Commentary: Patient Selection – Setting the Stage for Operant Treatment
In Chapter 7, Fordyce [18] defines the necessary patient selection criteria for operant treatment and describes key issues related to engagement of the spouse and involved family members. At the outset, he recognizes that all individuals whose pain has an operant component may not benefit from treatment. In order for operant treatment to be effective there are certain preconditions that must be met. But before going into detail about these, he comments on “the illusion of homogeneity in human behavior (p. 241). If an “organic” cause for pain cannot be identified then it must be “psychiatric, psychological, mental, non-physical, learned, conditioned or whatever” (p. 241). He is alluding to the tendency he observed among some health professionals to dichotomize pain as physiological or psychological, the organic reductionism* scientifically rejected with the advent of the Gate Control Theory of Pain [41]. Here and at other points in Behavioral Methods, Fordyce notes the limitations of this type of categorization and the barriers to effective treatment that it creates. Although operant treatment for individuals with chronic pain has been criticized [41,59], sometimes vociferously [42], Fordyce was careful to acknowledge that behavioral factors are but one aspect of an individual’s pain experience (p. 242)[18].
THE ILLUSION OF HOMOGENEITY
Using “psychotherapy” as an example, Fordyce points out the tendency of health professionals to assume that treatment approaches are to be delivered in a uniform manner, even though wide variations exist in both content and method of delivery. He uses an automobile mechanic analogy: “if the auto’s problem is one for which a crescent wrench is suitable, all is well. If not, little help is to be expected” (p. 242). It is fair to say that some progress has been made in this regard since 1976. For example, Turk and colleagues [53], recognize the “patient uniformity myth”. There is now ample evidence that patients with chronic pain are quite heterogeneous and have different combinations of physical, cognitive, behavioral and affective contributions to their experience of pain and disability. Two approaches have been proposed to address these differences: treatment based on patient characteristics (e.g., MPI subgroups [60], fear-avoidance [4], task persistence [63]) and tailoring treatment to individual patients (e.g., treatment preferences [33]). Despite recognition of the importance of individual differences and the initial steps taken to adapt our treatments, it is still unclear whether these efforts will improve treatment outcomes for those with chronic pain. Although it seems reasonable to assume that treatments that are tailored to individual patient needs will be more effective, treatment-matching research is at an early stage and the results are mixed [12,33].
Although our field has advanced in the sense that there is virtually universal acceptance among pain researchers that pain is and treatment should be multidimensional, most often we offer narrow (e.g., pharmaceutical, surgical) treatments based on the biomedical model. It is unclear why a clinical paradigm shift has not accompanied the scientific shift. Many explanations are possible, including inadequate education of health professionals, existing systems of provider remuneration (e.g., fee-for-service) or perceptions that multimodal assessment and treatment are more expensive. Certainly cost containment has been cited as a reason for closing multidisciplinary pain treatment programs worldwide over the past two decades, even though the available evidence suggest that such programs are cost-effective [20]. It is also fair to say that multidisciplinary programs usually take a “shotgun” approach in which patients get the entire treatment “package”, irrespective of their individual needs. Perhaps this can be attributed to a need for conceptual and administrative simplicity as well as the paucity of evidence for more idiographic approaches within programs. As Fordyce notes (p. 242)[18], there are no panaceas. There remain many patients for whom our treatments, in spite of their general effectiveness, do not produce clinically meaningful improvement. Strategies to better address individual differences offer the possibility of optimal treatment outcomes and deserve more attention [65]. However, a better understanding of how, when and for whom behavioral treatments are effective is unlikely to occur without drastic changes to our research methodologies [44], including single-case research designs originally developed for the purpose by applied behavioral analysts, e.g. [22], but now rarely used.
The remainder of Chapter 7 discusses how to establish the conditions for use of operant techniques, essentially the clinical context for behavioral analysis, as well as exploration of operant mechanisms related to the role of the patient’s significant others.
IDENTIFICATION OF BEHAVIORS FOR CHANGE
Fordyce was very clear in defining the patient selection criteria for operant treatment programs. Contingency management is only effective for treatment of patients whose level of disability is significantly influenced by environmental consequences. A point that seems to have often been missed by critics of operant programs is that he never intended that they be used for all patients who have chronic pain. It seems to me that he made every effort (I won’t say “took pains”) to differentiate operant from respondent pain and to indicate that although those with “mainly respondent pain” might achieve some benefit by increasing “well behaviors” such as exercise or reducing drug medication intake, operant treatment is not primarily meant for them.
In a 1985 response to critics of operant treatment, he pointed out [19] that:
“ … it is unreasonable to expect that every person who presents at a chronic pain clinic should be treated or might benefit from that treatment. Programs that are not selective appear to have larger dropout rates and lower success rates” (p. 117) and “In the case of pain behaviors occurring as a direct response to nociception, it is unlikely that pain, however defined, will be altered profoundly.” (p. 122).
In contrast to patients with respondent pain whose pain is a more likely a response to nociception (pain arising from an internal, physical stimulus), those whose behavior is significantly under the control of environmental contingencies are most likely to benefit from operant treatment programs. In such instances environmental reinforcement shapes and maintains maladaptive behaviors such as pain behaviors, inactivity and excessive use of narcotic analgesics. Although the use of operant treatments to decrease or eliminate narcotics has been somewhat controversial, the accumulated evidence suggests that their long-term use may only appropriate for a small subset of individuals, they should generally not be used as a first-line treatment, and their use must be balanced with the risk of adverse side effects [18]. The role of learning in relation to analgesic use has also found support in studies that have shown that pain-contingent analgesic administration is associated with poorer outcomes, including increased pain behavior [2], higher pain levels and worse mood [3,66], although some recent data challenges this view [54].
The process of identifying specific pain behaviors that can be targeted for extinction has changed little since Fordyce’s recommendations. However, there have been efforts to provide systematic evaluation of this domain. The first of these was that of Keefe and Block [26] who developed an observational system that assessed five pain behaviors (guarding, sighing, grimacing, rubbing, bracing) using videotapes of patients while they were sitting, standing, walking, and reclining. Trained observers could reliably code these behaviors and they were correlated with patient pain ratings. In addition, the frequency of these behaviors decreased significantly following behavioral treatment. Follick et al. [16] went a step further and developed an empirically based taxonomy of pain behaviors that reduced an initial group of 80 pain behaviors to a 16-category observational system. Of these categories, seven had sufficient frequencies and reliabilities for further analysis (guarding, bracing, position shifts, partial movement, grimacing, limitation statements, and sounds). The categories significantly discriminated chronic pain from non-pain controls. Such systems were useful additions to the literature from the point of view of informing the selection of pain behaviors in operant treatment programs, allowing one to systematically study the role of pain behaviors, and improving evaluation of treatment outcomes. An excellent, and still current, review of the assessment of pain behaviors has been provided by Keefe and colleagues [31].
FEAR-AVOIDANCE
Perhaps the most significant development in behavioral treatment approaches to disability in the last three decades has been the work of Johan Vlaeyen and colleagues [e.g., 64]. The substantial body of work investigating the fear-avoidance model has clearly demonstrated that learning in the context of pain goes beyond purely operant principles. For some individuals with persistent pain, fear of pain and reinjury can be more disabling than the pain itself [9, 38]. This model is cognitive-behavioral – integrating a cognitive component (catastrophizing) with affect (fear, anxiety) to produce avoidance of specific activities and consequently increased disability. Treatments such as exposure-based therapies, which reduce fear-avoidance significantly, benefit patients for whom this is a problem. However, as Nicholas [46] points out, a high percentage of patients do not return to work even after this type of treatment and many remain depressed. Other individual difference variables also impact the effectiveness of this treatment. For example, Cook, et al. [8] have shown that, on average, older patients with chronic pain have less pain-related fear than those who are middle-aged. Leeuw et al. [38] note that some individuals do not respond to fear of pain by efforts to avoid it. Rather, they persist in confronting the fear-eliciting situation. At present it is unclear why this type of behavior occurs. Nonetheless, given what is known about the role of fear-avoidance in chronic pain, a psychological assessment of those with chronic pain would not be complete without consideration of the role of this mechanism.
OVERACTIVITY AND ACTIVITY PACING
One aspect of Fordyce’s work that is often missed by those who have only a cursory understanding of the operant model was his early recognition of individuals who he viewed as overactive. Such individuals presumably find overactivity reinforcing, despite the fact that they may experience increased pain either at the time or some time later as a result. As Fordyce points out “A patient with a too-high activity level might then be selected for help in decreasing that level. Rest or some form of less vigorous activity then becomes the behavior to be increased. Periods of relatively unrestricted activity might be programmed as the reinforcer” (p. 243)[18]. For some patients, the increased pain associated with periods of overactivity causes a subsequent withdrawal from activity, due to increased pain, and perpetuates a “yo-yo” pattern [61]. This pattern may also produce fear-avoidance of activity following pain exacerbations and prolonged episodes of inactivity before returning to overactivity in order to “make up for lost time”. Recognition of this subset of patients led to inclusion of activity pacing as a component of many multidisciplinary chronic pain programs. However, until recently [48], activity pacing has not been the subject of much theoretical discussion or empirical research. Competing conceptual frameworks exist concerning the purpose, function, and methods that should be used to assist those who have an overactive pattern regulate their behavior more effectively. Advocates of the “energy conservation” approach assume that those with pain or other chronic illnesses have a limited amount of energy to expend each day and that they should conserve their energy so that they can participate in the activities that they value most. This approach is pain avoidant and in direct contrast with the operant approach to pacing which specifically seeks to reduce overactivity by setting time-contingent quotas for appropriate levels of activity and insuring that adherence is reinforced. Although some evidence supports the use of activity pacing for individuals with chronic pain [e.g., 63], this is an area that requires further research to determine treatment efficacy. Surprisingly, there has never been a randomized controlled trial of activity pacing techniques for chronic pain despite their widespread use. We (Racine, Jensen & Nielson) are currently completing a trial comparing operant and energy conservation approaches in patients with Fibromyalgia Syndrome.
MOTIVATION TO PARTICIPATE IN TREATMENT
Fordyce’s comments on p. 245 [18] foreshadow subsequent work on patient readiness to change. He notes that patients and/or spouses are often ambivalent about proceeding with treatment and skeptical about whether it will be helpful. Although not specifically discussed by Fordyce, in my experience there are individuals who are not compliant with treatment even when they initially agree to it and sign a behavioral contract. There are many reasons for this reluctance and the patient does not always openly voice them. However, some (a small minority) of individuals candidly admit that they do not wish to increase their activity level or report improvements in other areas because it might jeopardize their chances of obtaining compensation. In general, these individuals should not be admitted to pain treatment programs until the competing source(s) of reinforcement has/have been resolved. The much larger number of patients who express ambivalence should not be excluded from treatment. Indeed, prior to entering programmatic treatment patients typically have tried numerous interventions with little improvement. With that learning history, some ambivalence about yet another intervention strategy would seem appropriate and reasonable. Along these lines, many patients with a pending compensation claim will express a desire to increase their activity level and will benefit from treatment. Thus, a pending compensation claim in and of itself should not serve as a barrier to initiating treatment.
Since Fordyce first described his selection criteria, others have recommended similar admission criteria that address the issue of patient motivation. For example, Follick et al. [17] indicate that the patient and healthcare provider must agree that an operant program is the best approach and that patients must be motivated to participate in the program, including an explicit, public, commitment to learning to live with pain.
The use of a behavioral contract as suggested by Fordyce to further “solidify their [patient and spouse] commitment to proceed” (p. 246) has not, as far as I am aware, been subjected to empirical scrutiny (though behavioral contracts are widely used in the long-term treatment of pain using opioids (and I believe has been subject to experimental studies). Although providing this type of clarification for patients seems reasonable, it is unclear whether it improves motivation, adherence or treatment outcomes. A systematic review of the use of contracts for health behavior change, more generally [5], concluded that there is insufficient evidence to recommend their routine use. More specific contracting that includes goal-setting and regular review may have more potential [e.g., 7].
Within the operant model, motivation is considered to be a deprivation-satiation continuum: states of relative deprivation are associated with a higher reinforcement value for a given consequence. Complex schedules of reinforcement and related mechanisms [e.g., 49] are thought to determine whether behavioral responses do or do not occur under specific stimulus conditions. Patients with chronic pain are thought to use specific pain coping strategies not because they are maladaptive or adaptive in the long run, but because they are followed by reinforcement in the short run.
Recognizing that cognitive processes, particularly outcome expectancies, are also an important determinant of motivation for change, Jensen, Nielson, and Kerns proposed a motivational model of patient self-management [24]. This model incorporates current health behavior models and makes specific recommendations for interventions that are likely to promote patient motivation to adopt adaptive pain coping behaviors. From this perspective, patients will not stop using maladaptive pain management strategies and begin using adaptive self-management strategies “until and unless they believe that the maladaptive strategies will result in negative outcomes and the self-management strategies will result in positive outcomes” (p. 486). Promotion of self-efficacy and efforts to increase the perceived importance of behavior change (e.g., using motivational interviewing strategies) could be used to enhance engagement in both “pure” operant treatment programs as well as those that are multimodal [23]. Most research has supported the role of such motivational factors (30,32,25,45,37,11,21,58,35) but there are some exceptions [13]. Although assessment of patient readiness to engage in operant treatment likely influences treatment outcomes, no research has evaluated this possibility. If supported by research, it is also possible that instruments designed to assess readiness to engage in treatment (e.g., PSOCQ [36], MPRCQ2 [47]) could aid in patient selection. However, the multidimensionality of selection considerations and the inherently imperfect nature of psychometric instruments means that the decision to include or exclude a patient from treatment will remain a clinical judgment based on integration of all relevant assessment data. Anecdotally, even given a thorough assessment, my predictions of how much patients will benefit from treatment programs have been far from perfect and at times I have been surprised at how much better (or worse!) patients have done than I had expected. Patient’s attitudes and their level of commitment change during the treatment process, perhaps early on [6]. From an operant perspective immediate outcomes of their behavior change efforts are likely to shape their subsequent commitment to continuing treatment and to maintenance of the gains that have been achieved. Thus it may be especially important to set initial treatment goals (e.g., quotas) in a manner that ensures success (achievable, meaningful to the patient and supported by significant others). Both ethically and empirically it may be best to err on the side of including patients in treatment even if doubts about their level of commitment exist at the pre-treatment assessment.
THE ROLE OF SIGNIFICANT OTHERS
It is interesting that Fordyce allotted about 30% of his chapter on patient selection to the role of spouses and family members. This emphasis is a recognition that the success of operant (and other) treatment programs often hinges on the influence of the patient’s immediate environment: if key people in the patient’s life are not committed to change and engaged in a meaningful manner, behavior changes made during treatment may not be maintained later. This is especially true of outpatient programs, which even in the early stages of treatment rely very heavily on family involvement, especially the spouse or “one key person” (p. 381)[18]. Certainly, most subsequent research has strongly supported the impact of spousal responses to patient pain behaviors. Solicitous behavior on the part of the spouse is associated with decreased activity and higher levels of pain behavior [14,15, 50,51]. However, negative spousal responses present a mixed picture and other factors such as marital adjustment [56], marital satisfaction, patient depression [39] and gender [40] can moderate these relationships. Unfortunately, the few existing randomized, controlled trials of operant treatment for chronic pain have not isolated the treatment elements that are necessary and sufficient for positive patient outcomes [62]. Spousal participation generally appears to be associated with positive outcomes in operant treatment regimens [57], although one treatment study [1] found that such involvement “was not essential for a positive response to treatment” (p. 326). If one considers the literature beyond strict operant programs, the evidence indicates that systematic training of spouses to coach and reinforce patients’ use of coping skills can be highly effective [1,27,28,29]. Although there has been less research on the role of non-spousal family members it is thought that their involvement in treatment can also play an important role in patient adjustment to chronic pain [34,43,52]. In summary, the available data suggest that in selecting patients for operant programs clinicians should evaluate the reinforcing and punishing responses of significant others and the learning context in which these interactions occur.
CONCLUSION
The complexity of chronic pain (in all of its dimensions, including learned behavior) has only become increasingly obvious in the years since Fordyce’s work began. As with other chronic diseases, the goal must be management rather than cure and behavioral methods form a cornerstone of the contemporary pain management regimen. For some patients, following Fordyce’s recommendations for selection, operant methods form a key to successful rehabilitation. However, it is likely that relatively few patients would receive optimal treatment if only operant treatment were provided. General recognition of the complexity of chronic pain and the need to address the respondent component as well as vocational, social, and other psychological issues likely played a key role in the development of multidisciplinary treatment programs and the integration of operant strategies into the larger treatment “package”. The success of cognitive-behavioral treatments also suggests that for many patients a “pure” operant treatment approach may be insufficient. In general, it appears that operant treatments are most likely to improve pain behavior, functional impairment and solicitous spouse behaviors whereas cognitive-behavioral treatments are more effective in reducing affective distress and cognitive variables such as catastrophizing [1]. For patients with high pre-treatment levels of pain behavior, operant treatments are likely to be most effective; patients with low levels may be less likely to benefit [27].
REFERENCES
1. Abbasi M, Dehghani M, Keefe FJ, Jafari H, Behtash H, Shams J. Spouse-assisted training in pain coping skills and the outcome of multidisciplinary pain management for chronic low back pain treatment: a 1-year randomized controlled trial. Eur J Pain 2012;16:1033–43.
2. Ballantyne JC, Shin NS. Efficacy of opioids for chronic pain: a review of the evidence. Clin J Pain 2008;24:469–78.
3. Berntzen D, Gotestam KG. Effects of on-demand versus fixed-interval schedules in the treatment of chronic pain with analgesic compounds. J Consult Clin Psychol 1987;55:213–17.
4. Boersma K, Linton SJ. Psychological processes underlying the development of a chronic pain problem: a prospective study of the relationship between profiles of psychological variables in the fear-avoidance model and disability. Clin J Pain 2006;22:160–66.
5. Bosch-Capblanch X, Abba K, Prictor M, Garner P. Contracts between patients and healthcare practitioners for improving patients’ adherence to treatment, prevention and health promotion activities. Cochrane Database of Systematic Reviews 2007, Issue 2. Art. No.: CD004808. DOI: 10.1002/14651858.CD004808.pub3.
6. Burns JW, Nielson WR, Jensen MP, Heapy A, Czlapinski R, Kerns RD. Specific and General Therapeutic Mechanisms in Cognitive-Behavioral Treatment of Chronic Pain. J Consult Clin Psychol. In press. http://dx.doi.org/10.1037/a0037208.
7. Chisholm-Burns MA, Spivey CA, Graff Zivin J, Lee JK, Sredzinski E, Tolley EA. Improving outcomes of renal transplant recipients with behavioral adherence contracts: a randomized controlled trial. Am J Transplant 2013;13:2364–73.
8. Cook AJ, Brawer PA, Vowles KE. The fear-avoidance model of chronic pain: validation and age analysis using structural equation modeling. Pain 2006;121:195–206.
9. Crombez G, Vlaeyen JW, Heuts PH, Lysens R. Pain-related fear is more disabling than pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain 1999;80:329–39.
10. Duncan G. Mind-Body Dualism and the Biopsychosocial Model of Pain: What Did Descartes Really Say? J Med Philos 2000;25:485–513.
11. Elander J, Robinson G, Morris J. Randomized trial of a DVD intervention to improve readiness to self-manage joint pain. Pain 2011;152:2333–41.
12. Evers AWM, Kraaimaat FW, van Riel PLCM, de Jong AJL. Tailored cognitive-behavioral therapy in early rheumatoid arthritis for patients at risk: a randomized controlled trial. Pain 2002;100:141–53.
13. Finan PH, Burns JW Jensen MP, Nielson WR, Kerns RD. Pain Coping but not Readiness to Change is Associated with Pre-Treatment Pain-Related Functioning. Clin J Pain 2012;28:687–92.
14. Flor H, Kerns R D, Turk D C. The role of spouse reinforcement, perceived pain, and activity levels of chronic pain patients. J Psychosom Res 1987;31:251–59.
15. Flor H, Turk D C, Rudy T E. Relationship of pain impact and significant other reinforcement of pain behaviors: The mediating role of gender, marital status, and marital satisfaction. Pain 1989;38:45–50.
16. Follick MJ, Ahern DK, Aberger EW. Development of an audiovisual taxonomy of pain behavior: Reliability and discriminant validity. Health Psychology 1985;4:55–68.
17. Follick MJ, Zitter RE, Ahem DK. Failures in behavior therapy. In: EB Foa and P Emmelkamp (Eds.), Failures in Behavior Therapy, Wiley, New York, 1982.
18. Fordyce, W.E., Behavioral Methods for Chronic Pain and Illness, Mosby, St. Louis, MO, 1976.
19. Fordyce W, Roberts A, Sternbach R. The behavioral management of chronic pain: a response to critics. Pain 1985;22:113–25.
20. Gatchel RJ, Okifuji A. Evidence-based scientific data documenting the treatment and cost-effectiveness of comprehensive pain programs for chronic nonmalignant pain. J Pain 2006;7:779–93.
21. Gersh E, Arnold C, Gibson SJ. The relationship between the readiness for change and clinical outcomes in response to multidisciplinary pain management. Pain Med 2011;12:165–72.
22. Hersen M, Barlow, DH. Single-case Experimental Designs: Strategies for Studying Behavioral Change. New York: Pergamon Press, 1976.
23. Jensen MP. Enhancing motivation to change in pain treatment. In DC Turk & RJ Gatchel (Eds.), Psychological approaches to pain management: A practitioner’s handbook, 2nd ed. New York: Guilford Press, 2002.
24. Jensen MP, Nielson WR, Kerns RD. Toward the development of a motivational model of pain self-management. J Pain 2003;4:477–92.
25. Jensen MP, Nielson WR, Turner JA, Romano, JM, Hill ML. Readiness to self-manage pain is associated with coping and with psychological and physical functioning among patients with chronic pain. Pain 2003;104:529–37.
26. Keefe FJ, Block AR. Development of an observational method for assessing pain behavior in chronic low back pain patients. Behav Ther 1982;13:363–75.
27. Keefe FJ, Blumenthal J, Baucom D, Affleck G, Waugh R, Caldwell DS, Beaupre P, Kashikar-Zuck S, Wright K, Egert J, Lefebvre J. Effects of spouse-assisted coping skills training and exercise training in patients with osteoarthritic knee pain: a randomized controlled study. Pain 2004;110:539–49.
28. Keefe FJ, Caldwell DS, Baucom D, Salley A, Robinson E, Timmons K, Beaupre P, Weisberg J, Helms M. Spouse-assisted coping skills training in the management of osteoarthritis knee pain: long-term followup results. Arthritis Care Res 1999;12:101–11.
29. Keefe FJ, Caldwell DS, Baucom D, Salley A, Robinson E. Timmons K, Beaupre P, Weisberg J, Helms M. Spouse- assisted coping skills training in the management of osteoarthritic knee pain. Arthritis Care Res 1996;9:279–91.
30. Keefe FJ, Lefebvre JC, Kerns RD, Rosenberg R, Beaupre P, Prochaska J, Prochaska JO, Caldwell DS. Understanding the adoption of arthritis self-management: stages of change profiles among arthritis patients. Pain 2000;7:303–13.
31. Keefe FJ, Williams DA, Smith SJ. Assessment of pain behaviors. In DC Turk R. Melzack (Eds.), Handbook of pain assessment, 2nd ed. New York: Guilford Press, 2001.
32. Keller S, Herda C, Ridder K, Basler HD. Readiness to adopt adequate postural habits: an application of the Transtheoretical Model in the context of back pain prevention. Patient Educ Couns. 2001;42:175–84.
33. Kerns RD, Burns JW, Shulman M, Jensen MP, Nielson WR, Czlapinski R, Dallas M, Chatkoff D, Sellinger J, Heapy A, Rosenberger P. Can we improve chronic back pain treatment engagement and adherence? Controlled trial of tailored versus standard cognitive behavior therapy. Health Psychol 2014;33:938–947.
34. Kerns RD, Otis JD, Wise EA. Treatment families of chronic pain patients: Application of a cognitive-behavioral transactional model. In DC Turk & RJ Gatchel (Eds.), Psychological approaches to pain management: A practitioner’s handbook, 2nd ed. New York: Guilford Press, 2002.
35. Kerns RD, Rosenberg R. Predicting responses to self-management treatments for chronic pain: application of the pain stages of change model. Pain 2000;84:49–55.
36. Kerns RD, Rosenberg R, Jamison RN, Caudill MA, Haythornthwaite J. Readiness to adopt a self-management approach to chronic pain: the Pain Stages of Change Questionnaire (PSOCQ). Pain 1997;72:227–34.
37. Kratz AL, Molton IR, Jensen MP, Ehde DM, Nielson WR. Further evaluation of the Motivational Model of Pain Self-Management: coping with chronic pain in multiple sclerosis. Ann Behav Med 2011;41:391–400.
38. Leeuw M, Goossens ME, Linton SJ, Crombez G, Boersma K, Vlaeyen JW. The fear-avoidance model of musculoskeletal pain: current state of scientific evidence. J Behav Med 2007;30:77–94.
39. Leonard MT, Cano A, Johansen AB. Chronic pain in a couples context: a review and integration of theoretical models and empirical evidence. J Pain 2006;7:377–90.
40. Leong LE, Cano A, Johansen AB. Sequential and base rate analysis of emotional validation and invalidation in chronic pain couples: patient gender matters. J Pain 2011;12:1140–8.
41. Melzack R, Wall PD. The Challenge of Pain (updated 2nd ed). London: Penguin Books, 1982.
42. Merskey H. Re: Back pain in the workplace, W.E. Fordyce (ed.) (Letter to the Editor) Pain 1996;65:111–12.
43. Moore JE, Chaney EF. Outpatient Group Treatment of Chronic Pain: Effects of Spouse Involvement. J Consult Clin Psychol 1985;53:326–34.
44. Morley S, Williams A, Eccleston C. Examining the evidence about psychological treatments for chronic pain: time for a paradigm shift? Pain 2013;154:1929–31.
45. Molton IR, Jensen MP, Nielson WR, Cardenas D, Ehde DM. A preliminary evaluation of the motivational model of pain self-management in persons with spinal cord-related pain. J Pain 2008;2:606–12.
46. Nicholas MK. First things first: Reductions in catastrophising before fear of movement. Pain 2009;145:6–7.
47. Nielson WR, Armstrong JM, Jensen MP, Kerns RD. Two brief versions of the Multidimensional Pain Readiness to Change Questionnaire (MPRCQ2). Clin J Pain 2009;25:48–57.
48. Nielson WR, Jensen MP, Karsdorp PA, Vlaeyen JW. Activity pacing in chronic pain: concepts, evidence, and future directions. Clin J Pain 2013;29:461–8.
49. Premack D. Reinforcement theory. In Levine D (Ed.): Nebraska Symposium on Motivation: 1965. Lincoln, NE, University of Nebraska Press, 1965.
50. Riemsma R, Taal E, Wiegman O, Rasker JJ, Bruyn GAW, van Paassen HC. Problematic and postive support in relation to depression in people with rheumatoid arthritis. J Health Psychol 2000;5:221–30.
51. Romano JM, Jensen MP, Turner JA, Good AB, Hops H. Chronic Pain Patient-Partner Interactions: Further Support for a Behavioral Model of Chronic Pain. Behavior Therapy 2000;31:415–440.
52. Romano JM, Schmaling KB. Assessment of couples and families with chronic pain. In DC Turk R. Melzack (Eds.), Handbook of pain assessment, 2nd ed. New York: Guilford Press, 2001.
53. Rusu AC, Boersma K, Turk DC. Reviewing the concept of subgroups in subacute and chronic pain and the potential of customizing treatments. In MI Hasenbring, AC Rusu, DC Turk (Eds.), From acute to chronic back pain: Risk factors, mechanisms, and clinical implications. Oxford, UK: Oxford University Press;2012:485–511.
54. Sullivan M. Will data destroy our faith in long-acting opioids? Pain 2014;155:843–44.
55. Supreme Court of Canada. Nova Scotia WCB v Martin, 2 S.C.R. 504, 2003 SCC 542003.
56. Taylor SS, Davis MC, Zautra AJ. Relationship status and quality moderate daily pain-related changes in physical disability, affect, and cognitions in women with chronic pain. Pain 2013;154:147–53.
57. Thieme, K, Turk DC, Flor H. Responder criteria for operant and cognitive–behavioral treatment of Fibromyalgia Syndrome. Arthritis Care Research 2007;57: 830–836.
58. Tkachuk GA, Marshall JK, Mercado AC, McMurtry B, Stockdale-Winder F. Readiness for change predicts outcomes of functional rehabilitation following motor vehicle accident. J Occup Rehabil 2012;22:97–104.
59. Turk DC, Flor H. Pain greater than pain behaviors: the utility and limitations of the pain behavior construct. Pain 1987;31:277–95.
60. Turk DC, Okifuji A, Sinclair JD, Starz TW. Differential responses by psychosocial subgroups of fibromyalgia syndrome patients to an interdisciplinary treatment. Arthritis Care Res 1998;11:397–404.
61. Turk DC, Sherman JJ. Treatment of patients with Fibromyalgia Syndrome. In DC Turk DC & RJ Gatchel (Eds.), Psychological approaches to pain management: A practitioner’s handbook, 2nd ed. New York: Guilford Press, 2002.
62. Turner JA, Chapman CR. Psychological interventions for chronic pain: a critical review. II. Operant conditioning, hypnosis, and cognitive-behavioral therapy. Pain 1982;12:23–46.
63. van Koulil S, van Lankveld W, Kraaimaat FW, van Helmond T, Vedder A, van Hoorn H, Donders R, de Jong AJ, Haverman JF, Korff KJ, van Riel PL, Cats HA, Evers AW. Tailored cognitive-behavioral therapy and exercise training for high-risk patients with fibromyalgia. Arthritis Care Res 2010;62:1377–85.
64. Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain 2000;85:317–32.
65. Vlaeyen JWS, Morley S. Cognitive-behavioral treatments for chronic pain: What works for whom? Clin J Pain 2005;21:1–8.
66. White B, Sanders SH. Differential effects on pain and mood in chronic pain patients with time-versus pain-contingent medication delivery. Behav Ther 1985;16:28–38.
Commentary: Rigour and Principle Versus the Psychiatric Industry: Confusion in Patient Selection
Re-reading Fordyce’s book was a revelation (I have had it on book search databases for many years, but, understandably, none of those who had a copy was letting it go). Some forty years after it was published it seems fresh, strikingly patient-centered, and kind. He forbears from judging the choices people make in relation to their pain, or assigning responsibility for the difficult situations in which they find themselves. His focus is on propagating what he had shown could improve the situation, and in this chapter, identifying those whose behaviour and its context best fitted behavioural methods. Here I review how trials of behavioural methods, particularly in the decade following publication of his work, sometimes followed but in many cases traduced Fordyce’s clear principles by introducing psychiatric and psychopathological judgments. A generous reading of this would emphasize prevailing models at the time, psychodynamic psychology and psychosomatic medicine; and perhaps this is responsible for its persistence decades later, despite the rise and spread of psychology in the treatment of chronic pain. Perhaps re-reading Fordyce will enable us to reconnect to his rigor and his compassion.
Fordyce’s work gave psychologists a treatment that was specific, theory-based and integrated with pain science, and – most welcome to pain clinicians – directed towards the patients whom medicine found hard to manage. However, not all the therapeutic work that followed, in work or in practice, was so rigorous and value-free, as I detail below. Incorporation of behavioural methods and terminology into moralistic frameworks in medical and particularly psychiatric thinking, and frank pathologizing of behaviour using psychiatric terminology, muddied the waters. Even where psychiatric concepts conflicted directly with the tenets of learning theory, they were embraced. Remarkably, this confusion persists, emerging in different language such as ‘functional disorders’ (see [53]) or “medically unexplained behaviour” (see [66]), all of which rely on implicating psychological factors by the absence of medical findings, a profoundly dualistic model. It is therefore useful to examine in particular the decade or so of clinical activity which took Fordyce’s work as inspiration and example, before cognitive frameworks became the dominant explanatory framework for behaviour (for an early integration, see [43]), and to examine recent practice in selection of patients.
At the point of writing his book, there were relatively few demonstrations of the efficacy of operant methods applied to the problems of chronic pain: Fordyce’s own case examples, alongside an elaboration of the principles of treatment in 1968, and the effectiveness of social reinforcement of quota activity in a multiple baseline study of exercise performance [9]. In laboratory studies, White and Sanders [64] demonstrated that pain talk was sensitive to social reinforcement; Fordyce [16] and Cairns and Pasino [6] showed the effects of feedback on physical performance; Block et al. [5] compared nonverbal pain behaviour in different interpersonal conditions, often erroneously cited as showing that patients with solicitous spouses rated their pain higher when observed by their spouse rather than by a stranger (e.g., [58]). In fact, there was no difference for patients who rated their spouses as solicitous, but for those who rated their spouses as non-solicitous (and who incidentally had chronic pain of much shorter duration than those with solicitous spouses), their ratings of pain were lower when observed by spouse than when observed by a stranger. These mixed findings are hard to interpret, and are open to alternative explanations such as expectations of empathy.
In general, success of treatment which followed Fordyce’s methods was cited as evidence for the applicability of the behavioural model to chronic pain. It is common to find statements implying that pain behaviours are inevitably maladaptive, despite explicit challenge to this view by Turk and Matyas [60], Keefe and Gil [29] and others. The value-free discourse of Fordyce’s original work was thereby subsumed under non-contextual descriptions of behaviour which opened the doors to ad hominem explanations. One of the most widely propagated concepts was that of “abnormal illness behaviour” [44], defined by reference to “objective” medical findings, and thus reintroducing the dualistic framework which Fordyce had so clearly rejected, and which Keefe and colleagues [28,30] challenged on empirical grounds using findings from physical examination.
PSYCHOLOGICAL MODELS OF PAIN IN THE CONTEXT OF DISEASE MODELS
In a scholarly review in 1994, Gamsa [22] describes the evolution of psychological models of pain, particularly in relation to Melzack and Wall’s gate control model, and the contemporaneous decline of psychoanalytically-based theories of pain. At the time, behavioural models were explicitly contrasted with disease models of pain (see Fordyce chapter 2 in particular). For instance: “According to psychiatric theorists, maladaptive behaviours exhibited by a chronic pain patient provide a vehicle by which patients communicate their distress and unmet emotional needs. These psychiatric theories view anger, repressed hostility, anxiety, loss and depression as particularly important determinants of behaviour in chronic pain patients” (review: Keefe 1989, p. 407 [27]). Fordyce argued for a “behavioural science perspective” alongside the “medical science perspective” (Fordyce, 1990, p292 [17]) and, using gate control theory as his foundation, was critical of models of psychogenesis of pain with no supporting mechanistic model. He and his contemporaries were also critical of the “victim-blaming” [45], imputations of presumed etiology, including ‘hysteria’ and ‘hypochondriasis’, and, at a deeper level, of the dualism implicit in assigning pain to either the physical or mental domain [52].
At some points, psychiatric domination overwhelmed the attempts exemplified by Fordyce to refer to theory, to avoid pathologizing people with pain, and to focus on practical matters in selection. Sternbach himself later referred to distinguishing between psychogenic and somatogenic pain by the presence or absence of physical findings to account for pain using the classical disease model [51]. Others compromised, not challenging the label of psychogenic pain but equating its distress with that of medically-authenticated pain [8]. It is a serious weakness which persists to this day that on such secure foundations as the gate control theory and Fordyce’s principled behavioural stance, psychology abandoned the direction towards what has been described as a ‘normal psychology of pain’ [11] and capitulated wittingly or unwittingly to psychiatric models which had at best a weak scientific basis and little coherence to guide treatment development.
PATIENT SELECTION IN EARLY TREATMENT STUDIES AFTER FORDYCE
The operant treatment model was widely taken up after Fordyce et al. in 1968 [18], perhaps because of the growing recognition of the extent of the problem of unsuccessfully-treated chronic pain and its medical, personal and social costs. Among the earlier treatment studies, not all made their models and selection criteria explicit and, among those that did, there is evidence of divergence from a learning theory framework and from Fordyce’s rigorous attempt to avoid psychopathological formulation. Some researchers defined eligible patients by reference to ‘physical’ pain; some used an additive model, whereby the patient could suffer both physical and psychological problems, with the latter attributable to the former; and some used a more integrated model in which physical and psychological problems were assumed to be interrelated in multiple ways.
The extent of confusion, however, is exemplified by the use of ‘physical’ pain to decide eligibility. Some programs excluded patients with a “presumed organic cause” for pain (e.g., [32]): others defined patients’ pain by evidence of organic cause with failure to respond to medical treatment (e.g., [23,35]), a failure attributed by some (such as Gottlieb et al. [23]) to patient psychopathology. The additive model was similarly realized in heterogeneous ways: some programs assessed independently for physiological and psychological components of pain (e.g., [42]), while others proposed that psychological problems inevitably developed with prolonged pain and merely complicated assessment of pain problems (e.g., [24]).
Further confusion emerges from examining early studies. Seres and Newman (1976 [50]) proposed behavioural treatment methods to be applied to active physical therapy, relaxation, and drug use, but described patients in terms of hysterical conversion, secondary gain and “sociopathic or characterological personality disturbances”. A later publication from the same group [40] reported that none of these diagnoses was related to treatment success. Another confusing example [6] opens with a behavioural explanation of problems associated with chronic pain, but then describes the use of pentothal to uncover “significant psychogenic manifestation”. The authors discovered to their evident surprise that treatment response was unrelated to diagnosis, concluding that “The implication is that neither a “hysterical” process nor tissue pathology were controlling the patient’s pain behaviour and associated disability.”(Cairns & Pasino, p. 629 [6]). They also discovered that operant conditioning could be effective “regardless of the presence of psychogenic factors” and further, that “patients such as these have typically been labeled ‘unmotivated’ by the treatment staff when, in fact, the absence of reinforcement for attempts to increase physical function may contribute to the failure of rehabilitative efforts” (p. 629). Clearly, selection using psychiatric labels was not sustained by evidence.
In the decade of studies following Fordyce’s elaboration of treatment methods, there is no single coherent school of psychopathology from which descriptions of patients are drawn; models are implied, rather than specified, and based on theories of personality, of psychiatric nosology, and of psychodynamic motivation of behaviour. The exclusion of patients by the use of these terms and measures that arose from them, such as the MMPI, rarely described the basis for doing so. In rejecting the use of concepts of psychopathology, Fordyce et al. (1973)[19] wrote:
If the [operant] methods used produced conflict resolution or personality change, so much the better, though our data suggest that focus on constructs of conflict resolution or personality change are neither necessary nor sufficient to account for the results observed (p. 407).
Some authors referred (explicitly or implicitly) to models of adaptation, and to the learning of unhelpful behaviours, but not necessarily expressed in those terms. Remarkably few take the broad framework of operantly controlled behaviours as their treatment focus, rather selecting specific targets for contingency management (such as financial compensation after work accidents [25]; symptom and stimulus control [32]).
Specific practices and commentaries on selection from these historical studies show only one [4] which was entirely unselective in order to examine post hoc indications and contraindications for treatment. A few followed closely Fordyce’s identification of behaviours and relevant contingencies (e.g., [1]), even requiring availability of a partner or family member for reinforcement of well behaviours in the patient’s own environment although only the patient was directly treated (e.g., [46]). Assessment for selection and specification of treatment aims could be prolonged: up to a week’s inpatient stay for interview by each member of the multidisciplinary team and observation of operant behaviours (e.g., Anderson et al. [1], who subsequently rejected over half the patients assessed).
The Mayo group perhaps wrote most about selection in a series of publications with a high level of detail (see [54,56]). Uniquely, I believe, they investigated whether applying their (post hoc derived) predictors of success to selection improved attrition or outcome: it did not [33]. They explicitly encouraged patients to exclude themselves if they recognized a conflict between returning to an active life and their own choices, in particular, pursuit of medicolegal claims or drug ‘overuse’ [56]. Following Fordyce’s lead, these might rather have been seen as potential treatment targets. A fairly high rate (over 30%) of attrition from the program made it possible to compare baseline characteristics of completers and non-completers: the latter tended to have higher pain impact, longer duration of pain, more surgery and more drug ‘overuse’; they were also recorded as showing more pain behaviours during treatment. These findings led the team to require several days’ observation in a psychiatric unit before treatment [55], and, curiously, psychopathological terms are used to characterize difficulties in treatment. Strongly-held but unrealistic hopes of pain relief are described as “delusional systems”; distress when staff attempted to correct these beliefs is described as a “paranoid stance” [56].
EXCLUSION: COMPENSATION FOR DISABILITY
A minority of studies excluded potential patients on the grounds of pending litigation; most accepted them but recorded the fact as a patient characteristic with possible adverse impact on outcome. By contrast, while acknowledging the potential of payments (welfare or compensation) to reinforce disability, Fordyce (1976) drew attention to the risk of oversimplification of contingencies, pointing out that few propose that all salaried people can be described as working solely because of the reinforcement of monthly payments, and emphasized the more therapeutically useful focus of immediate and proximal contingencies whose effects could be observed.
Broad suspicion of secondary gain and malingering were invoked by many programs to exclude or even discharge patients, with many judging motivation by interview (e.g., [1, 23, 41]). But, for many, these decisions were made by third party funders before selection occurred [61], refusing funding to up to half those referred, and far less open about their decision-making than were the treatment programs. The few comparisons of untreated or refused patients with those treated from the same referral base reported that the latter at baseline were less disabled, less distressed, and had fewer other problems such as pending litigation [7,10,46].
EXCLUSION: DRUG USE
There is a distinct lack of theoretical development from Fordyce’s original proposals, and varied assumptions about whether analgesics are effective and therefore reinforcing (e.g. [3]) or ineffective (e.g., [62]). Drug dependence as the basis for exclusion was often defined by non-standard criteria (e.g., [46]), while fewer followed Fordyce in identifying it as a target of treatment (e.g., [15]). While drug withdrawal was common practice in programs, consistent with Fordyce, the rationale was not always provided, nor was it clear whether patients had presented drug intake as a problem for them. The terminology of addiction is surprisingly often used (e.g., [51]) as if synonymous with dependence, despite the lack of evidence of long term problems at that time (e.g., [34]). Some researchers took a pragmatic approach to integration of drug withdrawal alongside functional goals (e.g., [21,45]), while others noted the problematic cognitive and mood effects of opioids (e.g., [12]). There appear to be no observational accounts of use or of reduction of opioids in chronic pain patients within treatment programs, and trials only of different schedules of withdrawal [2,57, 64]; none isolating drug use from other pain behaviours and testing theoretically-grounded methods. Yet there is no suggestion that Fordyce considered unhelpful opioid use as anything but another problem to be examined using the framework of learning theory, and he and colleagues [18] published several case examples involving analgesic reduction, linking it with operant treatment, even incorporating drug delivery as a reinforcer for increased activity despite pain.
IDENTIFYING REINFORCERS
For those patients whose pain is identified as operant, Fordyce proposed the conditions under which treatment was likely to be effective, referring to chapter 6: that behaviours or functional impairments can be identified which are targets for reduction and for which change is feasible in their direct and/or indirect positive reinforcement; that well behaviours exist in the repertoire of the patient that are related to the lifestyle of the patient and spouse, and that punishment of or failure to reinforce these can be changed; and that the spouse or other family members who are important sources of reinforcement are willing to collaborate in treatment. Beyond that, he requires that early reinforcers be identified, in particular, rest, attention, medication, and indicators of progress.
The selection criteria therefore require discussion with the patient and his/her partner of broad targets to achieve, and of the partner’s availability and willingness to take part; this may be supplemented by observation of behaviour in context. Fordyce added substantial sections on both the ‘misguided spouse’, who is overprotective but open to more effective ways of helping the person with pain, and the spouse for whom the status quo is reinforcing and who therefore impedes the identification of behaviours to change for both patient and spouse, with less certainty about both feasibility and outcome.
Reference to reinforcement schedules in the patient’s own environment by early treatment studies is almost universal, if only with reference to the maintenance and generalisation of treatment gains. The most thoroughly addressed reinforcement opportunities were spouse and family interactions with the patient (also see Roy [48]). Adherence to this thinking in clinical trials and practice continued for a generation in remarkable isolation from parallel research on stress, health, and social support (noted by Block [3], Turk & Kerns [59], Newton-John & Williams [39], and others), and in apparent ignorance of failures to confirm them (e.g., see [39]), of differences according to gender and quality of the marital relationship (e.g., [13,47]), and counter-examples, where punishing rather than solicitous spouse responses were associated with higher reported pain and greater activity interference [31, and see 39].
CURRENT PRACTICE ON PATIENT SELECTION
What are the current practices in patient selection for behavioural and cognitive programs for chronic pain? There have been many attempts to determine predictors of outcome in order to inform selection: not only do these produce diverse findings and therefore recommendations, but such post hoc analyses are necessarily done only on those patients originally selected for treatment. They are therefore subject to multiple biases, from the broader health system to the specific criteria of that treatment program. There is no basis for expecting the populations from which such treatment programs recruit to be homogeneous, so the predictors from each program may only describe the particular population in which they were calculated (particularly where numbers are relatively low). A study [26] of four different treatment settings demonstrated marked differences between populations referred for treatment. Interestingly, the findings of outcome predictor studies often differ little from what emerges from studies of factors predicting spontaneous remission from low back pain [63].
Of the current major textbooks on chronic pain, some show Fordyce’s influence, describing selection using behavioural analysis in extended interview with the patient and those close to him or her (e.g., Flor & Turk 2011 [14]), although generally expressed more in cognitive than in behavioural dynamics, and often also using psychopathological or empirically-derived ways of defining suitability. Several pursue the issue further, aiming for patient-treatment matching based on a variety of typologies (for criticism of these methods, see [38]).
The patient selection criteria applied to the 42 treatment trials in a recent systematic review [65] give an impression of current practice. Almost a quarter excluded patients with pending litigation and/or receiving compensation or disability-related welfare payments. This is commonly described in terms of operants, whereby the more the patient makes gains in activity and function, the less payment s/he will receive.
Of more concern, given that it is nowhere mentioned by Fordyce as grounds for exclusion, is the use of psychiatric terms to describe patients’ current or past emotional state without placing it in the context of pain. A third of the trials in this review exclude patients with serious or major psychiatric conditions, many mentioning psychosis, and several specifying concern that the conditions would interfere with participation in treatment: these might be said to be pragmatic judgments from clinical experience rather than theory. But a handful exclude patients with a history of or with current depression, two exclude patients with ‘anger’ as a problem, one excludes anyone using antidepressants or anxiolytics, and several exclude “substance abusers” without defining it further.
Comparing these trial criteria with Fordyce’s description of selection, it is clear that examination of behaviour and of reinforcers is rarely if ever carried out: patients qualify by virtue of their diagnosis, sometimes by severity of pain or disability scores, and because no further medical treatment is planned or envisaged. The last is a negative criterion: nothing else can be done, so the problem is reformulated in psychological rather than medical terms. This is something of a contrast to the statement that many opportunities for operant learning occurred during the time during which unsuccessful medical interventions were tried, so “those pain behaviours are the point of attack” (Fordyce et al. 1968, p. 189 [18]). Crucially, in current selection practice, the patient him/herself is the entire focus of this reformulation of problems, making it all too easy to formulate in psychiatric or characterological terms, rather than in context.
In Fordyce, modifiable behaviours are clearly referenced to context and likely sources of reinforcement, most often the spouse, and it was clear that the spouse as far as possible should be engaged actively in treatment. The only trials of the 42 in our review [65] which required a spouse’s or partner’s availability were those few where partner involvement was a central feature of the trial. Another source of reinforcement highlighted in Fordyce’s exploration of likely early support for treatment gains is the patient’s own satisfaction in making progress, a variable which has been either taken for granted or completely lost: it certainly rarely appears in treatment studies. Instead, the behaviour identified as the predominant problem in treatment studies is avoidance, without specifying either the object of avoidance or the reinforcement contingencies believed to operate, and certainly without the observation which supported Fordyce’s treatment plans. Thus, stripped of context, it becomes another deficit in the patient.
PATIENT AS PARTNER IN TREATMENT
Fordyce determined targets for behavioural change from consideration of operant theory and observation of the behaviour of pain sufferers. Limitations of the rehabilitation ward as the setting for change also affected the intervention, and emphasis was placed on empowering the patient by a full explanation of the approach and methods. It is noticeable that Fordyce treats the patient as a partner in deciding on and designing treatment; his practice is fundamentally respectful in a way which was not necessarily emulated by behavioural therapists who followed. He relied for selection on exploration and discussion of the activities and contingencies in the patient’s life and that of his or her spouse, and little if at all on questionnaires. Having used his protocol for many hundreds of assessments, my experience was that patients felt heard as people. Some of his questions regularly elicited a pleased “Nobody has ever asked me that before”.
SUMMARY
It could be argued that we have abandoned some of the finest of Fordyce’s heritage, which is an excellent reason for re-reading his work. At best, his models are scientific, value-free, and socially integrated, and he made explicit the contrasts with the targets of change generated by disease models. Some of that remains in the field, but it has also become integrated – or entangled – in cognitivism, mysticism, and psychiatric disease frameworks. We criticize medical colleagues when they assume that the existence of a degenerative disc on imaging explains a patient’s persistent back or neck pain, but are we any more sophisticated when we assume that the presence of a caring spouse explains the pain? A wide range of behaviour related to pain and disability has been reduced to avoidance to the detriment of understanding context and other contingencies, and what reinforces pain behaviour is often assumed rather than observed. The issue of selection remains unresolved for multiple reasons, many of them extremely hard to tackle empirically [36,37]. We would do well to revisit Fordyce’s careful grounding of selection practice in theory and in sensitivity to the patient in his or her context.
REFERENCES
1. Anderson TP, Cole TM, Gullickson G, Hudgens A, Roberts AH. Behavior modification of chronic pain: a treatment program by a multidisciplinary team. Clin Orthop Relat Res 1977;129: 96–100.
2. Berntzen D, Gotestam KG. Effects of on-demand versus fixed-interval schedules in the treatment of chronic pain with analgesic compounds. J Cons Clin Psychol 1987;55:213–7.
3. Block AR. Multidisciplinary treatment of chronic low back pain: a review. Rehabil Psychol 1982;27:51–63.
4. Block AR, Kremer EG, Gaylor M. Behavioral treatment of chronic pain: variables affecting treatment efficacy. Pain 1980;8:367–75.
5. Block AR, Kremer EG, Gaylor M. Behavioral treatment of chronic pain: the spouse as a discriminative cue for pain behavior. Pain 1980;9:243–52.
6. Cairns D, Pasino JA. Comparison of verbal reinforcement and feedback in the operant treatment of disability due to chronic low back pain. Behav Ther 1977;8:621–30.
7. Cassisi JE, Sypert GW, Salamon A, Kapel L. Independent evaluation of a multidisciplinary rehabilitation program for chronic low back pain. Neurosurgery 1989;25:877–83.
8. Clark WC, Hunt HF. Pain. In: JA Downey, RC Darling, editors. Physiological basis of rehabilitation medicine Philadelphia: WB Saunders Company, 1971.
9. Doleys DM, Crocker M, Patton D. Response of patients with chronic pain to exercise quotas. Phys Ther 1982;62:1111–4.
10. Duckro PN, Margolis RB, Tait RC, Korytnyk N. Long-term follow-up of chronic pain patients: a preliminary study. Int J Psychiatry Med 1985;15:283–92.
11. Eccleston C. A normal psychology of chronic pain. The Psychologist 2011;26: 422–5.
12. Fishbain DA, Rosomoff HL, Rosomoff RS. Drug abuse, dependence, and addiction in chronic pain patients. Clin J Pain 1992;8:77–85.
13. Flor H, Kerns RD, Turk DC. The role of spouse reinforcement, perceived pain, and activity levels of chronic pain patients. J Psychosom Res 1987;31:251–9.
14. Flor H, Turk DC. Chronic pain: an integrated biobehavioral approach. Seattle: IASP Press, 2011.
15. Follick MJ, Zitter RE, Ahern DK. Failures in the operant treatment of chronic pain. In: EB Foa, PMG Emmelkamp, editors. Failures in behavior therapy. New York: John Wiley & Sons, 1983. p. 311–35.
16. Fordyce WE. A behavioral perspective on chronic pain. Brit J Clin Psychol 1982;21:313–20.
17. Fordyce WE. Learned pain: pain as behavior. In: JJ Bonica, JD Loeser, CR Chapman, WE Fordyce, editors. The management of pain, vol. 1, 2nd edition. Philadelphia: Lea & Febiger, 1990. p. 291–9.
18. Fordyce WE, Fowler RS, Lehmann JF, DeLateur BJ. Some implications of learning on problems of chronic pain. J Chronic Dis 1968;21:179–90.
19. Fordyce WE, Fowler RS, Lehmann JF, DeLateur BJ, Sand PL, Treischmann RB. Operant conditioning in the treatment of chronic pain. Arch Phys Med Rehabil 1973;54:399–408.
20. Fordyce WE, Roberts AH, Sternbach RA. Letter to the editor. re: The behavioral management of pain: a critique of a critique. Pain 1988;33:385–7.
21. France RD, Urban BJ, Keefe FJ. Long-term use of narcotic analgesics in chronic pain. Soc Sci Med 1984;19:1379–82.
22. Gamsa A. The role of psychological factors in chronic pain. II. A critical appraisal. Pain 1994;57:17–29.
23. Gottlieb H, Strite LC, Koller R, Madorsky A, Hockersmith V, Kleeman M, Wagner J. Comprehensive rehabilitation of patients having chronic low back pain. Arch Phys Med Rehabil 1977;58:101–8.
24. Greenhoot JH, Sternbach RA. Conjoint treatment of chronic pain. Adv Neurol 1974;4:595–603.
25. Hammonds W, Brena SF, Unikel IP. Compensation for work-related injuries and rehabilitation of patients with chronic pain. South Med J 1978;71:664–6.
26. Holzman AD, Rudy TE, Gerber KE, et al. Chronic pain: a multiple-setting comparison of patient characteristics. J Behav Med 1983;8:411–22.
27. Keefe FJ. Behavioral measurement of pain. In: CR Chapman, JD Loeser, editors. Issues in Pain Measurement. New York: Raven Press, 1989. p. 405–24.
28. Keefe FJ, Crisson JE, Maltbie A, Bradley L, Gil KM. Illness behavior as a predictor of pain and overt behavior patterns in chronic low back pain patients. J Psychosom Res 1986;30:543–51.
29. Keefe FJ, Gil KM. Behavioral concepts in the analysis of chronic pain syndromes. J Consult Clin Psychol 1986;54:776–83.
30. Keefe FJ, Wilkins RH, Cook WA. Direct observation of pain behavior in low back pain patients during physical examination. Pain 1984;20:59–68.
31. Kerns RD, Turk DC. Depression and chronic pain: the mediating role of the spouse. J Marriage Family 1984;Nov:845–52.
32. Khatami M, Rush AJ. A pilot study of the treatment of outpatients with chronic pain: symptom control, stimulus control and social system intervention. Pain 1978;5:163–72.
33. Maruta T, Swanson DW, McHardy MJ. Three year follow-up of patients with chronic pain who were treated in a multidisciplinary pain management center. Pain 1990;41:47–53.
34. McQuay HJ. Opioids in chronic pain. Brit J Anaesth 1989;63:213–26.
35. Morgan CD, Kremer E, Gaylor M. The behavioral medicine unit: a new facility. Compr Psychiatry1979;20:79–89.
36. Morley S, Williams A, Eccleston C. Examining the evidence of psychological treatments for chronic pain: time for a paradigm shift? Pain 2013;154:1929–31.
37. Morley S, Williams ACdeC. Editorial: RCTs of psychological treatments for chronic pain: progress and challenges. Pain 2006;121:171–72.
38. Morley S, Williams ACdeC. Conducting and evaluating treatment outcome studies. In: DC Turk, R Gatchel, editors. Psychological approaches to pain management: a practitioner’s handbook, 2nd Edition. New York: Guilford Press 2002. p. 52–68.
39. Newton-John TRO, Williams ACdeC. Chronic pain couples: perceived marital interactions and pain behaviors. Pain 2006;123:53–63.
40. Painter JR, Seres LJ, Newman RI. Assessing benefits of the pain center: why some patients regress. Pain 1980;8:101–13.
41. Parker JC, Frank RG, Beck NC, et al. Pain management in rheumatoid arthritis patients. A cognitive-behavioral approach. Arthritis Rheum 1988;31:593–601.
42. Parris WCV, Jamison RN, Vasterling JJ. Follow-up study of a multidisciplinary pain center. J Pain Symptom Manage 1987;2:145–51.
43. Philips HC. Thoughts provoked by pain. Behav Res Ther 1989;27:469–73.
44. Pilowsky I, Spence ND. Patterns of illness behaviour in patients with intractable pain. J Psychosom Res 1975;19:279–85.
45. Roberts AH. The operant approach to the management of pain and excess disability. In:AD Holzman, DC Turk, editors. Pain Management: A Handbook of Treatment Approaches. New York: Pergamon Press, 1986. p. 10–30.
46. Roberts AH, Reinhardt L. The behavioral management of chronic pain: long-term follow-up with comparison groups. Pain 1980;8:151–62.
47. Romano JM, Turner JA, Clancy SL. Sex differences in the relationship of pain patient dysfunction to spouse adjustment. Pain 1989;39:289–95.
48. Roy R. A problem-centered family systems approach in treating chronic pain. In: AD Holzman, DC Turk, editors. Pain Management: a handbook of psychological treatment approaches. New York: Pergamon Press, 1986. p. 113–130.
49. Schmidt AJM. The behavioral management of pain: a criticism of a response. Pain 1987;30:285–91.
50. Seres JL, Newman RI. Results of treatment of chronic low-back pain at the Portland Pain Center. J Neurosurg 1976;45:32–6.
51. Sternbach RA. Psychological factors in pain. In: JJ Bonica, D Albe-Fessard, editors. Advances in Pain Research and Therapy, Volume I. New York: Raven Press, 1976. p. 293–9.
52. Sternbach RA. Summary and common concepts. In: RA Sternbach, editor. Pain: a psychophysical analysis. New York: Academic Press, 1968. p. 147–64.
53. Stone J, Wojcik W, Durrance D, Carson A, Lewis S, MacKenzie L, Warlow CP, Sharpe M. What should we say to patients with symptoms unexplained by disease? The “number needed to offend”. BMJ 2002;325:1449–50.
54. Swanson DW, Floreen AC, Swenson WM. Program for managing chronic pain. II Short-term results. Mayo Clin Proc 1976;51:409–11.
55. Swanson DW, Maruta T, Swenson WM. Results of behavior modification in the treatment of chronic pain. Psychosom Med 1979;41:55–61.
56. Swanson DW, Swenson WM, Maruta T, McPhee MC. Program for managing chronic pain. I. Program description and characteristics of patients. Mayo Clin Proc 1976;51:401–8.
57. Taylor CB, Zlutnick SI, Corley MJ, Flora J. The effects of detoxification, relaxation, and brief supportive therapy on chronic pain. Pain 1980;8:319–29.
58. Tunks E. Behavioral interventions and their efficacy. In R Dubner, GF Gebhart, MR Bond, editors. Proceedings of the 5th World Congress on Pain. Amsterdam: Elsevier, 1988.
59. Turk DC, Kerns RD. The family in health and illness. In: DC Turk, RD Kerns, editors. Health, illness, and families: a life-span perspective. New York: John Wiley & Sons, 1986. p. 1–22.
60. Turk DC, Matyas TA. Pain-related behaviors: communication of pain. American Pain Society 1992;1:109–11.
61. Turk DC, Rudy TE. Neglected factors in chronic pain treatment outcome studies – referral patterns, failure to enter treatment, and attrition. Pain 1990;43:7–25.
62. Turner JA, Calsyn DA, Fordyce WE, Ready LB. Drug utilization patterns in chronic pain patients. Pain 1982;12:357–63.
63. Verkerk K, Luijsterberg PAJ, Miedema HS, Pool-Goodzward A, Koes B. Prognostic factors for recovery in chronic nonspecific low back pain: a systematic review. Phys Ther 2012;92:1093–8.
64. White B, Sanders SH. The influence on patients’ pain intensity ratings of antecedent reinforcement of pain talk or well talk. J Beh Ther Exper Psychiatry 1986;17:155–9.
65. Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Syst Rev 2012;11:CD007407.
66. Williams ACdeC, Johnson M. Debate and analysis. Persistent pain: not a medically unexplained symptom. Br J Gen Pract 2011;61:638–9.
* I prefer the term “organic reductionism” to the oft-used term “Cartesian dualism” as Descartes likely did not view the mind and body as separate entities or at least not in the manner that is often assumed [10]. Remarkably, organic reductionism with regard to pain has persisted even in the face of overwhelming evidence to the contrary. For example, in Canada it took a Supreme Court ruling in 2003[55] to finally reject the notion that pain can only exist in the presence of tissue damage and that disallowing compensation to individuals with chronic pain is a violation of the Charter of Rights and Freedom within the Canadian Constitution.
Full access? Get Clinical Tree
