- Email: [email protected]
Chronic pain: a reformulation of the cognitive-behavioural model
Behaviour Research and Therapy 39 (2001) 787–800 www.elsevier.com/locate/brat
Chronic pain: a reformulation of the cognitive-behavioural model Timothy J. Sharp
*
Suite 46, Level 9, 88 Pitt Street, Sydney, NSW 2000, Australia Accepted 12 April 2000
Abstract The last few decades has seen psychological theories of chronic pain dominated by behavioural concepts and particularly, by the operant model as described by Fordyce et al. (1968: Fordyce, W., Fowler, R., & DeLateur, B. (1968) An application of behavior modification technique to a problem of chronic pain. Behaviour Research and Therapy, 6, 105–107, 1976: Fordyce, W.E. (1976) Behavioral methods for chronic pain and illness. St Louis, MO: C. V Moseby). More recently, cognitive constructs have been included, giving the impression that the operant theory of chronic pain has been replaced by a cognitive-behavioural theory (CBT). Given the fact that CBTs were introduced essentially to overcome some inherent problems with operant theory, it is notable that they are still very much based on operant theory and continue to be founded upon operant principles. Further, the extent to which CBTs (at this stage) have included contemporary cognitive research has been limited. This paper argues that it is of questionable validity to continue to include problematic concepts (e.g., operant principles) into a revised theory (e.g., CBT). Instead, consistent with research from other areas (particularly the anxiety disorders), chronic pain and the problems associated with it may be better explained by a reformulated cognitive-behavioural theory which although not ignoring the observable behaviours of pain patients and their associations with social reinforcers, interprets these phenomena from a cognitive perspective. Thus, a modified CBT is proffered, focusing more directly on patients’ thoughts about, and appraisals of, their pain. Evidence in support of such a theory is provided, as are suggestions for further research and the implications such a theory has for treatment. 2001 Elsevier Science Ltd. All rights reserved. Keywords: Cognitive; Behavioural; Chronic pain
* Tel.: +61-2-9231-2522; fax: +61-2-9231-2533. E-mail address: [email protected] (T.J. Sharp).
0005-7967/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 5 - 7 9 6 7 ( 0 0 ) 0 0 0 6 1 - 9
788
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
1. Brief history of the presenting problem Definitions of pain have varied considerably over the years and despite significant advances in medical technology, including the development of sophisticated scanning techniques, general agreement on what pain is and what it means continues to evade clinicians and researchers (Merskey & Bogduk, 1994). Although popular and medical opinion typically links pain and tissue damage, the evidence for this association is questionable, at best. The link between reports of pain and identifiable pathology has been investigated and the findings have cast serious doubt on theories that attempt to understand pain as a purely physical phenomenon. Jensen et al. (1994), for example, found that many people with ‘abnormalities’ identified by MRI (e.g., disk bulges) did not report back pain. At the same time, Spitzer, LeBlanc et al. (1987; cited in Fordyce, 1995) report that many people complaining of back pain do not have identifiable pathology. In fact, the best evidence suggests that “fewer than 15% of persons with back pain can be assigned to one of the categories of specific low back pain” (Fordyce, 1995, p. 3). Thus, although they are no doubt related, the association between physical impairment and reports of pain is only moderate (Waddell & Main, 1984). The fact that physical pathology does not perfectly predict pain severity has long been recognised, as has the fact that different people respond in varying ways to chronic pain (Turk, 1996a). For example, not all pain patients are depressed and disabled to the same degree (Turner & Romano, 1984). Notably, the association between pain and depression and disability appears to be mediated by psychosocial variables such as perceived life-control (Fordyce, 1995; Lackner, Carosella & Feuerstein, 1996; Romano et al., 1995; Rudy, Kerns & Turk, 1988; Waddell, Newton, Henderson, Somerville & Main, 1993). Taken together, therefore, the evidence suggests that pathology alone cannot explain reports of pain, pain-related disability or depression. As a result, multidimensional models have been proffered to better account for the experimental findings and clinical phenomenology. The Gate Control Theory (Melzack & Wall, 1965), for example, significantly changed the way pain was conceptualised and contributed to the International Association for the Study of Pain (Merskey, 1979) defining pain as an “unpleasant sensory and emotional experience …”. The use of words such as ‘unpleasant’ and ‘emotional experience’ are crucial in this definition as they emphasise the influence of psychological variables. More recently, other multidimensional models have been suggested, with one of the most popular being attributed to Loeser (1982; see Waddell et al., 1993). Notably, this model makes the important distinction between sensory, cognitive, affective, behavioural and social variables. This multidimensional, biopsychosocial model drew heavily on the work of Fordyce et al. who introduced behavioural principles to the study of pain. 2. Operant theory Fordyce et al. (1968, 1976), drawing on the work of other behaviourists (e.g., Skinner, 1953) suggested a distinction be made between the original cause of pain and reports of, and displays of pain (known as pain behaviours). Although a complete review of the behavioural model is beyond the scope of this paper, its basic assumption is that pain behaviours (e.g., limping, grimacing) are subject to the same influences of conditioning as are any other behaviours (see
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
789
Fordyce, 1996). As such, if followed by reinforcing consequences, pain behaviours may persist for longer than the normal expected healing time (see Sanders, 1996). Although it has generally been accepted that there is good evidence to support treatments based on the operant model (American Psychological Association, 1993; Keefe, 1994; Morley, Eccleston & Williams, 1999) there is less agreement regarding how they work and the strength of the evidence to support the underlying theory upon which this intervention is based (Turk & Flor, 1984). Support of the operant theory is typified by studies that suggest behavioural interventions are effective (Cairns & Pasino, 1977), that reports of pain may be cued by the presence of a spouse (Block, Kremer & Gaylor, 1980), and that spouse responses are associated with pain behaviours (Romano et al., 1992). There are, however, several problems with these studies. Turk (1996b), for example, recently outlined a number of ‘concerns’ about the operant perspective which include (1) the questionable validity of the pain behaviour construct, (2) the lack of specificity of the pain behaviour construct, (3) the common assumption that observed pain behaviours are maladaptive, (4) the potential for pain to be underreported and the scope for this to be detrimental, (5) some patients’ lack of acceptance of operant treatment and/or dissatisfaction with operant treatment goals, (6) the focus on motor behaviour and (7) problems of maintenance of behaviours following treatment. In summary, the first few of these concerns relate to the fact that pain behaviours require interpretation to determine their raison d’eˆ tre, while the latter few refer to the findings that behavioural treatments are not effective for all patients and/or that even when they initially succeed, are often followed by relapse (Turk & Rudy, 1991). As these issues have each been discussed in considerable detail by Turk (1996b) they will not be re-stated here. Some additional problems, however, will be outlined. First, with respect to the operant theory of chronic pain, much weight has been given to the results of several key studies. These include those conducted by Cairns and Pasino (1977), Block et al. (1980) as well as the series of studies by Romano et al. (1991, 1992, 1995). The first (Cairns & Pasino, 1977) can be considered as an example of studies that essentially claim the reduction of ‘pain behaviours’ via the modification of environmental contingencies supports the principles of the operant model. This may well be true, but it neglects to acknowledge that there are other ways of interpreting these findings. The operant model, for example, does not acknowledge the possibility that patients’ interpretations of the environmental changes may be important. This possibility raises the prospect of cognition playing a role. It is similar to the argument that because antidepressants improve mood (in some patients), then depression is therefore a biological condition (Salkovskis, 1996a has applied a similar argument to theoretical problems in the anxiety disorders). Neither argument is necessarily logical, as there may well be multiple factors that influence behaviour. Simply because pain behaviours may change within pain programmes, does not necessarily mean that they are affected by the supposed contingencies ‘designed’ by the staff. Few, if any, investigators, for example, have reflected on the fact that many treatments have typically been effective regardless of the fact that they invariably devote little (or no) time to modifying spouse behaviours outside the clinic (Nicholas, Wilson & Goyen, 1992). Further, interventions that have specifically targeted spouses have not proven to be any more effective than standard patient interventions (Moore & Chaney, 1985). Either way, there is a logical flaw in the assumption that the operant model is supported simply because behavioural treatment has an effect on patients’ behaviour.
790
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
The Block et al. (1980) study has also been widely cited as providing support for the behavioural model. As has been noted by Paulsen and Altmaier (1995), however, a close reading of Block et al.’s data reveals findings not entirely consistent with their conclusions. Of most concern is the statement that patients systematically alter their report of pain depending on the presence of, and their perceptions of, their spouses. It is questionable whether such a conclusion can be made given that there was no significant difference between reports of pain in either condition for patients in the solicitous spouse group. Further, careful examination reveals that the reason for the interaction effect was not that patients in the solicitous spouse group reported more pain in the presence of their spouses, but that the patients in the non-solicitous spouse group reported less pain. Rather than indicating that solicitous responses increase the likelihood of higher pain intensity reports, this may suggest that non-supportive responses decrease the likelihood of reporting pain! One must also question the validity and generalisability of these results when so little descriptive data have been provided about the patients and the distribution in the two groups. The authors themselves acknowledge a significant group difference in pain duration. Taking these limitations into account, the degree to which this study provides support for the operant model must surely be questionable. Even the findings of a more recent study that attempted to re-test the conclusions of Block et al. (1980) were only partially consistent with hypotheses generated from an operant model (Lousberg, Schmidt & Groenman, 1992). In this investigation, Louseberg et al. found that patients’ perceptions of spouse solicitousness were not significantly associated with pain behaviours although interestingly, the partners’ perceptions were more closely related. Finally, a series of studies by Romano et al. (1991, 1992, 1995) has specifically assessed the relationship between patients’ pain behaviours and spouse responses using a sophisticated direct observation system. Although their methodology is in many ways beyond reproach, the validity and the generalisability of their findings are unclear due to the questionable representativeness of their sample. Romano et al. (1991) note that “Of the 233 chronic pain patients who met the study inclusion criteria and were approached to participate in the study, 50 agreed to participate”. Although they state that those who agreed to participate did not differ from those who refused in age or gender, they provide no information about pain severity, depression or disability. Thus, the extent to which this sample can be compared to a typical pain center population is unknown. Although ostensibly disappointing for the operant model, these findings are still promising from a psychological perspective as they are consistent with a cognitive model (to be described below) which among other things, hypothesises that if spouse responses were to have an influence they would be mediated by patients’ interpretations (Sharp, 1997). That is, if spouse responses are interpreted by two different patients as meaning “I can’t do anything without her”, or “With her help I will be able to do more”, they will have quite different effects. The implications of these interpretations for patients’ and for treatment will be discussed below. In summary, although there is good evidence to support the efficacy of behavioural treatments, empirical support for the operant theory may be more uncertain than is often acknowledged. In response to some of the problems described above, and given the developments and advances in other areas of psychological theory over the last decade or so (see Hawton, Salkovski, Kirk & Clark, 1989; Salkovskis, 1996b), many clinicians and researchers have begun to refer to ‘cognitive behavioural theories’ of pain.
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
791
3. Cognitive-behavioural theories Turk, Meichenbaum and Genest (1983) were among the first to develop a cognitive behavioural model of pain that drew on the work of Beck (1976), Beck, Rush, Shaw and Emery (1979) and Meichenbaum (1977). Beginning with the observation that behavioural treatments were not exclusively behavioural, the authors noted that even Fordyce (1976) referred to the importance of reaching a ‘shared conceptualisation’ with patients. They also noted that the founders of behaviour therapy (e.g., Wolpe, 1958; Wolpe & Lazarus, 1966) emphasised the need to ‘remove patients mistaken beliefs’. Consistent with these sentiments, there is now good evidence to support the influence of cognitive factors. Cognitions have been found to influence reports of pain intensity, attempts to cope (Jensen, Turner & Romano, 1991), mood and pain-related disability (Turk & Rudy, 1992; Williams & Keefe, 1991; Wilson, Henry & Nicholas, 1993). A number of key themes have emerged from the research into pain and cognitions. Patients who interpret their pain as being indicative of more damage tend to be more avoidant and more disabled (Vlaeyen, Kole-Snijders, Boeren & van Eek, 1995). Similarly, catastrophic interpretations of pain have been found to be associated with higher levels of depression (Rosensteil & Keefe, 1983) and the affective component of pain (Geisser, Robinson, Keefe & Weiner, 1994). Further, predictions of pain, self-efficacy, and perceptions of control have all been found to be related to depression and behavioural performance (Arntz & Peters, 1995; Dolce, Crocker, Molettiere & Doleys, 1986; Rudy et al., 1988). Taken together, these studies provide strong support for the inclusion of cognitive factors in a model of chronic pain. Although not suggesting cognitions cause pain, there is strong support for the notion that pain related cognitions and beliefs are associated with the utilisation of helpful or unhelpful coping strategies and thereby affect the extent to which pain impacts on patients’ lives. At the same time, however, behavioural constructs have continued to dominate theory and treatment. While describing cognitive behavioural interventions, both Turk (1996a, 1996b) and Bradley (1996), for example, specifically refer to the fact that “environmental factors can also influence behaviour” (Turk, 1996a, p. 93). Bradley’s intervention outlines 15 weeks of treatment, and describes attending to patients’ and spouses’ behaviours, relaxation training, biofeedback, goal setting and relapse prevention. Notably, the identification and evaluation of unhelpful cognitions or beliefs is not explicitly mentioned. Turk et al. (1983) describe a very similar programme and also do not include specific strategies for the modification of cognitions. These examples seem to exemplify the relative failure of cognitive behaviour therapy of chronic pain to adequately incorporate cognitive constructs. At this stage, no theory has attempted to explain (or even noted) the persistence of catastrophic thoughts despite patients undergoing repeated experiences that should have challenged their validity. For example, the catastrophic thought ‘I can’t cope’ should have undergone re-evaluation following ‘survival’ of numerous pain flare-ups. Yet such a thought is common in patients many years after the onset of their pain. Further, despite the questions raised earlier, cognitive behavioural theory has continued to be founded on behavioural principles and cognitive behavioural interventions have continued to be dominated by behavioural components. It is debatable whether adding on a session of cognitive therapy really makes a treatment cognitive (Salkovskis, 1986; 1987). The failure to adequately modify pain management interventions is particularly important given the risk that it may be
792
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
jeopardising treatment gains. This is most notable given the advances cognitive treatments have recently made in the treatment of the anxiety disorders (see Salkovskis, 1996b). At the same time, however, questioning the evidence for the operant model does not necessarily imply that behaviour is not important. Anxiety provides a fascinating example of the link between cognition and behaviour and of how behaviour is still important in a cognitive framework (Salkovskis, 1991). 4. Lessons from the anxiety research The Oxford Group (e.g., Salkovskis, Clark & Gelder, 1996) have provided empirical support for the notion that behaviours maintaining anxiety can be viewed from within a cognitive model (Salkovskis, 1991). Salkovskis describes the relationship between ‘threat cognitions’ and ‘safety seeking behaviours’, arguing that certain behaviours are understandable given a belief that danger is imminent. Avoidance, for example, is not conceptualised as a response to (conditioned) anxiety that is maintained by negative reinforcement (Rachman, 1977). Instead, avoidance is conceptualised as a response that “arises out of, and is logically linked to, the perception of serious threat. [Further] because safety seeking behaviour is perceived to be preventative, and focused on especially negative consequences … disconfirmation of threat is made particularly unlikely” (Salkovskis, 1991, p. 6). Behaviours are thus seen as an integral component in the maintenance of anxiety, within a cognitive framework. Given that the associated cognitive therapies have been found to be more effective, in certain cases, than the more traditional behavioural treatments (Salkovskis & Warwick, 1985; Salkovskis & Westbrook, 1989), it is possible that their application to pain may similarly improve outcomes which although impressive in some cases (e.g., Williams et al., 1996), could indubitably be improved (Turk & Rudy, 1991). 5. Towards a reformulation of the cognitive-behavioural conceptualisation The modified cognitive-behavioural model begins with the proposition that problems associated with chronic pain originate in the way patients react to their pain. Reactions are defined as including all forms of cognition (e.g., pain-related thoughts and imagery), not simply observable behaviours. It is proposed that the difference between patients who are distressed and/or disabled, and those for whom pain is not markedly problematic, lies not necessarily in the sensory activity, but rather in the patients’ appraisals and interpretations of the pain. Fig. 1 represents the reformulated cognitive-behavioural model in which patients’ appraisals of their pain (the meaning ascribed to the pain by the individual) are ascribed a primary role. As indicated in the model, the various interactions can become self-maintaining such that distress and disability, as well as physiological arousal, may continue despite the absence of the original sensory input. Avoidance and/or safety behaviours may interfere with disconfirmation of and thereby maintain certain cognitions or beliefs. At the same time, anxiety and distress may maintain autonomic arousal that may confirm beliefs that some underlying condition exists. Anxiety and depression may then increase the likelihood of patients making cognitive errors or negative appraisals, as well as potentially maintaining avoidance, and the cycle would thereby continue. The model also takes into account the likelihood that stress and iatrogenic factors may exacer-
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
793
Fig. 1. The reformulated cognitive-behavioural model of chronic pain.
bate anxiety and arousal, and/or maintain certain behaviours. Borrowing from models of hypochondriasis (Salkovskis & Bass, 1997), the model emphasises patients’ beliefs about their symptoms and the tendency to seek reassurance. Many pain patients, for example, express fears about experiencing more damage and believe that more investigations or treatments are needed. According to the reformulated model, the ways doctors respond to this may inadvertently reinforce patients’ anxieties and may unwittingly reinforce excessive disability and a passive approach (Kouyanou, Pither, Rabe-Hesketh & Wessley, 1998; Pither & Nicholas, 1991).
794
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
Given that this conceptualisation borrows considerably from other cognitive models (e.g., Clark, 1986; Salkovskis & Warwick, 1986), the parallel can be taken further by adapting hypotheses related to patients’ ‘meta-cognitions’ (Wells, 1995). The current model hypothesises that if patients interpret the presence of pain itself or of pain-relevant thoughts as indicating something negative about their condition or persona (e.g., “Thinking about pain means my pain is serious” or “It is so horrible thinking about the pain all the time … if I can’t stop it I’ll go crazy” or “All I can think about is the pain, I can’t focus on anything else, I’m so hopeless”) then they may attempt to suppress or neutralise such thoughts (note: this can be seen as a form of safety behaviour). The model further assumes that attempts to suppress (e.g., “not think about the pain”) or neutralise (e.g., attempts to protect oneself from going crazy by controlling one’s mind or by using distraction) pain-related thoughts may lead to an increase in their frequency and in their perceived aversiveness (Salkovskis & Campbell, 1994; Salkovskis, Westbrook, Davis, Jeavons & Gledhill, 1997). If, as is frequently the case, pain persists and more investigations and treatments prove unhelpful, it is hypothesised that a form of learned helplessness may develop. That is, patients could start to believe that ‘nothing has worked so far so why would any future treatment help?’. Patients’ holding such a belief may not try as hard in treatment, which may increase the chances of the treatment not helping, which may in turn reinforce their belief that ‘nothing is going to help’. Patients’ could then be labeled as ‘not motivated’. This has important treatment implications, because labeling patients in this way may affect the way they are assessed and/or the treatments they are offered. Alternatively, if viewed from within the proposed model (i.e., as patients with unhelpful and unrealistic cognitions), these patients could be seen as appropriate candidates for cognitive therapy, and therefore as salvageable. It should be noted that these pain-related cognitions and beliefs are assumed to develop as a result of patients’ early learning histories and cultural backgrounds. That is, depending upon their prior experiences with pain, or knowledge of other people with pain, patients will have learned that pain has certain meanings. Depending upon how they have seen others respond to pain, or on how pain is perceived within their particular culture, they may learn to interpret and/or to evaluate it in different ways (see, for example, Bates, 1987; Bates, Edwards & Anderson, 1993; Craig, 1978). Research investigating the role of significant other responses can also be reinterpreted within the modified cognitive-behavioural model. Spouse responses could be conceptualised as providing evidence for, or against, patients’ beliefs. If patients believe they are disabled, and their spouse is overly solicitous, this may confirm patients’ beliefs while also preventing patients from disconfirming their own perceptions of disability. Similarly, medication use which has traditionally been understood as a behaviour maintained by negative reinforcement can be reinterpreted as a safety behaviour in response to the beliefs “I can’t cope … this pain is unbearable”. Real or perceived relief may interfere with effective evaluation of the belief. Importantly, the model incorporates the possibility of learning without awareness. The model would also predict that these cognitions and the associated behaviours are inextricably linked and are mutually maintaining. Patients who avoid engaging in tasks because they are afraid of further damage, will never gather disconfirming information and will continue to believe their pain is indicative of danger. Alternatively, if they do attempt tasks their beliefs may still be maintained if they engage in protective behaviours (such as lifting a box but only while
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
795
wearing a back brace). Rather than disproving the danger of lifting, such experiences may simply confirm that lifting is dangerous and would have lead to more damage if only the brace had not provided the necessary protection. 6. Support for the reformulated cognitive-behavioural model There are several advantages in viewing chronic pain from within a more cognitive framework. First, it is suggested that this model better explains the existing evidence. Second, the model makes clear and testable predictions, including predictions about treatment and about what should work for individual patients. Theoretically, identifying and targeting key cognitions should lead to more effective interventions than simply using the ‘blunderbuss’ approach typically offered (Turk, 1990). Most importantly, the new cognitive behavioural model is well supported by, and consistent with a range of recent empirical findings. Of particular interest are a series of studies by Eccleston (1995; Eccleston, Crombez, Aldrich & Stannard, 1997) and Crombez (Crombez, Eccleston, Baeyens & Eelen, 1998a, 1998b). In brief, these studies tend to support the notion that if patients are worried about their pain (referred to as ‘catastrophisers’), they are more likely to be hypervigilant (and be more somatically aware), which not surprisingly interferes with attention but which also “amplifies somatosensory information and primes fear mechanisms” (Crombez et al., 1998a, p. 187). This supports the hypothesised relationship between anxiety and hypervigilance. Further, anxiety has been found to predict ‘non-specific physical complaints’ in pain patients, a finding consistent with patients being hypervigilant (Asmundson, Kuperos & Norton, 1997; McCracken, Faber & Janeck, 1998). Fear of pain has also been found to be a better predictor of avoidance than pain severity or physical pathology (Crombez Vervaet, Lysens, Baeyens & Eelen, 1998). The authors interpreted their findings as being more consistent with a cognitive model than an operant one and there is a growing body of support for such a model (Asmundson, Norton & Allerdings, 1997; Vlaeyen, Kole-Snijders, Rottveel, Ruesink & Heuts, 1995). There is also some evidence to suggest that attempts to suppress pain-related thoughts may be associated with heightened pain experience (Sullivan, Rouse, Bishop & Johnston, 1997). This research, however, was conducted on a non-clinical population and so the findings do need to be treated with some caution. Nevertheless, there is good evidence from the anxiety research that attempts to suppress thoughts are not helpful in the long term, are likely to exacerbate and maintain distress, and do hold true in clinical populations (Harvey & Bryant, 1998; Rutledge, 1998). The role of autonomic arousal in the model has also received relatively strong support. In a comprehensive review of the literature on psychophysiology of pain Flor and Turk (1989) concluded that although chronic pain patients do not necessarily differ from normals on physiological measures at baseline, they do tend to differ from non-pain patients in response to pain and/or stress. Flor, Turk and Birbaumer (1985) asked subjects with chronic low back pain to discuss personally stressful (and/or painful) situations. The investigators measured subjects’ heart rate and skin resistance level changes. Among other things, they found that compared to neutral tasks (such as mental arithmetic), stressful tasks were associated with significant increases in paraspinal EMG levels in pain patients, but not in normals. Unfortunately, the methodologies of other studies
796
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
investigating these factors have been criticised (Flor & Turk, 1989) and it is therefore necessary for these findings to be replicated. Nevertheless, these findings are consistent with a model in which the relationship between pain and arousal is mediated by appraisal (Flor et al., 1985; Turner, Clancy & Vitaliano, 1987). Finally, the role of iatrogenic factors has received some support from research that, among other things, has found that inappropriate and excessive use of medications and investigations may be associated with patients’ reports of symptomatology and with disability (Kouyanou et al., 1998; Pither and Nicholas, 1991). Kouyanou et al. compared chronic pain patients whose symptoms were considered to be medically explained with those whose symptoms could not be explained. They found significant differences on several variables including a higher prevalence of over-investigation and over-prescribing in the group whose symptoms were not explicable. They hypothesised that the management of these patients was often unhelpful and that by frequently ‘disconfirming’ patients’ pain and by suggesting it was ‘all in the mind’ they were, in fact, maintaining the patients’ sick role. The authors suggested that this could be attributed to patients’ determination to legitimise their pain by seeking more investigations and treatments. The salient point for the proposed model is that the responses of the medical profession can influence patients’ beliefs about their pain and subsequently, their pain-related treatment-seeking behaviours. Taken together, a substantial body of research now exists to support the conceptualisation of chronic pain in a manner similar to that which is generally accepted for anxiety. If one equates pain with intrusive thoughts, appraisals with cognitions, and attempts to neutralise with pain safety behaviours such as avoidance then chronic pain can be understood using similar constructs as those applied to obsessional problems (Salkovskis, 1985). This suggests that pain management could be improved as has the treatment of obsessions (Salkovskis & Westbrook, 1989). There is, however, undoubtedly a need for more research. Most notably, the concept of ‘safety behaviours’ needs to be clearly and specifically established within pain patients. Although the potential role of safety seeking behaviours has been hypothesised (Sharp, in press), investigations examining the specific predictions of this part of the model are yet to be conducted. In addition, individually tailored cognitive interventions need to be tested against the more general programmes now on offer, to assess their relative efficacy. (Note: as this paper was in preparation, a preliminary investigation along these lines was completed by Vlaeyen, de Jong, Geilen, Heuts & van Breukelen, 2000.) 7. Conclusions This paper has attempted to trace the development of psychological theories of chronic pain starting with the operant model, through to a reformulated cognitive behavioural model that parallels models proffered for the anxiety disorders (Salkovskis, 1996a, 1996b). In reviewing the relevant evidence, it appears that the operant model is not as well supported as has been generally thought and that even some cognitive-behavioural models suffer many of the problems their predecessors do because they have continued to rely on outdated behavioural principles. Accordingly, a reformulated cognitive behavioural theory is outlined which gives substantially more attention to patients’ thoughts about and appraisals of their pain. Patients’ behaviours are
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
797
still considered important, but from the point of view of their relationship to cognitions and particularly with regards to the ways they may maintain pain-related cognitions. It is argued that this new model is relatively well supported by the empirical findings and that it explains contemporary research better than does the operant theory. Although more research is needed, the cognitive behavioural model presented here proffers specific hypotheses and provides clear directions for future investigations.
References American Psychological Association Division of Clinical Psychology (Division 12) (1993, October). Task Force Report on the Promotion and Dissemination of Psychological Procedures: Final Report Arntz, A., & Peters, M. (1995). Chronic low back pain and inaccurate predictions of pain: is being too tough a risk factor for the development and maintenance of chronic pain? Behaviour Research and Therapy, 33, 49–53. Asmundson, G., Kuperos, J., & Norton, G. (1997). Do patients with chronic pain selectively attend to pain-related information? Preliminary evidence for the mediating role of fear. Pain, 72, 27–32. Asmundson, G., Norton, G., & Allerdings, M. (1997). Fear and avoidance in dysfunctional chronic back pain patients. Pain, 69, 231–236. Bates, M. (1987). Ethnicity and pain: a biocultural model. Social Science Medicine, 24, 47–50. Bates, M., Edwards, W., & Anderson, K. (1993). Ethnocultural influences on variation in chronic pain perception. Pain, 52, 101–112. Beck, A. (1976). Cognitive therapy and the emotional disorders. New York: International University Press. Beck, A., Rush, J., Shaw, B., & Emery, G. (1979). Cognitive therapy of depression. New York: The Guilford Press. Block, A., Kremer, E., & Gaylor, M. (1980). Behavioral treatment of chronic pain: the spouse as a discriminative cue for pain behavior. Pain, 9, 243–252. Bradley, L. (1996). Cognitive-behavioral therapy for chronic pain. In R. Gatchel, & D. Turk, Psychological approaches to pain management: a practitioner’s guide. New York: The Guilford Press. Cairns, D., & Pasino, J. (1977). Comparison of verbal reinforcement and feedback in the operant treatment of disability due to chronic low back pain. Behaviour Therapy, 8, 621–630. Clark, D. (1986). A cognitive approach to panic. Behaviour Research and Therapy, 24, 461–470. Craig, K. (1978). Social modeling influences on pain. In R. Sternbach, The psychology of pain. Raven Press: New York. Crombez, G., Eccleston, C., Baeyens, F., & Eelen, P. (1998a). When somatic information threatens, catastrophic thinking enhances attentional interference. Pain, 75, 187–198. Crombez, G., Eccleston, C., Baeyens, F., & Eelen, P. (1998b). Attentional disruption is enhanced by the threat of pain. Behaviour Research and Therapy, 36, 195–204. Crombez, G., Vervaet, L., Lysens, R., Baeyens, F., & Eelen, P. (1998). Avoidance and confrontation of painful, backstraining movements in chronic back pain patients. Behavior Modification, 22, 62–77. Dolce, J., Crocker, M., Molettiere, C., & Doleys, D. (1986). Exercise quotas, anticipatory concern and self-efficacy expectations in chronic pain: a preliminary report. Pain, 24, 365–372. Eccleston, C. (1995). Chronic pain and distraction: an experimental investigation into the role of sustained and shifting attention in the processing of chronic persisting pain. Behaviour Research and Therapy, 33, 391–405. Eccleston, C., Crombez, G., Aldrich, S., & Stannard, C. (1997). Attention and somatic awareness in chronic pain. Pain, 72, 209–215. Flor, H., & Turk, D. (1989). Psychophysiology of chronic pain: do chronic pain patients exhibit symptom-specific psychophysiological responses? Psychological Bulletin, 105, 215–259. Flor, H., Turk, D., & Birbaumer, N. (1985). Assessment of stress-related psychophsiological reactions in chronic back pain patients. Journal of Consulting and Clinical Psychology, 53, 354–364. Fordyce, W., Fowler, R., & DeLateur, B. (1968). An application of behavior modification technique to a problem of chronic pain. Behaviour Research and Therapy, 6, 105–107. Fordyce, W. E. (1976). Behavioral methods for chronic pain and illness. St Louis, MO: C. V. Moseby.
798
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
Fordyce, W. E. (1995). Back pain in the workplace: management of disability in nonspecific conditions. Seattle: IASP Press. Fordyce, W. E. (1996). Chronic pain: the behavioral perspective. In M. Cohen, & J. Campbell, Pain treatment centers at a crossroads: a practical and conceptual reappraisal. Seattle: IASP Press. Geisser, M., Robinson, M., Keefe, F., & Weiner, M. (1994). Catastrophising, depression and the sensory, affective and evaluative aspects of chronic pain. Pain, 59, 79–83. Harvey, A., & Bryant, R. (1998). The effect of attempted thought suppression in acute stress disorder. Behaviour Research and Therapy, 36, 583–590. Hawton, K., Salkovskis, P., Kirk, J., & Clark, D. (1989). Cognitive behaviour therapy for psychiatric problems: a practical guide. Oxford: Oxford University Press. Jensen, M., Brant-Zawadzki, M., Obuchowski, N., Modic, M., Malkasian, D., & Ross, J. (1994). Magnetic resonance imaging of the lumbar spine in people without back pain. The New England Journal of Medicine, 331, 69–73. Jensen, M., Turner, J., & Romano, J. (1991). Self-efficacy and outcome expectancies: relationships to chronic pain coping strategies and adjustment. Pain, 44, 263–269. Keefe, F., Crisson, J., Urban, B., & Williams, D. (1990). Analyzing chronic low back pain: the relative contribution of pain coping strategies. Pain, 40, 293–301. Kouyanou, K., Pither, C., Rabe-Hesketh, S., & Wessley, S. (1998). A comparative study of iatrogenesis, medication abuse, and psychiatric comorbidity in chronic pain patients with and without medically explained symptoms. Pain, 76, 417–426. Lackner, J., Carosella, A., & Feuerstein, M. (1996). Pain expectancies, pain, and functional self-efficacy expectancies as determinants of disability in patients with chronic low back disorders. Journal of Consulting and Clinical Psychology, 64, 212–220. Loeser, J. (1982). Concepts of pain. In M. Stanton-Hicks, & R. Boas, Chronic low back pain. New York: Raven Press. Lousberg, R., Schmidt, A., & Groenman, N. (1992). The relationship between spouse solicitousness and pain behaviour: Searching for more experimental evidence. Pain, 51, 75–79. McCracken, L., Faber, S., & Janeck, A. (1998). Pain-related anxiety predicts non-specific physical complaints in persons with chronic pain. Behavior Research and Therapy, 36, 621–630. Meichenbaum, D. (1977). Cognitive behavior modification: an integrative approach. New York: Plenum. Melzack, R., & Wall, P. (1965). Pain mechanisms: a new theory. Science, 50, 971–979. Merskey, H. (1979). Pain terms: a list with definitions and notes on usage. Recommended by the IASP subcommittee on taxonomy. Pain, 6, 249–252. Merskey, H., & Bogduk, N. (1994). Classification of chronic pain — descriptions of chronic pain syndromes and definitions of pain terms. (2nd ed.). Seattle: IASP Press. Moore, J., & Chaney, E. (1985). Outpatient group treatment of chronic pain: effects of spouse involvement. Journal of Consulting and Clinical Psychology, 53, 326–334. Morley, S., Eccleston, C., & Williams, A. (1999). Systematic review and meta-analysis of randomised controlled trials of cognitive behaviour therapy and behaviour therapy for chronic pain in adults, excluding headache. Pain, 80, 1–14. Nicholas, M., Wilson, P., & Goyen, J. (1992). Comparison of cognitive-behavioural group treatment and an alternative non-psychological treatment for chronic low back pain. Pain, 48, 339–347. Paulsen, J., & Altmaier, E. (1995). The effects of perceived versus enacted social support on the discriminative cue function of spouses for pain behaviours. Pain, 60, 103–110. Pither, C., & Nicholas, M. (1991). The identification of iatrogenic factors in the development of chronic pain syndromes: abnormal treatment behaviour? In Bond, Charlton, & Woolf, Proceedings of the Fifth World Congress on Pain. Elsevier Science Publishers. Rachman, S. (1977). The conditioning theory of fear acquisition: a critical examination. Behaviour Research and Therapy, 15, 375–387. Romano, J., Turner, J., Friedman, L., Bulcroft, R., Jensen, M., & Hops, H. (1991). Observational assessment of chronic pain patient-spouse behavioural interactions. Behavior Therapy, 22, 549–567. Romano, J., Turner, J., Friedman, L., Bulcroft, R., Jensen, M., Hops, H., & Wright, S. (1992). Sequential analysis of chronic pain behaviours and spouse responses. Journal of Consulting and Clinical Psychology, 60, 777–782. Romano, J., Turner, J., Jensen, M., Friedman, L., Bulcroft, R., Hops, H., & Wright, S. (1995). Chronic pain patientspouse behavioral interactions predict patient disability. Pain, 63, 353–360.
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
799
Rosensteil, A., & Keefe, F. (1983). The use of coping strategies in chronic low back pain patients: relationship to patient characteristics and current adjustment. Pain, 17, 33–44. Rudy, T., Kerns, R., & Turk, D. (1988). Chronic pain and depression: toward a cognitive-behavioral mediation model. Pain, 35, 129–140. Rutledge, P. (1998). Obsessionality and the attempted suppression of unpleasant personal intrusive thoughts. Behavior Research and Therapy, 36, 403–416. Salkovskis, P. (1985). Obsessional-compulsive problems: a cognitive-behavioural analysis. Behaviour Research and Therapy, 23, 571–583. Salkovskis, P. (1986). The cognitive revolution: new way forward, backward somersault or full circle? Behavioural Psychotherapy, 14, 278–282. Salkovskis, P. (1987). Cognitive therapy: is it always cognitive, is it always therapy? In H. Dent, Clinical psychology: research and development. London: Croom Helm. Salkovskis, P. (1991). The importance of behaviour in the maintenance of anxiety and panic: a cognitive account. Behavioural Psychotherapy, 19, 6–19. Salkovskis, P. (1996a). Avoidance behaviour is motivated by threat beliefs: a possible resolution of the cognitionbehaviour debate. In P. Salkovskis, Trends in cognitive and behavioural therapies. Chicester: John Wiley & Sons. Salkovskis, P. (1996b). The cognitive approach to anxiety: threat beliefs, satety-seeking behaviour, and the special case of health anxiety and obsessions. In P. Salkovskis, Frontiers of cognitive therapy (pp. 48–74). London: The Guilford Press. Salkovskis, P., & Bass, C. (1997). Hypochondriasis. In D. Clark, & C. Fairburn, Cognitive behavioural therapy: science and practice. Oxford: Oxford University Press. Salkovskis, P., & Campbell, P. (1994). Thought suppression in naturally occuring negative intrusive thoughts. Behaviour Research and Therapy, 32, 1–8. Salkovskis, P., Clark, D., & Gelder, M. (1996). Cognition-behaviour links in the persistence of panic. Behaviour Research and Therapy, 34, 453–458. Salkovskis, P., & Warwick, H. (1985). Cognitive therapy of Obsessive-compulsive disorder: treating treatment failures. Behavioural Psychotherapy, 13, 243–255. Salkovskis, P., & Warwick, H. (1986). Morbid preoccupations, health anxiety and reassurance: a cognitive-behavioural approach to hypochondriasis. Behaviour, Research and Therapy, 24, 597–602. Salkovskis, P., & Westbrook, D. (1989). Behaviour therapy and obsessional ruminations: can failure be turned into success? Behaviour, Research and Therapy, 27, 149–160. Salkovskis, P., Westbrook, D., Davis, J., Jeavons, A., & Gledhill, A. (1997). Effects of neutralising on intrusive thoughts: an experiment investigating the etiology of obsessive-compulsive disorder. Behaviour, Research and Therapy, 35, 211–219. Sanders, S. (1996). Operant conditioning with chronic pain: back to basics. In R. Gatchel, & D. Turk, Psychological approaches to pain management: a practitioner’s guide. New York: The Guilford Press. Sharp, T. (1997). Chronic pain patients and their spouses: an analysis of their interactions. Unpublished Doctoral Dissertation Sharp, T. The safety behaviours construct and its application to chronic pain. Behavioural and Cognitive Psychotherapy (in press) Skinner, B. (1953). Science and human behavior. New York: MacMillan. Spitzer, W., LeBlanc, F. et al. (1987). Scientific approach to the assessment and management of activity-related spinal disorders. Report of the Quebec Task Force on Spinal Disorders, Spine, 12:7 S (European ed. Suppl. 1), S1–S59 Sullivan, M. J., Rouse, D., Bishop, S., & Johnston, S. (1997). Thought suppression, catastrophising and pain. Cognitive Therapy and Research, 21, 555–568. Turk, D. (1990). Customising treatment for chronic pain patients: who, what, and why? The Clinical Journal of Pain, 6, 255–270. Turk, D. (1996a). Biopsychosocial perspective on chronic pain. In R. Gatchel, & D. Turk, Psychological approaches to pain management — a practitioner’s handbook. New York: The Guilford Press. Turk, D. (1996b). Cognitive factors in chronic pain and disability. In K. Dobson, & K. Craig, Advances in cognitivebehavioral therapy. Newbury Park, California: Sage Publications.
800
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
Turk, D., & Flor, H. (1984). Etiological theories and treatments for chronic back pain. II. Psychological models and interventions. Pain, 19, 209–233. Turk, D., Meichenbaum, D., & Genest, M. (1983). Pain and behavioral medicine: a cognitive-behavioral perspective. New York: The Guilford Press. Turk, D., & Rudy, T. (1991). Neglected topics in the treatment of chronic pain patients — relapse, noncompliance, and adherence enhancement. Pain, 44, 5–28. Turk, D., & Rudy, T. (1992). Cognitive factors and persistent pain: a glimpse into Pandora’s box. Cognitive Therapy and Research, 16, 99–122. Turner, J., Clancy, S., & Vitaliano, P. (1987). Relationships of stress, appraisal and coping, to chronic low back pain. Behavior Research and Therapy, 25 (4), 281–288. Turner, J., & Romano, J. (1984). Self-report screening measures for depression in chronic pain patients. Journal of Clinical Psychology, 40, 909–913. Vlaeyen, J., de Jong, J., Geilen, M., Heuts, P., & van Breukelen, G. (2000). Graded exposure in vivo in the treatment of pain-related fear: a replicated single-case experimental design in four patients with chronic low back pain. Behaviour Research and Therapy (in press) Valeyen, J., Kole-Snijders, A., Rottveel, A., Ruesnik, R., & Heuts, P. (1995). The role of fear of movement/(re)injury in pain disability. Journal of Occupational Rehabilitation, 5, 235–252. Vlaeyen, J., Kole-Snijders, A., Boeren, R., & van Eek, H. (1995). Fear of movement/(re)injury in chronic low back pain and its relation to behavioural performance. Pain, 62, 363–372. Waddell, G., & Main, C. (1984). Assessment of severity of low back disorders. Spine, 9, 204–208. Waddell, G., Newton, M., Henderson, I., Somerville, D., & Main, C. (1993). A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain, 52, 157–168. Wells, A. (1995). Meta-cognition and worry: a cognitive model of generalised anxiety disorder. Behavioural and Cognitive Psychotherapy, 23, 265–280. Williams, A., Richardson, P., Nicholas, M., Pither, C., Harding, V., Ridout, K., Ralphs, J., Richardson, I., Justins, D., & Chamberlainc, J. (1996). Inpatient vs outpatient pain management: results of a randomised controlled trial. Pain, 66, 13–22. Williams, D., & Keefe, F. (1991). Pain beliefs and the use of cognitive-behavioural coping strategies. Pain, 46, 185–190. Wilson, P., Henry, J., & Nicholas, M. (1993). Cognitive methods in the management of chronic pain and tinnitus. Australian Psychologist, 28 (3), 172–180. Wolpe, J. (1958). Psychotherapy by reciprocal inhibition. Stanford, California: Stanford University Press. Wolpe, J., & Lazarus, A. (1966). Behavior therapy techniques: a guide to the treatment of neuroses. London: Pergamon Press.
Chronic pain: a reformulation of the cognitive-behavioural model Timothy J. Sharp
*
Suite 46, Level 9, 88 Pitt Street, Sydney, NSW 2000, Australia Accepted 12 April 2000
Abstract The last few decades has seen psychological theories of chronic pain dominated by behavioural concepts and particularly, by the operant model as described by Fordyce et al. (1968: Fordyce, W., Fowler, R., & DeLateur, B. (1968) An application of behavior modification technique to a problem of chronic pain. Behaviour Research and Therapy, 6, 105–107, 1976: Fordyce, W.E. (1976) Behavioral methods for chronic pain and illness. St Louis, MO: C. V Moseby). More recently, cognitive constructs have been included, giving the impression that the operant theory of chronic pain has been replaced by a cognitive-behavioural theory (CBT). Given the fact that CBTs were introduced essentially to overcome some inherent problems with operant theory, it is notable that they are still very much based on operant theory and continue to be founded upon operant principles. Further, the extent to which CBTs (at this stage) have included contemporary cognitive research has been limited. This paper argues that it is of questionable validity to continue to include problematic concepts (e.g., operant principles) into a revised theory (e.g., CBT). Instead, consistent with research from other areas (particularly the anxiety disorders), chronic pain and the problems associated with it may be better explained by a reformulated cognitive-behavioural theory which although not ignoring the observable behaviours of pain patients and their associations with social reinforcers, interprets these phenomena from a cognitive perspective. Thus, a modified CBT is proffered, focusing more directly on patients’ thoughts about, and appraisals of, their pain. Evidence in support of such a theory is provided, as are suggestions for further research and the implications such a theory has for treatment. 2001 Elsevier Science Ltd. All rights reserved. Keywords: Cognitive; Behavioural; Chronic pain
* Tel.: +61-2-9231-2522; fax: +61-2-9231-2533. E-mail address: [email protected] (T.J. Sharp).
0005-7967/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 5 - 7 9 6 7 ( 0 0 ) 0 0 0 6 1 - 9
788
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
1. Brief history of the presenting problem Definitions of pain have varied considerably over the years and despite significant advances in medical technology, including the development of sophisticated scanning techniques, general agreement on what pain is and what it means continues to evade clinicians and researchers (Merskey & Bogduk, 1994). Although popular and medical opinion typically links pain and tissue damage, the evidence for this association is questionable, at best. The link between reports of pain and identifiable pathology has been investigated and the findings have cast serious doubt on theories that attempt to understand pain as a purely physical phenomenon. Jensen et al. (1994), for example, found that many people with ‘abnormalities’ identified by MRI (e.g., disk bulges) did not report back pain. At the same time, Spitzer, LeBlanc et al. (1987; cited in Fordyce, 1995) report that many people complaining of back pain do not have identifiable pathology. In fact, the best evidence suggests that “fewer than 15% of persons with back pain can be assigned to one of the categories of specific low back pain” (Fordyce, 1995, p. 3). Thus, although they are no doubt related, the association between physical impairment and reports of pain is only moderate (Waddell & Main, 1984). The fact that physical pathology does not perfectly predict pain severity has long been recognised, as has the fact that different people respond in varying ways to chronic pain (Turk, 1996a). For example, not all pain patients are depressed and disabled to the same degree (Turner & Romano, 1984). Notably, the association between pain and depression and disability appears to be mediated by psychosocial variables such as perceived life-control (Fordyce, 1995; Lackner, Carosella & Feuerstein, 1996; Romano et al., 1995; Rudy, Kerns & Turk, 1988; Waddell, Newton, Henderson, Somerville & Main, 1993). Taken together, therefore, the evidence suggests that pathology alone cannot explain reports of pain, pain-related disability or depression. As a result, multidimensional models have been proffered to better account for the experimental findings and clinical phenomenology. The Gate Control Theory (Melzack & Wall, 1965), for example, significantly changed the way pain was conceptualised and contributed to the International Association for the Study of Pain (Merskey, 1979) defining pain as an “unpleasant sensory and emotional experience …”. The use of words such as ‘unpleasant’ and ‘emotional experience’ are crucial in this definition as they emphasise the influence of psychological variables. More recently, other multidimensional models have been suggested, with one of the most popular being attributed to Loeser (1982; see Waddell et al., 1993). Notably, this model makes the important distinction between sensory, cognitive, affective, behavioural and social variables. This multidimensional, biopsychosocial model drew heavily on the work of Fordyce et al. who introduced behavioural principles to the study of pain. 2. Operant theory Fordyce et al. (1968, 1976), drawing on the work of other behaviourists (e.g., Skinner, 1953) suggested a distinction be made between the original cause of pain and reports of, and displays of pain (known as pain behaviours). Although a complete review of the behavioural model is beyond the scope of this paper, its basic assumption is that pain behaviours (e.g., limping, grimacing) are subject to the same influences of conditioning as are any other behaviours (see
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
789
Fordyce, 1996). As such, if followed by reinforcing consequences, pain behaviours may persist for longer than the normal expected healing time (see Sanders, 1996). Although it has generally been accepted that there is good evidence to support treatments based on the operant model (American Psychological Association, 1993; Keefe, 1994; Morley, Eccleston & Williams, 1999) there is less agreement regarding how they work and the strength of the evidence to support the underlying theory upon which this intervention is based (Turk & Flor, 1984). Support of the operant theory is typified by studies that suggest behavioural interventions are effective (Cairns & Pasino, 1977), that reports of pain may be cued by the presence of a spouse (Block, Kremer & Gaylor, 1980), and that spouse responses are associated with pain behaviours (Romano et al., 1992). There are, however, several problems with these studies. Turk (1996b), for example, recently outlined a number of ‘concerns’ about the operant perspective which include (1) the questionable validity of the pain behaviour construct, (2) the lack of specificity of the pain behaviour construct, (3) the common assumption that observed pain behaviours are maladaptive, (4) the potential for pain to be underreported and the scope for this to be detrimental, (5) some patients’ lack of acceptance of operant treatment and/or dissatisfaction with operant treatment goals, (6) the focus on motor behaviour and (7) problems of maintenance of behaviours following treatment. In summary, the first few of these concerns relate to the fact that pain behaviours require interpretation to determine their raison d’eˆ tre, while the latter few refer to the findings that behavioural treatments are not effective for all patients and/or that even when they initially succeed, are often followed by relapse (Turk & Rudy, 1991). As these issues have each been discussed in considerable detail by Turk (1996b) they will not be re-stated here. Some additional problems, however, will be outlined. First, with respect to the operant theory of chronic pain, much weight has been given to the results of several key studies. These include those conducted by Cairns and Pasino (1977), Block et al. (1980) as well as the series of studies by Romano et al. (1991, 1992, 1995). The first (Cairns & Pasino, 1977) can be considered as an example of studies that essentially claim the reduction of ‘pain behaviours’ via the modification of environmental contingencies supports the principles of the operant model. This may well be true, but it neglects to acknowledge that there are other ways of interpreting these findings. The operant model, for example, does not acknowledge the possibility that patients’ interpretations of the environmental changes may be important. This possibility raises the prospect of cognition playing a role. It is similar to the argument that because antidepressants improve mood (in some patients), then depression is therefore a biological condition (Salkovskis, 1996a has applied a similar argument to theoretical problems in the anxiety disorders). Neither argument is necessarily logical, as there may well be multiple factors that influence behaviour. Simply because pain behaviours may change within pain programmes, does not necessarily mean that they are affected by the supposed contingencies ‘designed’ by the staff. Few, if any, investigators, for example, have reflected on the fact that many treatments have typically been effective regardless of the fact that they invariably devote little (or no) time to modifying spouse behaviours outside the clinic (Nicholas, Wilson & Goyen, 1992). Further, interventions that have specifically targeted spouses have not proven to be any more effective than standard patient interventions (Moore & Chaney, 1985). Either way, there is a logical flaw in the assumption that the operant model is supported simply because behavioural treatment has an effect on patients’ behaviour.
790
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
The Block et al. (1980) study has also been widely cited as providing support for the behavioural model. As has been noted by Paulsen and Altmaier (1995), however, a close reading of Block et al.’s data reveals findings not entirely consistent with their conclusions. Of most concern is the statement that patients systematically alter their report of pain depending on the presence of, and their perceptions of, their spouses. It is questionable whether such a conclusion can be made given that there was no significant difference between reports of pain in either condition for patients in the solicitous spouse group. Further, careful examination reveals that the reason for the interaction effect was not that patients in the solicitous spouse group reported more pain in the presence of their spouses, but that the patients in the non-solicitous spouse group reported less pain. Rather than indicating that solicitous responses increase the likelihood of higher pain intensity reports, this may suggest that non-supportive responses decrease the likelihood of reporting pain! One must also question the validity and generalisability of these results when so little descriptive data have been provided about the patients and the distribution in the two groups. The authors themselves acknowledge a significant group difference in pain duration. Taking these limitations into account, the degree to which this study provides support for the operant model must surely be questionable. Even the findings of a more recent study that attempted to re-test the conclusions of Block et al. (1980) were only partially consistent with hypotheses generated from an operant model (Lousberg, Schmidt & Groenman, 1992). In this investigation, Louseberg et al. found that patients’ perceptions of spouse solicitousness were not significantly associated with pain behaviours although interestingly, the partners’ perceptions were more closely related. Finally, a series of studies by Romano et al. (1991, 1992, 1995) has specifically assessed the relationship between patients’ pain behaviours and spouse responses using a sophisticated direct observation system. Although their methodology is in many ways beyond reproach, the validity and the generalisability of their findings are unclear due to the questionable representativeness of their sample. Romano et al. (1991) note that “Of the 233 chronic pain patients who met the study inclusion criteria and were approached to participate in the study, 50 agreed to participate”. Although they state that those who agreed to participate did not differ from those who refused in age or gender, they provide no information about pain severity, depression or disability. Thus, the extent to which this sample can be compared to a typical pain center population is unknown. Although ostensibly disappointing for the operant model, these findings are still promising from a psychological perspective as they are consistent with a cognitive model (to be described below) which among other things, hypothesises that if spouse responses were to have an influence they would be mediated by patients’ interpretations (Sharp, 1997). That is, if spouse responses are interpreted by two different patients as meaning “I can’t do anything without her”, or “With her help I will be able to do more”, they will have quite different effects. The implications of these interpretations for patients’ and for treatment will be discussed below. In summary, although there is good evidence to support the efficacy of behavioural treatments, empirical support for the operant theory may be more uncertain than is often acknowledged. In response to some of the problems described above, and given the developments and advances in other areas of psychological theory over the last decade or so (see Hawton, Salkovski, Kirk & Clark, 1989; Salkovskis, 1996b), many clinicians and researchers have begun to refer to ‘cognitive behavioural theories’ of pain.
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
791
3. Cognitive-behavioural theories Turk, Meichenbaum and Genest (1983) were among the first to develop a cognitive behavioural model of pain that drew on the work of Beck (1976), Beck, Rush, Shaw and Emery (1979) and Meichenbaum (1977). Beginning with the observation that behavioural treatments were not exclusively behavioural, the authors noted that even Fordyce (1976) referred to the importance of reaching a ‘shared conceptualisation’ with patients. They also noted that the founders of behaviour therapy (e.g., Wolpe, 1958; Wolpe & Lazarus, 1966) emphasised the need to ‘remove patients mistaken beliefs’. Consistent with these sentiments, there is now good evidence to support the influence of cognitive factors. Cognitions have been found to influence reports of pain intensity, attempts to cope (Jensen, Turner & Romano, 1991), mood and pain-related disability (Turk & Rudy, 1992; Williams & Keefe, 1991; Wilson, Henry & Nicholas, 1993). A number of key themes have emerged from the research into pain and cognitions. Patients who interpret their pain as being indicative of more damage tend to be more avoidant and more disabled (Vlaeyen, Kole-Snijders, Boeren & van Eek, 1995). Similarly, catastrophic interpretations of pain have been found to be associated with higher levels of depression (Rosensteil & Keefe, 1983) and the affective component of pain (Geisser, Robinson, Keefe & Weiner, 1994). Further, predictions of pain, self-efficacy, and perceptions of control have all been found to be related to depression and behavioural performance (Arntz & Peters, 1995; Dolce, Crocker, Molettiere & Doleys, 1986; Rudy et al., 1988). Taken together, these studies provide strong support for the inclusion of cognitive factors in a model of chronic pain. Although not suggesting cognitions cause pain, there is strong support for the notion that pain related cognitions and beliefs are associated with the utilisation of helpful or unhelpful coping strategies and thereby affect the extent to which pain impacts on patients’ lives. At the same time, however, behavioural constructs have continued to dominate theory and treatment. While describing cognitive behavioural interventions, both Turk (1996a, 1996b) and Bradley (1996), for example, specifically refer to the fact that “environmental factors can also influence behaviour” (Turk, 1996a, p. 93). Bradley’s intervention outlines 15 weeks of treatment, and describes attending to patients’ and spouses’ behaviours, relaxation training, biofeedback, goal setting and relapse prevention. Notably, the identification and evaluation of unhelpful cognitions or beliefs is not explicitly mentioned. Turk et al. (1983) describe a very similar programme and also do not include specific strategies for the modification of cognitions. These examples seem to exemplify the relative failure of cognitive behaviour therapy of chronic pain to adequately incorporate cognitive constructs. At this stage, no theory has attempted to explain (or even noted) the persistence of catastrophic thoughts despite patients undergoing repeated experiences that should have challenged their validity. For example, the catastrophic thought ‘I can’t cope’ should have undergone re-evaluation following ‘survival’ of numerous pain flare-ups. Yet such a thought is common in patients many years after the onset of their pain. Further, despite the questions raised earlier, cognitive behavioural theory has continued to be founded on behavioural principles and cognitive behavioural interventions have continued to be dominated by behavioural components. It is debatable whether adding on a session of cognitive therapy really makes a treatment cognitive (Salkovskis, 1986; 1987). The failure to adequately modify pain management interventions is particularly important given the risk that it may be
792
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
jeopardising treatment gains. This is most notable given the advances cognitive treatments have recently made in the treatment of the anxiety disorders (see Salkovskis, 1996b). At the same time, however, questioning the evidence for the operant model does not necessarily imply that behaviour is not important. Anxiety provides a fascinating example of the link between cognition and behaviour and of how behaviour is still important in a cognitive framework (Salkovskis, 1991). 4. Lessons from the anxiety research The Oxford Group (e.g., Salkovskis, Clark & Gelder, 1996) have provided empirical support for the notion that behaviours maintaining anxiety can be viewed from within a cognitive model (Salkovskis, 1991). Salkovskis describes the relationship between ‘threat cognitions’ and ‘safety seeking behaviours’, arguing that certain behaviours are understandable given a belief that danger is imminent. Avoidance, for example, is not conceptualised as a response to (conditioned) anxiety that is maintained by negative reinforcement (Rachman, 1977). Instead, avoidance is conceptualised as a response that “arises out of, and is logically linked to, the perception of serious threat. [Further] because safety seeking behaviour is perceived to be preventative, and focused on especially negative consequences … disconfirmation of threat is made particularly unlikely” (Salkovskis, 1991, p. 6). Behaviours are thus seen as an integral component in the maintenance of anxiety, within a cognitive framework. Given that the associated cognitive therapies have been found to be more effective, in certain cases, than the more traditional behavioural treatments (Salkovskis & Warwick, 1985; Salkovskis & Westbrook, 1989), it is possible that their application to pain may similarly improve outcomes which although impressive in some cases (e.g., Williams et al., 1996), could indubitably be improved (Turk & Rudy, 1991). 5. Towards a reformulation of the cognitive-behavioural conceptualisation The modified cognitive-behavioural model begins with the proposition that problems associated with chronic pain originate in the way patients react to their pain. Reactions are defined as including all forms of cognition (e.g., pain-related thoughts and imagery), not simply observable behaviours. It is proposed that the difference between patients who are distressed and/or disabled, and those for whom pain is not markedly problematic, lies not necessarily in the sensory activity, but rather in the patients’ appraisals and interpretations of the pain. Fig. 1 represents the reformulated cognitive-behavioural model in which patients’ appraisals of their pain (the meaning ascribed to the pain by the individual) are ascribed a primary role. As indicated in the model, the various interactions can become self-maintaining such that distress and disability, as well as physiological arousal, may continue despite the absence of the original sensory input. Avoidance and/or safety behaviours may interfere with disconfirmation of and thereby maintain certain cognitions or beliefs. At the same time, anxiety and distress may maintain autonomic arousal that may confirm beliefs that some underlying condition exists. Anxiety and depression may then increase the likelihood of patients making cognitive errors or negative appraisals, as well as potentially maintaining avoidance, and the cycle would thereby continue. The model also takes into account the likelihood that stress and iatrogenic factors may exacer-
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
793
Fig. 1. The reformulated cognitive-behavioural model of chronic pain.
bate anxiety and arousal, and/or maintain certain behaviours. Borrowing from models of hypochondriasis (Salkovskis & Bass, 1997), the model emphasises patients’ beliefs about their symptoms and the tendency to seek reassurance. Many pain patients, for example, express fears about experiencing more damage and believe that more investigations or treatments are needed. According to the reformulated model, the ways doctors respond to this may inadvertently reinforce patients’ anxieties and may unwittingly reinforce excessive disability and a passive approach (Kouyanou, Pither, Rabe-Hesketh & Wessley, 1998; Pither & Nicholas, 1991).
794
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
Given that this conceptualisation borrows considerably from other cognitive models (e.g., Clark, 1986; Salkovskis & Warwick, 1986), the parallel can be taken further by adapting hypotheses related to patients’ ‘meta-cognitions’ (Wells, 1995). The current model hypothesises that if patients interpret the presence of pain itself or of pain-relevant thoughts as indicating something negative about their condition or persona (e.g., “Thinking about pain means my pain is serious” or “It is so horrible thinking about the pain all the time … if I can’t stop it I’ll go crazy” or “All I can think about is the pain, I can’t focus on anything else, I’m so hopeless”) then they may attempt to suppress or neutralise such thoughts (note: this can be seen as a form of safety behaviour). The model further assumes that attempts to suppress (e.g., “not think about the pain”) or neutralise (e.g., attempts to protect oneself from going crazy by controlling one’s mind or by using distraction) pain-related thoughts may lead to an increase in their frequency and in their perceived aversiveness (Salkovskis & Campbell, 1994; Salkovskis, Westbrook, Davis, Jeavons & Gledhill, 1997). If, as is frequently the case, pain persists and more investigations and treatments prove unhelpful, it is hypothesised that a form of learned helplessness may develop. That is, patients could start to believe that ‘nothing has worked so far so why would any future treatment help?’. Patients’ holding such a belief may not try as hard in treatment, which may increase the chances of the treatment not helping, which may in turn reinforce their belief that ‘nothing is going to help’. Patients’ could then be labeled as ‘not motivated’. This has important treatment implications, because labeling patients in this way may affect the way they are assessed and/or the treatments they are offered. Alternatively, if viewed from within the proposed model (i.e., as patients with unhelpful and unrealistic cognitions), these patients could be seen as appropriate candidates for cognitive therapy, and therefore as salvageable. It should be noted that these pain-related cognitions and beliefs are assumed to develop as a result of patients’ early learning histories and cultural backgrounds. That is, depending upon their prior experiences with pain, or knowledge of other people with pain, patients will have learned that pain has certain meanings. Depending upon how they have seen others respond to pain, or on how pain is perceived within their particular culture, they may learn to interpret and/or to evaluate it in different ways (see, for example, Bates, 1987; Bates, Edwards & Anderson, 1993; Craig, 1978). Research investigating the role of significant other responses can also be reinterpreted within the modified cognitive-behavioural model. Spouse responses could be conceptualised as providing evidence for, or against, patients’ beliefs. If patients believe they are disabled, and their spouse is overly solicitous, this may confirm patients’ beliefs while also preventing patients from disconfirming their own perceptions of disability. Similarly, medication use which has traditionally been understood as a behaviour maintained by negative reinforcement can be reinterpreted as a safety behaviour in response to the beliefs “I can’t cope … this pain is unbearable”. Real or perceived relief may interfere with effective evaluation of the belief. Importantly, the model incorporates the possibility of learning without awareness. The model would also predict that these cognitions and the associated behaviours are inextricably linked and are mutually maintaining. Patients who avoid engaging in tasks because they are afraid of further damage, will never gather disconfirming information and will continue to believe their pain is indicative of danger. Alternatively, if they do attempt tasks their beliefs may still be maintained if they engage in protective behaviours (such as lifting a box but only while
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
795
wearing a back brace). Rather than disproving the danger of lifting, such experiences may simply confirm that lifting is dangerous and would have lead to more damage if only the brace had not provided the necessary protection. 6. Support for the reformulated cognitive-behavioural model There are several advantages in viewing chronic pain from within a more cognitive framework. First, it is suggested that this model better explains the existing evidence. Second, the model makes clear and testable predictions, including predictions about treatment and about what should work for individual patients. Theoretically, identifying and targeting key cognitions should lead to more effective interventions than simply using the ‘blunderbuss’ approach typically offered (Turk, 1990). Most importantly, the new cognitive behavioural model is well supported by, and consistent with a range of recent empirical findings. Of particular interest are a series of studies by Eccleston (1995; Eccleston, Crombez, Aldrich & Stannard, 1997) and Crombez (Crombez, Eccleston, Baeyens & Eelen, 1998a, 1998b). In brief, these studies tend to support the notion that if patients are worried about their pain (referred to as ‘catastrophisers’), they are more likely to be hypervigilant (and be more somatically aware), which not surprisingly interferes with attention but which also “amplifies somatosensory information and primes fear mechanisms” (Crombez et al., 1998a, p. 187). This supports the hypothesised relationship between anxiety and hypervigilance. Further, anxiety has been found to predict ‘non-specific physical complaints’ in pain patients, a finding consistent with patients being hypervigilant (Asmundson, Kuperos & Norton, 1997; McCracken, Faber & Janeck, 1998). Fear of pain has also been found to be a better predictor of avoidance than pain severity or physical pathology (Crombez Vervaet, Lysens, Baeyens & Eelen, 1998). The authors interpreted their findings as being more consistent with a cognitive model than an operant one and there is a growing body of support for such a model (Asmundson, Norton & Allerdings, 1997; Vlaeyen, Kole-Snijders, Rottveel, Ruesink & Heuts, 1995). There is also some evidence to suggest that attempts to suppress pain-related thoughts may be associated with heightened pain experience (Sullivan, Rouse, Bishop & Johnston, 1997). This research, however, was conducted on a non-clinical population and so the findings do need to be treated with some caution. Nevertheless, there is good evidence from the anxiety research that attempts to suppress thoughts are not helpful in the long term, are likely to exacerbate and maintain distress, and do hold true in clinical populations (Harvey & Bryant, 1998; Rutledge, 1998). The role of autonomic arousal in the model has also received relatively strong support. In a comprehensive review of the literature on psychophysiology of pain Flor and Turk (1989) concluded that although chronic pain patients do not necessarily differ from normals on physiological measures at baseline, they do tend to differ from non-pain patients in response to pain and/or stress. Flor, Turk and Birbaumer (1985) asked subjects with chronic low back pain to discuss personally stressful (and/or painful) situations. The investigators measured subjects’ heart rate and skin resistance level changes. Among other things, they found that compared to neutral tasks (such as mental arithmetic), stressful tasks were associated with significant increases in paraspinal EMG levels in pain patients, but not in normals. Unfortunately, the methodologies of other studies
796
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
investigating these factors have been criticised (Flor & Turk, 1989) and it is therefore necessary for these findings to be replicated. Nevertheless, these findings are consistent with a model in which the relationship between pain and arousal is mediated by appraisal (Flor et al., 1985; Turner, Clancy & Vitaliano, 1987). Finally, the role of iatrogenic factors has received some support from research that, among other things, has found that inappropriate and excessive use of medications and investigations may be associated with patients’ reports of symptomatology and with disability (Kouyanou et al., 1998; Pither and Nicholas, 1991). Kouyanou et al. compared chronic pain patients whose symptoms were considered to be medically explained with those whose symptoms could not be explained. They found significant differences on several variables including a higher prevalence of over-investigation and over-prescribing in the group whose symptoms were not explicable. They hypothesised that the management of these patients was often unhelpful and that by frequently ‘disconfirming’ patients’ pain and by suggesting it was ‘all in the mind’ they were, in fact, maintaining the patients’ sick role. The authors suggested that this could be attributed to patients’ determination to legitimise their pain by seeking more investigations and treatments. The salient point for the proposed model is that the responses of the medical profession can influence patients’ beliefs about their pain and subsequently, their pain-related treatment-seeking behaviours. Taken together, a substantial body of research now exists to support the conceptualisation of chronic pain in a manner similar to that which is generally accepted for anxiety. If one equates pain with intrusive thoughts, appraisals with cognitions, and attempts to neutralise with pain safety behaviours such as avoidance then chronic pain can be understood using similar constructs as those applied to obsessional problems (Salkovskis, 1985). This suggests that pain management could be improved as has the treatment of obsessions (Salkovskis & Westbrook, 1989). There is, however, undoubtedly a need for more research. Most notably, the concept of ‘safety behaviours’ needs to be clearly and specifically established within pain patients. Although the potential role of safety seeking behaviours has been hypothesised (Sharp, in press), investigations examining the specific predictions of this part of the model are yet to be conducted. In addition, individually tailored cognitive interventions need to be tested against the more general programmes now on offer, to assess their relative efficacy. (Note: as this paper was in preparation, a preliminary investigation along these lines was completed by Vlaeyen, de Jong, Geilen, Heuts & van Breukelen, 2000.) 7. Conclusions This paper has attempted to trace the development of psychological theories of chronic pain starting with the operant model, through to a reformulated cognitive behavioural model that parallels models proffered for the anxiety disorders (Salkovskis, 1996a, 1996b). In reviewing the relevant evidence, it appears that the operant model is not as well supported as has been generally thought and that even some cognitive-behavioural models suffer many of the problems their predecessors do because they have continued to rely on outdated behavioural principles. Accordingly, a reformulated cognitive behavioural theory is outlined which gives substantially more attention to patients’ thoughts about and appraisals of their pain. Patients’ behaviours are
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
797
still considered important, but from the point of view of their relationship to cognitions and particularly with regards to the ways they may maintain pain-related cognitions. It is argued that this new model is relatively well supported by the empirical findings and that it explains contemporary research better than does the operant theory. Although more research is needed, the cognitive behavioural model presented here proffers specific hypotheses and provides clear directions for future investigations.
References American Psychological Association Division of Clinical Psychology (Division 12) (1993, October). Task Force Report on the Promotion and Dissemination of Psychological Procedures: Final Report Arntz, A., & Peters, M. (1995). Chronic low back pain and inaccurate predictions of pain: is being too tough a risk factor for the development and maintenance of chronic pain? Behaviour Research and Therapy, 33, 49–53. Asmundson, G., Kuperos, J., & Norton, G. (1997). Do patients with chronic pain selectively attend to pain-related information? Preliminary evidence for the mediating role of fear. Pain, 72, 27–32. Asmundson, G., Norton, G., & Allerdings, M. (1997). Fear and avoidance in dysfunctional chronic back pain patients. Pain, 69, 231–236. Bates, M. (1987). Ethnicity and pain: a biocultural model. Social Science Medicine, 24, 47–50. Bates, M., Edwards, W., & Anderson, K. (1993). Ethnocultural influences on variation in chronic pain perception. Pain, 52, 101–112. Beck, A. (1976). Cognitive therapy and the emotional disorders. New York: International University Press. Beck, A., Rush, J., Shaw, B., & Emery, G. (1979). Cognitive therapy of depression. New York: The Guilford Press. Block, A., Kremer, E., & Gaylor, M. (1980). Behavioral treatment of chronic pain: the spouse as a discriminative cue for pain behavior. Pain, 9, 243–252. Bradley, L. (1996). Cognitive-behavioral therapy for chronic pain. In R. Gatchel, & D. Turk, Psychological approaches to pain management: a practitioner’s guide. New York: The Guilford Press. Cairns, D., & Pasino, J. (1977). Comparison of verbal reinforcement and feedback in the operant treatment of disability due to chronic low back pain. Behaviour Therapy, 8, 621–630. Clark, D. (1986). A cognitive approach to panic. Behaviour Research and Therapy, 24, 461–470. Craig, K. (1978). Social modeling influences on pain. In R. Sternbach, The psychology of pain. Raven Press: New York. Crombez, G., Eccleston, C., Baeyens, F., & Eelen, P. (1998a). When somatic information threatens, catastrophic thinking enhances attentional interference. Pain, 75, 187–198. Crombez, G., Eccleston, C., Baeyens, F., & Eelen, P. (1998b). Attentional disruption is enhanced by the threat of pain. Behaviour Research and Therapy, 36, 195–204. Crombez, G., Vervaet, L., Lysens, R., Baeyens, F., & Eelen, P. (1998). Avoidance and confrontation of painful, backstraining movements in chronic back pain patients. Behavior Modification, 22, 62–77. Dolce, J., Crocker, M., Molettiere, C., & Doleys, D. (1986). Exercise quotas, anticipatory concern and self-efficacy expectations in chronic pain: a preliminary report. Pain, 24, 365–372. Eccleston, C. (1995). Chronic pain and distraction: an experimental investigation into the role of sustained and shifting attention in the processing of chronic persisting pain. Behaviour Research and Therapy, 33, 391–405. Eccleston, C., Crombez, G., Aldrich, S., & Stannard, C. (1997). Attention and somatic awareness in chronic pain. Pain, 72, 209–215. Flor, H., & Turk, D. (1989). Psychophysiology of chronic pain: do chronic pain patients exhibit symptom-specific psychophysiological responses? Psychological Bulletin, 105, 215–259. Flor, H., Turk, D., & Birbaumer, N. (1985). Assessment of stress-related psychophsiological reactions in chronic back pain patients. Journal of Consulting and Clinical Psychology, 53, 354–364. Fordyce, W., Fowler, R., & DeLateur, B. (1968). An application of behavior modification technique to a problem of chronic pain. Behaviour Research and Therapy, 6, 105–107. Fordyce, W. E. (1976). Behavioral methods for chronic pain and illness. St Louis, MO: C. V. Moseby.
798
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
Fordyce, W. E. (1995). Back pain in the workplace: management of disability in nonspecific conditions. Seattle: IASP Press. Fordyce, W. E. (1996). Chronic pain: the behavioral perspective. In M. Cohen, & J. Campbell, Pain treatment centers at a crossroads: a practical and conceptual reappraisal. Seattle: IASP Press. Geisser, M., Robinson, M., Keefe, F., & Weiner, M. (1994). Catastrophising, depression and the sensory, affective and evaluative aspects of chronic pain. Pain, 59, 79–83. Harvey, A., & Bryant, R. (1998). The effect of attempted thought suppression in acute stress disorder. Behaviour Research and Therapy, 36, 583–590. Hawton, K., Salkovskis, P., Kirk, J., & Clark, D. (1989). Cognitive behaviour therapy for psychiatric problems: a practical guide. Oxford: Oxford University Press. Jensen, M., Brant-Zawadzki, M., Obuchowski, N., Modic, M., Malkasian, D., & Ross, J. (1994). Magnetic resonance imaging of the lumbar spine in people without back pain. The New England Journal of Medicine, 331, 69–73. Jensen, M., Turner, J., & Romano, J. (1991). Self-efficacy and outcome expectancies: relationships to chronic pain coping strategies and adjustment. Pain, 44, 263–269. Keefe, F., Crisson, J., Urban, B., & Williams, D. (1990). Analyzing chronic low back pain: the relative contribution of pain coping strategies. Pain, 40, 293–301. Kouyanou, K., Pither, C., Rabe-Hesketh, S., & Wessley, S. (1998). A comparative study of iatrogenesis, medication abuse, and psychiatric comorbidity in chronic pain patients with and without medically explained symptoms. Pain, 76, 417–426. Lackner, J., Carosella, A., & Feuerstein, M. (1996). Pain expectancies, pain, and functional self-efficacy expectancies as determinants of disability in patients with chronic low back disorders. Journal of Consulting and Clinical Psychology, 64, 212–220. Loeser, J. (1982). Concepts of pain. In M. Stanton-Hicks, & R. Boas, Chronic low back pain. New York: Raven Press. Lousberg, R., Schmidt, A., & Groenman, N. (1992). The relationship between spouse solicitousness and pain behaviour: Searching for more experimental evidence. Pain, 51, 75–79. McCracken, L., Faber, S., & Janeck, A. (1998). Pain-related anxiety predicts non-specific physical complaints in persons with chronic pain. Behavior Research and Therapy, 36, 621–630. Meichenbaum, D. (1977). Cognitive behavior modification: an integrative approach. New York: Plenum. Melzack, R., & Wall, P. (1965). Pain mechanisms: a new theory. Science, 50, 971–979. Merskey, H. (1979). Pain terms: a list with definitions and notes on usage. Recommended by the IASP subcommittee on taxonomy. Pain, 6, 249–252. Merskey, H., & Bogduk, N. (1994). Classification of chronic pain — descriptions of chronic pain syndromes and definitions of pain terms. (2nd ed.). Seattle: IASP Press. Moore, J., & Chaney, E. (1985). Outpatient group treatment of chronic pain: effects of spouse involvement. Journal of Consulting and Clinical Psychology, 53, 326–334. Morley, S., Eccleston, C., & Williams, A. (1999). Systematic review and meta-analysis of randomised controlled trials of cognitive behaviour therapy and behaviour therapy for chronic pain in adults, excluding headache. Pain, 80, 1–14. Nicholas, M., Wilson, P., & Goyen, J. (1992). Comparison of cognitive-behavioural group treatment and an alternative non-psychological treatment for chronic low back pain. Pain, 48, 339–347. Paulsen, J., & Altmaier, E. (1995). The effects of perceived versus enacted social support on the discriminative cue function of spouses for pain behaviours. Pain, 60, 103–110. Pither, C., & Nicholas, M. (1991). The identification of iatrogenic factors in the development of chronic pain syndromes: abnormal treatment behaviour? In Bond, Charlton, & Woolf, Proceedings of the Fifth World Congress on Pain. Elsevier Science Publishers. Rachman, S. (1977). The conditioning theory of fear acquisition: a critical examination. Behaviour Research and Therapy, 15, 375–387. Romano, J., Turner, J., Friedman, L., Bulcroft, R., Jensen, M., & Hops, H. (1991). Observational assessment of chronic pain patient-spouse behavioural interactions. Behavior Therapy, 22, 549–567. Romano, J., Turner, J., Friedman, L., Bulcroft, R., Jensen, M., Hops, H., & Wright, S. (1992). Sequential analysis of chronic pain behaviours and spouse responses. Journal of Consulting and Clinical Psychology, 60, 777–782. Romano, J., Turner, J., Jensen, M., Friedman, L., Bulcroft, R., Hops, H., & Wright, S. (1995). Chronic pain patientspouse behavioral interactions predict patient disability. Pain, 63, 353–360.
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
799
Rosensteil, A., & Keefe, F. (1983). The use of coping strategies in chronic low back pain patients: relationship to patient characteristics and current adjustment. Pain, 17, 33–44. Rudy, T., Kerns, R., & Turk, D. (1988). Chronic pain and depression: toward a cognitive-behavioral mediation model. Pain, 35, 129–140. Rutledge, P. (1998). Obsessionality and the attempted suppression of unpleasant personal intrusive thoughts. Behavior Research and Therapy, 36, 403–416. Salkovskis, P. (1985). Obsessional-compulsive problems: a cognitive-behavioural analysis. Behaviour Research and Therapy, 23, 571–583. Salkovskis, P. (1986). The cognitive revolution: new way forward, backward somersault or full circle? Behavioural Psychotherapy, 14, 278–282. Salkovskis, P. (1987). Cognitive therapy: is it always cognitive, is it always therapy? In H. Dent, Clinical psychology: research and development. London: Croom Helm. Salkovskis, P. (1991). The importance of behaviour in the maintenance of anxiety and panic: a cognitive account. Behavioural Psychotherapy, 19, 6–19. Salkovskis, P. (1996a). Avoidance behaviour is motivated by threat beliefs: a possible resolution of the cognitionbehaviour debate. In P. Salkovskis, Trends in cognitive and behavioural therapies. Chicester: John Wiley & Sons. Salkovskis, P. (1996b). The cognitive approach to anxiety: threat beliefs, satety-seeking behaviour, and the special case of health anxiety and obsessions. In P. Salkovskis, Frontiers of cognitive therapy (pp. 48–74). London: The Guilford Press. Salkovskis, P., & Bass, C. (1997). Hypochondriasis. In D. Clark, & C. Fairburn, Cognitive behavioural therapy: science and practice. Oxford: Oxford University Press. Salkovskis, P., & Campbell, P. (1994). Thought suppression in naturally occuring negative intrusive thoughts. Behaviour Research and Therapy, 32, 1–8. Salkovskis, P., Clark, D., & Gelder, M. (1996). Cognition-behaviour links in the persistence of panic. Behaviour Research and Therapy, 34, 453–458. Salkovskis, P., & Warwick, H. (1985). Cognitive therapy of Obsessive-compulsive disorder: treating treatment failures. Behavioural Psychotherapy, 13, 243–255. Salkovskis, P., & Warwick, H. (1986). Morbid preoccupations, health anxiety and reassurance: a cognitive-behavioural approach to hypochondriasis. Behaviour, Research and Therapy, 24, 597–602. Salkovskis, P., & Westbrook, D. (1989). Behaviour therapy and obsessional ruminations: can failure be turned into success? Behaviour, Research and Therapy, 27, 149–160. Salkovskis, P., Westbrook, D., Davis, J., Jeavons, A., & Gledhill, A. (1997). Effects of neutralising on intrusive thoughts: an experiment investigating the etiology of obsessive-compulsive disorder. Behaviour, Research and Therapy, 35, 211–219. Sanders, S. (1996). Operant conditioning with chronic pain: back to basics. In R. Gatchel, & D. Turk, Psychological approaches to pain management: a practitioner’s guide. New York: The Guilford Press. Sharp, T. (1997). Chronic pain patients and their spouses: an analysis of their interactions. Unpublished Doctoral Dissertation Sharp, T. The safety behaviours construct and its application to chronic pain. Behavioural and Cognitive Psychotherapy (in press) Skinner, B. (1953). Science and human behavior. New York: MacMillan. Spitzer, W., LeBlanc, F. et al. (1987). Scientific approach to the assessment and management of activity-related spinal disorders. Report of the Quebec Task Force on Spinal Disorders, Spine, 12:7 S (European ed. Suppl. 1), S1–S59 Sullivan, M. J., Rouse, D., Bishop, S., & Johnston, S. (1997). Thought suppression, catastrophising and pain. Cognitive Therapy and Research, 21, 555–568. Turk, D. (1990). Customising treatment for chronic pain patients: who, what, and why? The Clinical Journal of Pain, 6, 255–270. Turk, D. (1996a). Biopsychosocial perspective on chronic pain. In R. Gatchel, & D. Turk, Psychological approaches to pain management — a practitioner’s handbook. New York: The Guilford Press. Turk, D. (1996b). Cognitive factors in chronic pain and disability. In K. Dobson, & K. Craig, Advances in cognitivebehavioral therapy. Newbury Park, California: Sage Publications.
800
T.J. Sharp / Behaviour Research and Therapy 39 (2001) 787–800
Turk, D., & Flor, H. (1984). Etiological theories and treatments for chronic back pain. II. Psychological models and interventions. Pain, 19, 209–233. Turk, D., Meichenbaum, D., & Genest, M. (1983). Pain and behavioral medicine: a cognitive-behavioral perspective. New York: The Guilford Press. Turk, D., & Rudy, T. (1991). Neglected topics in the treatment of chronic pain patients — relapse, noncompliance, and adherence enhancement. Pain, 44, 5–28. Turk, D., & Rudy, T. (1992). Cognitive factors and persistent pain: a glimpse into Pandora’s box. Cognitive Therapy and Research, 16, 99–122. Turner, J., Clancy, S., & Vitaliano, P. (1987). Relationships of stress, appraisal and coping, to chronic low back pain. Behavior Research and Therapy, 25 (4), 281–288. Turner, J., & Romano, J. (1984). Self-report screening measures for depression in chronic pain patients. Journal of Clinical Psychology, 40, 909–913. Vlaeyen, J., de Jong, J., Geilen, M., Heuts, P., & van Breukelen, G. (2000). Graded exposure in vivo in the treatment of pain-related fear: a replicated single-case experimental design in four patients with chronic low back pain. Behaviour Research and Therapy (in press) Valeyen, J., Kole-Snijders, A., Rottveel, A., Ruesnik, R., & Heuts, P. (1995). The role of fear of movement/(re)injury in pain disability. Journal of Occupational Rehabilitation, 5, 235–252. Vlaeyen, J., Kole-Snijders, A., Boeren, R., & van Eek, H. (1995). Fear of movement/(re)injury in chronic low back pain and its relation to behavioural performance. Pain, 62, 363–372. Waddell, G., & Main, C. (1984). Assessment of severity of low back disorders. Spine, 9, 204–208. Waddell, G., Newton, M., Henderson, I., Somerville, D., & Main, C. (1993). A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain, 52, 157–168. Wells, A. (1995). Meta-cognition and worry: a cognitive model of generalised anxiety disorder. Behavioural and Cognitive Psychotherapy, 23, 265–280. Williams, A., Richardson, P., Nicholas, M., Pither, C., Harding, V., Ridout, K., Ralphs, J., Richardson, I., Justins, D., & Chamberlainc, J. (1996). Inpatient vs outpatient pain management: results of a randomised controlled trial. Pain, 66, 13–22. Williams, D., & Keefe, F. (1991). Pain beliefs and the use of cognitive-behavioural coping strategies. Pain, 46, 185–190. Wilson, P., Henry, J., & Nicholas, M. (1993). Cognitive methods in the management of chronic pain and tinnitus. Australian Psychologist, 28 (3), 172–180. Wolpe, J. (1958). Psychotherapy by reciprocal inhibition. Stanford, California: Stanford University Press. Wolpe, J., & Lazarus, A. (1966). Behavior therapy techniques: a guide to the treatment of neuroses. London: Pergamon Press.