Behavioral interventions for arthritis have frequently been “omnibus” in nature. Omnibus interventions are comprised of many different treatment elements and require the acquisition of several skills on the part of patients. Interventions for pain, for example, have included such elements as rational restructuring, activity pacing, relaxation training, increasing social support, and engaging in pleasurable activities and events (Dixon et al., 2007; Radojevic, Nicassio, & Weisman, 1992). The inclusion of several treatment elements in clinical trials research may reflect the expectation that teaching patients several skills is a more effective way of enhancing the potency of interventions instead of relying on more streamlined, targeted approaches that focus on a limited set of treatment skills or strategies. However, the risk of omnibus approaches is that interventions may become saturated, making it problematic for patients to master a range of skills and apply them on a consistent basis. The uptake or learning of aspects of the interventions and resulting impacts becomes more difficult as complexity of the interventions increases (Ingersoll & Cohen, 2008; Stone et al., 2001; Vermeire, Hearnshaw, Van Royen, & Denekens, 2001). Adherence to, and performance of, learned skills may also be compromised as patient demands increase (Barlett 2002). Furthermore, the inclusion of several elements in behavioral treatment interventions has made it difficult to isolate the specific components that are responsible for clinical improvement. Future research directed at identifying active treatment elements in multifaceted treatment packages would fill an important gap in the behavioral intervention literature. Doing so will also contribute to the efficiency of delivering psychosocial care.

There is also a need for understanding the mechanisms of action underlying the efficacy of behavioral treatment approaches for arthritis pain and disease activity. Figure 16.3 illustrates potential mechanisms that may contribute to clinical change. Behavioral interventions may work through a number of potential pathways either individually or in concert with one another. They may also be correlated or causally related with one another.

Behavioral interventions may contribute to improvements in patients’ beliefs or appraisals of their health. They may promote increases in self-efficacy or reductions in negative beliefs such as helplessness or catastrophizing. Most behavioral interventions based on social learning theory principles have attempted to foster healthier illness appraisals that promote a sense of optimism and agency in patients. Recently, investigators have shown increased interest in mindfulness approaches that foster acceptance and a non-judgmental awareness about health (Grossman, Niemann, Schmidt, & Walach, 2004; Hofmann, Sawyer, Witt, & Oh, 2010; Nyklíček, Hoogwegt, & Westgeest, 2015; Rosenzweig, Greeson, Reibel, Green, & Jasser, 2009). Changes in patients’ beliefs have been linked to a range of outcomes, including less pain, improved mood, and decreased disease activity (Carrico & Antoni, 2008; Irwin et al., 2015; McCracken & Turk, 2002; McCracken & Vowles, 2008; Turk & Okifuji, 2002). Reductions in helplessness and catastrophizing, in turn, have been shown to contribute to adaptive coping (Cabib, Campus, & Colelli, 2012; Evers, Kraaimaat, Van Riel, & De Jong, 2002; Hannibal & Bishop, 2014; Hewlett et al., 2011).

Importantly, behavioral interventions may promote changes in patients’ coping mechanisms for their pain by increasing active coping and reducing passive or avoidant coping strategies. Greater active coping and less passive coping have been shown to be consistently related to reductions in pain and improvements in mood in a range of pain populations (Higgins, Bailey, LaChapelle, Harman, & Hadjistavropoulos, 2014; McCracken & Turk, 2002).

Behavioral interventions also may target mechanisms for increasing the social network size and quality of support of patients. Social support has been associated with positive health outcomes in both OA and RA patient populations (Ethgen et al., 2004; Evers et al., 2002; Holtzman, Newth, & Delongis, 2004; Krol, Sanderman, & Suurmeijer, 1993; Zyrianova, 2006). Many behavioral interventions are rendered in a group format that promotes a supportive, interactive atmosphere. Social support interventions build on that foundation by incorporating specific strategies for increasing support and effective social integration. Some behavioral interventions have addressed social support by including a family member in the intervention (Radojevic et al., 1992) to facilitate the performance of behavioral skills in the patient or by enhancing communication skills around pain and related issues (Cegela, McClure, Marinelli, & Post, 2000; Keefe et al., 2004).

Behavioral interventions have also targeted mood disturbance in arthritis populations since depression is a common symptom of arthritis sufferers (Lin et al., 2003; Sharpe et al., 2001; Zautra et al., 2008). Depression is commonly linked with greater pain, disability, and disease activity in patients with arthritis (Baqar & Moore, 1990; Fifield, Tennen, Reisine, & McQuillan, 1998; Peck, Smith, Ward, & Milano, 1989; Sharpe, Sensky, & Allard, 2001). Interventions to improve mood include such strategies as cognitive restructuring to promote healthier beliefs, increasing daily pleasurable activities, or savoring positive emotions to promote gains in positive affect through mindfulness practices.

Interventions for arthritis may also include strategies to increase activity level and energy expenditure through structured aerobic exercise, walking, or by reducing sedentary behavior (see Chap. 13; Knittle et al., 2010). Behavioral interventions for pain often include strategies for pacing of activity that combines periods of energy exertion and rest (Gatchel & Turk, 1996; Murphy, Lyden, Smith, Dong, & Koliba, 2010). Activity pacing can be integrated with other approaches such as relaxation during periods of rest or emotional savoring while patients engage in pleasurable activity.

Inflammation is the underlying cause of joint pain and swelling in RA and also may exacerbate fatigue, sleep disturbance, and depression (Belza, Henke, Yelin, Epstein, & Gilliss, 1993; Bergman et al., 2009; Davis et al., 2008; Drewes, 1999; Lee et al., 2009; Stebbings, Herbison, Doyle, Treharne, & Highton, 2010). There is some evidence that behavioral interventions in other chronic diseases may lead to reduced inflammation and positive alterations in immune functioning (Antoni et al., 2000; Balagopal et al., 2005; Irwin, Wang, Campomayor, Collado-Hidalgo, & Cole, 2006; Miller, Ancoli-Israel, Bower, Capuron, & Irwin, 2008; Miller & Cohen, 2001; Thornton, Andersen, Schuler, & Carson, 2009; Zautra et al., 2008). Theoretically, behavioral interventions for arthritis should have similar salutary effects on inflammation. Evidence attesting to the effects of behavioral interventions on inflammation, however, is limited (Zautra et al., 2008).

However, despite the considerable evidence that behavioral interventions may promote positive changes in potential mediators , it is unclear at this juncture whether such changes are responsible for the improvements in clinical outcomes. A major reason for this gap in knowledge is that clinical trials have not incorporated mediational statistical methodologies that are necessary to address this question. In order for a mediator of a behavioral intervention to be identified, the following conditions must be met: (1) The behavioral intervention has to affect the clinical outcome, (2) The behavioral intervention has to lead to change in the mediator, (3) The mediator must be related to the clinical outcome, and (4) for full mediation, the effect of the behavioral intervention becomes non-significant after the contribution of the mediating variable is taken into account (Baron & Kenny, 1986). Partial mediation occurs when the pathway from the behavioral intervention to the clinical outcome remains significant after controlling for the mediating variable. Frameworks also can be adopted that incorporate a multiple mediation approach in which the effects of two or more mediators are analyzed simultaneously (Preacher & Hayes, 2008a, 2008b). Multiple mediation strategies may be particularly appropriate for analyzing the effects of behavioral interventions since they commonly target more than one mechanism for improvement.

While identifying mechanisms provides an explanation of why behavioral interventions are effective, there are other benefits to identifying mechanisms as well. A key advantage is that behavioral interventions may increase their efficiency by focusing more directly on mediators that are critical to change. Potential mediators that do not show a relationship with outcomes may be deemphasized in favor of those that have a clinical impact. Another advantage is that other approaches may be developed or implemented that focus on the mediator. Rather than fostering allegiance to a particular theoretical approach or model, new treatments can be developed that focus directly on altering the underlying mechanisms involved. For example, independent of traditional behavioral interventions that are primarily based on social learning theory, there could be other strategies (e.g., complementary medicine techniques, exercise, physical therapy) that target self-efficacy, increase active coping, foster social support, or improve mood. Greater knowledge and application of these mediational frameworks may thus lead to increased treatment efficiencies by promoting rational integration between treatments, mediators, and clinical outcomes.

The Council of Academic Health Centers for Integrative Medicine has advocated for the adoption of an integrative medicine model in the management of chronic disease. Kligler and Chesney (2014) note that this organization has defined integrative medicine in the following way:

“Integrative medicine is the practice of medicine that reaffirms the importance of the relationship between practitioner and patient, focuses on the whole person, is informed by evidence, and makes use of all therapeutic approaches, healthcare professionals, and disciplines to achieve optimal health and healing.”

Behavioral interventions fall under the integrative medicine rubric to the extent that they are evidence-based, are holistic in nature, and promote a collaborative relationship between patients and practitioners. The implementation of this framework in arthritis care represents both a challenge and opportunity in the effort to improve the quality of life of arthritis sufferers. Key questions emerge regarding the dissemination of behavioral treatments and their integration into clinical practice.

The RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) model borrowed from public health has provided impetus for examining the translational impact of behavioral interventions in clinical and community settings (Gaglio, Shoup, & Glasgow, 2013). Reach refers to the absolute number of individuals who are willing to participate in an initiative; effectiveness refers to the impact of the initiative on clinical outcomes; adoption involves the number of settings and intervention agents who are willing to initiate a program; implementation refers to the degree of fidelity on the part of intervention agents to elements of a program protocol; maintenance involves the degree to which a program becomes institutionalized or part of the institution’s routine organizational practices. While the RE-AIM framework has had a major impact on the dissemination of interventions to increase physical activity (Dzewaltowski, Estabrooks, & Glasgow, 2004), it also is relevant to examining the potential expansion and integration of behavioral interventions in arthritis clinics and community settings. The RE-AIM framework provides a conceptual background for addressing the logistical issues involved in integrative care. In many instances, the mechanisms for integration can be challenging and difficult due to obstacles that are inherent in the healthcare system. Access to care problems, time constraints during medical visits, and lack of personnel resources in the clinic setting to incorporate behavioral interventions may all be operative to various degrees and interfere with integrated treatment (Nicassio, 2008).

There are different models for examining the integration of behavioral treatments in arthritis care. An essential prerequisite is that members of the healthcare team (e.g., rheumatologists, allied health professionals, behavioral medicine specialists) have a mutual understanding of the need to adopt a biopsychosocial model of care and to embrace its relevance in working with clinic patients. Agreement on the importance of the model will facilitate communication among healthcare providers and between healthcare providers and patients. It will provide a foundation for the coordination and implementation of behavioral interventions and their role in medical care. After embracing the importance of behavioral interventions and their clinical value, healthcare providers face the challenge of how to integrate them in the clinic.

Referral-based frameworks involve the provision of behavioral medicine services by a specialist who functions in a different environment and is independent of the rheumatology clinic. Referrals are made to a specialist with expertise in pain management, the treatment of depression, or other psychosocial issues. This traditional model of providing psychosocial care makes integration more difficult since patients may be reticent to see an outside specialist. Also, communication between the rheumatologist and behavioral medicine professional may become problematic and interfere with coordination of treatments. This model still may achieve some success, however, if patients do, in fact, receive effective care from the specialist. Referral-based models are common in medical practice with its increased specialization and compartmentalization of care. Despite its drawbacks, it is the model that has the greatest familiarity among rheumatologists and other medical providers.

A superior model for integration involves the provision of behavioral treatments in the rheumatology setting. In this form of integration, behavioral treatments are “embedded” into the clinical environment. This could be accomplished in the following ways. First, the behavioral medicine specialist may function as part of the rheumatology team. The specialist could perform psychosocial evaluations and screenings, and work with individual patients to manage their pain and other problems that impact their arthritis. The specialist could provide feedback to the rheumatologist that would facilitate treatment planning and medical decision-making. This model of care has had a positive impact in primary care and oncology settings (Fisher & Dickinson, 2014; Guo et al., 2013; Holland, 2004; Jacobsen & Wagner, 2012; Kearney, Post, Pomerantz, & Zeiss, 2014; Miller, Petterson, Burke, Phillips, & Green, 2014; Villareal et al., 2006).

If it isn’t feasible to have a behavioral medicine specialist in the rheumatology clinic, other alternatives may still be possible for the integration of behavioral medicine services. Allied health professionals such as nurses or physical therapists who work in the same setting as rheumatologists have the potential to implement behavioral treatments during the course of their interactions with patients. Behavioral treatments for pain , for example, can be brief and efficiently implemented. Manualized treatment applications exist that are suitable for use by professionals without formal training in behavioral medicine. Another alternative is that rheumatologists themselves take responsibility for implementing behavioral interventions with support from the allied health professionals. While time constraints may interfere with the feasibility of this option, behavioral treatments may have high credibility to patients when rendered by rheumatologists. The two aforementioned approaches would require expansion of the roles of allied health professionals and rheumatologists. Education and training on behavioral approaches would be necessary through continuing education and support from professional organizations. Increasing the capacity of rheumatology health professionals to render behavioral treatments could also be addressed through more frequent consultation with behavioral medicine specialists.

There are significant opportunities for the growth and relevance of behavioral treatments for arthritis. Behavioral treatments have an established history in the treatment of chronic disease and the amelioration of patient suffering. Clinical trials research has established their efficacy in reducing pain, psychological distress, and disability. Further research is needed to address their mechanisms of action, shedding light on why they are effective. The increased impact of behavioral treatments for arthritis will depend on the adoption of models that will facilitate their dissemination and application in clinical care in order to expand their reach. Behavioral medicine specialists and rheumatology health professionals are both invested in providing optimal care to arthritis patients. They must form a closer academic and clinical partnership in order to achieve this important goal.