Let me take a moment to apply those rules to pain problems. That way you can see how operant conditioning can work to reduce the problem.
Take the example of back pain and doing sit-ups. The person with the pain who is doing the exercise usually works until the pain begins or becomes severe, and then he or she stops to rest. In that arrangement, the action or behavior to be decreased is pain and the behavior to be increased is exercise, in this case, sit-ups. We can diagram it like this (Fig. 1). Notice also that rest—stopping the exercise—is a pleasant or favorable consequence. The person doing the exercise, by working until the pain began or became severe, had arranged things so that the favorable consequence, rest, occurred immediately after pain. That was the thing he wanted to decrease, not increase. He was reinforcing the very thing he wanted to get rid of. A simple rearrangement can change that. Instead of exercising until the pain begins, work to a certain point before pain begins or becomes severe, and then stop to rest. In that arrangement, rest, the positive reinforcer, now immediately follows the behavior we are trying to strengthen: activity or exercise. Rest, the reinforcer, arrives before the pain but immediately after the exercise. Doing it that way and gradually increasing the amount of exercise done before rest begins helps most patients to increase considerably how much they can do before pain or weakness become a major issue.
That is the basic format of operant conditioning. Essentially the same principles are used in the treatment of chronic pain. The main emphasis is on exercise and activity, on the use of pain medications, and on how those around the patient respond or react to pain and to activity.
The nice part of the program is that it is likely to be helpful in increasing activity or decreasing the amount of interference in your life from the pain problem, regardless of whether your pain is now predominantly learned pain or whether the original or other causes are still the major factor. To the extent that your pain is now controlled mainly by learning factors, we can probably help your system to unlearn it and thereby to get rid of the surplus pain. Even if your pain is not now primarily learned, these procedures can probably help you to get along with the problem and have a decrease in the amount of interference from it.
INCREASING ACTIVITY OR REDUCING MEDICATION ADDICTION/HABITUATION IN PROBLEMS OF RESPONDENT PAIN
The evaluation process is now complete. The results indicate that we are not going to be able to do much about the physical problem from which your pain originates. There is a real chance, however, that we can help to reduce how much it hurts and to reduce the amount of limitation or functional impairment you have because of the pain. We may also be able to reduce the amount of medication and medical treatment you will need for it now and in the future and help keep the problem from getting worse. Pain problems such as you have tend not to remain the same. If some form of treatment doesn’t solve the problem, usually the pain and the amount of interference in daily activity gradually increase. If much pain medication is being taken, that problem tends also to get worse. A person’s system becomes adjusted to the medication so that it takes more and more to produce the same amount of relief. You may become addicted, and your system may become toxic or poisoned by the amount of medication you take. One way or another, even if the physical damage producing the pain can’t be changed much, there are probably some things we can do to make it easier to live with the problem and to help keep you from more difficulty in the future.
It is possible for you to train your system not to feel the pain so much. That is, you may be able to train your system to shut down, partly or even substantially, the pain impulses coming from the source of your pain. Your mind is potentially capable of doing that, but it takes a systematic effort to bring it about. A program of selected exercises and activity can be set up and a strict sequence is followed. That is, the amount of activity or exercise done is set by the sequence, and not by the pain. Neither you nor anyone around you is to respond to or pay attention to the pain in any way. The amount of exercise done will start at easy levels, which will gradually increase. Eventually, they will reach amounts that previously triggered considerable pain. Once those levels are reached, you are to continue with the assigned amount of exercise no matter how it makes you feel. In some cases exercises may be held at that level for long periods before increasing further. In other cases it may be judged wiser to increase the amounts done. If you persist faithfully with the program, both a reduction in how much you hurt and an increase in how much you can do have a good chance of occurring.
PAIN MEDICATIONS AND THE PAIN COCKTAIL—FOR ADDICTED/HABITUATED PATIENTS, RESPONDENT OR OPERANT
The procedure we will follow to help you reduce the amount of pain medicine you take and need is simple. First let me say that in this part of the program, as in every other part, we will never ask you to do something you cannot do.
For about a week we are going to let you demonstrate how much medication you need. You will decide how often and how much you need. The only restriction is that it must be taken orally. If you are taking medication by injection, we will shift that to oral form, although we will make it strong enough to cover you well. All you must do is to record what you take, how much of it, and when. We will give the medications to the nurses (in the case of an inpatient) or to your husband/wife (in the case of an outpatient) to help keep track of how much is taken. The medication will be there for you whenever you want it. After about a week, we will change to the pain cocktail. That is a procedure in which al1 of your medications will be incorporated into a single dose, the pain cocktail. The cocktail will contain all of your medications in the approximate amounts you have been taking them. It will also contain a color- and taste-masking substance, perhaps cherry syrup. Once we know how often you need and take the medications, we will put you on a fixed time schedule, every few hours, around the clock. The time between doses will be less than what you used to take. The result is that the pain medication will reach you before the pain begins or before it becomes severe. In that way we will break the hookup between hurting and taking medication. Now the medication will come without the pain.
We will gradually reduce the amount of medication in the cocktail, taking it down as far as we can, perhaps to zero. If and when we get to zero, we will tell you. We won’t tell you exactly when we have reduced the dosage level because we think it will be easier for your system to decondition if it doesn’t know that. You can be certain, however, that we will start you with as much or even slightly more medication than you are now taking. We know that amount will cover you. We will work from there. It usually takes about 7 to 9 weeks to get down to the level we are aiming for.
USE OF ATTENTION OR SOCIAL REINFORCEMENT
Those are the major elements of the program. The plan will be essentially the same for medication and for exercise and rest. We shall also be concerned with how others respond or react to your pain. Make no mistake about it, that makes a difference for everyone. We work with the responses of others—we can call it a form of attention—in two ways. One is that the staff here will tend not to pay attention to pain. We won’t ignore it; that could be dangerous. We will be socially nonresponsive: shrugs of the shoulders and that sort of thing.
After about 2 weeks patients in this program sometimes feel that no one around here cares about them. We are more interested in what you think of us after the total program than after 2 weeks. With some people, staff nonresponsiveness is important; with others it plays only a small part. Either way, it can help and so we work with it.
The second way we work with attention is to help your family to respond in ways that will be of the greatest help to you. That is such an important part of the program that we won’t agree to work with you unless your husband/wife agrees to participate. We will want to see them perhaps as often as twice a week, for brief sessions. We will work on helping to develop the best ways of responding to pain behavior and to activity. Your husband/wife will have homework to do, just as you will have treatment work to do.
If this program proves helpful to you, it means that at the end of treatment you will be taking little if any pain medication. It also means that there may be little interference from pain in what you do or how you spend your time. We hope and expect you wi1l be able to do much more than you now can do. All of us need to be concerned with how you plan to spend your time. You and our team together must be concerned about that, not because we wish to tell you what to do, but because working with you to plan activities can help you to maintain and even expand treatment gains. We will meet with you and your husband/wife about those matters as the program moves along.
REDUCTION OF GAPS IN WELL BEHAVIOR
Anyone who has been on the sidelines with a pain problem as long as you have is bound to find it difficult to get back into the swing of things. During the course of treatment, we will involve you increasingly in activities similar to things you will be doing after treatment. That will provide you with a chance to prepare yourself at your own pace, instead of all at once. There are probably other things you cannot do so well or problems that make it harder for you to engage in a full range of normal activities. We shall be concerned about those, too. We may work on them directly with you, or it may prove better to let you work with others on those problems. (If evaluation has indicated a clear problem pattern—something relating to avoidance learning—that can be spelled out.)
GENERAL ORIENTATION ISSUES
This concludes a sample of patient/spouse orientation. The specificity with which it has been presented is both to illustrate how it might be done and to emphasize the value of candor. There is a myth having some prevalence in human service settings that implies that knowing about behavior-consequence relationships, as they apply to oneself, somehow interferes with their impact. That is not the case. Knowing about the pain cocktail procedure, for example, or of the programming of attention, does not thereby reduce effectiveness of the method. Moreover, patient and spouse participation seems to profit considerably from the candor and completeness of explanation given at the outset. Such an atmosphere serves, among other functions, to help dissipate the suspiciousness and anger that usually accompany long-frustrated efforts at gaining help for a pain problem.
It is often helpful to add to the foregoing orientation a brief description of a typical treatment day and the plan to use passes as a programmed part of treatment.
There should also be a detailed consideration of the question of posttreatment employment, if that is a viable alternative in the picture. The issue of posttreatment employment cannot receive proper attention until there has been clarification with third-party sponsors or with sources of wage replacement or compensation funds as to their plans and expectations. This is often a delicate and complicated issue. It cannot be avoided, and its consideration cannot be delayed beyond the point of considering and offering treatment. If evaluation indicates a reasonable likelihood that the patient will be physically capable of resuming full or part-time employment, a decision is needed as to whether continued compensation is contingent on continued pain behavior and associated impairment. That is, of course, mainly a legal or policy decision to be made by third-party agencies. When employment is concluded to be a viable alternative, all factors considered, it should be made explicit to the patient and spouse that successful treatment will mean that the treatment facility will then certify that the patient is ready for employment and not a candidate for continued compensation. Only when that is made explicit and is convincingly agreed to by patient and spouse should treatment proceed. Alternatively, if the realities of long-time employment, restrictions in employability relating to skill deficits or to medical liability constraints, or the labor market suggest that employment is not a viable alternative—however successful is treatment—patient, spouse, and funding source need to decide whether to proceed. If the decision is then to proceed, presumably it will be based on the intent to legitimate retirement or, at the least, to strive to reduce anticipated medical costs relating to the pain problem. In either case, it will have been agreed that compensation is not contingent on continued pain behavior and associated functional impairment. These issues have considered earlier in relation to the genesis of operant pain. The point for now is to clarify the issue in relation to each patient before treatment begins.
For years, Fordyce’s  Behavioral Methods for Chronic Illness has served as a foundational text for any researcher who works in the field of chronic illness, and chronic pain specifically. In particular, Fordyce’s legacy is clearly seen in research that investigates the social context of pain. As Fordyce notes time and again throughout his seminal book, learning occurs when conditions are favorable, including when there are reinforcers present to alter the frequency of behavior. The social environment is particularly important in this regard. Fordyce offers clinicians and researchers several avenues for research and clinical applications. Indeed, research that followed the original publication of his treatise supported some of Fordyce’s suppositions regarding the reinforcing value of some spousal behaviors by applying rigorous observational methods [19,21,22]. However, some of his ideas have yet to be fully explored, especially when considering the role of the spouse or other loved ones in the maintenance of pain behaviors that could limit function and contribute to functional impairment. Specifically, there is a need for research that can better characterize the bidirectional nature of patient-spouse conditioning and the roles of spouses’ pain learning history in patient-spouse interactions. In addition, there is much to learn about the important variations in the functions of both patient and spouse behaviors across couples and the manner in which new contingencies could be offered to couples to improve health.
Fordyce mentions in this chapter and elsewhere that conditioning is not unidirectional. A behavior-response cycle shapes the expectations and behaviors of the person with pain and their significant others. Most of the basic and intervention research has focused on the role of the spouse on patient behaviors [15,17]. These valuable findings support Fordyce’s claims that the social environment plays a key role in the development, maintenance, and treatment of pain. However, very little is known about how spouses’ behaviors are conditioned. With the exception of one study that provided evidence for the shaping of solicitous spouse responses in response to patients’ verbal and non-verbal pain behaviors , no other research of which we are aware has examined how patients’ behaviors shape spouses’ behaviors. It is understandable that the focus has been on how the social environment affects patients’ behaviors. Yet, Fordyce acknowledges that there is a dynamic learning process between patients and their loved ones. Other researchers have argued similarly, that the history of relationship interactions shape one’s current interactions having to do with pain and that current family dynamics are important to consider when intervening to improve health [1,9]. Thus, an important area of research that was early identified by Fordyce and still much needed concerns the manner in which patients and spouses affect each other and how these interaction patterns contribute to health in both partners.
Similarly, another aspect of learning history that has not received much attention involves the spouse’s personal history of pain. It appears that there is a presumption that the partners of people with chronic pain do not have chronic pain because few researchers report the rates of chronic pain in partners. However, it has been demonstrated that up to fifty percent of partners report chronic pain problems of their own [14,16]. Given the importance of learning history in patients with pain, it is also important for clinicians and researchers to investigate how the partner’s learning history with acute and chronic pain may affect interactions with a partner with pain. Goubert et al.’s  pain empathy model includes a role for observers’ personal experiences of pain that might influence one’s understanding of and reactions to pain. Research testing this model has shown some interesting findings when comparing couples in which one partner versus both partners report chronic pain. For instance, there is no difference between the extent to which spouses with or without chronic pain can accurately estimate the thoughts and feelings of patients (i.e., empathic accuracy or cognitive empathy) . However, there is evidence showing that spouses are more emotionally validating (i.e., attending to patients’ emotional disclosure, expressing empathy) to patients when the former do not report chronic pain of their own . This finding is in line with an fMRI study showing that participants exposed to experimental pain showed lower activation in brain regions associated with pain empathy when viewing photos of people in pain than participants without pain experience . Spouses with chronic pain may not express as many empathic behaviors for at least two reasons. In these studies, if both partners reported chronic pain, the spouse with the most severe pain as reported by both partners, was designated as the patient. Thus, although spouses reported chronic pain, it was less severe than the designated patients. These spouses likely decoded the pain behaviors expressed by patients through the lens of their own pain experiences and may not have responded with empathy because they interpreted the pain behaviors as exaggerated expressions of pain. Alternatively, patients may have dismissed spouses’ pain as being less severe and not requiring empathy on the patients’ part. In this case, spouses may have been punished for their pain behaviors and may be responding in kind. Both of these explanations point to the role of personal experience with pain in shaping empathic responses.
Fordyce also emphasized that the functionality of behavior is extremely important when evaluating the role of the social environment on patients. He suggested ways to extinguish patients’ verbal pain behaviors but is also clear that a spousal behavior that is reinforcing of pain behavior for one person might have no effect on another person. Although conclusions about the behaviors that are universally reinforcing or punishing cannot be made, research has provided some clues about patient-spouse interaction. This is especially true when sequential analysis is used because this type of analysis provides information about the temporal association between one behavior and another. For instance, Romano and colleagues demonstrated that solicitous spouse responses, including offers of physical assistance or actual help, are reinforcing of pain behaviors expressed during household tasks [21,22].
To investigate further, it may be helpful for researchers to consider two of Fordyce’s points concerning verbal behavior: 1) the same verbal pain behavior may serve different functions for different people and 2) the same verbal pain behavior may serve different functions for the same person in different situations. As an example, one patient may find that disclosing his or her anger about having pain to the partner leads to increased intimacy within the couple, which may be gratifying to the patient. Another patient may find that the same verbal behavior is aversive, perhaps being met with pity or contempt, which may decrease the patient’s likelihood of expressing similar emotions in the future. In a similar vein, a single individual may sometimes express anger and sadness about the pain when s/he needs reassurance or comforting but not express the same emotions when s/he already feels supported by the partner. These examples suggest that similar verbal pain behaviors may be expressed for different reasons. This line of reasoning has been taken up by more recent conceptual models [4,23], which suggest that some pain behaviors may be aimed at expressing emotions to loved ones and securing social support. Indeed, some verbal pain behaviors appear to be attempts to build intimacy with one’s partner . Fordyce would have likely argued that verbal pain behaviors may be differentially reinforced depending on the spouses’ interpretations, expectations, and current emotional state. Thus, in the context of a close relationship such as marriage, it is possible that while indifference to verbal pain behaviors extinguishes those behaviors, it may also increase feelings of resentment and a loss of intimacy. It is also possible that emotional validation of pain-related distress would increase intimacy while also reinforcing verbal pain behaviors. That is, a reinforcement strategy may be beneficial in one area of functioning while a detriment to another area of functioning. There are hints of an opposing pattern in early self-report research, which shows that solicitous spouse responses are positively associated with both marital satisfaction and pain severity in patients (e.g., [6,13]). However, a reliance on self-report measures of spouses’ responses to test operant models, whether made by the patient or the spouse, can be problematic because self-reports are subject to selective attention and memory biases as well as appraisal processes. Single subject, or in this case, single couple designs using observational methods are still needed to provide key information about which types of verbal pain behaviors should be a target of extinction, how the use of extinction procedures might impact other areas of functioning, and whether certain couples would benefit from this approach.
Finally, Fordyce made the excellent point that spouses need to attend to and reinforce well behavior in people with pain. In other words, a different behavioral target (well behavior) and pattern of contingencies need to be introduced to patients and their family members to disrupt the maladaptive pain behavior reinforcing interactions. Fordyce recommended that significant others should reinforce well behaviors such as exercise and activity. One way to accomplish this task is to have spouses provide praise and encouragement when patients engage in well behavior. Perhaps, a more effective method would be to help couples identify and commit to joint activities that involve physical activity and promote health in both partners such as going for a walk or cycling together. Clinicians may also need to help couples create new opportunities to engage in activities that have positive consequences more generally. That is, the couple should be encouraged to focus on life beyond pain. Pain catastrophizing and worry characterizes some couples with chronic pain [2,14] and this cognitive style may narrow the focus of couples’ attention, making it hard for them to think about and engage in other valued activities. Acceptance and Commitment Therapy  and other contextual cognitive-behavioral therapies (e.g., Integrative Behavioral Couples Therapy [5,12], Contextual Cognitive-Behavior Therapy for Chronic Pain ) have shown that identifying valued activities and committing to those behaviors is necessary to promote healthy adjustment. However, these approaches often target one domain of functioning (e.g., pain interference, couple functioning). One could argue that couples who experience both relationship distress and inflexible ways of coping with pain may benefit from an approach that addresses pain and relationship difficulties to maximize the ability of couples to work together to cope with pain and communicate and interact in mutually beneficial, rewarding, and health-promoting ways. In addition, helping couples identify shared values and meaningful activities may introduce opportunities for mutually positively reinforcing interactions and behaviors that would improve functioning more generally to include pain, well-being, and relationship satisfaction. For instance, some couples may find that they do not spend much time together even though they value having a close and intimate relationship with each other. They may also value living a healthy lifestyle but may not be engaging in activities to fulfill this value. In these cases, finding a shared activity that could address these values for both partners may be helpful. A shared yoga or exercise practice may not only improve strength and flexibility and reduce stress but also may facilitate positive interactions between partners which, in turn, may contribute to greater intimacy and relationship well-being.
In sum, Fordyce’s book has certainly spurred a great deal of research examining the role of patient-spouse interactions in chronic pain. Most of this research has focused on how these interactions have affected patients but there are many outstanding questions that have yet to be answered. This book, though published almost 40 years ago, will continue to inform research endeavors for decades to come.
1. Baucom D, Kirby J, Kelly J. Couple-based interventions to assist partners with psychological and medical problems. In: K Halwehg, M Grawe-Gerber, D Baucom, editors. Enhancing Couples: The Shape of Couple Therapy to Come. Cambridge, MA: Hogrefe Publishing; 2009. p. 79–94.
2. Cano A, Leonard MT, Franz A. The Significant Other Version of the Pain Catastrophizing Scale (PCS-S): Preliminary validation. Pain 2005;119:26–37.
3. Cano A, Leong LM, Williams AM, May DK, Lutz JR. Correlates and consequences of the disclosure of pain-related distress to one’s spouse. Pain 2012;153:2441–7.
4. Cano A, Williams ACdC. Social interaction in pain: Reinforcing pain behaviors or building intimacy? Pain 2010;149:9–11.
5. Christensen A, Jacobson N, Babcock J. Integrative behavioral couple therapy. Clinical Handbook of Marital Therapy. New York: Guilford Press, 1995. p. 31–64.
6. 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.
7. Fordyce WE. Behavioral methods for chronic pain and illness. Saint Louis: The C.V. Mosby Company; 1976.
8. Goubert L, Craig K, Vervoort T, Morley S, Sullivan MJL, Williams ACdC, Cano A, Crombez G. Facing others in pain: the effects of empathy. Pain 2005;118:285–8.
9. Hadjistavropoulos T, Craig KD, Duck S, Cano A, Goubert L, Jackson PL, Mogil JS, Rainville P, Sullivan MJL, de C. Williams AC, Vervoort T, Fitzgerald TD. A biopsychosocial formulation of pain communication. Psychol Bull 2011;137:910–939.
10. Hayes SC, Luoma JB, Bond FW, Masuda A, Lillis J. Acceptance and commitment therapy: Model, processes and outcomes. Behav Res Ther 2006;44:1–25.
11. Issner JB, Cano A, Leonard MT, Williams AM. How do I empathize with you? Let me count the ways: Relations between facets of pain-related empathy. J Pain 2012;13:167–175.
12. Jacobson N, Christensen A, Prince S, Cordova J, Eldridge K. Integrative behavioral couple therapy: An acceptance-based, promising new treatment for couple discord. J Consult Clin Psychol 2000;68:351–5.
13. Kerns RD, Haythornthwaite J, Southwick S, Giller EL. The role of marital interaction in chronic pain and depressive symptom severity. J Psychosom Res 1990;34:401–408.
14. Leonard MT, Cano A. Pain affects spouses too: Personal experience with pain and catastrophizing as correlates of spouse distress. Pain 2006;126:139–146.
15. 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.
16. Leonard MT, Issner, J. B., Cano, A., Williams, A. M. Correlates of spousal empathic accuracy for pain-related thoughts and feelings. Clin J Pain 2013;29:324–33.
17. Martire LM, Schulz R, Helgeson VS, Small BJ, Saghafi EM. Review and meta-analysis of couple-oriented interventions for chronic illness. Ann Behav Med 2010;40(3):325–42.
18. McCracken LM. Contextual cognitive-behavioral therapy for chronic pain. Seattle: International Association for the Study of Pain; 2005.
19. Paulsen JS, Altmaier EM. The effects of perceived versus enacted social support on the discriminative cue function of spouses for pain behaviors. Pain 1995;60:103–10.
20. Preis MA, Schmidt-Samoa C, Dechent P, Kroener-Herwig B. The effects of prior pain experience on neural correlates of empathy for pain: An fMRI study. Pain 2013;154:411–8.
21. Romano JM, Turner JA, Friedman LS, Bulcroft RA, Jensen MP, Hops H, Wright SF. Sequential analysis of chronic pain behaviors and spouse responses. J Consult Clin Psychol 1992;60:777–82.
22. Romano JM, Turner JA, Jensen MP, Friedman LS, Bulcroft RA, Hops H, Wright SF. Chronic pain patient-spouse behavioral interactions predict patient disability. Pain 1995;63:353–60.
23. Sullivan MJL, Thorn B, Haythornthwaite JA, Keefe F, Martin M, Bradley LA, Lefebvre JC. Theoretical perspectives on the relation between catastrophizing and pain. Clin J Pain 2001;17:52–64.
REFLECTIONS ON PATIENT AND SPOUSE ORIENTATION
The behavioral model espoused by Fordyce maintains that patients with persistent pain do not change their behavior independent of their environment. Consistent with this model, the behavioral treatment program that Fordyce outlines required that spouses be involved in treatment. At the time, requiring a spouse or partner to be involved in the treatment of chronic pain was unusual. However, as evidence for the efficacy of multidisciplinary programs based on behavioral principles grew, many treatment programs for chronic pain either required or strongly encouraged spouse/partner involvement.
Patients and spouses are often confused about why health care providers might want a spouse to be part of a pain treatment program. This chapter provides an excellent example of a rationale that can be provided to a patient and spouse for involving the spouse as part of the treatment process. The detailed rationale presented has several key messages. First, patients and spouses are told that behavioral adjustments to pain, including patterns of pain and well behavior, develop over time through a process of learning. This message is an optimistic one. If maladaptive behavior patterns (such as excessive rest, overdependence on others, or high levels of pain medication intake) are learned, then they can be unlearned and new, more adaptive behaviors can be learned in their place (gradually increasing activity level, doing more on one’s own, and decreasing one’s pain medication). Second, patients and spouses are asked to carefully consider the role that consequences can play in how one learns to respond to pain. In particular they are asked to focus what happens when the immediate consequences of pain behavior are positive (e.g. when reports of high levels of pain are followed immediately by rest, pain medication, or solicitous attention from a spouse). Fordyce often said that positive consequences are agnostic, and that they will reinforce whatever comes just before them. For many patients and spouses, this rationale may be the first time they have considered the power that immediate responses to pain can have on pain and the possibility that such short-term responses increase the likelihood of longer term, entrenched patterns of excessive disability. At the time, Fordyce’s encouragement of patients and spouses to consider the effects of learning upon the pain experience represented a novel therapeutic strategy.
Third, with acknowledgement of the role that positive consequences can play in rewarding pain behavior, patients and spouses are asked to consider the possibility that the same rewards can be used to foster adaptive behavioral responses to pain. Although the idea of intentionally programming consequences for chronic pain such as rest, pain medication, or the attention of others is now accepted by many health professionals, for many couples it remains a radical and somewhat unsettling idea. The detailed rationale provided by Fordyce with its emphasis on learning and the benefits/downsides related to the immediate consequences of pain shows how one can prepare patients and spouses to better accept the notion that by controlling environmental responses to pain, they can make these consequences work for, rather than against the patient. In Fordyce’s program these consequences were initially controlled as part of an inpatient treatment program, following discharge from which the patient and spouse assumed control of them. Along these lines, Fordyce routinely taught spouses about the importance that their own social reinforcement had in maintaining pain behavior and how to use social reinforcement methods (providing attention and praise) in the patient’s home environment so as to increase well behaviors (activity, exercise, and a full range of daily activities) and to reduce pain behaviors. Finally, implicit in the rationale that Fordyce provides is the notion that a behavioral treatment program is an experiment and that a patient and their spouse can best judge the value of rationale provided only after observing the effects of the treatment it describes. This early behavioral approach fits well with the more recent emphasis on patient centered care where the focus is on evaluating outcomes not just based on biological outcomes but also based on what patients (and their caregivers) view as important.
One of the most striking features of the detailed rationale provided by Fordyce in this chapter is its candor. Fordyce is upfront and honest with patients and their spouses about the problems he observes. He frankly describes how patient –interaction patterns might contribute to those problems, the fact that continued medical and surgical treatments are not likely to help, and the role that patient learning and spousal responses can play in reducing disability and suffering. Over the years, behaviorally oriented health care providers have used rationales for involving spouses that are similar in terms of their level of detail and candor.
FORDYCE’S’ IMPACT: OBSERVATIONAL STUDIES OF PATIENTS HAVING CHRONIC PAIN AND THEIR SPOUSES
One of the most important contributions of Fordyce’s work is that it stimulated behavioral research on patients with chronic pain and their spouses. The literature in this area is now quite large and a comprehensive review is beyond the scope of this commentary. In this section we focus on a number of key studies, many of them from the 1980s and 1990s, that used the kind of direct measures Fordyce favored (i.e. either lab-based or direct observational methods) to examine key aspects of the behavioral model that he advocated. The findings of these studies highlight the role of the spouse in influencing pain behavior and thus strengthen the rationale for involving spouses in behavioral treatment programs.
Andrew Block conducted some of the earliest research on spousal responses to pain behavior. His first study examined the effects that the presence of a solicitous spouse has on patient ratings of pain . Based on questionnaire responses, patients in this study were divided into two groups, those who reported their spouse tended to respond to their pain in a solicitous fashion and those who reported their spouse was not solicitous. Patients were then interviewed in a room that had a one way mirror. Over the course of the interview they were asked to provide two pain ratings, one when they were told their spouse was behind the one way mirror and a second when they were told a ward clerk who they did not know was behind the mirror. The data analyses revealed that, compared to when they were observed by a ward clerk, the patients who reported having a solicitous spouse tended to report more severe pain when they believed their spouse was observing them. Patients who reported their spouse was not solicitous reported significantly less pain when they believed their spouse was observing them. This pattern of results was replicated in a subsequent study by Flor et al. . Taken together, these two studies underscore the influence that solicitous spouse responding can have on pain and suggest that the mere presence of a spouse who is solicitous may serve to alter pain reports.
Block  was interested in why spouses might respond in a solicitous fashion and conducted a study to test if solicitousness might be related to spouses’ emotional responsiveness to pain behavior displays. He hypothesized that spouses who reported high levels of marital satisfaction might be particularly likely to show such emotional responsiveness. To test this, he had spouses of chronic pain patients watch videotapes of facial expressions their patient/partners showed when they were having high levels of pain. Throughout the time they watched the videotapes, a physiological response indicative of emotional arousal (galvanic skin response) was recorded. Results showed that spouses who reported high levels of marital satisfaction were much more likely to show high GSR reactivity in response to facial pain displays. Block speculated that, because of their tendency to be more emotionally/physiologically reactive, some spouses were more likely to respond to pain in a solicitous fashion. Interestingly, there is more recent evidence that individuals who are highly empathic (something that may well be true of a solicitous spouse) are more likely to respond to facial displays of pain with activation of mirror neurons in the brain, i.e. neurons that are similar to those activated during pain .
Fordyce would argue that the most valid and reliable way of studying interactions between chronic pain patients and their spouses is to directly observe them. Building on Keefe and Block’s  observational method for recording pain behavior in patients, Romano, et al.  developed an observational system to record patient pain behaviors (verbal and non-verbal) and spousal responses (solicitous responses, negative responses) to those behaviors. Patient-spouse interactions were coded while couples engaged in a series of tasks (sweeping a floor, picking up and folding clothing, and moving logs from one side of the room to the other). For each task, the patient was instructed to take the lead and the spouse to assist as needed. The first study using this observation method (1), found that patients with chronic pain showed significantly higher levels of verbal and nonverbal pain behaviors when interacting with their spouses than healthy control participants. Consistent with Fordyce’s behavioral model, the spouses of chronic pain patients also were much more likely to engage solicitous behavior than the spouses of control participants.
The behavioral model also predicts that spousal responses that occur just after (or just before) pain behaviors are likely to be especially important. To examine such sequences of behavior Romano et al.  conducted yet another study using their observational methodology. As might be expected, the study’s findings showed that spouses of chronic pain patients indeed were much more likely to engage in solicitous behavior both immediately after and immediately before the patient engaged in pain. This pattern was not observed in the control group participants and their partners.
We conducted a study of sequences of patient-spouse interactions in a sample of osteoarthritis patients . In contrast to earlier studies that highlighted the negative effects of solicitous spousal responding, we found spousal facilitative behaviors (expressing approval, agreement, or support, all of which are generally thought to foster well behaviors) were linked to high pain behaviors. Specifically, facilitative spousal responses were much more likely to occur both after and before patient behaviors than spouse solicitous behavior. In this case, facilitative behavior appeared to be functioning as a positive reinforcer. Again, this study supports Fordyce’s claim that one cannot assume a particular spousal response is an adaptive one without directly observing its effects, a point echoed in other commentaries in this volume (see Main’s commentary on chapter 2). In this instance a spousal response thought to be adaptive was actually significantly associated with higher levels of patient pain behavior.
In addition to the observational studies above, numerous questionnaire and interview studies of patients with chronic pain and their spouses have been conducted. These studies are reviewed in other commentaries in this book. Suffice it to say that the observational studies of patient-spouse interactions that flowed out of Fordyce’s work lent strong support to the importance of including spouses in behavioral treatment programs for pain.
FORDYCE’S IMPACT: BEHAVIORAL TREATMENT PROGRAMS THAT INVOLVE SPOUSES
Fordyce and his colleagues’ writings and subsequent observational research supporting the behavioral model led to a much greater emphasis on involving spouses in pain treatment. Over the years, three approaches to involving spouses in behaviorally oriented treatment programs have emerged: partner-assisted behavioral treatment, partner-assisted cognitive behavioral treatment, and couples-based interventions. In this section, we describe these approaches, review studies testing the, and highlight some of the issues involved.
Partner-assisted behavioral treatment programs—In partner-assisted behavioral treatment, the focus is on the patient with the partner’s role being primarily to use social reinforcement principles (e.g. providing attention to well behaviors, minimizing attention to pain behavior) to support the patient ‘s behavior change efforts. The approach is quite similar to that described by Fordyce with the exception that currently many of these partner-assisted behavioral programs are conducted entirely on an outpatient basis. Turner and Clancy  provided one of the first descriptions and empirical tests of an outpatient group protocol for spouse-assisted behavioral treatment. This protocol, delivered to small groups of 5 to 10 couples over eight, two hour sessions, included several key components: a) training of patients and spouses in the concepts of pain and well behaviors, b) instructing the patient and spouse in how to avoid reinforcing pain behaviors and how to positively reinforce well behaviors, c) a quota system to increase aerobic exercise, d) training in communication skills, and e) behavioral goal setting. The protocol produced significant reductions in physical and psychosocial disability. Interestingly, the partner-assisted behavioral treatment led to larger initial improvements than a cognitive-behavioral treatment program that did not involve spouses, though both treatments showed similar effects at 12 months follow up. This suggests that spouse involvement may be particularly important in producing immediate gains but not as important when it comes to maintenance of treatment gains.
A subsequent study by Turner et al.  compared the effects of the same spouse-assisted behavioral treatment protocol either with or without the aerobic exercise component included. Interestingly, analysis of data collected pre- and post-treatment showed that the treatment condition that combined spouse-assisted behavioral training with exercise was the most effective. At 6 and 12 months however all three conditions (spouse-assisted behavioral treatment with exercise, spouse-assisted behavioral treatment alone, exercise) produced significant treatment gains. These findings support Fordyce’s contention that spouses can play a key role, particularly in initiating behavior change efforts. They also indicate that gains achieved either with spouse-assisted behavioral treatment or exercise protocol can be maintained over a one year period.
More recently, Thieme et al.  conducted a randomized clinical trial to compare the effects of a family-assisted behavioral treatment protocol to a physical therapy intervention. Like Fordyce’s original program, the behavioral treatment was conducted on an inpatient unit and used treatment methods such as staff social reinforcement, behavioral goal setting, and time contingent reduction of pain medication. Patients with fibromyalgia were randomly assigned to either receive the family-assisted behavioral treatment program or a medical program that emphasized physical therapy. Family members visited the programs on weekends and received training in behavioral principles. Although role playing is often used in couples-based pain treatment this program used a novel approach to teach patients and family members to identify and reinforce pain and well behavior. During the role play sessions, participants held up a red card whenever a pain behavior was displayed and a green card when well behaviors were displayed. Data analysis showed that 65% of the patients in the behavioral treatment group showed clinically significant improvements in treatment outcome, whereas none of those in the physical therapy intervention did so. A subsequent study by Thieme, et al.  found that family-assisted behavioral treatment was as effective as cognitive-behavioral treatment in reducing pain. However, improvements in physical functioning and behavioral outcomes (self-report measures of pain behavior, and pain-related spouse solicitous responding) were only found in patients receiving the family-assisted treatment, not the cognitive-behavioral treatment.
Studies of partner-assisted behavioral treatment suggest that this approach can yield positive outcomes. The focus of spousal involvement in such programs, however, is usually a limited one, i.e. to provide social reinforcement for behavior change. Other approaches to including spouses involve more active spouse involvement.
Partner-assisted cognitive-behavioral treatment—Cognitive behavioral approaches to pain management maintain that one must not only focus on behavior change but also directly focus on changes in patient’s cognitions, i.e., their pain beliefs, memories, and expectations. The rationale is that patients often show maladaptive behavioral and cognitive responses to pain and that focusing is necessary and also more efficient. In contrast Fordyce would have argued that the most efficient approach is to focus on behavior change and that other responses, e.g., changes in cognitions, emotions, etc. would ensue.
Having spouses assist patients undergoing cognitive-behavioral treatment can be helpful for several reasons . First, through being included in treatment, spouses may learn more about pain and the important role that the patient’s own coping efforts play in adjusting to pain. Understanding the problem can lead to changes in spouses own thoughts and beliefs and enable them to better support and assist the patient. Second, spouses may find that many of the pain management skills that patients learn can be used to manage their own stress. Finally, when patients and spouses actively practice coping skills together it can increase their sense of intimacy and improve their relationship.
Over the past 20, years we have conducted a series of studies testing the efficacy of partner-assisted cognitive-behavioral treatment protocols. In our earliest study of osteoarthritis patients  we compared the effects of a spouse-assisted pain coping skills based on cognitive behavioral principles, a conventional pain coping skills training protocol (which only included patients), and an arthritis education protocol that involved spousal support. The partner –assisted protocol trained patients and spouses in a variety of pain coping skills (relaxation, imagery, activity pacing, goal setting, and cognitive restructuring) and also taught patients and spouses communication skills, strategies for practicing coping skills together (e.g. setting aside time for home practice and setting goals for joint practice in situations where pain is likely to be problematic), and how to set mutually agreed upon short- and long-term goals. The overall pattern of results revealed that patients in the spouse-assisted treatment condition showed the best outcomes with conventional pain coping skills training the next best outcome (though not significantly different from spouse-assisted training). The spouse-assisted training was significantly more effective than the arthritis education-spousal support condition suggesting that simply involving a spouse in educational/informational sessions is not enough. A follow up study  of this sample found that patients receiving the spouse-assisted training were able to maintain initial gains in self-efficacy and physical disability 12 months after treatment. A later study tested the effects of spouse-assisted training with and without exercise training . It found that spouse-assisted training, either alone or combined with exercise training produced significant improvements in coping and self-efficacy, whereas exercise training either alone or combined with spouse-assisted training produced improvements in physical fitness and strength.
We have also examined the utility of a partner-assisted pain coping skills training protocol for patients having cancer pain at end of life . A 3 session home based protocol taught partners and patients coping skills for managing the challenges of cancer pain in their home setting. Data analyses revealed that, compared to usual care, the training protocol produced significant improvements in spouse’s self-efficacy for helping the patient manage pain and self-efficacy for helping the patient manage other symptoms. Spouses also showed a trend towards reduced caregiver strain.
Couples-based approaches to pain and symptom management—in couples based behavioral interventions the focus is on the couples’ relationship and how that affects pain [1,10]. In these interventions, patients and spouses are equally involved in behavior change efforts. The target of an intervention focuses on the couple’s skills, such as how the patient and spouse can more effectively communicate about pain and/or engage in problem-solving efforts. Couples-based interventions have shown efficacy for persons suffering from agoraphobia, depression, and alcohol abuse and dependence . Although several investigators have called for research on couples-based interventions for pain [1,10], very few controlled studies of this approach have been conducted. We conducted a small randomized study testing the effects of a couples-based relationship enhancement intervention to treatment as usual in the management of symptoms reported by breast cancer survivors . Couples in the relationship enhancement condition were taught problem solving skills, emotional expressiveness skills, and prompted to use these skills to better manage the challenges that both the patient and spouse experienced in coping with cancer. Results showed that patients in the relationship enhanced condition not only reported significant decreases in pain and other symptoms, but also reported improvements in relationship functioning and psychological distress. Effects sizes were in the moderate to large range both immediately after treatment and at one year follow up. These preliminary findings suggest that a couples-based approach may have merit. However, because so few studies have examined the couples-based approach to pain management we know very little about which couples are most likely to benefit vs not from this treatment.
The key points from this commentary are presented in Box 1
1. Providing a compelling rationale for spouse or partner involvement in behavioral treatment is important since patients and spouses are often confused about why health care providers might want a spouse or partner to be integrated into pain management efforts.
2. Key reasons for involving spouses/partners include: a) how patients adjust to pain is learned and the way that spouses respond to patients’ pain behavior can influence pain behavior, b) by using their ability to provide positive reinforcement, spouses/partners can help patients learn new and more adaptive ways of responding to pain, c) behavioral treatment is an experiment and patients and spouses should judge its effects by observing and recording its effects on the patient’s behavior.
3. Research studies have shown that, if a spouse responds to pain behavior in a solicitous fashion, then patients are more likely to engage in pain behavior. Further, there is evidence that the presence of a solicitous spouse is associated with higher ratings of pain by the patient.
4. Research suggests that partner-assisted or couples based training in behavioral pain management can help patients acquire and maintain more adaptive pain coping skills.