Sociodemographic, disease, and symptom correlates of fatigue in systemic sclerosis: Evidence from a sample of 659 Canadian Scleroderma Research Group Registry patients




To assess fatigue levels and demographic, socioeconomic, disease, and psychosocial correlates of fatigue in patients with systemic sclerosis (SSc).


We conducted a cross-sectional, multicenter study of 659 patients with SSc from the Canadian Scleroderma Research Group Registry. Fatigue was assessed during annual Registry visits with the Short Form 36 (SF-36) health survey vitality subscale. Patients completed measures of depressive symptoms and pain and underwent clinical histories and medical examinations. Kendall's tau was used to assess bivariate association of sociodemographic, medical, and psychosocial variables with fatigue. Multivariable associations of demographic (step 1), socioeconomic (step 2), global disease (step 3), specific disease and lifestyle (step 4), and psychosocial (step 5) factors with fatigue were assessed using hierarchical multiple linear regression.


The mean ± SD score of the patients on the SF-36 vitality subscale was 45.6 ± 10.8, substantially lower (indicating more fatigue) than the mean ± SD score for the Canadian general population (65.8 ± 18.0). In multivariate analysis, higher fatigue was significantly associated with the number of medical comorbidities (standardized β = −0.11, P = 0.004), breathing problems (standardized β = −0.23, P < 0.001), the number of gastrointestinal (GI) symptoms (standardized β = −0.27, P < 0.001), and current smoking (standardized β = −0.08, P = 0.018). As a group, specific symptom and lifestyle variables predicted the most incremental variance in fatigue (▵R2 = 21.6%, P < 0.001), despite being added to the model after demographic, socioeconomic, and global disease duration/severity indicators. Symptoms of depression (β = −0.42) and pain (β = −0.21) were also independently associated with fatigue (P < 0.001).


High levels of fatigue are common in patients with SSc and are independently associated with clinical variables, including number of comorbidities, breathing problems, GI symptoms, and smoking.


Persistent fatigue from chronic illness involves ongoing exhaustion that is disproportionate to exertion and not alleviated by rest (1). Although patients with chronic disease rate fatigue as one of the most important factors impacting quality of life (QOL) (1), it is often overlooked by clinicians and researchers (1). In one study, almost 90% of rheumatologists reported that they never assess fatigue (2). Across disease groups, fatigue decreases QOL by diminishing a person's ability to engage in meaningful personal and social activities and has important implications for employment, compliance with prescribed medical treatments, and use of health care services (1, 3).

Systemic sclerosis (SSc; scleroderma) is a chronic, multisystem disorder of connective tissue characterized by thickening and fibrosis of the skin, involvement of internal organs, substantially reduced QOL, and significant morbidity and mortality (4). Common SSc symptoms associated with reduced QOL include pain, gastrointestinal (GI) symptoms, joint deformity, limited mobility, dyspnea, and sleep difficulties (5, 6). Few studies have addressed the impact of fatigue in SSc. One study (n = 49) found the most frequent symptoms of SSc to be stiff joints (79%), pain (75%), and fatigue (75%) (7), and another study reported that pain and fatigue were mentioned more than any other symptoms by focus group participants (8). A third study (n = 123) reported that patients rated fatigue as more bothersome than any other symptom (9). These studies, however, all used single-item or counting methods with unknown measurement characteristics. A recent study compared fatigue measured by the Multidimensional Fatigue Inventory in 106 patients with SSc with fatigue measured in other chronic disease samples identified through systematic review, and found that the level of fatigue in patients with SSc was significantly higher than general population levels, and was comparable with levels reported by patients with other rheumatic diseases and by patients with cancer in active treatment (10).

The pathophysiology of fatigue related to chronic illness is not well understood, although numerous possible factors have been identified, including anemia and malnutrition, interactions of cytokines and serotonin, nausea and other GI problems, dyspnea, sleep disturbance, inactivity and deconditioning, lifestyle factors such as smoking, excessive alcohol consumption, and exercise, and psychological factors such as depression, anxiety, and specific cognitions (e.g., catastrophizing) (11). Wagner and Cella (11) classified causes of fatigue in cancer as direct effects of the cancer, side effects of the treatments (e.g., chemotherapy), comorbid medical conditions (e.g., malnutrition), exacerbating comorbid symptoms (e.g., chronic pain, deconditioning), and psychosocial factors (e.g., depression). There is currently no published research on etiologic factors of fatigue in patients with SSc, but possible causes may similarly include 1) direct effects of the SSc, including skin tightening limiting movement and chest expansion, interstitial lung disease and pulmonary hypertension leading to dyspnea, GI symptoms such as chronic diarrhea, arthralgias/arthritis impairing mobility, and inflammatory muscle disease causing weakness; 2) comorbid medical conditions, including anemia, malnutrition, and other comorbidities (e.g., unrelated heart, lung, and thyroid disease); 3) exacerbating comorbid symptoms, such as chronic pain, sleep disturbances, and deconditioning; 4) treatment side effects of immunosuppressive drugs used to treat skin and lung disease (e.g. methotrexate or cyclophosphamide associated with nausea, diarrhea, and malaise) and other symptomatic treatments (e.g., antihypertensives causing hypotension); 5) lifestyle factors such as exercise, smoking, and excessive alcohol consumption; and 6) psychosocial factors, including overextended coping resources and stress, depression, anxiety, and specific fatigue-related cognitions (e.g., catastrophizing) (8). Possible etiologic factors for fatigue in patients with SSc are shown in Figure 1.

Figure 1.

Hypothesized etiologic factors for fatigue in systemic sclerosis (SSc; scleroderma). IL-1β = interleukin-1β; IL-6 = interleukin-6; INFγ = interferon-γ; TNF α = tumor necrosis factor α.

The objective of this study was to assess fatigue levels in a large sample of patients with SSc and to identify important demographic, socioeconomic, disease, lifestyle, and patient outcome correlates (depressive symptoms, pain).


Patient sample.

The sample consisted of patients enrolled in the Canadian Scleroderma Research Group (CSRG) Registry from September 2004 through June 2008 for whom there were complete data on study measures. Patients in this Registry were recruited from 15 centers across Canada. To be eligible for the Registry, patients must have a diagnosis of SSc made by a Registry rheumatologist, be age ≥18 years, and be fluent in English or French. There are no additional inclusion or exclusion criteria. Registry patients undergo extensive clinical history, physical evaluation, and laboratory investigations, and they complete a series of self-report questionnaires that includes sociodemographic variables, lifestyle variables (e.g., smoking history), other health problems, environmental exposures, family history of autoimmune diseases, SSc symptoms, disability, QOL, pain, symptoms of depression, and medical resource utilization. Patients from all sites provided informed consent, and the research ethics board of each study site approved the data collection protocol.


Fatigue was measured using the vitality subscale of the Short Form 36 (SF-36) health survey (12). Analyses of possible predictors of fatigue included self-reported sociodemographic data (age, sex, education, and marital status) and variables measuring direct effects of SSc (duration, global severity, number of tender joints, number of GI symptoms, skin involvement, dyspnea, and inflammatory muscle disease), medical comorbidities, exacerbating comorbid symptoms (pain, deconditioning/capacity for exercise), lifestyle factors (smoking, alcohol consumption), and psychosocial variables (symptoms of depression).


The SF-36 vitality subscale (12) includes 4 Likert items with 5 response options each (with anchors at all of the time and none of the time) that assess a patient's level of fatigue during the previous 4 weeks. The SF-36 vitality subscale has been used to measure fatigue in general population samples and in patients with medical illness and injury. A recent systematic review concluded that the SF-36 vitality subscale has good evidence for validity, reliability, sensitivity to change, and feasibility in rheumatoid arthritis (RA) (13). Cronbach's α = 0.84 in the current sample of patients with SSc.

Disease-related variables.

SSc disease duration was established as the time from onset of the first non-Raynaud's disease symptoms based on a clinical history obtained by the study physicians. SSc global disease severity was rated by the study physicians on a 0–10 numerical rating scale, which has been shown to be a valid measure of severity in SSc (14). Limited skin disease was defined as skin involvement distal to the elbows and knees with or without face involvement. Skin involvement was also assessed using the modified Rodnan skin thickness score, range 0–51 (15). Tender joint count was recorded by the study physicians using a 28-joint count (16). Shortness of breath was rated by the patient on a 0–10 numerical rating scale (17). The number of GI symptoms was determined by patient report from a checklist that included weight loss, anorexia, dysphagia, reflux, pyrexia, choking at night, early satiety, bloating, nausea/vomiting, constipation, diarrhea, malabsorption, fecal incontinence, antibiotics for bacterial overgrowth, and hyperalimentation. Patients were considered to have inflammatory muscle disease if physicians indicated the presence of inflammatory myositis, polymyositis, or dermatomyositis.

The number of medical comorbidities was based on a patient self-report version of the medical record–based Charlson comorbidity index (18). The Charlson comorbidity index includes myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, hemiplegia, chronic obstructive pulmonary disease, ulcer disease, diabetes mellitus, kidney problems, eye problems, renal problems, RA, Alzheimer's disease or dementia, cirrhosis or liver damage, leukemia, lymphoma, skin cancer, other cancer, metastatic tumor, and acquired immunodeficiency syndrome (AIDS). The patient-reported comorbidities questionnaire has been found to have good agreement with the medical record–based Charlson comorbidity index (18).

The New York Heart Association (NYHA) classification was used as a proxy variable for exercise capacity or deconditioning. The NYHA is a subjective, physician-rated measure of physical activity limitation commonly used in patients with heart failure to assess disease severity related to exercise tolerance. Patients are rated in 4 categories: I, no symptoms or limitation; II, mild symptoms and slight limitation during ordinary activity, comfortable at rest; III, marked limitation in activity, comfortable only at rest; and IV, severe limitations with symptoms even at rest.

The data collected on medications included calcium channel blockers, promotility agents, corticosteroids, D-penicillamine, methotrexate, angiotensin-converting enzyme inhibitors, antiplatelet agents, warfarin, nonsteroidal antiinflammatory drugs, narcotics, hormone replacement therapy, thyroid supplements, and gastroprotective agents. Anticentromere antibodies were detected by screening indirect immunofluorescence on HEp-2 cells and confirmed by a line assay (InnoLIA; Innogenetics, Gent, Belgium), and antitopoisomerase antibodies were detected by an addressable laser bead immunoassay (QuantaPlex 8, Inova, San Diego, CA) in a central laboratory (Dr. M. Fritzler, Advanced Diagnostics Laboratory, Calgary, Alberta, Canada).

Current smoking was assessed by patient report. Alcohol consumption was assessed with a single patient-reported item, “How many alcoholic drinks do you have per week?” Weight loss in kilograms in the last year was assessed by patient report.

Symptoms of depression.

The Center for Epidemiological Studies Depression Scale (CES-D) (19) is a 20-item measure designed to assess the presence and severity of depressive symptomatology. The frequency of occurrence of each symptom during the past week is rated on a 0–3 Likert-type scale (anchored at rarely or none of the time and most or all of the time), and total scores range 0–60. Standard cutoffs are ≥16 for possible depression and ≥23 for probable depression (19).


The Scleroderma Health Assessment Questionnaire (SHAQ) consists of the HAQ disability index and visual analog scales (VAS) to measure the severity of symptoms specific for SSc, including pain in the past week (17). Unlike the VAS that was originally used for the SHAQ, the pain assessments in this study were made using 11-point numerical rating scales (range 0–10). The numerical rating scale has been found to have excellent psychometric properties and to perform similarly to the VAS for pain (20).

Statistical analyses.

Kendall's tau correlations were used to assess the bivariate association between sociodemographic, medical, and psychosocial variables with fatigue (SF-36 vitality subscale scores). Multivariable associations between demographic (step 1), socioeconomic (step 2), global disease (step 3), specific disease and lifestyle factors (step 4), and depression and pain (step 5) with the SF-36 vitality scale were assessed using a hierarchical multiple linear regression. Step 1 of the regression consisted of age and sex. Marital status (married or living as married) and education (more than high school) were added during step 2. Disease duration and physician-rated global severity were entered during step 3. Diffuse/limited status, number of tender joints, number of GI symptoms, breathing problems, inflammatory muscle disease, medical comorbidities, exercise capacity, and current smoking status were entered during step 4. Depression and pain were entered during step 5. However, there are bidirectional pathways between symptoms of fatigue, depression, and pain, and it is not clear to what degree fatigue would precede or be influenced by depression and pain (21, 22). To avoid overinterpreting a possibly misspecified model, step 5 was considered exploratory (23). All analyses were conducted using SPSS, version 15.0 (SPSS, Chicago, IL). All statistical tests were 2-sided with P values less than 0.05 considered significant.


Sample characteristics.

A total of 659 patients were included in the study. Patient demographics and medical and symptom outcome variables are shown in Table 1. The mean age of the sample was 55.2 years, 87% of the sample was female, and 91% of the sample was white. The mean time since onset of the first non-Raynaud's symptoms was 10.5 years, and the mean time since diagnosis of SSc was 8.2 years. Patient comorbidities are shown in Table 2. Sample characteristics were similar to those reported from other large North American and European cohorts (4). The mean ± SD score on the vitality subscale of the SF-36 was 45.6 ± 10.8, which is consistent with other studies of patients with SSc (range 39–50) (24–28), and substantially lower (indicating more fatigue) than the normal mean ± SD score for the general Canadian population (65.8 ± 18.0) (29).

Table 1. Sociodemographic variables, medical variables, and correlations with fatigue as measured by the SF-36 vitality subscale (n = 659)*
VariablesValueKendall's tau coefficientP
  • *

    Values are the mean ± SD unless otherwise indicated. SF-36 = Short Form 36 health survey; SSc = systemic sclerosis; GI = gastrointestinal; NYHA = New York Heart Association; CES-D = Center for Epidemiological Studies Depression Scale.

  • n = 635.

  • n = 647.

  • §

    Lower numbers indicate worse fatigue.

 Age, years55.2 ± 12.30.0270.324
 Female sex, no. (%)571 (86.6)−0.0450.177
 Race/ethnicity white, no. (%)635 (90.6)0.0010.988
Socioeconomic, no. (%)   
 Education level less than high school310 (47.0)0.0330.312
 Married or living as married465 (70.6)0.0370.259
 Time since onset of first non-Raynaud's symptoms, years10.5 ± 8.70.0250.346
 Time since diagnosis of SSc, years8.2 ± 7.70.0160.557
 Physican-rated global disease severity (1–10)2.8 ± 2.5−0.148< 0.001
 Diffuse systemic sclerosis, no. (%)267 (40.5)−0.0550.093
 Modified Rodnan total skin score (0–51)10.8 ± 9.8−0.0490.083
 Comorbidity score0.8 ± 1.1−0.156< 0.001
 Tender joint count (0–28)1.7 ± 4.2−0.110< 0.001
 Inflammatory muscle disease, no. (%)78 (11.8)−0.0740.025
 Breathing problems (0–10)2.0 ± 2.6−0.305< 0.001
 Number of GI symptoms (0–14)4.0 ± 2.9−0.319< 0.001
 NYHA III or IV, no. (%)61 (9.2)−0.198< 0.001
 Current smoker, no. (%)105 (15.9)−0.1050.001
 Alcoholic drinks per week, no. (%) 0.0470.114
  <1305 (47.1)  
  1–7280 (43.3)  
  8–1447 (7.3)  
  >1415 (2.3)  
 Pain numerical rating scale (0–10)3.6 ± 2.8−0.361< 0.001
 Depressive symptoms (CES-D) (0–60)14.2 ± 10.5−0.454< 0.001
 Fatigue (SF-36 vitality subscale) (0–100)§45.6 ± 10.8
Table 2. Patient-reported medical comorbidities (n = 659)*
ComorbidityNo. (%)
  • *

    CHF = congestive heart failure; COPD = chronic obstructive pulmonary disease; RA = rheumatoid arthritis; AIDS = acquired immunodeficiency syndrome.

Myocardial infarction29 (3.8)
CHF39 (5.9)
Peripheral vascular disease17 (2.6)
Cerebrovascular disease26 (3.9)
Hemiplegia4 (0.6)
COPD95 (14.4)
Ulcer disease75 (11.4)
Diabetes mellitus40 (6.1)
Kidney problems3 (0.5)
Eye problems3 (0.5)
Renal problems30 (4.6)
RA109 (16.5)
Alzheimer's disease/dementia2 (0.3)
Cirrhosis/liver damage11 (1.7)
Leukemia1 (0.2)
Lymphoma1 (0.2)
Skin cancer14 (2.1)
Cancer, other12 (1.8)
Metastatic tumor2 (0.3)
AIDS0 (0.0)

The patients' mean ± SD score on the CES-D was 14.2 ± 10.5; 241 (36.6%) patients scored ≥16, and 135 patients scored ≥23, which exceeds rates typically reported for patients with other chronic conditions using the CES-D with the same scoring cutoffs (6). The patients' mean ± SD pain rating was 3.61 ± 2.76, which is equivalent to 1.20 ± 0.92 on a 0–3 scale. This is similar to pain ratings (on a 0–3 scale) reported in a previous study in 43 patients with SSc (mean ± SD 1.37 ± 0.90), is slightly higher than ratings from 42 patients with RA (mean ± SD 1.01 ± 0.73), and is substantially higher than ratings from 60 healthy controls (mean ± SD 0.27 ± 0.57) (26).

Predictors of fatigue.

Significant bivariate predictors of fatigue included global disease severity, number of comorbidities, tender joint count, inflammatory muscle disease, dyspnea, number of GI symptoms, NYHA classification III or IV, current smoking, pain, and depressive symptoms (Table 1). Results from the hierarchical multiple linear regression for sociodemographic and medical predictors are shown in Table 3. Age, sex, marital status, and level of education were not significant predictors of fatigue, nor were the combination of the variables in step 1 (P = 0.347) and step 2 (P = 0.340). Demographic variables accounted for 0.3% of the variance, and socioeconomic variables accounted for another 0.4% of the variance. In step 3, only physician-rated disease severity (P < 0.001) was a significant predictor of fatigue. Global disease duration and severity factors accounted for an additional 4.1% of the variance. Specific disease and lifestyle factors added in step 4 accounted for the largest proportion of the variance (an additional 21.6%). Of the specific disease and lifestyle factors, breathing problems (P < 0.001), the number of GI symptoms (P < 0.001), comorbid health problems (P = 0.004), and smoking (P = 0.018) significantly predicted fatigue. When specific disease factors were included in the equation, physician-rated global disease severity was no longer significant (P = 0.183). Both symptoms of depression (standardized regression coefficient [β] = −0.42) and pain (β = −0.21) were significantly related to fatigue (P < 0.001) when added in step 5 (R2 = 45.2%; not shown). However, the magnitude of the regression coefficients for symptoms of depression and pain, which were approximately 3 times and 2 times greater than the next largest predictor (i.e., number of GI symptoms), suggests that these were not appropriately specified as precursors to fatigue, but rather likely reflected bidirectional relationships or shared method influences. The relationship of breathing problems, GI symptoms, and comorbid health problems with fatigue did not change substantively in step 5. Medications, antibodies, weight loss in the past year, and alcohol consumption were not independently associated with fatigue.

Table 3. Hierarchical linear regression of the relationship between demographic variables, socioeconomic variables, global disease duration/severity, specific disease factors, and fatigue as measured by the SF-36 vitality subscale*
VariablesIndividual variable parametersOverall model fit statistics
BSE of BβPdfR2Adjusted R2ΔR2P
  • *

    B = raw regression coefficients; β = standardized regression coefficients; ΔR2= variance accounted for. See Table 1 for additional definitions.

Step 1: demographic variables    2, 6560.0030.0000.0030.347
 Female sex−1.6721.234−0.0530.179     
Step 2: socioeconomic variables    4, 6540.0070.0010.0040.340
 Female sex−1.6911.244−0.0530.174     
 More than high school education0.7100.8650.0330.412     
 Married or living as married1.2250.9250.0520.186     
Step 3: global disease duration/severity    6, 6520.0480.0390.041< 0.001
 Female sex−0.9681.231−0.0300.432     
 More than high school education0.4370.8500.0200.607     
 Married or living as married0.5850.9160.0250.523     
 Time since onset of first non-Raynaud's  symptoms−0.0410.049−0.0330.406     
 Physician-rated disease severity (0–10)−1.0000.189−0.208< 0.001     
Step 4: specific disease and lifestyle factors    14, 6440.2640.2480.216< 0.001
 Female sex−0.5201.109−0.0160.639     
 More than high school education−0.8060.765−0.0370.293     
 Married or living as married0.1620.8180.0070.843     
 Time since onset of first non-Raynaud's  symptoms0.0040.0040.0370.296     
 Physician-rated disease severity (0–10)−0.2570.192−0.0530.183     
 Comorbidities/other health problems−1.0420.357−0.1060.004     
 Diffuse scleroderma−1.0170.696−0.0520.144     
 Tender joint count−0.1530.090−0.0590.090     
 Inflammatory muscle disease−0.7301.171−0.0220.533     
 Breathing problems (0–10)−0.9660.186−0.229< 0.001     
 Number of GI symptoms−0.9910.139−0.271< 0.001     
 NYHA III or IV−0.3600.700−0.0230.607     
 Current smoker−2.4851.045−0.0840.018     


This study demonstrated that fatigue levels are high in patients with SSc compared with the general population, which is consistent with the results of a recent systematic review that found that fatigue related to SSc was significantly higher than in general population samples and similar to patients with other rheumatic diseases or patients with cancer in active treatment (10). In multivariate analysis, statistically significant correlates of fatigue included number of medical comorbidities, patient-reported breathing problems, patient-reported number of GI symptoms, and smoking. Tender joint count also approached statistical significance. Physician-rated global disease severity was a robust predictor of fatigue on a bivariate basis and after controlling for sociodemographic factors, but was no longer statistically significant when medical comorbidities and specific disease factors were included in the model. Depressive symptoms and pain were robust predictors of fatigue, but sociodemographic variables (age, sex, marital status, and education) were not associated with fatigue.

No studies have been published on fatigue management in SSc. Research in cancer and other rheumatic diseases shows that fatigue is amenable to intervention. Guidelines for cancer-related fatigue (30) suggest a 2-stage approach of first identifying and treating contributing factors (e.g., pain, emotional distress, sleep disturbance, and deconditioning), followed by management of residual fatigue. Among nonpharmacologic management strategies (e.g., exercise, sleep hygiene, and psychosocial interventions), empirical support is strongest for exercise and psychosocial interventions, including education and stress management groups and cognitive–behavioral therapy (31). Pharmacologic interventions have been used successfully to reduce fatigue in cancer (32), Parkinson's disease (33), and the human immunodeficiency virus/AIDS (34). In the rheumatic diseases, standard pharmacologic treatments including anti–tumor necrosis factor (anti-TNF) and adjunctive therapies reduce fatigue (35–39). A similar approach targeting depression, pain, and smoking among patients with SSc is supported by the findings from this study. Addressing other lifestyle factors, such as sleep and exercise, is also promising, although those variables are not collected in the CSRG Registry and were not used in this study.

Two important factors that make addressing fatigue in SSc difficult are the lack of an agreed-upon standard for identifying clinically significant fatigue, and a need for greater understanding of fatigue etiology, including the relationship between depressive symptoms, pain, and fatigue. A number of measurement tools have been validated to assess symptoms of fatigue, but these tools are not designed for accurate case detection, are not benchmarked to a case-definition standard, and do not necessarily identify clinically significant fatigue levels that warrant investigation and treatment. Establishing a unified case-definition approach to assessing fatigue would enhance comparability of research results, as well as the ability to detect important differences across groups that could help explain etiology. In addition, a standard case-definition approach would facilitate the benchmarking of cutoff scores on fatigue questionnaires and the development of brief, easily used screening tools to improve clinical management. Research in cancer has laid the groundwork for this by developing case-definition criteria for cancer-related fatigue (40) that appear in the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision. Four studies that have used the cancer-related fatigue case definition and structured diagnostic interview have reported consistent prevalence estimates (17–26%) (3, 41–43). Work is currently underway by investigators of the CSRG to determine the degree to which this case-definition protocol may be viable as a unifying framework for fatigue assessment across patient groups, including SSc.

From an etiologic perspective, high levels of both fatigue (1) and depression (44) are common among patients with chronic medical diseases, including SSc (6, 45, 46). There is substantial overlap between fatigue and depression, and measures of each were highly correlated in this study. Behavioral and cognitive factors link fatigue and depressive symptoms. In addition, basic neuroscience research has demonstrated that proinflammatory cytokines can signal the central nervous system to induce symptoms of fatigue and other sickness behaviors (e.g., malaise, listlessness, loss of appetite, and depressed mood) (47). Several proinflammatory mediators, including the acute response cytokines interleukin-1β (IL-1β) and TNF-α, as well as IL-6 and interferon-γ (INF-γ), have been linked to altered central nervous system activity and symptoms of depression and fatigue (48). The presence of chronic inflammation enhances cytokine production, and is one of the 3 major pathways that cause most of the organ damage in SSc (48). The proinflammatory cytokines IL-1β, IL-6, INFγ, and TNF-α are key components in this process and may also play a role in the pathogenesis of fatigue and depression in SSc (48). Research is needed to clarify the relationship of depression to fatigue in SSc.

Limitations should be considered in interpreting the results of this study. This study did not evaluate factors related to proinflammatory cytokines and the inflammation process. In addition, potentially important variables related to malnutrition, anemia, sleep, exercise, and cognitive factors were not available. Other limitations of this study include its cross-sectional design and the concurrent assessment of both predictor and outcome variables, which did not allow for the evaluation of pathways of influence. Indeed, it was not possible to determine the nature of the relationship between depression, pain, and fatigue. Similar to fatigue, each of these variables reflects the life experiences of patients with SSc. Therefore, although there are causal pathways and linkages among these variables, they are all potential outcomes related to the demographic, socioeconomic, and disease factors that were the focus of this study. In addition, method overlap related to self-reporting of these variables can inflate associations between these variables. It is important to note that self-report method overlap may also explain the relative predictive importance of specific indicators of disease factors because breathing difficulties, number of GI symptoms, and tender joint count were all based on patient report. Although tender joint count was recorded by a physician, it was based on patient-report of tenderness/pain in each joint.

In summary, findings from this study are consistent with prior evidence that fatigue is an important, albeit largely ignored, problem for patients with SSc. Patients with more difficulty breathing, higher numbers of GI symptoms, more medical comorbidities, and who are current smokers are the most vulnerable. More research is needed on fatigue in patients with SSc, including work on assessment and case identification and the development and testing of interventions to reduce fatigue.


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. Thombs 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. Thombs, Hudson, Bassel, Taillefer, Baron.

Acquisition of data. Thombs, Hudson, Bassel, Taillefer, Baron.

Analysis and interpretation of data. Thombs, Hudson, Bassel, Taillefer, Baron.


Actelion and Pfizer had no role in the design of the study, analysis of the data, preparation of the manuscript, or decision to submit the manuscript for publication.



Additional members of the Canadian Scleroderma Research Group not listed as coauthors include J. Markland: Saskatoon, Saskatchewan; J. Pope: London, Ontario; D. Robinson: Winnipeg, Manitoba; N. Jones: Edmonton, Alberta; P. Docherty: Moncton, New Brunswick; M. Abu-Hakima: Calgary, Alberta; N. A. Khalidi: Hamilton, Ontario; S. Le Clercq: Calgary, Alberta; E. Sutton: Halifax, Nova Scotia; C. D. Smith: Ottawa, Ontario; E. Kaminska: Hamilton, Ontario; J. P. Mathieu: Montreal, Quebec; P. Rahman: St. John's, Newfoundland; and S. Ligier: Montreal, Quebec.