Members of the Ad Hoc Committee on SLE Response Criteria for Fatigue are as follows: J. Antonio Aviña-Zubieta, MD, MSc: Arthritis Research Centre of Canada, Vancouver, British Columbia, Canada; Graciela S. Alarcón, MD: University of Alabama at Birmingham; Heike A. Bischoff-Ferrari, MD, MPH: Boston University School of Medicine, Boston, Massachusetts; Rebecca Fischer-Betz, MD, Matthias Schneider, MD: Heinrich-Heine-University, Düsseldorf, Germany; Victoria Gall, MD, Matthew H. Liang, MD, MPH, Peter Schur, MD: Brigham and Women's Hospital, Boston, Massachusetts; Gabor Illei, MD: NIH, NIDCR, Bethesda, Maryland; Jamal Mikdashi, MD: University of Maryland, College Park; Michelle Petri, MD, MPH: Johns Hopkins School of Medicine, Baltimore, Maryland; Charlotte Phillips, RN, MPH, CCRC: New England Baptist Hospital, Boston, Massachusetts; Jacques Pouchot, MD: European Georges Pompidou Hospital, Paris, France; William St.Clair, MD: Duke University Medical Center, Durham, North Carolina.
Measurement of fatigue in systemic lupus erythematosus: A systematic review
Article first published online: 29 NOV 2007
Copyright © 2007 by the American College of Rheumatology
Arthritis Care & Research
Volume 57, Issue 8, pages 1348–1357, 15 December 2007
How to Cite
Ad Hoc Committee on Systemic Lupus Erythematosus Response Criteria for Fatigue (2007), Measurement of fatigue in systemic lupus erythematosus: A systematic review. Arthritis & Rheumatism, 57: 1348–1357. doi: 10.1002/art.23113
- Issue published online: 29 NOV 2007
- Article first published online: 29 NOV 2007
- Manuscript Accepted: 7 JUN 2007
- Manuscript Received: 21 DEC 2006
- American College of Rheumatology
- Kirkland Scholar Award
- SLE Foundation of New York
- The Lupus Erythematodes Selbsthilfegemeinschaft e.V. Germany
- NIH. Grant Numbers: AR-47782, R13-AR-47584-01
- Robert B. Brigham Arthritis and Musculoskeletal Diseases Clinical Research Center (Harvard University)
- Heinrich-Heine-University in Düsseldorf
- Arthritis Research Centre of Canada (University of British Columbia)
- Massachusetts Veterans Epidemiology Research and Information Center
- Center for Advanced Methodological Support for Innovative SLE Clinical Trials
- CONACYT (Mexico)
- Canadian Arthritis Network
- Michael Smith Foundation for Health Research
- Canadian Institutes for Health Research
- Systemic lupus erythematosus;
- Fatigue Severity Scale;
- Systemic Lupus Activity Measure;
- Systemic Lupus Erythematosus Disease Activity Index
To systematically review fatigue instruments used in patients with systemic lupus erythematosus (SLE) and to seek consensus on which instruments should be recommended for future studies and which data on comorbid conditions should be reported when evaluating fatigue in patients with SLE.
We used data from Medline and EMBase databases (from January 1970 to June 2006), clinical experts, and bibliographies. Data were extracted independently by 4 authors and reviewed by a working group and larger expert panel to produce a recommendation. Instruments were examined for construct validity, reliability, and responsiveness to change. Correlations between fatigue and some components of the Medical Outcomes Study Short Form 36 (SF-36), disease activity, and comorbidities were reviewed.
We identified 34 studies that used 15 fatigue instruments in patients with SLE. The Fatigue Severity Scale (FSS) was used in 56% of the studies. The Systemic Lupus Activity Measure was significantly correlated with fatigue, but the Systemic Lupus Erythematosus Disease Activity Index was not. Fatigue was also correlated with pain, poor quality of sleep, depression, and with each subscale of the SF-36.
The working group and expert panel recommend the 9-item FSS for evaluating fatigue in SLE patients. Responsiveness to change of fatigue instruments has not been well established in SLE and needs further study. The panel suggested that an important improvement or response could be a 15% decrease in the FSS. The effect of several important confounding factors of fatigue such as sleep disorders, depression, fibromyalgia, and anemia needs to be collected and reported.
Systemic lupus erythematosus (SLE) can affect any organ, including skin, kidney, lung, brain, heart, and joints. Therefore, clinical manifestations are broad and variable in severity. In clinical management and research of patients with SLE, objective parameters are used to describe disease activity and treatment efficacy. In addition, there are some subjective parameters that are also important for the patients since these have an impact on their quality of life. Among these subjective parameters, fatigue is the most prevalent symptom in SLE, since it is present in up to 90% of the patients (1). Moreover, ∼50% of the patients consider fatigue the most disabling disease symptom (2). Despite its high prevalence and its impact on quality of life, fatigue has not been well studied in patients with SLE.
An important problem with fatigue is its definition and therefore its measurement. In the literature there are numerous ambiguous and inconsistent definitions for fatigue, which vary between research studies, including those evaluating fatigue in a single condition such as cancer-related fatigue (3). This is a major problem since some instruments only measure physical fatigue through limitation on daily living activities. Others also consider mental fatigue and its associated distress. Thus, level of fatigue and its impact on quality of life depends on the instrument used to measure it and the conceptualization of fatigue.
In SLE, the origin of fatigue is often multifactorial and could be mediated through comorbid conditions (depression, sleep disorder, fibromyalgia [FM]), behavioral factors (physical activity), and disease-related factors (disease activity, treatment). Among comorbid conditions, depression has been consistently demonstrated to be associated with fatigue in patients with SLE (2, 4–6). The coexistence of FM has been proposed as a contributor to persistent fatigue in patients with SLE, or at least as an influence to the perception of fatigue (7). Other studies have demonstrated that the association between fatigue and FM is related to the number of tender points, with higher fatigue scores seen in patients who satisfy the American College of Rheumatology criteria for FM (8) than in those who do not (4, 9–11).
Behavioral factors such as poor exercise participation have been associated with fatigue, and interventions in this area have demonstrated a positive impact on fatigue (12, 13). Finally, disease activity has been one of the most widely studied contributors of fatigue in SLE. However, the results are conflicting; some studies showed a significant correlation between disease activity and fatigue (4, 6, 14, 15) while others showed either no or weak associations (1, 5, 16).
There are several validated instruments to measure fatigue and many have been used in SLE patients (2,3,17–27). To date, there has been no consensus on which instrument should be used in the assessment of fatigue in SLE, and how to favorably affect it. Our working group and expert panel convened an Ad Hoc Committee on SLE Response Criteria for major organ involvement and key manifestations, to make recommendations for clinical trials in patients with SLE.
The aim of this ad hoc committee was to perform a systematic review of the literature on fatigue instruments used in SLE. Based on quality criteria, including construct validity, reliability, and responsiveness, the panel recommended instruments or a combination of instruments for use in future studies of patients with SLE, and also made recommendations on whether to develop new instruments. A consensus on which instrument or instruments should be used to measure fatigue in SLE would allow improved efficiency in future research by allowing comparisons across studies. The panel also identified important covariates that are strong correlates of fatigue in patients with SLE, and made recommendations on what factors should be measured and reported.
MATERIALS AND METHODS
We conducted a systematic review of all English language publications using the Medline and EMBase electronic databases from their inception (1966 and 1980, respectively) to July 2006. We included all intervention studies and observational studies in which a self-reported measurement of fatigue was applied to patients with SLE. In addition, clinical experts were contacted and bibliographies of existing publications were reviewed. MeSH terms (medical subject headings) included fatigue (fatigue/or fatigue syndrome, chronic/or mental fatigue/or muscle fatigue) and lupus (lupus/or lupus erythematosus, systemic/or lupus erythematosus, discoid/or lupus nephritis). Data abstraction was conducted by 4 investigators (JAA-Z, CP, HAB-F, and JP).
Analysis and consensus on scale validation.
The consensus process involved 3 steps. First, the data were extracted and summarized by 2 investigators (CP and HAB-F). Second, the working group and expert panel committee met at Heinrich-Heine-University, Düsseldorf, Germany from May 9–12, 2002. The working group discussed the collected data on fatigue instruments in SLE. Third, once the working group reached a consensus, their findings were presented to the larger expert panel. The final consensus is presented below.
The main psychometric properties (construct validity, reliability, and responsiveness to change) were evaluated using the original article describing the validity of the instrument. Psychometric properties were described as either present or absent. We sought to identify a single fatigue instrument that was optimal, based on these criteria. Studies that assessed fatigue in SLE were screened for additional comorbid factors that needed to be evaluated in the context of fatigue in SLE. These factors were sleep disorder, depression, FM, and anemia, as well as medications that may cause fatigue. Of further interest, we looked for the correlations between fatigue and some components of the Medical Outcomes Study Short Form 36 (SF-36), an extensively used measure of health-related quality of life (23). The SF-36 covers 8 domains of health status, including physical functioning, role physical, role emotional, social functioning, bodily pain, mental health, vitality, and general health. Scores in each domain range from 0–100, where higher scores reflect better performance or better quality of life. We calculated pooled summary estimates of correlation coefficients between the Fatigue Severity Scale (FSS) (2) and several domains of the SF-36 using the random effects model, and tested for heterogeneity using the bootstrap version of the Q statistic. All analyses were done using HePIMA software (28) with adjustment for correlations.
Ad hoc committee participants.
The Ad Hoc Committee consisted of clinicians and trial methodologists from the American College of Rheumatology, Systemic Lupus International Collaborating Clinics, The European League Against Rheumatism, Pan American League of Associations for Rheumatology, International League of Associations for Rheumatology, Food and Drug Administration, and Outcome Measures in Rheumatoid Arthritis Clinical Trials. Most of the authors of this article were members of the working group on fatigue in SLE.
We reviewed 15 fatigue instruments that were used in 34 studies with SLE patients (1, 2, 4–7, 9–17, 19, 21, 22, 26, 27, 29–43). Table 1 shows the distribution of all identified fatigue instruments applied in patients with SLE. The majority of the studies (19 [56%] of 34) used the FSS (2). One study used the Fatigue Assessment Instrument (FAI), which is an expanded version of the FSS and includes 29 items instead of the original 9 (26). In contrast with the FSS, the FAI is a multidimensional instrument that measures and differentiates normal fatigue from fatigue related to medical disorders, and clearly defines fatigue in order to promote consistency across respondents. In addition to being the most frequently used, the FSS was also the instrument with the highest number of citations among all fatigue instruments. The FSS was developed and tested on patients with SLE and differentiates patients from controls. Finally, the FSS is one of the few instruments where responsiveness has been evaluated in patients with SLE. These were some of the criteria that led the panel to recommend the FSS for future trials in SLE patients.
|Instrument (ref)||Comments||Rating/scale direction†||No. of SLE studies‡||No. of citations§|
|FSS (2)||Measures fatigue impact on specific types of functioning||1–7 per item/negative||19||474|
|ChFS (18)||Measures severity of fatigue in 2 dimensions (mental, physical)||0–4 per item (0=never, 1=rarely, 2=sometimes, 3=often, 4=always)/positive||4||343|
|MAC-FS (19)||Assesses stamina, energy, and tiredness||0–10 per item/positive||3||32|
|PFS (20)||Measures fatigue in 4 subscales (temporal, severity, affective, sensory)||0–100 per item/negative||1||72|
|SF20+1 (21)||SF-20 plus 1 added fatigue question||0–100/negative||1||82|
|FSES (22)||Assesses confidence in controlling fatigue||1–4 per item/positive||1||20|
|SF-36-V (23)||4-item measure of vitality (energy level and fatigue)||1–6 per item/positive for items 1, 2 and negative for items 3, 4||1||NA¶|
|MAF (24)||Measures 5 domains of degree, severity, distress, impact on activities of daily living, and timing||0–500/positive||2||77|
|MFI-20 (44)||20-item self-report that generates 5 dimensions of 4 items each (general, physical, reduced activity, reduced motivation, mental fatigue)||1–5 per item (from agreement to disagreement)/positive||1||287|
|Fatigue rating (15)||Self-report of fatigue after strenuous, regular, or mild activity||0–3/positive||1||25|
|VAS for fatigue (25)||4 VAS to measure different aspects of fatigue (mental, physical, exertional mental, exertional physical)||0–100 per item/positive||3||40|
|FAI (26)||Expanded version of the FSS evaluating 4 dimensions of fatigue (severity, situation-specificity, psychological consequences, response to rest/sleep)||1–7 per item/negative||1||110|
|VAS-fatigue (41)||100-mm line to measure severity of fatigue||0–100||6||1|
|SBPI (27)||3 dimensions (somatic fatigue, mental fatigue, and general discomfort) with several facets in each one. Designed to measure fatigue in Sjögren's syndrome||1–7 per item/positive||1||8|
|Single question for fatigue (10)||Presence or absence of fatigue daily in the last 3 months||Yes/no||1||11|
Table 2 provides a descriptive summary of all fatigue instruments used in patients with SLE and also shows some of the general psychometric properties for each instrument. Construct validity was not evaluated in all studies, especially those that used a single item to measure fatigue. The FSS, FAI, and Robert B. Brigham Multipurpose Arthritis Center-Fatigue Scale were the only instruments that have been evaluated for reliability and responsiveness in patients with SLE (Table 2).
|Instrument||No. items||Dimensions||Population used to develop/no.||Internal consistency Cronbach's α||Construct validity (correlations)†||Responsiveness|
|FSS||9||Fatigue||MS/25, SLE/29, healthy/20, lyme/6||0.89||VAS (0.68), CES-D (0.46)||Paired t-test, P < 0.01|
|ChFS||14||Physical fatigue (PhF) Mental fatigue (MF)||NR/274, CA/100||0.84 for PhF, 0.82 for MF||CIS-R, ROC analysis||NA|
|MAC-FS||4||Fatigue||SLE/23||0.89||SF-36, CES-D||Paired t-test, P < 0.05|
|PFS-R||22||Sensory, affective meaning, cognitive/mood, behavioral/severity||Breast cancer surviors/328||0.96||NA||NA|
|SF-20+1||1 for fatigue||Fatigue (1 item)||SLE/150||NA||SF-36 (0.60)||NA|
|FSES||5||Self-efficacy for controlling fatigue||SLE/58||0.75||FSS (−0.55), AIMS2 physical scale (−0.40), affect (−0.44), PPCS (0.07)||NA|
|SF-36-V||4 for vitality||Vitality (energy, fatigue)||General population/2,450||From 0.86 for vitality||Factor analysis SF-36 physical (0.59), SF-36 mental (0.57), general health (0.65), quality of life (0.45), NHP (−0.33, −0.58), SF20+1 (0.6)||NAV|
|MAF||14||Severity, distress, timing, interference||RA/133||0.93||POMS-F (0.78), POMS-V (0.60)||NA|
|Fatigue rating 1 item||Fatigue||SLE/83||NA||NA||NA|
|MFI-20||General fatigue, PhF, reduced activity, reduced motivation, MF||RP/111, CFS/357, PS/481, MD/158, AR/326||0.84||VAS-fatigue (0.23, 0.77)||Paired t-test, P = < 0.0001|
|VAS-fatigue||4||Fatigue (pervasive and postexertional PhF and MF)||FSS (0.6), ChFS (0.6)|
|FAI||29||Fatigue severity, fatigue situation-specificity, fatigue psychological consequences, fatigue response to rest/sleep||Lyme/35, CFS/46, SLE/47, MS/40, dysthymia/13, healthy/37||Factor 1 (0.92), Factor 2 (0.76), Factor 3 (0.70), Factor 4 (0.85)||DFA RVI (0.4, 0.7) ES (0.4)||NA|
|VAS-fatigue||1||Fatigue||SLE/127||NA||FSS, ChFS, HAD||NA|
|SBPI||8||Somatic fatigue, MF, general discomfort||PSS/18, RA/18, SLE/11||NA||SF-36 (−0.53, 0.70 among domains) HAD (0.36, 0.45) WHOQOL-BREF (−0.60, 0.69)||NA|
|Single question for fatigue||Fatigue||SLE/216||NA||NA||NA|
The 9-item FSS measures the impact of fatigue on specific types of functioning in the previous 2 weeks (Table 3). This instrument was originally developed with SLE patients, and a normative range has been established (an FSS score <4 is considered normal). The FSS has been translated from English (US) and adapted into several languages and populations, including English-speaking countries (UK, Australia, New Zealand, Canada), Spanish (Mexico, Spain), French (France, Canada), German (Germany, Switzerland), Chinese (Taiwan), and Portuguese (Brazil).
|1. My motivation is lower when I am fatigued.|
|2. Exercise brings on my fatigue.|
|3. I am easily fatigued.|
|4. Fatigue interferes with my physical functioning.|
|5. Fatigue causes frequent problems for me.|
|6. My fatigue prevents sustained physical functioning.|
|7. Fatigue interferes with carrying out certain duties and responsibilities.|
|8. Fatigue is among my three most disabling symptoms.|
|9. Fatigue interferes with my work, family, or social life.|
Table 4 shows the construct validity for most of the fatigue instruments that have been evaluated in SLE patients. The majority of studies included a validated disease activity index. The specific instruments for disease activity were the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) (45), Systemic Lupus Activity Measure (SLAM) (46), European Consensus Lupus Activity Measure (47), British Isles Lupus Assessment Group (48), and a Mexican version of the SLEDAI (49). The depression scales used were The Center for Epidemiological Studies Depression Scale (50), the Beck Depression Inventory (51), the Hospital Anxiety and Depression Scale (52), the Minnesota Multiphasic Personality Inventory (53), and the Pittsburgh Sleep Quality Index to evaluate sleep (54).
|Instrument (ref)||Disease activity (instrument)||Depression (instrument)||Pain (instrument)||Sleep (instrument)||Anemia|
|FSS (6)||NA||0.22 (BDI)‡||NA||NA||NA|
|FSS (14)||0.53 (SLAM)†||0.59 (CES-D)†||NA||NA||NA|
|FSS (4)||0.37 (SLAM)† 0.40 (ECLAM)†||0.49 (HAD)‡||NA||0.46 (PSQI)‡||NA|
|FSS (1)||0.27 (SLAM)‡||NA||0.35 (VAS-pain)||NA||NA|
|FSS (16)||0.26 (SLAM)† 0.22 (SLEDAI)||NA||NA||NA||NA|
|FSS (22)||NA||NA||0.54 (AIMS2-pain)†||NA||NA|
|FSS (5)||0.16 (SLEDAI)||0.28 (CES-D)†||NA||NA||NA|
|FSS (40)||0.3 (physician rating)||0.46 (CES-D)‡||NA||NA||NA|
|Fatigue rating (15)||0.49 (BILAG)‡||NA||NA||NA||NA|
|MAC-FS (19)||NA||0.14 (CES-D)||NA||NA||NA|
|SF-20+1 (21)||0.31 (BILAG)‡||NA||NA||NA||NA|
|SF-36 (24)||0.30||NA||0.74 (VAS-pain)†||−0.60 (SPI)||NA|
|Physical||0.26 (SLAM)†||0.47 (CES-D)‡||NA||0.45 (PSQI)‡|
|Mental||0.02 (SLAM)||0.52 (CES-D)‡|
|VAS (7)||0.13 (SLEDAI)||0.49 (BDI)‡||0.47‡||NA||NA|
Some studies evaluated the correlations between fatigue and comorbidities, including disease activity, depression, pain, sleep disorders, and anemia (Table 4). The SLAM (without fatigue) was significantly correlated with fatigue in all studies. On the contrary, the SLEDAI did not correlate with fatigue. All but 1 study identified a positive correlation between fatigue and depression. In addition, fatigue was also correlated with pain. Sleep disorder was assessed in only 15% of the studies, however all agreed that patients with poor quality of sleep have increased levels of fatigue (4, 6, 11, 34, 39). Moreover, SLE patients with FM had sleep fragmentation and poorer quality of sleep compared with patients without FM. In addition, SLE patients with FM experienced higher levels of fatigue (11).
The single and pooled correlations between the FSS and depression, helplessness, pain from SLE, and the SF-36 domains are shown in Figure 1. Fatigue was correlated with each domain in all studies. When testing for heterogeneity among the different pooled correlations, the depression, general health, mental functioning, and vitality subscales were statistically significant, suggesting that combining the studies would not be appropriate. However, the pattern and direction of the studies measuring the same outcomes were homogeneous, suggesting that the construct of the FSS is valid.
No studies evaluating response criteria or patients' perspectives on fatigue were identified. The panel proposed by consensus that an improvement or response of ≤15% decrease in the FSS should be considered clinically important, but further research in this area will be needed to test this recommendation.
Fatigue is usually clinically important, although it is a subjective and nonspecific characteristic in patients with SLE as well as those with other chronic diseases. The word fatigue originated based on the experience of healthy individuals, but interview-based studies have revealed that even though patients label their sensation as fatigue, they often find it qualitatively very different from the fatigue they experienced before they became sick (55). This suggests that fatigue may not always be accurately described on a simple continuum from no fatigue to severe fatigue.
Dittner et al described and evaluated 30 different scales and concluded that further validation is needed for all scales and no single scale is appropriate for measuring fatigue in all disease groups (56). A recent bibliographic study of fatigue instruments (57) identified 252 different ways to measure fatigue, of which 150 ways were used only once, and only 71 scales were specifically designed to measure fatigue.
Since there is little consensus on which scales possess the most attractive properties, it may be useful to know which scales are actually used to measure fatigue in patients with SLE. Our purpose was to evaluate the available literature on the topic in order to provide a recommendation for future use in patients with SLE.
The high prevalence and relative importance of fatigue in patients with SLE is one argument for the need for its systematic assessment. However, fatigue has not been assessed in a standardized way using a common instrument nor have cormobid conditions that may mediate fatigue been assessed. We identified 15 different instruments used to measure fatigue in SLE. The most commonly used and cited was the FSS. Moderate and consistent correlations of fatigue with the 8 subscales of the SF-36, including physical functioning, role physical, role emotional, social functioning, bodily pain, mental health, vitality, and general health, have been documented (1). The pattern and direction of the studies measuring these outcomes were homogeneous, suggesting that the construct of the FSS is valid and best evaluated in patients with SLE. In addition, several associations of fatigue with some comorbidities have also been reported, including disease activity, depression, pain, and sleep disorders. However, the relationship between fatigue and disease activity is controversial and research in this area is complicated by the fact that SLAM, 1 of the 2 commonly used disease activity measures, includes fatigue as a component score. None of the studies that used the SLEDAI found an association with fatigue. In contrast with the SLAM, the SLEDAI was designed to measure only inflammatory activity in patients with SLE, therefore items that may include patient perception (e.g., fatigue) are not included, and this could explain the observed differences. Nevertheless, when the fatigue component of the SLAM was removed, the association became weaker, suggesting that the relationship between fatigue and disease activity is neither strong nor linear (4).
The most common potential source of fatigue, sleep disorder, has been assessed in 15% of the SLE studies that measured fatigue. Medications that may cause fatigue were not assessed in any study, FM was assessed in only 5 (14%), and depression assessed in 9 (26%) of the studies. This is a limitation that the panel thought should be addressed in future studies. The panel believed that a future assessment of comorbid conditions associated with fatigue in patients with SLE was necessary to gain a complete understanding of the pathogenesis of fatigue in order to develop effective treatment strategies.
There are some limitations of this study. First, our current understanding of fatigue's construct in patients with SLE has not been well defined and, therefore, it is not clear which clinical, biologic, or social factors are the most important contributors to or predictors of fatigue. Second, although correlations among different fatigue instruments have been reported in SLE patients (4, 13, 34), not all instruments measure the same attribute, and direct comparison among instruments was not directly evaluated. Third, duration of fatigue is not measured uniformly across all fatigue instruments; therefore, memory bias could play a role in the scores of fatigue in the different instruments. Fourth, medications that may cause fatigue were not assessed in any study. The latter is a limitation that the panel felt important to be addressed in future studies. The panel agreed with Dittner et al that there is no gold standard for fatigue, nor is there ever likely to be (56).
The panel recommended the use of the FSS in clinical trials and observational studies in SLE patients. The arguments of the panel that led to this recommendation were that the instrument was developed in patients with SLE, it is frequently used in SLE studies, and it has valid psychometric properties (including internal consistency, responsiveness, and construct validity. No other instrument met all these criteria. Furthermore, its availability in several languages was also considered an advantage suggesting worldwide acceptance from the scientific community. The panel further emphasized that the use of a single common instrument will allow improved efficiency in future research on fatigue in patients with SLE by enabling results to be compared across studies and interventions.
Gaps identified by the panel include a closer examination of several dimensions of fatigue in patients with SLE (severity, impact, trends, affective, and evaluative components), and the identification of factors other than depression, disease activity, sleep quality, and FM. Other important variables such as emotional stressors, environmental factors, and effect of age, sex, culture, and disease duration on perception of fatigue need to be addressed. Furthermore, patients' perspectives on fatigue are critical and need to be evaluated, especially the ones related to what they consider the minimal clinically important difference (MCID; the smallest difference in fatigue score that patients perceive as beneficial and that would mandate a change in the patient's management). The MCID will be essential to interpret the magnitude of longitudinal changes or differences when comparing therapeutic strategies. This will also be necessary to estimate sample sizes for future trials using fatigue as an outcome measure.
The panel felt that a recommendation based on experts' opinion was needed in the absence of evidence; we considered that this was better than nothing at all. Therefore, we recommended that an important improvement or response could be a 15% decrease in the FSS score. It can be argued that expert opinion is not a valid approach; however, this strategy has been used as a proxy in the absence of available information (e.g., to calculate sample sizes). Furthermore, a recent study from several authors participating in this study evaluated the MCID from the patients' perspective for several fatigue instruments including the FSS, and found that a 10% response (95% confidence interval 4.9–14.6) will be considered important (58).
In summary, we suggest that in future studies, the FSS is the best available instrument to measure fatigue in patients with SLE based on its validated psychometric properties, the fact that it is the most commonly used instrument to date, and it has been used and validated in several languages. Future research areas of interest are longitudinal studies of fatigue in early lupus, identification of clinical and psychosocial factors as determinants of fatigue in diverse patient populations, and the estimation of the MCID for fatigue from the patient's perspective.
Dr. Aviña-Zubieta 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 design. Bischoff-Ferrari, Fischer-Betz, Liang, Phillips, Schneider, Schur, St.Clair.
Acquisition of data. Aviña-Zubieta, Bischoff-Ferrari, Illei, Liang, Phillips, Pouchot, St.Clair.
Analysis and interpretation of data. Aviña-Zubieta, Alarcón, Bischoff-Ferrari, Fischer-Betz, Illei, Liang, Mikdashi, Petri, Phillips, Pouchot, Schneider, Schur, St.Clair.
Manuscript preparation. Aviña-Zubieta, Alarcón, Bischoff-Ferrari, Fischer-Betz, Gall, Liang, Mikdashi, Petri, Phillips, Pouchot, Schneider, Schur, St.Clair.
Statistical analysis. Aviña-Zubieta, Bischoff-Ferrari, Liang.
- 23SF-36 health survey: manual and interpretation guide. Boston: The Health Institute, New England Medical Center; 1983., , , .
- 47European Consensus Study Group for Disease Activity in SLE. Disease activity in systemic lupus erythematosus: report of the Consensus Study Group of the European Workshop for Rheumatology Research. II. Identification of the variables indicative of disease activity and their use in the development of an activity score. Clin Exp Rheumatol 1992; 10: 541–7., , , , , , et al, and The
- 51Beck Depression Inventory manual. San Antonio: The Psychological Corporation; 1987., .
- 53The MMPI-2/MMPI: an interpretative manual. Boston: Allyn and Bacon; 1991..
- 58The minimal clinical important difference for seven measures of fatigue in patients with systemic lupus erythematosus. 2007. (Submitted), , , , , , et al.