains including physical functioning, role limitation because of physical health problems, bodily pain, social functioning, general mental health, role limitations because of emotional problems, vitality, and general health perceptions.⁷ These eight domains can be summarized into a physical component score (PCS) and a mental component score (MCS). A final score for each domain is provided between 0 and 100, with 0 being the worst possible health and 100 the best possible health. The score is standardized against normative population data, where the mean score ± SD of 50 ± 10 is the population normative score.

HAQ-DI score

The Health Assessment Questionnaire Disability Index (HAQ-DI) is a patient-reported measure of function, where a higher score indicates a worse functional status.⁸ Participants are asked 20 questions about their ability to perform activities of daily living. Each item is rated on a scale from 0 to 3, where “3” represents being unable to complete the task.⁸ Visual analog scales (VAS) to evaluate SSc organ system symptoms including Raynaud’s phenomenon, gastrointestinal (GI) tract, lung involvement, pain, and overall disease severity have been added to the HAQ-DI and the combination of those scales and the original HAQ-DI is commonly referred to as the Scleroderma HAQ (S-HAQ).⁹

Statistics

We assessed the relationship between the four anchor categories and the change in DI over 1 year by analysis of variance (ANOVA) and performed a Tukey test to compare paired group means.

Anchor methods: We used four anchor methods.^4,10–18

Receiver operating characteristic (ROC) curve. We plotted sensitivity and specificity for worsen versus not-worsen group at each change of DI. The best cut point was calculated with Youden’s index.

Change difference (CD). MCID was calculated as the average score change of worsen versus not-worsen patients.

Regression analysis (REG). MCID was the beta coefficient change in the anchor regressor.

Average change (AC). This was the average score change in patients considered worse by the anchor.

Results

We determined the MCID value in a subgroup of 1672 CSRG and ASIG patients who met entry criteria, that is, those who completed SCTC-DI scores at baseline and at the first full follow-up visit at 0.75–1.25 years, as well as completed the anchor question in the SF-36 questionnaire at the first full follow-up visit. Characteristics of the patients studied are shown in Table 1.

Table 1.

Patient characteristics and change of characteristics of total sample according to individual MCID of ⩾ 1.

Variable	Whole cohort (n = 1672)	Change of DI ⩾ 1 (n = 502)	Change of DI = 0 (n = 1170)	p-value of change of DI ⩾ 1 versus 0
N (% of total)	1672 (100.0%)	502 (100.0%)	1170 (100.0%)	—
Age, mean (SD)	56.8 (12.22)	57.3 (11.96)	56.6 (12.32)	0.279
Female, n (%)	1458 (87.2%)	421 (83.9%)	1037 (88.6%)	0.007
Disease duration at first visit, years (mean, SD)	11.4 (10.02)	10.9 (10.52)	11.6 (9.80)	0.214
lcSSc, n (%)	1037 (62.5%)	264 (52.8%)	773 (66.6%)	< 0.001
dcSSc, n (%)	623 (37.5%)	236 (47.2%)	387 (33.4%)	< 0.001
Baseline DI, mean (SD)	5.3 (4.21)	5.2 (4.31)	5.4 (4.17)	0.573
Baseline DI, median (IQR)	4.00 (3.00–7.00)	4.00 (3.00–7.00)	4.00 (3.00–7.00)	0.528
Follow up DI (Mean, SD)	6.2 (4.56)	8.3 (4.77)	5.4 (4.17)	< 0.001
Follow up DI, median (IQR)	5.00 (3.00–8.00)	7.00 (5.00–11.00)	4.00 (3.00–7.00)	< 0.001
Change DI, mean (SD)	0.9 (1.75)	3.1 (1.91)	0.0 (0.00)	< 0.001
Patients who at baseline indicated on anchor that they felt worse than previous year, n (%)	557 (33.3%)	196 (39.0%)	361 (30.9%)	0.001
Baseline HAQ DI, mean (SD)	0.7 (0.68)	0.9 (0.71)	0.7 (0.66)	< 0.001
Change HAQ DI, mean (SD)	0.0 (0.37)	0.1 (0.42)	0.0 (0.35)	0.002
Baseline mRSS, mean (SD)	8.3 (8.41)	10.7 (9.72)	7.3 (7.55)	< 0.001
Change mRSS, mean (SD)	−0.4 (4.65)	−0.0 (5.80)	−0.5 (4.03)	0.051
Baseline SF-36 PCS, mean (SD)^a	39.1 (11.31)	36.8 (11.26)	40.1 (11.19)	< 0.001
Change SF-36 PCS, mean (SD)^a	−0.2 (7.24)	−0.4 (7.35)	−0.1 (7.20)	0.415
Baseline SF-36 MCS, mean (SD)^a	47.8 (11.85)	47.4 (12.24)	47.9 (11.68)	0.439
Change SF-36 MCS, mean (SD)^a	0.3 (9.43)	0.1 (10.20)	0.4 (9.09)	0.593
Baseline overall health, mean (SD)^b	3.2 (2.27)	3.6 (2.35)	3.0 (2.22)	< 0.001
Change overall health, mean (SD)^b	0.0 (2.14)	0.2 (2.28)	−0.1 (2.07)	0.085
Baseline pain severity, mean (SD)^b	3.3 (2.74)	3.6 (2.75)	3.2 (2.72)	0.005
Change pain severity, mean (SD)^b	−0.0 (2.36)	0.2 (2.42)	−0.1 (2.33)	0.017
Baseline Raynaud’s severity, mean (SD)^b	2.7 (2.82)	2.9 (2.91)	2.6 (2.78)	0.066
Change Raynaud’s severity, mean (SD)^b	0.1 (2.55)	−0.1 (2.80)	0.1 (2.43)	0.245
Baseline finger ulcer severity, mean (SD)^b	1.5 (2.68)	1.9 (3.01)	1.4 (2.52)	< 0.001
Change finger ulcer severity, mean (SD)^b	0.0 (2.42)	−0.1 (2.68)	0.1 (2.29)	0.108
Baseline intestinal problems severity, mean (SD)^b	1.8 (2.52)	1.98 (2.61)	1.7 (2.47)	0.027
Change intestinal problems severity, mean (SD)^b	0.0 (2.34)	0.1 (2.67)	0.0 (2.18)	0.418
Baseline breathing problems severity, mean (SD)^b	1.9 (2.44)	2.2 (2.59)	1.8 (2.37)	0.001
Change breathing problems severity, mean (SD)^b	0.1 (2.02)	0.2 (2.17)	−0.0 (1.96)	0.031
Baseline limitation activity severity, mean (SD)^c	3.4 (2.55)	3.9 (2.55)	3.2 (2.52)	< 0.001
Change limitation activity severity, mean (SD)^c	−0.0 (2.08)	0.0 (2.20)	−0.0 (2.03)	0.427

DI: damage index; lcSSc: limited cutaneous systemic sclerosis; dcSSc: diffuse cutaneous systemic sclerosis; IQR: interquartile range; HAQ DI: health assessment questionnaire disability index; mRSS: modified Rodnan skin score; SF-36: 36-item short form health survey; PCS: physical component score; MCS: mental component score.

Change refers to change over 1 year. For categorical variables, a number (percentage) is presented with a p-value calculated by chi-square test. For normally distributed continuous variables, the mean (standard deviation) is presented with a p-value calculated by Student’s t-test. For non-normally distributed continuous variables, the median (interquartile range) is presented with a p-value calculated by Wilcoxon rank-sum test.

SF-36 physical and mental component scores.

0–10 numerical rating scales from S-HAQ.⁹

Response to the question “Overall, considering pain, discomfort, limitations in your daily life, and other changes in your body and life, how would you rate your disease in the past week?”

Table 2 shows the relationship between anchor categories after 1 year and change in DI scores. ANOVA found that at least one group mean was different from others (F = 14.11, < 0.001). A Tukey test to compare paired group means found that the “much worse” group was significantly different from all other groups; the “somewhat worse” group was significantly different from the “better” or “no change” groups. Groups of “much better,” “somewhat better,” and “about the same” were not significantly different from each other. This confirmed the expected relationship between the change in DI scores and the anchors.

Table 2.

Relationship between anchor categories after 1 year and change in DI scores.

	Anchor score after 1 year
	Much better (n = 96)	Somewhat better (n = 239)	About the same (n = 866)	Somewhat worse (n = 409)	Much worse (n = 62)
Change DI over 1 year, mean (SD)	1.0 (1.84)	0.8 (1.48)	0.7 (1.45)	1.3 (2.01)	2.0 (3.19)

DI: damage index.

Several different anchor-based methods were used to calculate the MCID. A histogram of the number of subjects with each amount of change and the number who felt worse with that change is depicted in Figure 1.

ROC curve: Supplementary Figure 1 shows the ROC curve for sensitivity and specificity of multiple change of DI values for patient-reported worsening versus not worsening by the anchor over 1 year. The area under the curve (AUC) is only 0.58, which is poor. The best cut-off from Youden analysis is MCID = 1 at which value the sensitivity is 0.399 and specificity is 0.739 for feeling worse.

CD: Table 3 shows the MCID calculated as the difference between the ACs of DI in the worsen group and in the not-worsen group was 0.6250, (95% confidence interval (CI) = 0.4411–0.8089). The values differed by subsets of patients. For example, incident patients (those with disease duration < 7 years) have a higher MCID as do those with diffuse cutaneous systemic sclerosis (dcSSc).

REG: A logistic regression model was fitted to estimate the beta coefficient for change in DI and the result is shown in Supplementary Table 1. Change in DI is significantly related to the anchor (p < 0.0001) and the estimate MCID using the beta coefficient is 0.1879 (95% CI: 0.1291–0.2466)

AC: The mean change of DI in the worsen group is 1.37 units (95% CI: 1.17–1.56)

Summary of results by different methods: A summary of all results for MCID determination is displayed in Table 4.

Table 3.

MCID calculated as the difference between average changes of DI in the worsen group and in the not-worsen group.

Subtype	All patients (95% CI)	Prevalent patients (95% CI)	Incident patients (95% CI)
All patients	0.6250 (0.4411–0.8089)	0.5687 (0.3217–0.8156)	0.8878 (0.5534–1.2222)
lcSSc patients	0.3472 (0.1267–0.5677)	0.5015 (0.1994–0.8035)	0.2386 (–0.1432–0.6204)
dcSSc patients	1.0825 (0.7628–1.4022)	0.6859 (0.2659–1.1059)	1.6479 (1.0857–2.2101)

CI: confidence interval; lcSSc: limited cutaneous systemic sclerosis; dcSSc: diffuse cutaneous systemic sclerosis.

Table 4.

Summary of the results for MCID estimation using various anchor-based methods.

	Anchor Based Method
	ROC	CD	REG	AC
MCID	1	0.6250	0.1879	1.37
95% CI	NA	0.4411–0.8089	0.1291–0.2466	1.17–1.56

ROC: receiver operating characteristic curve; CD: change difference; REG: regression analysis; AC: average change; MCID: minimal clinically important difference; CI: confidence interval.

We found different MCID values depending on the method used. However, the mean MCID over all four methods was 0.8. As the regression method produced a result that was clearly an outlier, without that value the mean of the remaining three methods was 1.0. An MCID can be applied to individuals, that is, what proportion of individuals change by the MCID or greater, or to a group, that is, did a group mean change by the MCID or greater. As the SCTC-DI in an individual can only change by a whole number, we therefore feel confident that the most appropriate MCID for both a group and an individual would be a change of ⩾ 1.0.

We therefore assessed how subjects with a change of ⩾ 1 would fare. As mentioned above under “ROC curve,” the sensitivity and specificity of an MCID of ⩾ 1 for worsening versus not worsening anchors were 0.399 and 0.739, respectively. Supplementary Table 2 shows the numbers for that calculation.

Table 1 demonstrates the baseline characteristics and the changes in variables over 1 year for those who met and who did not meet the individual MCID of ⩾ 1. Those who worsened by the individual MCID or greater were more likely to die within the next 12 months or during follow-up. The HAQ DI, modified Rodnan skin score (mRSS), SF-36 PCS, and ratings of overall health, pain severity, finger ulcer severity, intestinal problems severity, breathing problems severity, and limitation of activity severity were worse at baseline in those with MCID of ⩾ 1. At baseline, more patients in the group whose DI worsened stated that they felt worse than in the previous year. It can be seen that those with a change of 1 or greater had more worsening over 1 year of the HAQ, pain severity, and breathing severity, and borderline less improvement in mRSS. There was no significant change in the SF-36 PCS or SF-36 MCS or other S-HAQ items between the two groups.

Discussion

The SCTC-DI is a new measure of disease-related damage in patients with SSc.²

First, we verified that worsening of the anchor from the SF-36 moves in parallel with worsening of the DI score.

Then we used four different anchor-based methods to determine an MCID. These methods were recently reviewed.¹⁸ With this approach, we determined that the best MCID to determine if an individual had worsened or an MCID for a group change would be a change of ⩾ 1.0.

We showed that individuals with a change of DI ⩾ 1 were more likely to die at 1 year and during follow-up. Their SF-36 PCS was worse at baseline and over 1 year their HAQ DI, pain severity and breathing severity worsened more and their mRSS improved less.

We also found that if the MCID was calculated as the difference between ACs of DI in the worsen group and in the not-worsen group, the MCID values were different with prevalent or incident patients and if patients had lcSSc or dcSSc (see Table 3). These variations imply that such differences should be taken into account when studying specific subsets of cases.

When we examined the baseline scores of how patients felt about the severity of individual organ system involvement, we found that those who accrued damage over the following year rated the degree of severity of disease of the individual organs as worse than those who accrued no damage (see Table 1). We also noted that more patients who accrued damage over follow-up also indicated on the same anchor question at baseline that they felt worse than the previous year (see Table 1). As the DI baseline values were about the same in those who accrued damage over the subsequent year and those who did not, we think that perhaps those who accrued damage may have been experiencing more disease activity at baseline and in the year prior. This explanation would fit with the hypothesis that disease activity eventually leads to disease damage.

From Table 1, we can also see that patients who felt globally worse after 1 year and did accrue more damage did not notice much change in many individual organ or system assessments. We are not certain why this is so but it is possible that these questions were not sensitive enough to detect change or that perhaps individual organ changes were small but overall there was still some global deterioration in how they felt compared to 1 year prior.

Our study had some limitations. Although the SF-36 Health Transition Item as anchor has been used in many MCID studies, including for many different types of outcome measures including those used for spine surgery,^{11,12,15,19–26} not only disease damage may affect this item. For example, there may be more disease activity at the follow-up visit and without more damage that item might reflect the worsening of disease activity. This same issue applies to all studies using an anchor that does not necessarily represent only the measure being studied. Indeed, we can see from our data in Figure 1 that a considerable proportion of patients indicated a worsening in the anchor who did not even have the minimal possible change of one unit in the SCTC-DI. This is a definite limitation of the use of a generic health transition item when assessing the MCID of almost any new outcome measure. However, we did show in Table 2 that the degree of worsening in the SCTC-DI increases progressively with patient’s feeing progressively worse. Showing that there is a relationship between the anchor and the measure being studied is a pre-requisite for using the anchor in such a study and suggests that even though there may be other factors affecting the anchor, this relationship supports its use.^18,27

Perhaps it would have been more accurate to use an anchor in which the patient rates the degree of disease damage only. In unpublished work that we have done interviewing our patients in preparation for determining if we could use patient self-reports for studying disease activity and damage, we discovered that patients could not differentiate between the concepts of disease damage and disease activity. We did introduce the following questions in our data collection for a brief time:

1. Overall, considering how much pain, discomfort, limitation in your daily life and other changes in your body and life, how “active” would you rate your scleroderma in the past week?

2. Overall, considering how much pain, discomfort, limitation in your daily life and other changes in your body and life, how would you rate the “damage” you have incurred to date from your scleroderma?

We withdrew those questions when we understood the difficulty patients had differentiating activity from damage. So if we were to do a prospective study with an anchor to determine the MCID of the SCTC-DI, we would still most likely use the SF-36 health transition question.

The use of a subjective anchor is common in such MCID studies. In a recent review of the design of anchors in studies of MCID of a measure, of 340 articles found in a systematic review, 99.12% used subjective anchors.²⁷ Most studies (73.24%) used a single anchor as we did.²⁷ One practical reason for choosing this anchor was that this study used an existing cohort of patients and we needed an anchor that was included in our data and the SF-36 transition item was available.

Another possible limitation is that our analysis using the ROC curve does not “appear” successful in that the AUC is not large. However, the best cut-off was still an MCID of 1, which is similar to two of the three other methods and its specificity was good.

In general, anchor-based methods have a major limitation in their recall bias. A recall bias occurs when a patient is asked to recall a memory that happened a long time before, which may result in an effect of recall bias.

Distribution-based methods, on the contrary, have the disadvantage of losing information that reflect patients’ perspective because patients’ feedback is not considered, and may result in a lack of clinical interpretability. Moreover, the distribution-based methods used to estimate MCID can be sample-specific and are sensitive to sample characteristics, such as skewness in the data. In any case, we were not able to use these methods because they apply to normally distributed changes of improvement and worsening over time and the SCTC DI can only worsen.

There is also some controversy about the term MCID. According to the Outcome Measures in Rheumatoid Arthritis Clinical Trials (OMERACT) handbook, “OMERACT has traditionally used the term MCID and more recently has adopted the term minimal important difference (MID) to represent change that is important for example to patients but may not have been generated in the context of its clinical relevance for clinical decision making.” ²⁸ The handbook, however, also states that “clinical meaningfulness refers to an amount of change that is thought to be relevant to patients . . . and that some have reserved the acronym MCID for this type of change score.”²⁸ As the term MCID is still most commonly used and as we have assessed clinical meaningfulness to patients, we have retained that terminology.

A strength of our study is the very large size of the cohort studied and the use of multiple anchor-based methods. In addition, finding an MCID of ⩾ 1 is intuitively correct in that all items in the DI, even those which only add 1 point to the score, are serious enough to expect the patient to feel worse. The scale is built on discreet data where each increase is significant. This suggests that treatment should be aimed at preventing any increase in damage, even if only preventing a score increase of 1.

In conclusion, we found that an MCID for the SCTC-DI for both individuals with SSc and for groups of patients was a change of ⩾ 1. In addition, we showed that the MCID differs in different subsets of patients and adjustments could be considered for prevalent versus incident cases or lcSSc versus dcSSc cases.

Supplemental Material

sj-docx-1-jso-10.1177_23971983251327808 – Supplemental material for The minimal clinically important difference of the scleroderma clinical trials consortium damage index

Supplemental material, sj-docx-1-jso-10.1177_23971983251327808 for The minimal clinically important difference of the scleroderma clinical trials consortium damage index by Murray Baron, Dylan Hansen, Susanna Proudman, Wendy Stevens, Mianbo Wang and Mandana Nikpour in Journal of Scleroderma and Related Disorders

Footnotes

Acknowledgements

The following are members of the Australian Scleroderma Interest Group (ASIG) who were not named as authors but all of whom collected data from this paper and reviewed the article: Joanne Sahhar, Monash University; Nava Ferdowsi, Victoria Australia; Kathleen Morrisroe, University of Melbourne; Laura Ross, University of Melbourne; Gene Siew Ngian, University of Melbourne; Jennifer Walker, Flinders University; Janet Roddy, Fiona Stanely Hospital; Lauren Host, Sir Charles Gairdner Hospital.

The following are members of the Canadian Scleroderma Research Group (CSRG) who were not named as authors but all of whom collected data from this paper and reviewed the article: Mohammed Osman, University of Alberta; Peter Docherty, The Moncton Hospital; Paul R. Fortin, Centre de Recherche Arthrite, Division of Rheumatology, Department of Medicine, CHU de Québec—Université Laval; Marvin J. Fritzler, University of Calgary; Genevieve Gyger, McGill University; Alena Ikic, Université Laval; Niall Jones, University of Alberta; Elzbieta Kaminska, University of Calgary; Nader A. Khalidi, St Joseph’s Healthcare, McMaster University; Maggie Larche, St Joseph’s Healthcare, McMaster University; Sophie Ligier, Hopital Maisonneuve-Rosemont; Ada Man, University of Manitoba; Ariel Masetto, University of Sherbrooke; Janet E. Pope, Western University; David Robinson, University of Manitoba; Tatiana S. Rodriguez-Reyna, Instituto Nacional de Ciencias Médicas y Nutrición; Evelyn Sutton, Dalhouse University; Carter Thorne, The Arthritis Program Research Group, University of Toronto.

Data availability

The datasets generated during and/or analyzed during this study are available from the corresponding author on reasonable request.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Canadian Scleroderma Research Group (CSRG) is funded by the Canadian Institutes of Health Research (CIHR), Scleroderma Canada, the Scleroderma Society of Ontario, the Scleroderma Association of Saskatchewan, Scleroderma Manitoba, the Scleroderma Society of Nova Scotia, the Scleroderma Association of British Columbia, the Cure Scleroderma Foundation, the Canadian Blood and Marrow Transplant Group, and the Lady Davis Institute of Medical Research of the Jewish General Hospital, Montreal, QC. The CSRG has also received educational grants from Actelion pharmaceuticals and Mallinckrodt.

Dr. Nikpour is recipient of a National Health and Medical Research Council of Australia Emerging Leadership Investigator Grant (GNT 1176538).

The Australian Scleroderma Cohort Study is supported by research grants from Janssen, Boehringer Ingelheim, Australian Rheumatology Association and Arthritis Australia.

Ethical approval

This study was approved by the ethics committee at the Jewish General Hospital in Montreal, Canada.

Informed consent

Written informed consent was obtained from all participants at recruitment.

ORCID iD

Murray Baron

Supplemental material

Supplemental material for this article is available online.

References

Johnson

Fransen

Khanna

, et al. There is a need for new systemic sclerosis subset criteria. A content analytic approach (abstract). Conference abstract. Ann Rheum Dis 2015; 74: 1141. DOI: 10.1136/annrheumdis-2015-eular.2277.

Ferdowsi

Huq

Stevens

, et al. Development and validation of the scleroderma clinical trials consortium damage index (SCTC-DI): a novel instrument to quantify organ damage in systemic sclerosis. Ann Rheum Dis 2019; 78: 807–816.

Barbacki

Baron

Wang

, et al. Damage trajectories in systemic sclerosis using group-based trajectory modeling. Arthritis Care Res (Hoboken) 2023; 75: 640–647. DOI: 10.1002/acr.24873.

Jaeschke

Singer

Guyatt

GH.

Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989; 10(4): 407–415. DOI: 10.1016/0197-2456(89)90005-6.

McGlothlin

Lewis

RJ.

Minimal clinically important difference: defining what really matters to patients. JAMA 2014; 312: 1342–1343. DOI: 10.1001/jama.2014.13128.

van den Hoogen

Khanna

Fransen

, et al. 2013 classification criteria for systemic sclerosis: an American college of rheumatology/European league against rheumatism collaborative initiative. Ann Rheum Dis 2013; 72: 1747–1755. DOI: 10.1136/annrheumdis-2013-204424.

Danieli

Airò

Bettoni

, et al. Health-related quality of life measured by the Short Form 36 (SF-36) in systemic sclerosis: correlations with indexes of disease activity and severity, disability, and depressive symptoms. Clin Rheumatol 2005; 24(1): 48–54. DOI: 10.1007/s10067-004-0970-z.

Pope

Measures of systemic sclerosis (scleroderma): Health Assessment Questionnaire (HAQ) and Scleroderma HAQ (SHAQ), physician- and patient-rated global assessments, Symptom Burden Index (SBI), University of California, Los Angeles, Scleroderma Clinical Trials Consortium Gastrointestinal Scale (UCLA SCTC GIT) 2.0, Baseline Dyspnea Index (BDI) and Transition Dyspnea Index (TDI) (Mahler’s Index), Cambridge Pulmonary Hypertension Outcome Review (CAMPHOR), and Raynaud’s Condition Score (RCS). Arthritis Care Res (Hoboken) 2011; 63(suppl 11): S98–S111. DOI: 10.1002/acr.20598.

Steen

Medsger

Jr.

The value of the Health Assessment Questionnaire and special patient-generated scales to demonstrate change in systemic sclerosis patients over time. Arthritis Rheum 1997; 40(11): 1984–1991.

10.

Guyatt

Walter

Norman

Measuring change over time: assessing the usefulness of evaluative instruments. J Chronic Dis 1987; 40(2): 171–178. DOI: 10.1016/0021-9681(87)90069-5.

11.

Kaleth

Slaven

Ang

DC.

Determining the Minimal Clinically Important Difference for 6-Minute Walk Distance in Fibromyalgia. Am J Phys Med Rehabil 2016; 95(10): 738–745. DOI: 10.1097/phm.0000000000000485.

12.

Kang

Veeraraghavan

Martin

, et al. An updated approach to determine minimal clinically important differences in idiopathic pulmonary fibrosis. ERJ Open Res 2021; 7(4): 00142–2021. DOI: 10.1183/23120541.00142-2021.

13.

Rai

Yazdany

Fortin

, et al. Approaches for estimating minimal clinically important differences in systemic lupus erythematosus. Arthritis Res Ther 2015; 17: 143. DOI: 10.1186/s13075-015-0658-6.

14.

Revicki

Hays

Cella

, et al. Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes. J Clin Epidemiol 2008; 61(2): 102–109. DOI: 10.1016/j.jclinepi.2007.03.012.

15.

Steinhaus

Iyer

Lovecchio

, et al. Minimal clinically important difference and substantial clinical benefit using PROMIS CAT in cervical spine surgery. Clin Spine Surg 2019; 32(9): 392–397. DOI: 10.1097/bsd.0000000000000895.

16.

Wells

Beaton

Shea

, et al. Minimal clinically important differences: review of methods. J Rheumatol 2001; 28: 406–412.

17.

Wright

Hannon

Hegedus

, et al. Clinimetrics corner: a closer look at the minimal clinically important difference (MCID). J Man Manip Ther 2012; 20(3): 160–166. DOI: 10.1179/2042618612Y.0000000001.

18.

Mouelhi

Jouve

Castelli

, et al. How is the minimal clinically important difference established in health-related quality of life instruments? Review of anchors and methods. Health Qual Life Outcomes 2020; 18: 136. DOI: 10.1186/s12955-020-01344-w.

19.

Copay

Cher

DJ.

Is the Oswestry Disability Index a valid measure of response to sacroiliac joint treatment?

Qual Life Res 2016; 25(2): 283–292. DOI: 10.1007/s11136-015-1095-3.

20.

Kafaja

Clements

Wilhalme

, et al. Reliability and minimal clinically important differences of forced vital capacity: results from the scleroderma lung studies (SLS-I and SLS-II). Am J Respir Crit Care Med 2018; 197: 644–652. DOI: 10.1164/rccm.201709-1845OC.

21.

Carreon

Glassman

Campbell

, et al. Neck disability index, short form-36 physical component summary, and pain scales for neck and arm pain: the minimum clinically important difference and substantial clinical benefit after cervical spine fusion. Spine J 2010; 10(6): 469–474. DOI: 10.1016/j.spinee.2010.02.007.

22.

Parker

Mendenhall

Shau

, et al. Determination of minimum clinically important difference in pain, disability, and quality of life after extension of fusion for adjacent-segment disease. J Neurosurg Spine 2012; 16(1): 61–67. DOI: 10.3171/2011.8.Spine1194.

23.

Gum

Glassman

Carreon

LY.

Clinically important deterioration in patients undergoing lumbar spine surgery: a choice of evaluation methods using the Oswestry Disability Index, 36-Item Short Form Health Survey, and pain scales: clinical article. J Neurosurg Spine 2013; 19(5): 564–568. DOI: 10.3171/2013.8.Spine12804.

24.

Auffinger

Lam

Shen

, et al. Measuring surgical outcomes in subaxial degenerative cervical spine disease patients: minimum clinically important difference as a tool for determining meaningful clinical improvement. Neurosurgery 2014; 74(2): 206–213; discussion 213–204. DOI: 10.1227/neu.0000000000000247.

25.

Conijn

Bipat

Reekers

, et al. Determining the minimally important difference for the vascuqol sumscore and its domains in patients with intermittent claudication. Eur J Vasc Endovasc Surg 2016; 51(4): 550–556. DOI: 10.1016/j.ejvs.2015.12.012.

26.

Joelson

Nerelius

Sigmundsson

, et al. The minimal important change for the EQ VAS based on the SF-36 health transition item: observations from 25772 spine surgery procedures. Qual Life Res 2022; 31(12): 3459–3466. DOI: 10.1007/s11136-022-03182-3.

27.

Zhang

Huang

The anchor design of anchor-based method to determine the minimal clinically important difference: a systematic review. Health Qual Life Outcomes 2023; 21: 74. DOI: 10.1186/s12955-023-02157-3.

28.

Beaton

Maxwell

Grosskleg

, et al. Chapter 5, The OMERACT Handbook Version 2.1. Updated April 2021. https://omeract.org/handbook/

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.03 MB