Interpersonal problems as a predictor of pain catastrophizing in patients with chronic pain

1 Introduction

Pain catastrophizing may be defined as an exaggerated negative orientation to actual or anticipated pain [1], and has been found to be a robust predictor of adverse outcomes with chronic pain conditions [2], [3]. Whereas the concept of catastrophizing, originating within a cognitive-behavioral framework, has increased our understanding of chronic pain as in the fear-avoidance model (FAM) [4], [5], it has nonetheless been criticized for being overly simplistic, ignoring interpersonal factors [1], [6]. According to this perspective, pain catastrophizing may be a strategy to elicit social support from others. While there is abundant research demonstrating psychosocial factors (negative affect, childhood trauma, social support, etc.) to be related to the development and maintenance of chronic pain [7], data specifically supporting the notion of pain catastrophizing as a strategy to elicit support has been more mixed (e.g. [8], [9], [10]). This may reflect differences in research design, measurements as well as lack of control of confounding factors. Hence, the aim of the present study was to examine interpersonal problems as a predictor of pain catastrophizing in a sample of patients with chronic pain, while controlling for relevant confounding variables such as gender, age, pain severity, psychiatric symptoms, fatigue, insomnia, adverse life experiences and social support.

Essentially, the communal coping model (CCM) suggests that pain catastrophizing may be a coping strategy to “maximize proximity or to solicit assistance or empathic responses from others” [1], and serve a strategic purpose similar to other communicative, pain-related behaviors such as vocalizations or facial displays of pain. However, whereas some studies have found support for a positive association between catastrophizing and solicitousness from significant others [11], [12], [13], other studies have been negative [9], [14], [15]. This mixed support may reflect a fundamental misconception in relying too heavily on principles from operant conditioning, where the presence or absence of pain catastrophizing is predicted to be contingent upon reward or punishment from the environment (e.g. attention, support) [8]. Research on anxiety disorders, for example, illustrates how avoidance may be an ineffective coping strategy that maintain or even increase anxiety symptoms over time, although commonly used and believed to be effective [16]. In a similar way, maladaptive coping behaviors of a more interpersonal nature may also persist even in the absence of any positive reinforcement from the environment. Pain catastrophizing may elicit attention and support from significant others in the short term, but the long-term outcomes may be reduced support or even punishing responses from significant others, especially if expressed in aversive, demanding and controlling ways causing others to react negatively [17], [18].

In this regard, interpersonal theory [19], [20] provides a useful paradigm for examining the role of maladaptive interpersonal behaviours in relation to catastrophizing. According to the interpersonal circumplex model (IIP; [21], [22]), interpersonal behaviors may be organized along two orthogonal dimensions labelled affiliation (horizontal axis, ranging from hostile to friendly behavior) and dominance (vertical axis, ranging from submissive to dominating behaviors). Four specific types of interpersonal problems may be derived by combining the two axes, representing ways of relating to others that may be either over- or underdeveloped: Hostile-Dominant, Hostile-Submissive, Friendly-Submissive and Friendly-Dominant. This allows for an examination of how specific types of interpersonal problems relates to pain catastrophizing, and some studies suggests a possible association between dominant, needy and controlling behaviors (i.e. Hostile-Dominant and/or Friendly-Dominant) and pain catastrophizing [10], [17], [18].

Chronic pain patients, however, in general typically report high levels of Hostile-Submissive and Friendly-Submissive interpersonal problems [23], associated with being overly submissive, exploitable and avoidant. This implies that patients with chronic pain generally score low on the IIP axis “dominance”, and one previous study using the IIP reported self-sacrificing interpersonal problems to predict pain catastrophizing even when controlling for general distress [8]. Although informative, the study utilized a sample of patients with irritable bowel syndrome, and a subscale from the coping strategies questionnaire (CSQ) [24] to measure pain catastrophizing. It did thus not target the broader population of chronic pain patients, and the CSQ taps mostly into pessimism whereas a broader conceptualization of catastrophizing is found in the pain catastrophizing scale (PCS) [25]. Furthermore, a more rigorous test of the CCM of pain catastrophizing should control for a larger number of potentially confounding variables, as pain catastrophizing is positively correlated with numerous variables such as pain intensity, general psychological distress and traumatic experience [7], as well as fatigue [26].

Utilizing a sample of patients with chronic pain entering a cognitive-behavioral rehabilitation program, we were thus interested in examining the unique influence of interpersonal problems (Hostile-Dominant, Hostile-Submissive, Friendly-Submissive, Friendly-Dominant) on pain catastrophizing in a cross-sectional study. After controlling for confounding variables such as demographics (age, gender), pain severity, comorbid symptoms (anxiety/depression, fatigue, insomnia), adverse life experiences and perceived social support, we hypothesized that higher levels of interpersonal problems would predict higher levels of pain catastrophizing in this patient sample.

2 Methods

3 Results

4 Discussion

The aim of the present study was to examine interpersonal problems as a predictor of pain catastrophizing in a sample of patients with chronic pain, controlling for confounding variables such as demographics (gender, age) pain severity (BPI), comorbid symptoms (anxiety/depression, fatigue, insomnia), adverse life experiences (ALE) and social support. A series of separate hierarchical regression analyses with each of the IIP Quadrants as predictors revealed that only levels of Hostile-Dominant interpersonal problems predicted levels of pain catastrophizing (PCS). That is, patients who reported more cold, vindictive, controlling and demanding interpersonal problems, also reported higher levels of pain catastrophizing. Hostile-Dominant interpersonal problems explained 6% of the unique variance in pain catastrophizing (d=0.59), not better accounted for by patient’s overall level of psychological distress (anxiety, depression, fatigue or insomnia), the severity of pain or social risk- or protective factors.

The results add support for a central tenet in the CCM, namely that pain catastrophizing may serve a communicative purpose aimed at maximizing the probability that distress is being managed (or attempted to be managed) within a social/interpersonal context [1], [43]. A Hostile-Dominant style, characterized by dominant, vindictive, self-centered and controlling behaviors, may be helpful in eliciting attention and support in the short-term, but is likely to backfire over time [15]. This may also help explain why research on the role of solicitousness, such as the display of pain behaviors and caregiving responses from the environment, has been mixed (e.g. [9], [11]). Indeed, as noted by Lackner and Gurtman [8], the CCM may draw too heavily on principles from operant conditioning [44], linking pain processes such as catastrophizing to reinforcing environmental variables such as attention, support, assistance or care.

In comparison, interpersonal theory suggests that people interact with their environment in a more transactional manner [33], and is theoretically more akin to attachment theory [45]. Both theories aim to describe and explain individual’s more enduring patterns of relating to others, such as the regulation of autonomy and attachment, and how interpersonal strategies aimed to meet these needs may be over- or underdeveloped. In this light, it is perhaps not surprising that the results from the present study are also in accordance with previous studies demonstrating pain catastrophizing to be associated with a support- and caretaking-demanding interpersonal style [8], [13], as well as with insecure- and/or anxious attachment styles [46], [47], [48]. Taken together, the results indicate that the interpersonal context of the individual is of importance and may help shed light on the functional role of pain catastrophizing amongst chronic pain patients.

The results from this study may be somewhat at odds with an earlier study [8], which found interpersonal problems in the Friendly-Submissive quadrant to predict pain catastrophizing. Bearing in mind that the two studies utilized different diagnostic groups (chronic pain vs. irritable bowel syndrome) and controlled for an unequal number of variables (8 vs. 1), a possible explanation for this discrepancy is that the studies utilized different measures of pain catastrophizing. The subscale from the CSQ taps primarily into the dimensions of helplessness and pessimism in the context of pain, which may be associated with a more passive, submissive and dependent interpersonal style. In comparison, the PCS is a broader measure of pain catastrophizing, tapping into dimensions of rumination and magnification (and not only helplessness), which may be more closely associated with an interpersonal “pull” for attention and support. Adding to this, of the four IIP quadrants, only the Friendly-Submissive quadrants did not demonstrate a significant zero-order correlation with the PCS (see Table 1).

Pain severity (BPI) remained a significant predictor in the final analysis of pain catastrophizing (see step 9 in Table 2), but the lack of a direct relationship between pain severity and each of the four IIP Quadrants (see Table 1) suggests that Hostile-Dominant interpersonal problems may be unrelated to the interpretation or experience of pain. This finding is interesting, and in accordance with the notion that people may primarily catastrophize to convey distress and elicit support from others; that is, to manage and cope with pain within an interpersonal context. Whereas this study did not examine to what extent catastrophizing de facto is an effective strategy to elicit support or attention, the negative correlation between social support, on the one hand, and pain catastrophizing (r=−0.22) as well as Hostile-Dominant interpersonal problems (r=−0.30) on the other hand (see Table 1), suggest that catastrophizing may be an ineffective strategy in this regard. It is also noteworthy that Hostile-Dominant interpersonal problems predicted pain catastrophizing even after controlling for any eventual protective effect of social support (see Table 2).

Whereas the sample consisted of patients with chronic pain, who mostly reported comorbid symptoms in the clinical range (psychiatric symptoms, pain intensity, insomnia, etc.), the level of pain catastrophizing (16.55) was below ≥23 which has been proposed as a cut-off differentiating “high” from “low” catastrophizers [49]. However, it is important to bear in mind that this study did not primarily target “high catastrophizers”, and a large proportion of patients with chronic pain are typically not classified as “high catastrophizers” [50]. Moreover, all participants in the present study were on sick leave with the goal of returning to work, whereas many studies on chronic pain conditions typically include large proportions of patients on disability/state pension.

Although interpersonal theory may add to our understanding of catastrophizing as a strategic coping strategy in chronic pain patients, it is important to note that this does not imply that an intrapersonal conceptualization of pain catastrophizing is without merit. Importantly, this study did not examine the central tenet in the CB model, namely that people catastrophize when they appraise a stressor as posing a threat that exceeds their coping capacity. An intrapersonal- and interpersonal conceptualization of pain catastrophizing are not mutually exclusive [10], and their relative emphasizes may differ between individuals. Moreover, the cumulative empirical evidence suggests that interpersonal problems, such as being demanding and controlling, as well as having an insecure/anxious attachment, may not be uniquely associated with the management of pain. Rather, they may reflect a more pervasive and habitual way of responding to emotional distress. In accordance with this (see Table 1), all IIP Quadrants correlated significantly with general psychiatric symptoms, fatigue and insomnia in the present study. However, it is of both theoretical and clinical importance that the effects of general psychiatric symptoms, fatigue and insomnia, were no longer statistically significant when all variables were entered in the final model (see Table 2), implying that interpersonal factors are perhaps more important than correlated symptoms.

The clinical implication of this study is that a reliance exclusively on an intrapersonal conceptualization of pain catastrophizing, risks missing out on an important piece of information, such as the patients’ way of relating to other people. High catastrophizers may have a controlling, demanding and vindictive style, which at least may be a maintaining factor for pain catastrophizing as well as the experience of pain and daily function. Depending on the particular patient, therapeutic interventions may need to take into account the social context of the individual, including spouse and/or family members, and there is a risk that treatment may be compromised by a controlling, interpersonal style on the part of the patient.

A Hostile-Dominant interpersonal style explained 6% of the variance in pain catastrophizing, whereas all included variables roughly explained 30% of the variance (see Table 2). As such, it is important to be cognizant of the fact that the larger share of the variance in pain catastrophizing was unexplained in the present study.