Systemic Lupus Erythematosus
Self-efficacy and pain catastrophizing in systemic lupus erythematosus: Relationship to pain, stiffness, fatigue, and psychological distress
To determine how self-efficacy for pain control and pain catastrophizing, both potentially modifiable pain coping cognitions, are related to pain, stiffness, fatigue, and psychological distress in patients with systemic lupus erythematosus (SLE).
We conducted a cross-sectional study of patients with SLE who completed measures of pain coping cognitions (i.e., self-efficacy for pain control, pain catastrophizing), symptom ratings (i.e., pain, stiffness, fatigue), and psychological distress.
Correlational analyses revealed that self-efficacy for pain control and pain catastrophizing were associated with the patients' physical symptom reports and psychological distress. After controlling for age, race, and disease activity, patients with lower levels of self-efficacy for pain control reported much higher levels of pain, stiffness, and fatigue. Patients with higher levels of pain catastrophizing reported much lower positive mood. SLE activity as assessed by the rheumatologist was not associated with physical symptoms, psychological distress, self-efficacy for pain control, or pain catastrophizing.
These results suggest that pain coping cognitions (i.e., either self-efficacy for pain control or pain catastrophizing) are significantly related to physical symptoms and psychological distress in patients with SLE. This finding is important because the results of studies from other samples of patients with persistent pain conditions have shown that these pain coping variables can be modified using psychological interventions, and that such treatment-related changes in pain cognitions are related to improved patient outcomes.
Individuals with systemic lupus erythematosus (SLE) frequently experience distressing physical and psychological symptoms. High pain levels that interfere with quality of life are a primary concern for patients with SLE (1, 2). In addition to high levels of pain, patients with SLE also experience significant stiffness, fatigue, and psychological distress (3–5); all are symptoms that are inadequately affected by current immunomodulatory therapies for SLE.
The judgments people make about their ability to cope with pain are critical in understanding individuals' adjustment to persistent pain. No study to date has examined how coping cognitions specific to pain may influence important outcomes, such as pain, stiffness, fatigue, and mood, in SLE patients. Past studies in patients with SLE have demonstrated that patients' general coping strategies and perceptions of control (i.e., not specifically related to SLE pain) are related to SLE disease activity (2, 6–8), physical functioning (9), psychological functioning (2, 10), and other measures of quality of life (11). Two pain coping cognitions that may be particularly important to consider in patients with SLE are self-efficacy for pain control and pain catastrophizing. Self-efficacy for pain control refers to a person's confidence in their ability to control pain related to rheumatic disease (12). Pain catastrophizing refers to an individual's tendency to ruminate about pain, magnify pain sensations, and feel helpless when dealing with pain (13). These 2 pain coping cognition variables have been shown to be related to pain, stiffness, fatigue, and psychological disability in other samples of patients with persistent pain conditions (14–16). It has been shown that, even in patients with high levels of pain, behavioral interventions can enhance a patient's self-efficacy in pain control, and that this can lead to lower levels of pain (17, 18). Furthermore, there is evidence that behavioral interventions that systematically teach patients with persistent pain how to decrease catastrophic thinking about pain can produce significant decreases in pain (19).
People with SLE often face a chronic and unpredictable disease course that may challenge their ability to cope with disease symptoms such as pain. Pain coping cognitions (i.e., self-efficacy for pain control and pain catastrophizing) may particularly influence symptoms such as pain, stiffness, fatigue, and psychological distress in individuals with SLE. Understanding the relationships between pain coping cognitions and pain in SLE may help to focus the application of behavioral interventions for these patients. To our knowledge, no study has examined how these potentially important pain coping cognitions relate to the adjustment to disease in people with SLE.
In this study, we aimed to examine how the pain coping cognitions of self-efficacy for pain control and pain catastrophizing were related to physical symptoms and psychological distress in patients with SLE after controlling for demographic variables and disease activity. Based on prior studies, we expected that patients who report low levels of self-efficacy for pain control and higher levels of pain catastrophizing would report more physical symptoms and psychological distress, including pain, stiffness, fatigue, low positive mood, and high negative mood. We also expected that controlling for age, race, and SLE disease activity would not diminish these relationships.
Significance & Innovations
This is one of the first studies to examine the relationship between pain coping cognition variables and important physical and psychological outcomes in patients with systemic lupus erythematosus (SLE).
Self-efficacy for pain control and pain catastrophizing are related to the patients' SLE physical symptoms and psychological symptoms in ways that are not explained by standard disease activity measures.
PATIENTS AND METHODS
We conducted a cross-sectional study of consecutive patients of the Duke Lupus Clinic who met the American College of Rheumatology 1997 criteria for a diagnosis of SLE (20). Patients were enrolled from July 2010 to February 2011. The Institutional Review Board at Duke University approved all study procedures, all participants provided informed consent, and all research was carried out in compliance with the Declaration of Helsinki. The participants completed a questionnaire that included items assessing self-efficacy for pain control, pain catastrophizing, pain, stiffness, fatigue, and psychological distress (i.e., positive and negative mood). The treating rheumatologists (LC-S and MEBC) assessed SLE activity using the Safety of Estrogens in Lupus Erythematosus National Assessment (SELENA) version of the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) (21).
For the assessment of SLE disease activity, the SELENA–SLEDAI (21) is a physician-completed tool to assess SLE disease activity with a rating scale consisting of 24 descriptors associated with 9 organ systems. The SELENA–SLEDAI includes clinical and laboratory measures of SLE activity, weighted according to severity. The weighted items are summed to obtain an overall score, with a possible total score ranging from 0–105. The SELENA–SLEDAI has been validated and shown to be sensitive to changes over time (21).
For self-efficacy for pain control, we measured patients' confidence in their ability to control their pain with the Chronic Pain Self-Efficacy Scale (12). The scale consists of 5 questions inquiring about the participants' certainty in their ability to control pain. The participants circled the number that best corresponded with their certainty on a 10-point scale, where 10 = very uncertain and 100 = very certain. Numerically high scores indicate high levels of self-efficacy for pain control. This tool has proven reliability in other samples of patients with persistent pain conditions and had excellent reliability in this sample (Cronbach's α = 0.91).
Pain catastrophizing was assessed using the Short-Form Catastrophizing Scale of the Coping Strategies Questionnaire (CSQ) (22, 23). The scale contains 2 items that ask patients to rate the frequency with which they engage in catastrophizing when they experience pain on a 6-point scale, where 0 = never and 6 = always. In prior studies, the CSQ had good internal reliability (22) and was highly stable over time (23, 24). The internal reliability of this scale was good in this study (Cronbach's α = 0.82).
The participants rated their average pain, morning stiffness, and fatigue in the last week on three 10-point visual analog scales (VAS), where 0 = no symptom and 10 = symptom as bad as you can imagine. VAS symptom rating assessments are widely used in rheumatic disease research with high demonstrated reliability (25, 26).
Positive and negative mood were measured with the 10-item Positive and Negative Affect Scale (PANAS) (27). Patients indicated the degree of their positive and negative affective states (e.g., excited, distressed, scared, inspired) by rating items from 1 (very slightly or not at all) to 5 (extremely). The summary scores for negative affect and positive affect were averaged separately, yielding 2 indices of affect. A high score on positive affect reflects enthusiasm, energy, and alertness, and a high score on negative affect reflects distress and aversive mood states. The PANAS has demonstrated good internal consistency in samples of patients with persistent pain (28, 29); Cronbach's alpha in this sample indicated good reliability (positive α = 0.79, negative α = 0.85).
Table 1 shows the descriptive data for the demographic variables (i.e., age, sex, race), disease activity, and measures of self-efficacy for pain control, pain catastrophizing, pain, stiffness, fatigue, positive mood, and negative mood. Sex was not included in the analyses because the majority (95%) of the sample were women. For all analyses, race was coded as 0 = white and 1 = African American/other. Correlational analyses were performed to examine the associations between demographic variables, disease activity, and other study variables. Age, race, and disease activity were retained as control variables in further analyses because past studies suggested that they are each related to other study variables in patients with SLE (30–32). Hierarchical linear regression (22) was used to assess the contribution of self-efficacy for pain control and pain catastrophizing to the outcomes after controlling for age, race, and SLE disease activity. In each hierarchical linear regression, self-efficacy for pain control and pain catastrophizing were entered on the final step (step 3) after controlling for age and race on step 1 and disease activity on step 2. Tolerance statistics for multicollinearity were examined to ensure that there was adequate unique variance available for each predictor.
Table 1. Demographic and descriptive variables (n = 74)
|Age, years||39.39 ± 13.30|| || |
|Female|| ||70 (95)|| |
|Race|| || || |
| White|| ||28 (38)|| |
| African American|| ||42 (57)|| |
| Other|| ||4 (5)|| |
|Disease activity||4.60 ± 4.53|| ||0–20 (0–105)|
|Pain self-efficacy||48.02 ± 24.67|| ||10–100 (10–100)|
|Pain catastrophizing||3.99 ± 3.23|| ||0–12 (0–12)|
|Pain||4.51 ± 2.79|| ||0–9 (0–10)|
|Stiffness||4.55 ± 3.11|| ||0–10 (0–10)|
|Fatigue||6.00 ± 2.54|| ||0–10 (0–10)|
|Positive mood||15.17 ± 4.88|| ||5–25 (5–25)|
|Negative mood||8.37 ± 3.90|| ||5–19 (5–25)|
A total of 74 patients with SLE participated in this study. Our sample was 95% women, and the mean ± SD age was 39.39 ± 13.3 years. The racial composition was 38% white (n = 28), 57% African American (n = 42), and 5% other groups (n = 4). The very low level of self-efficacy for pain control (mean ± SD 48.02 ± 24.67) reported by patients with SLE in this study was striking, even when compared to patients referred to specialty pain clinics with low back pain and other persistent pain conditions (mean ± SD 51.30 ± 22.13) (33). Also, the levels of pain catastrophizing (mean ± SD 3.99 ± 3.23) in this sample were quite high; a sample of veterans referred to a specialty pain management clinic due to persistent pain conditions scored similarly on the same assessment tool (mean ± SD 3.93 ± 1.45) (34) (Table 1).
Age and race did not strongly correlate with pain, fatigue, stiffness, or negative mood, but older age correlated with a high degree of positive mood (r = 0.33, P < 0.01). Neither race nor disease activity was correlated with pain, stiffness, fatigue, or positive mood. Higher levels of disease activity were correlated with higher negative mood (r = 0.26, P < 0.05), but not positive mood. Both dependent variables (self-efficacy for pain control and pain catastrophizing) correlated with pain, stiffness, fatigue, and positive and negative mood (r = 0.23 to 0.65; P for all < 0.05) (Table 2).
Table 2. Correlations between study variables
|Age, years||–|| || || || || || || || || |
|Race*||−0.03||–|| || || || || || || || |
|Disease activity||−0.22||0.28†||–|| || || || || || || |
|Pain self-efficacy||−0.04||−0.03||−0.16||–|| || || || || || |
|Pain catastrophizing||−0.14||−0.14||0.11||−0.62‡||–|| || || || || |
|Pain||0.05||−0.09||0.04||−0.60‡||0.50‡||–|| || || || |
|Stiffness||0.23||−0.02||−0.10||−0.65‡||0.46‡||0.79‡||–|| || || |
|Fatigue||0.13||−0.14||−0.06||−0.52‡||0.39‡||0.66‡||0.66‡||–|| || |
|Positive mood||0.33‡||0.16||0.01||0.23†||−0.42‡||−0.19||−0.19||−0.11||–|| |
Hierarchical linear regression.
Table 3 shows the hierarchical linear regression results of the association between self-efficacy for pain control and pain catastrophizing and the outcomes pain, fatigue, stiffness, and mood after controlling for age, race, and disease activity. The smallest multicollinearity tolerance statistic in any of our models was 0.62, suggesting that there was more than an adequate percent of variance in each predictor that could not be accounted for by the other predictors.
Table 3. Hierarchical linear regression analyses for symptoms, psychological distress, and memory*
|Pain|| || || || || |
| 1, Age|| || ||−0.03||−0.24||0.81|
| Race||0.01|| ||0.06||−0.49||0.63|
| 2, Disease activity||0.01||0.01||−0.07||−0.58||0.57|
| 3, Pain self-efficacy|| || ||−0.52||3.93||< 0.001†|
| Pain catastrophizing||0.37||0.36||0.14||1.02||0.31|
|Stiffness|| || || || || |
| 1, Age|| || ||0.16||1.63||0.10|
| 2, Disease activity||0.07||0.01||−0.21||−2.11||0.04†|
| 3, Pain self-efficacy|| || ||−0.57||−4.83||< 0.001†|
| Pain catastrophizing||0.51||0.44||0.17||1.45||0.14|
|Fatigue|| || || || || |
| 1, Age|| || ||0.05||0.40||0.69|
| 2, Disease activity||0.04||0.01||−0.08||−0.70||0.49|
| 3, Pain self-efficacy|| || ||−0.48||−3.47||0.001†|
| Pain catastrophizing||0.31||0.27||0.08||0.59||0.56|
|Positive mood|| || || || || |
| 1, Age|| || ||0.30||2.60||0.01†|
| 2, Disease activity||0.11||0.01||0.18||1.52||0.14|
| 3, Pain self-efficacy|| || ||0.07||0.46||0.64|
| Pain catastrophizing||0.32||0.21||−0.43||−3.10||0.003†|
|Negative mood|| || || || || |
| 1, Age|| || ||−0.16||−1.30||0.20|
| 2, Disease activity||0.07||0.04||0.15||1.14||0.26|
| 3, Pain self-efficacy|| || ||−0.24||−1.56||0.13|
| Pain catastrophizing||0.17||0.10||0.11||0.72||0.47|
The overall model with pain as an outcome was significant (F[5,57] = 6.54, P < 0.001). As shown in Table 3, demographic variables and disease activity each accounted for 1% of the variance in pain. Self-efficacy for pain control and pain catastrophizing accounted for 36% of the variance in pain. Of these 2 pain coping cognitions, self-efficacy for pain control was a significant independent predictor of pain severity (β = −0.52, t = 3.93, P < 0.001).
The overall model with stiffness as the outcome was significant (F[5,57] = 11.79, P < 0.001). Age and race accounted for 7% of the variance in stiffness. Disease activity accounted for an additional 1% of the variance in stiffness. Self-efficacy for pain control and pain catastrophizing accounted for an additional 44% variance in stiffness. Only self-efficacy for pain control was a significant independent predictor of stiffness (β = −0.57, t = −4.83, P < 0.001).
The overall model with fatigue as the outcome was significant (F[5,57] = 5.00, P = 0.001). Age and race explained 3% of the variance in fatigue. Disease activity accounted for an additional 1% of the variance in fatigue. Self-efficacy for pain control and pain catastrophizing accounted for an additional 27% of the variance in fatigue. Only self-efficacy for pain control was a significant predictor of fatigue (β = −0.48, t = −3.47, P = 0.001).
For psychological distress, the overall model with positive mood as the outcome was significant (F[5,56] = 5.15, P = 0.001). Age and race accounted for 10% of the variance in positive mood. Disease activity accounted for an additional 1% of the variance in positive mood. Self-efficacy for pain control and pain catastrophizing accounted for an additional 21% of the variance in positive mood. Only pain catastrophizing was a significant individual predictor (β = −0.43, t = −3.10, P = 0.003). When examining negative mood, the overall model was not significant (F[5,57] = 2.29, P = 0.06).
The goal of this study was to examine how the pain coping cognitions of self-efficacy for pain control and pain catastrophizing related to pain, stiffness, fatigue, and psychological symptoms in patients with SLE even after controlling for demographic and medical variables. In our SLE sample, in contrast with other studies of patients with persistent pain (16, 35), the patients' scores on an objective disease measure (i.e., the SELENA–SLEDAI) were not associated with our study outcomes of pain, fatigue, or mood. However, pain coping cognitions were related to our outcomes, emphasizing the potential importance of these constructs in patients with SLE. Participants who reported low levels of self-efficacy for pain control and/or high levels of pain catastrophizing reported more physical symptoms and psychological distress than patients with lower levels of pain catastrophizing. Our results suggest that participants with low self-efficacy for pain control were much more likely to report the symptoms of pain, stiffness, and fatigue. Patients with higher levels of pain catastrophizing were much more likely to report increased psychological distress.
To our knowledge, this is one of the first studies to examine the relationship between pain coping cognitions (i.e., self-efficacy for pain control, pain catastrophizing) and physical and psychological outcomes in people with SLE. An important feature of this study is that ∼60% of our sample was African American; examining these factors in a sample as diverse as ours is important, since SLE affects African Americans at a rate of 2:1 compared to whites (36). Individuals with SLE commonly report symptoms, including pain, stiffness, and fatigue, even in periods of disease quiescence as assessed by the physician (37), resulting in an ongoing compromised quality of life. In our sample, disease activity was not associated with pain, stiffness, or fatigue. However, in our study, patients who reported low levels of self-efficacy for pain control also reported frequent distressing symptoms of SLE, including high levels of pain, morning stiffness, and fatigue. Our findings support the findings of Jolly and Utset (38) that suggested that traditional disease-specific assessments (i.e., the SLEDAI, the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index) for SLE do not adequately reflect the compromised quality of life SLE patients often experience. Our findings, moreover, suggest that the patients' level of confidence in their ability (i.e., self-efficacy) to control their pain may be an important variable explaining health-related quality of life, such as pain, stiffness, and fatigue.
We also examined how pain catastrophizing is related to pain and other important SLE-related outcomes (i.e., stiffness, fatigue, psychological distress). Pain catastrophizing has been shown to be one of the most important psychological constructs in predicting outcomes in patients with persistent pain. Our results suggest that patients with SLE who engage in pain catastrophizing are at a high risk for psychological distress (i.e., lower positive mood). This finding is particularly important because it persists even after controlling for age, race, and disease status. Low positive mood has been closely related to depression in past studies (39). Depression is common in patients with SLE (40, 41); these results suggest that pain catastrophizing may be an important construct to consider in SLE patients who have depression.
High levels of pain catastrophizing were also correlated with high levels of pain, morning stiffness, and fatigue. Our regression analyses suggested that self-efficacy for pain control accounts for more of the variance in these physical symptoms than pain catastrophizing does.
This study is limited by the use of cross-sectional data, which diminishes our ability to suggest causality. Increased pain, physical, and psychological symptoms may lead to maladaptive patterns of coping, including higher pain catastrophizing and lower self-efficacy for pain control. On the other hand, prior interventional studies in other disease populations have documented that attention to these pain coping cognitions can lead to significant changes in the pain, physical, and psychological symptoms that a patient experiences (17–19). Given the limited success in improving these symptoms through traditional immunosuppressive therapy for SLE and the lack of a relationship between disease activity and pain, attention to pain coping cognitions could be a welcome and valuable addition to the treatment of patients with SLE. Another limitation of this study is the small percentage of men (5%), which limited our ability to statistically examine potential sex differences in pain coping cognitons (i.e., self-efficacy for pain control, pain catastrophizing). Future studies should examine the potential sex differences in pain coping cognitions in SLE patients.
Overall, our results suggest that pain coping strategies are related to patients' SLE physical symptoms and psychological symptoms in ways that are not explained by standard disease activity measures, such as the SELENA–SLEDAI (21, 38). Disease activity is often associated with pain and symptom variables in samples of rheumatoid arthritis or osteoarthritis patients (16, 35). Our results support the importance of assessing coping constructs in SLE patients, such as self-efficacy for pain control and pain catastrophizing, which are two pain coping cognition variables shown to be modifiable through the application of psychological interventions in other samples of patients with chronic pain conditions (17–19). Shifting treatment from a traditional medication approach (e.g., immunosuppressive therapy) to a multidisciplinary approach that includes addressing coping cognitions, such as self-efficacy and pain catastrophizing, as well as other psychosocial constructs, may have important clinical and economic importance. The patients who participated in this study reported being very interested in a psychological intervention that would teach them to cope with pain and other SLE symptoms. Future studies should consider designing psychological interventions to address these pain coping cognitions in patients with SLE, especially when self-efficacy for pain control is low or pain catastrophizing is high.
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 published. Dr. Somers 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. Somers, Criscione-Schreiber, Keefe, Clowse.
Acquisition of data. Kurakula, Criscione-Schreiber, Clowse.
Analysis and interpretation of data. Somers, Criscione-Schreiber, Keefe, Clowse.