Original article
Comparison of the responsiveness of lupus disease activity measures to changes in systemic lupus erythematosus activity relevant to patients and physicians

https://doi.org/10.1016/S0895-4356(01)00509-1Get rights and content

Abstract

Both the revised Systemic Lupus Activity Measure (SLAM-R) and the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) are valid and reliable measures of disease activity in systemic lupus erythematosus (SLE). However, more study of their responsiveness is needed. The purpose of this study was to compare the responsiveness of SLAM-R and SLEDAI to disease activity changes relevant to physicians and patients. Patients were evaluated monthly for up to 12 months. At each visit, the physician completed SLAM-R and SLEDAI. Patients and physicians assessed whether relevant improvement or worsening of disease activity had occurred since the previous visit. Based on repeated measurements, effect size (ES), standardized response mean (SRM), and control-standardized response mean (CSRM) were calculated for each response category, with bootstrap-based 95% confidence intervals (CIs). Seventy-six patients contributed 471 score changes. For physicians' responses, the CSRMs for SLAM-R and SLEDAI were −0.47 versus −0.42 for improvement, 0.04 versus 0.003 for no change, and 0.65 versus 0.66 for deterioration. For patients, the CSRMs for SLAM-R and SLEDAI were −0.31 versus −0.18 for improvement, −0.08 versus 0.06 for no change, and 0.48 versus 0.05 for deterioration. Only for SLAM-R did the 95% CIs exclude zero when improvement or deterioration were detected. Similar results were found for ES and SRM. Both SLAM-R and SLEDAI are responsive to changes in SLE disease activity important to physicians. Only SLAM-R is responsive to changes important to patients. These differences may result from the inclusion of subjective SLE manifestations in SLAM-R.

Introduction

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized both by unpredictable flares and remissions and by the great variety of manifestations that may be observed both in different patients and in the same patients over time. Although it can affect people of all ages and both sexes, it occurs primarily in women of childbearing age [1]. Improved 5- and 10-year survival rates [2] have led to increased interest in studying outcomes other than death, such as disease activity. Over 60 different measures of SLE activity with varying psychometric properties exist [3]. Two commonly used measures in North America are the revised Systemic Lupus Activity Measure (SLAM-R) [4] and the SLE Disease Activity Index (SLEDAI) [5]. Both are valid and reliable 4, 5, 6, 7, but studies on their responsiveness to changes in SLE activity have produced conflicting results. Gladman et al. [8] and Brunner et al. [9] found SLEDAI to be more sensitive to change than the original SLAM, an instrument that differs only slightly from SLAM-R, whereas Ward et al. [7] and Fortin et al. [10] found SLAM-R to be more responsive than SLEDAI. This inconsistency may have been due in part to differences in analytical approaches. Moreover, because the sampling distributions of the responsiveness measures used are unknown, the studies were not able to assess to what extent the observed differences between the responsiveness of the two instruments might have been due simply to chance.

Few studies on SLAM-R and SLEDAI have evaluated their responsiveness to changes in disease activity reported by patients. There is evidence that patients and physicians do not always agree on assessments of disease activity at a single point in time [11], and it is possible that they also differ in their judgment of the importance of changes in activity. Because objectively scored instruments may not reflect patient assessments in the same manner that they reflect physician perceptions, it is important to evaluate their responsiveness to patient-reported changes in disease activity separately.

A major problem for many SLE studies is that the rareness of the disease makes recruitment of large numbers of patients unfeasible, resulting in low statistical power and limited precision of the estimates. We have overcome this problem by incorporating repeated measures of patients in our analysis while accounting for intrapatient dependence of observations.

Our objectives in this study are to evaluate the responsiveness of SLAM-R and SLEDAI in terms of their sensitivity to change relevant to patients and/or physicians and to assess the precision of these estimates to enable assessment of the statistical significance of the differences between the responsiveness of the two instruments.

Section snippets

Design of the Study of Methotrexate in Lupus Erythematosus

A secondary analysis was performed on data obtained from the Study of Methotrexate in Lupus Erythematosus (SMILE), a Canadian multi-centre, randomized controlled trial (RCT) involving SLE patients. Approval for the RCT was granted by the research ethics boards of all participating centers, and informed consent was given by enrolled patients. The participants were blindly assigned to one of two treatment arms for 1 year and followed up at monthly visits. Because of the inclusion criteria, all

Study population

Eighty-six patients enrolled in SMILE at the time the data were obtained contributed 761 postbaseline visit pairs during the blinded phase of the study. Visit pairs were omitted from the analyses if the physician or patient transition score for the second visit in the pair was missing (n = 58) or if SLAM-R and SLEDAI scores at the first visit were missing, either because the visit had happened at home (n = 27) or because none of the items had been scored (n = 3). For our analysis of

Discussion

We evaluated the ability of SLAM-R and SLEDAI to detect changes in disease activity relevant to patients and physicians. Responsiveness to change was measured by effect size, standardized response mean, and control-standardized response mean. We found that SLAM-R and SLEDAI were equally responsive to changes relevant to physicians. However, SLAM-R was systematically more responsive than SLEDAI to both improvement and deterioration relevant to patients, a finding consistent with that of Ward et

Conclusion

We found that both SLAM-R and SLEDAI were responsive to physician assessment of relevant change in SLE activity. However, both instruments, particularly SLEDAI, tended to be less responsive to changes in disease activity important to patients. Although either SLAM-R or SLEDAI could be used to reflect physician perception of change in disease activity in SLE trials, SLAM-R is better able to reflect patient-reported changes. This may be due in part to the inclusion in SLAM-R but not SLEDAI of a

Acknowledgements

The authors would like to thank Roxane du Berger and Marielle Olivier for their assistance with SAS. This study was supported in part by grants from the Canadian Arthritis Network, The Arthritis Society (Grant #95072), and the Natural Sciences and Engineering Research Council (NSERC) of Canada (Grant #105521-98). Dr. Abrahamowicz is a Health Scientist of the Canadian Institutes of Health Research (CIHR). Dr. Fortin is a Senior Research Scholar of The Arthritis Society and Director of Clinical

References (21)

  • M.H Liang et al.

    Systemic lupus erythematosus activityan operational definition

    Rheum Dis Clin North Am

    (1988)
  • G Guyatt et al.

    Measuring change over timeassessing the usefulness of evaluative instruments

    J Chronic Dis

    (1987)
  • M Hochberg

    The epidemiology of systemic lupus erythematosus

  • J.O Schroeder et al.

    Recognition and management of systemic lupus erythematosus

    Drugs

    (1997)
  • M.H Liang et al.

    Reliability and validity of six systems for the clinical assessment of disease activity in systemic lupus erythematosus

    Arthritis Rheum

    (1989)
  • C Bombardier et al.

    Derivation of the SLEDAIa disease activity index for lupus patients. The Committee on Prognosis Studies in SLE

    Arthritis Rheum

    (1992)
  • M Petri et al.

    Validity and reliability of lupus activity measures in the routine clinic setting

    J Rheumatol

    (1992)
  • M.M Ward et al.

    Comparison of the validity and sensitivity to change of 5 activity indices in systemic lupus erythematosus

    J Rheumatol

    (2000)
  • D.D Gladman et al.

    Sensitivity to change of 3 Systemic Lupus Erythematosus Disease Activity Indicesinternational validation

    J Rheumatol

    (1994)
  • H.I Brunner et al.

    Sensitivity of the Systemic Lupus Erythematosus Disease Activity Index, British Isles Lupus Assessment Group Index, and Systemic Lupus Activity Measure in the evaluation of clinical change in childhood-onset systemic lupus erythematosus

    Arthritis Rheum

    (1999)
There are more references available in the full text version of this article.

Cited by (42)

  • Clinician-reported outcome measures in lupus trials: a problem worth solving

    2021, The Lancet Rheumatology
    Citation Excerpt :

    A single global score has the benefit of providing a summary measure of disease activity, but also has the drawback that different disease activity patterns can result in the same total score, and concomitant improvement and worsening in different organ systems might not be reflected. Importantly, in relation to trials, dichotomous scoring means that the SLEDAI does not fully account for variations in the severity of manifestations, which introduces floor and ceiling effects and reduces sensitivity to change when used to assess treatment response.29–31 In one study of 76 patients,29 more than half (54%) had improved as assessed by the physician but did not have a corresponding improvement in SLEDAI, most likely because a near-complete resolution of a manifestation is needed to be reflected as an improvement in the SLEDAI score.

  • Performance of an item response theory-based computer adaptive test in identifying functional decline

    2012, Archives of Physical Medicine and Rehabilitation
    Citation Excerpt :

    Additionally, our analytic approach using GEEs offers a means of adjusting for the inevitably correlated nature of data collected through longitudinal assessments. Prior investigators have generally handled this correlation by reporting separate responsiveness statistics for each assessment interval (eg, 0–3 and 0–6mo),43,47 or by using bootstrap or jackknife techniques.30,48 While GEEs are methodologically similar to the jackknife approach, they permit inclusion of unlimited assessment intervals, afford greater flexibility in specifying the correlation between repeated measures, and are more computationally efficient.

  • Two systemic lupus erythematosus (SLE) global disease activity indexes-the SLE Disease Activity Index and the Systemic Lupus Activity Measure-demonstrate different correlations with activation of peripheral blood CD4 <sup>+</sup> T cells

    2011, Human Immunology
    Citation Excerpt :

    Most published works discuss the problem from the clinical viewpoint and ask questions about assessing the patient's clinical status. According to those papers, the SLAM scale seems to be more sensitive than the SLEDAI scale [2,20–22]. The main reason for those observations could be associated with the fact that the SLAM accounts for the severity gradient for organ involvement, whereas the SLEDAI measures only the presence versus the absence of organ involvement [2,22].

View all citing articles on Scopus
View full text