Catastrophizing and pain in arthritis, fibromyalgia, and other rheumatic diseases

Abstract

Objective

Pain is among the most frequently reported, bothersome, and disabling symptoms described by patients with osteoarthritis, rheumatoid arthritis, fibromyalgia, and other musculoskeletal conditions. This review describes a growing body of literature relating catastrophizing, a set of cognitive and emotional processes encompassing magnification of pain-related stimuli, feelings of helplessness, and a generally pessimistic orientation, to the experience of pain and pain-related sequelae across several rheumatic diseases.

Methods

We reviewed published articles in which pain-related catastrophizing was assessed in the context of one or more rheumatic conditions. Because much of the available information on catastrophizing is derived from the more general chronic pain literature, seminal studies in other disease states were also considered.

Results

Catastrophizing is positively related, in both cross-sectional and prospective studies across different musculoskeletal conditions, to the reported severity of pain, affective distress, muscle and joint tenderness, pain-related disability, poor outcomes of pain treatment, and, potentially, to inflammatory disease activity. Moreover, these associations generally persist after controlling for symptoms of depression. There appear to be multiple mechanisms by which catastrophizing exerts its harmful effects, from maladaptive influences on the social environment to direct amplification of the central nervous system's processing of pain.

Conclusion

Catastrophizing is a critically important variable in understanding the experience of pain in rheumatologic disorders as well as other chronic pain conditions. Pain-related catastrophizing may be an important target for both psychosocial and pharmacologic treatment of pain.

Introduction

Pain is a nearly ubiquitous experience, and a cardinal symptom of many rheumatologic conditions. Catastrophizing, a set of negative emotional and cognitive processes (1), is increasingly implicated in the experience of pain in rheumatoid arthritis (RA), osteoarthritis (OA), and fibromyalgia (FM). The construct of catastrophizing incorporates magnification of pain-related symptoms, rumination about pain, feelings of helplessness, and pessimism about pain-related outcomes. The recognition that pain is a consistent risk factor for mortality (2–5) highlights the importance of better understanding the biopsychosocial nature of pain and identifying groups at high risk for adverse pain-related consequences. This review summarizes evidence that catastrophizing represents an important target for investigation and intervention in the rheumatic diseases.

Catastrophizing

Catastrophizing is typically measured using a self-report inventory: the 6-item catastrophizing subscale of the Coping Strategies Questionnaire (CSQ) (6) or the Pain Catastrophizing Scale (PCS) (7), which expanded the original 6 CSQ items to include 7 others. Individuals rate the extent to which they experience (when they are in pain) the thought or feeling described by each item (Figure 1). Each scale has good psychometric characteristics (1); the PCS has 3 subscales, magnification, rumination, and helplessness (7), which have similar psychometric properties in patients with FM and controls (8).

Figure 1. The Pain Catastrophizing Scale (7). Total scores range from 0 to 52.

Although individuals are sometimes dichotomized as catastrophizers and noncatastrophizers, most research treats catastrophizing as a continuous, normally distributed variable (1). In our database of patients with pain, there is wide variability around the mean catastrophizing score (Figure 2). Catastrophizing also exists on a continuum in healthy, pain-free individuals (9); indeed, higher catastrophizing, assessed in pain-free adults, predicts the future development of chronic pain and pain-related health care utilization (10, 11). A rich area of debate has centered on whether catastrophizing is best conceptualized as a stable and enduring trait, such as a dimension of personality, or as a modifiable characteristic (1, 12), with some evidence supporting both positions. Several studies report a high test–retest stability of catastrophizing measured over time frames of up to a year in patients with RA and in other samples (7, 13). In contrast, catastrophizing often decreases when patients undergo cognitive-behavioral therapy (CBT; a set of psychologist-delivered interventions designed to facilitate the development of self-management skills, including regulating one's thoughts, emotions, and behaviors) (14, 15), indicating that catastrophizing can be altered by treatment.

Figure 2. Distribution of Coping Strategies Questionnaire (CSQ) catastrophizing subscale scores in a sample of 2,257 patients experiencing heterogeneous chronic pain syndromes (unpublished data). Scores on the CSQ catastrophizing subscale range from 0 to 6.

Adverse Outcomes of Catastrophizing

Pain severity.

Cross-sectionally, catastrophizing relates to higher pain severity among patients with RA (16–18) and OA (19, 20). High levels of catastrophizing are also associated with more severe and widespread pain and more emotional disturbance among individuals with FM (21–24) and scleroderma (25). In general, these associations persist even after statistically controlling for depression, anxiety, or neuroticism (20, 26). Several prospective studies have illustrated the longitudinal association of catastrophizing with pain in RA. In daily diary studies, patients with RA who exhibited greater catastrophizing reported more day-to-day pain and attention to pain than low catastrophizers (26, 27). Findings from another prospective study suggested that baseline catastrophizing scores predicted enhanced pain and depression in patients with RA at 1-year followup (28). Catastrophizing may also influence the success of pain-related treatments in patients with musculoskeletal disease. In studies of patients with OA recovering from knee surgery (29, 30), higher preoperative levels of catastrophizing were associated with more pain and disability up to 6 months postoperatively. Whether catastrophizing predicts the onset of painful rheumatic conditions is not known, although high catastrophizing was shown to be a risk factor for the onset of low back pain and disability in a population-based study (10, 11).

Pain sensitivity.

Catastrophizing shows positive associations with tender point counts in both population studies of musculoskeletal tenderness and clinic-based samples of patients with FM (24, 31–33). Hyperalgesia, or enhanced responsiveness to painful stimuli, is a defining feature of FM but has also been noted in patients with RA (34–39) and OA (40–42). Catastrophizing may be correlated with some of these hyperalgesic responses. For example, catastrophizing was associated with decreased heat pain threshold and tolerance in women with FM (43), reduced pain tolerance during a cold pressor test in patients with juvenile rheumatoid arthritis (JRA) (44, 45), and lower pain threshold and tolerance in response to electrical stimulation among patients with OA (46). Recent evidence from our laboratory suggests that higher catastrophizing relates to greater central nervous system (CNS) sensitization during sustained pain (47), which may account for the consistent positive relationship between catastrophizing and pain sensitivity.

Depression.

In general, catastrophizing is strongly associated with measures of negative affect (1). Multiple investigators have documented positive associations between catastrophizing and depressive symptoms in FM (21, 23, 43). Similar findings have been reported in patients with RA (16), and we have observed a significant Pearson's correlation (r = 0.65) between catastrophizing and scores on the Beck Depression Inventory in several hundred individuals with scleroderma. Prospective studies have documented the association of high catastrophizing at baseline with increases in depressive symptoms over periods of up to 1 year in patients with RA (13) and FM (28). In a recent diary study of patients with OA, catastrophizing showed concurrent and prospective relationships with more intense negative mood (i.e., increases in catastrophizing on a given day related to worsened mood that same day and on the next day) (19). Taken together, these findings suggest that in the context of chronic pain, catastrophizing may contribute to depressed mood on a short- and long-term basis. Interestingly, virtually no research to date has examined associations between pain-related catastrophizing and formally assessed (e.g., by structured interview) psychiatric diagnoses such as posttraumatic stress disorder, generalized anxiety disorder, etc., which represents an important avenue for future catastrophizing research.

Disability.

Catastrophizing shows robust associations with self-reported disability and with more objective indices of function such as returning to work. In patients with OA, catastrophizing relates to higher levels of observed pain behaviors and functional limitations during standardized activity tests (20). Among patients with OA undergoing knee surgery, catastrophizing prospectively predicted postsurgical disability, even after controlling for other psychosocial factors (29). Importantly, although pain severity is often a primary determinant of disability, RA studies have established that catastrophizing predicts disability even after controlling for pain severity (16, 48). Finally, catastrophizing and other indices of poor pain coping are prospectively associated with reductions in objectively measured mobility and muscle strength over periods of up to 5 years in patients with RA (49, 50). These findings are consistent with studies of low back pain (51–53) in which catastrophizers reported more pain and reduced function during standardized physical tasks (e.g., range-of-motion exercises, etc.).

Physiologic indices of disease activity in RA.

Multiple RA studies have reported positive relationships between catastrophizing (or helplessness, one component of catastrophizing) and elevated disease activity (54–58). Although much of this research is cross-sectional, at least 1 longitudinal RA study has shown that catastrophizing prospectively predicted worsening disease activity (defined by erythrocyte sedimentation rate [ESR] and joint counts) (26). Among patients with JRA, catastrophizing directly influences physicians' global assessments of disease (59), with higher catastrophizing predicting more severe disease assessment. Whereas most of these studies did not control for symptoms of anxiety or depression, a recent RA study concluded that although helplessness was strongly positively associated (i.e., 15% shared variance) with elevated high-sensitivity C-reactive protein (CRP) levels, anxiety and depression were either unrelated or only modestly related to CRP (60). Finally, prospective RA research has also revealed that baseline helplessness predicts future increases in ESR (61) as well as mortality (62, 63), even when controlling for baseline disease severity. Whether catastrophizing directly impacts other physiologic systems such as the sympathetic nervous system or the hypothalamic–pituitary–adrenal axis is uncertain; cold pressor studies have demonstrated that high catastrophizing does not predict cortisol reactivity to pain (64), but does predict sustained increases in myocardial contractility (65), a potential index of sympathetic activity. Although it is unclear how catastrophizing influences disease severity, helplessness does correlate with less effective medication use (66) and less positive health behavior such as exercise (67), suggesting several plausible pathways by which catastrophizing could enhance disease, reduce physical health, and promote mortality.

Additional outcomes.

The impact of catastrophizing on outcomes can be fairly broad (i.e., not limited to pain). For example, catastrophizing is related to greater reports of fatigue among women with breast cancer (68, 69), increased constitutional symptoms such as nausea in individuals with infections (70), reduced maternal social involvement among new mothers (71), and dissatisfaction with treatment among patients being treated for gastrointestinal symptoms (72). From a societal perspective, catastrophizing is an important variable to understand because it relates to greater health care utilization and use of pain-related medications in the general population (11, 73), even after controlling for pain intensity.

Hypothesized Mechanisms of Action

Catastrophizing interferes with pain-coping and beneficial health behaviors.

Perceptions of helplessness and pessimism may diminish the likelihood that high catastrophizers anticipate positive outcomes from other coping efforts, which may therefore be underutilized (1). In 2 experimental pain studies, individuals who were high catastrophizers reported using fewer active coping strategies (e.g., distraction, relaxation, etc.) during a cold pressor test (7, 74). Catastrophizing also relates to lower coping efficacy in patients with OA (19) and RA (18); indeed, in this latter study, higher levels of catastrophizing were associated with reduced perceptions of coping self efficacy on the part of both the patient and his or her spouse. Finally, as noted above, helplessness (one component of catastrophizing) correlates with reduced adherence to medication regimens (66) and less positive health behaviors such as exercise (67), each of which could potentially lead to increases in musculoskeletal pain symptoms.

Catastrophizing increases attention to pain.

Some research has also examined the hypothesis that catastrophizing enhances the experience of pain via its effects on attentional processes. That is, high levels of catastrophizing may lead individuals to attend selectively and intensely to pain-related stimuli. Catastrophizers experience more difficulty controlling or suppressing pain-related thoughts than do noncatastrophizers, they ruminate more about their pain, and their cognitive and physical performance are more disrupted by anticipation of pain (51, 75–77). In patients with FM, catastrophizing is strongly correlated with increased attention to pain (78) and greater vigilance to bodily sensations (36, 79).

Catastrophizing amplifies pain processing in the CNS.

Incoming signals in the CNS are subject to modulation at a variety of sites from the spinal cord to the cortex (80). One hypothesized mechanism by which catastrophizing impacts the experience of pain promotes sensitization or interfering with pain inhibition in the CNS (1, 21, 43, 47). An early study suggested that reducing catastrophizing resulted in the activation of descending endogenous opioid systems that inhibited nociception (81). A more recent functional magnetic resonance imaging study of patients with FM indicated that high catastrophizers showed enhanced activity in cortical regions involved in the affective processing of pain, such as the anterior cingulate cortex and insular cortex, during the experience of acute pain (21). Recent data from our laboratory also suggest that catastrophizing may be directly associated with CNS pain-facilitatory processes in the spinal cord (82). Overall, preliminary evidence indicates that catastrophizing may amplify pain processing at multiple CNS loci, with some researchers postulating that bidirectional relationships between catastrophizing and nociceptive processing may contribute to the chronicity of many pain conditions (1).

Catastrophizing has a maladaptive impact on the social environment.

The communal coping model of catastrophizing postulates that expressions of catastrophizing function to maximize access to supportive responses from others, and that these social responses may then reinforce displays of pain and expressions of catastrophizing (1, 83, 84). In support of the model, catastrophizers are perceived by others as less able to manage pain, and are more likely to seek social support (18, 48, 85, 86). It is interesting to consider that daily diary studies of patients with RA suggest that expressing emotions and seeking social support, which may reflect catastrophizing, are prospectively associated with greater arthritis pain (87–90). Also noteworthy is the finding that high levels of catastrophizing are associated with greater perceived stress and less non–pain-related social support within the social network (86, 91), suggesting that more frequent catastrophizing may have paradoxical effects by both eliciting more solicitous responses to pain (92) and reducing the general availability of support, potentially by enhancing distress in others (93).

Implications for Treatment

Screening for psychosocial risk factors that predict poor treatment outcome is not widespread, but it may hold promise as a low-cost means to identify individuals who would benefit most from adjunctive treatments. For example, given that catastrophizing is a risk factor for poor surgical outcomes (29, 30), long-term treatment success may be maximized by offering CBT either presurgery or concurrently with surgery to those who exhibit high levels of catastrophizing. We should also note that reducing catastrophizing is particularly important in the context of preventing disability. As several RA studies have demonstrated, the degree of physical disability exhibited by patients is a function not solely of pain frequency or intensity, but also of the degree of catastrophizing, suggesting that simple pain reduction is not an adequate treatment goal. In this regard, exposure paradigms are an important part of behavioral treatments for painful conditions such as FM, RA, and OA; interventions designed to increase physical activity levels may result in greater pain in the short term, but by reducing catastrophizing and enhancing self efficacy, these interventions are likely to reduce long-term pain and disability (94, 95).

Chronic pain and disability are increasingly managed by multidisciplinary means; analgesic regimens are frequently supplemented by physical therapy or psychological interventions for individuals experiencing persistent pain from rheumatic diseases (96–98). Indeed, for patients with FM, nonpharmacologic adjunctive therapies may demonstrate benefits superior to those provided by analgesic medications (96, 99). Emerging evidence indicates that catastrophizing may be an important mediating variable contributing directly to the outcomes of such treatments. In 2 previous studies, decreases in catastrophizing correlated with reductions in levels of depression and pain behaviors (such as distorted mobility and verbal and nonverbal complaints) among patients undergoing pain treatment (14, 100). Subsequent work using more sophisticated analytic techniques has extended these findings; during multidisciplinary pain treatment, early reductions in catastrophizing are associated with greater improvements in pain later in treatment, whereas individuals whose catastrophizing does not change demonstrate few or no benefits from multidisciplinary interventions (15, 101–103). Previous studies of cognitive and behavioral interventions for pain suggest that these methods are effective in decreasing pain-related catastrophizing (14, 15, 104), and based on these findings, it may be of great importance to target catastrophizing early in the multidisciplinary management of pain. Future treatment studies in patients with rheumatic disease may benefit from the assessment of catastrophizing pre- and posttreatment, the consideration of catastrophizing as a mediator or moderator of treatment effects, and a more fine-grained analysis of the pathways by which catastrophizing impacts important outcomes.

A crucial unanswered question is whether catastrophizing is a cause or consequence of chronic pain (1, 12). Some indirect evidence bears on this issue, although no longitudinal studies have yet examined whether catastrophizing changes following the development of chronic pain. First, catastrophizing tends to be stable over time in both healthy adults and patients with pain, showing high test–retest reliability measured over weeks or months (7, 13). Recent data have also suggested that catastrophizing, measured initially when patients were experiencing acute pain and remeasured several weeks later when they were pain free, did not change when patients' pain was alleviated (47). In contrast, as noted above, self report of catastrophizing often decreases when patients undergo CBT (14, 15), indicating that it may be substantially modifiable. Finally, studies using daily diary methodologies offer a unique opportunity to examine the dynamic properties of catastrophizing because these methodologies allow assessment of variability both within persons (i.e., variation from time point to time point) and between persons. A recent diary study in patients with chronic pain highlighted the short-term stability of catastrophizing: individuals differed substantially in how much they catastrophized, but a given person was likely to show similar levels of catastrophizing across the 2-week period, despite fluctuations in pain (105). Collectively, catastrophizing appears to develop relatively early in life (106, 107) and to possess many stable, trait-like characteristics, but it is clearly also amenable to reduction by certain types of psychosocial treatment.

Conclusions

Catastrophizing shows strong influences on many pain-related outcomes in patients with rheumatic disease, with multiple mechanisms of action accounting for its effects (Figure 3). Because it is robustly correlated with treatment success, catastrophizing represents an appealing target for multidisciplinary pain-management interventions. Because catastrophizing may act via numerous pathways, multimodal treatments incorporating pharmacologic, cognitive, behavioral, and potentially social interventions are perhaps most likely to succeed in ameliorating its effects. Unfortunately, no published studies have evaluated the efficacy of pharmacologic interventions in the reduction of catastrophizing, which represents an important area for future research. However, CBT and multidisciplinary treatments consistently reduce catastrophizing, even in samples of patients with long-standing complaints. For example, in patients reporting chronic pain for >12 years, a 4-week group CBT intervention delivered by a psychologist improved PCS scores by ∼40% from pre- to posttreatment (108). At present, little information is available on the management of catastrophizing in patients with rheumatic disease, although based on the broader chronic pain literature, patients with high levels of catastrophizing are likely to benefit from referrals to a pain psychologist or to other health professionals with expertise in CBT. Finally, given that catastrophizing relates to enhanced inflammatory processes and disease activity in RA, the refinement and application of cognitive, behavioral, and other interventions designed to diminish catastrophizing in patients with arthritis would potentially represent an important development in disease management.

Figure 3. Diagram of the putative associations of catastrophizing with pain outcomes.