Integrating cognitive behavioral therapy into primary care settings

This article serves as an introduction to the first half of our behavioral medicine two-part special series on cognitive-behavioral practice in medical settings. This first set of articles focuses on strategies and recommendations for integrating cognitive behavioral therapy (CBT) into primary care settings, and the unique challenges primary care in particular presents. Our subsequent issue will focus on the implementation of CBT in other more specialized forms of medical care, including cancer treatment and HIV care.

Why Focus on Behavioral Health in Primary Care?

In recent years, the movement to integrate behavioral health into primary care has been rapidly growing. The passing of the Affordable Care Act in 2010, the wide adoption of a patient-centered medical home model in primary care, and the advent of the Accountable Care Organization (ACO) Medicare Shared Savings Program have led primary care sites throughout the U.S. to further consider comprehensive, “whole person” (American Academy of Family Physicians, 2008) care of their patients, including a focus on behavioral health. This policy focus on behavioral health care in primary care settings reflects the current needs of primary care patients and providers. Data shows that the majority of individuals seeking mental health services turn to primary care as their first or only source of treatment (Cauce et al, 2002; Wang, et al., 2005; Wang, et al., 2006). As a result, more than half of common mental health problems are treated exclusively in primary care (Bea & Tesar, 2002). In 2007, in the U.S., nearly half of all prescriptions for antidepressants and anxiolytics were written in primary care settings (Schappert & Rechtsteiner, 2011).

Importantly though, these estimates do not include the proportion of patients who present in need of behavioral modification of health risk lifestyle factors, or whom might benefit from the assistance of behavioral health provider in learning how to adjust to and manage chronic physical illness. As readers of Cognitive and Behavioral Practice know well, there are numerous empirically supported cognitive and behavioral interventions for chronic disease management and lifestyle modification, including programs targeting chronic pain (e.g., Allen et al., 2012; Hoffman et al., 2007; Thorn et al., 2002), diabetes (e.g., Safren, et al., 2014), insomnia (e.g., Bélanger et al., 2012; Espie, et al., 2012), obesity (e.g., DiLillo et al., 2003; Unich et al., 2013), smoking cessation (e.g., Stanton & Grimshaw, 2013), and adherence to treatment regimens (e.g., Demonceau et al., 2013; Newcomb et al., 2014). Traditionally, mental health treatment in primary care settings has focused on provision of psycho-pharmacotherapy and psychosocial treatments have been managed via referrals to providers in the community. However, research shows that one-third to one-half of primary care patients referred to mental health specialists do not attend even a first visit (Fisher & Ransom, 1997). Primary care patients cite inaccessible offices, inconvenient office hours, difficulty finding providers who take their insurance, and/or the fact they do not have insurance, as some of the key barriers for not following up on these referrals (Fisher & Ransom, 1997). As a result, many patients who may benefit from psychosocial treatments do not receive this care.

For example, one of us found that nearly half (47%) of a sample of primary care patients in New England who had a current anxiety disorder diagnosis were not receiving any mental health treatment. Only 32% had received any form of psychotherapy in the past 3 months (Weisberg, et al., 2007). In this study, only 14% had received psychotherapy that reportedly contained key components of CBT such as exposure and/or cognitive restructuring. When we followed the primary care patients for up to five years, we found that under 37% ever, at any time during the follow-period, received psychotherapy containing CBT elements (Weisberg et al., 2013). Thus, though efficacious CBT treatments have existed for many years for the treatment of anxiety disorders, approximately two-thirds of primary care patients with anxiety disorders in our sample did not receive this type of care at any time over a 5 year period.

Embedding mental health services within the primary care site may help foster receipt of behavioral health treatment. There are a number of models for bringing behavioral health into primary care settings. In co-location, the behavioral health provider functions as an independent professional, but is geographically located within the same service as the primary care team and may have basic or close consultation and collaboration with the primary care providers. In integration, the behavioral health provider is part of the primary care team. There is close collaboration within an integrated system. All providers are part of the business of the practice, attend team meetings, and use the same medical record system. As noted below, the articles in this series present treatments that primarily are co-located, in that the treatment research team brought in services from their home organizations and performed them on-site in collaboration with the primary care team. In a few of these papers, the work approached integration, in that the behavioral health staff was part of the same organization and a collaborative treatment team (e.g. Corso et al., this issue; Gomez et al., this issue; Goodie & Hunter, this issue) and/or used a shared electronic medical records (Pigeon & Funderburk, this issue).

Whether co-located or more fully integrated, a growing body of research indicates that collaborative behavioral-primary care results in improved patient outcomes (e.g., Archer et al., 2012; Craske et al., 2011; Bower et al., 2006; Gilbody et al., 2006, Rollman et al., 2005; Roy-Byrne et al., 2010). However, collaborative care does not always include the provision of psychotherapy. The key defining elements of collaborative care are that health professionals work with primary care providers to serve as care managers and/or behavioral health clinicians, and that these professionals monitor patient treatment adherence and outcomes over time in a systematic manner, and provide feedback to the primary care provider. The health professional may be a nurse who monitors adherence to hypertension medications and home blood pressure checks and reports problems to the provider, or they may be a psychologist who provides brief psychotherapy to primary care patients with depression, while also monitoring outcomes and reporting these to the PCPs. This is important to note, because recent meta-analyses of collaborative care for depression—the disorder / problem with the largest primary care collaboration research base—found that while collaborative care was overall associated with decreased depressive symptoms, and care managers with a mental health background were associated with better outcomes than those without such education, whether or not the collaboration included psychotherapy services was not predictive of outcomes (Bower et al., 2006; Gilbody et al., 2006).

As cognitive-behavioral therapy researchers, we find this information troubling. A wealth of data from controlled trials in tertiary care shows that we have efficacious psychotherapies for the treatment of depression, so it is puzzling that the addition of psychotherapy to other collaborative care (primarily care management) was not associated with improved outcomes for depressed patients. However, further examination of the meta-analyses cited above (Bower et al., 2006; Gilbody et al., 2006) shows that psychotherapy was considered as one broad variable in these analyses. That is, there was no differentiation between cognitive behavioral therapy and other therapies with less of an evidence base. Similarly, Funderburk and colleagues (2011) examined the chart notes of primary care patients who had received behavioral health services as part of an integrated behavioral health-primary care program in the Veteran’s Administration Medical Centers in upstate New York. A random sample of 10% of the 1,870 patients who had at least one visit with a behavioral health provider was reviewed. Although the behavioral health providers were all trained in the functions of their role and in the idea of providing brief, co-located interventions, the behavioral health providers did not receive training or guidance as to the specific interventions to use during these brief therapy sessions. The authors found that chart notes made infrequent mention of the use of CBT techniques. Within the VA medical centers under study, only 18% of primary care patients seen by a behavioral health provider for depression received psychotherapy that included cognitive therapy techniques, and only approximately 25% received behavioral activation. Patient education and supportive treatments were commonly used. Thus, it is possible that that the provision of psychotherapy has not been found to predict outcomes in meta-analyses of collaborative care in part because the specific interventions used in the psychotherapy are not always those with an evidence-base. Improving the integration of CBT into primary care settings may therefore be crucial for improving patient outcomes and demonstrating the important role of brief psychotherapies in this context.

Providing CBT in primary care settings is challenging. As Blount (2009) points out, there are a great number of differences between working as a psychotherapist in a specialty mental health setting and being a behavioral health provider in primary care. As a behavioral health provider, you are the ancillary, rather than primary, health care provider. The goal is typically to manage the needs of the population of patients, and thus to effectively manage resources, short-term or episodic treatment is often provided, and each session is often only 30 minutes. The goal of treatment is thus often improved functioning, rather full recovery and, importantly, mental/behavioral health is the specialty, rather than being a depression, anxiety, HIV, ADHD specialist. These differences make it challenging for behavioral health providerss to adapt and apply our CBT evidence base to patients in primary care.

Moreover, guidance for behavioral health providers of how to adapt and apply CBT interventions to primary care patients across range of presenting problems is hugely lacking. Additional resources that are accessible and applicable to real-world primary care settings are needed for behavioral health providers to shift current practices and create a sustainable integration of evidence-based CBT approaches in these settings. As such, this special issue aims to begin to fill in these gaps by providing specific examples and practical guidelines for using CBT strategies across a range of commonly presenting behavioral health and psychological symptoms in primary care settings.

Overview of the Special Issue

The papers in this issue provide a range of recommendations for integrating cognitive behavioral approaches in primary care to address insomnia, depressive symptoms, and child externalizing symptoms, to manage suicide risk, and to improve medication adherence. Pigeon & Funderburk present a brief cognitive behavioral treatment for insomnia (CBT-I) tailored for the VA primary care setting for veterans who present with co-occurring insomnia and depression. The intervention described is a four-session protocol consisting of brief 15–30 minute sessions, two of which were telephone sessions, delivered by doctoral students. This work has important implications for increasing access to behavioral sleep interventions for depressed veterans. Goodie & Hunter also provide guidance for addressing insomnia in the primary care setting. In line with an integrated model, they present an approach in which behavorial health providers assess and treat insomnia using brief, evidence-based, cognitive behavioral techniques. They use a case example to illustrate practical recommendations for assessment and delivery of brief evidence-based CBT techniques in this context.

Bryan, Corso, & Macalanda present a model for managing suicide risk in the context of a patient-centered medical home (PCMH), a setting in which behavioral health care is integrated into primary care, and where behavioral health is a critical component of the treatment model. Despite the explicit integration of behavioral health care into this model, there have been few efforts to present guidelines for managing suicide risk in this setting. The authors describe a model for managing suicide risk in this setting, which includes screening and targeted assessment, delivery of brief CBT interventions, and ongoing collaboration between treatment team providers, including the PCP, nurse, and an integrated behavioral health consultant. Specific guidelines and example case material is provided for readers interested in implementing these practice in a PCMH.

Sheldon and colleagues present a model for addressing depression and psychotropic medication adherence in the context of a large, urban public health system of eight primary care clinics treating an underserved, largely low-income minority population. Their model, the Telephonic Assessment, Support and Counseling Program (TASC), uses evidence-based assessment methods and CBT strategies, including behavioral activation (BA) and motivational interviewing (MI), to address depressive symptoms and medication adherence, delivered over five brief telephone sessions by masters- or doctoral-level mental health providers. Telephone-delivery may have important implications for feasibility, access, and cost effectiveness for addressing depression and medication non-adherence, particularly in low-resource settings.

As an example of integrating evidence-based principles into primary care for pediatric populations, Gomez et al. describe the delivery of parent management training (PMT) strategies by behavioral health consultants (BHCs) in primary care to address externalizing symptoms. Approximately 15–16% of children present to their pediatrician with emotion or externalizing symptoms (Briggs-Gowan et al., 2003; Polaha, Dalton, & Allen, 2011; Williams, Klinepeter, Palmes, Pulley, & Foy, 2004), and integration of CBT into primary care settings may improve access to evidence-based treatments for these symptoms. Gomez and colleagues describe a truly integrated model using BHCs in primary care delivering evidence-based approaches for PMT (mean 2.38 visits). Their paper illustrates the process from triage to intervention for these children, and pilot data presented provides initial support for the acceptability, feasibility, and effects of the intervention on reducing children’s levels of distress.

Conclusion

In summary, the articles in this series offer behavioral providers a guide of how to deliver evidence-based treatments within primary care settings, and also include pilot data to add to the growing literature showing that when CBT is delivered in primary care, it improves patient outcomes. This data is important both at a clinical and a policy level. As more and more mental and behavioral health care will be provided in these settings in the near future, insurance companies, policy makers, and other stakeholders will soon be determining what the standard of this care should be, and what will be reimbursed. There is thus a strong need for the development and empirical evaluation of CBT, evidence-based interventions in primary care settings. Further, improving access to resources and trainings for behavioral health providers in primary care is important to promote adoption of evidence-based resources in this setting. We hope this Issue is a first step in illustrating delivery of evidence-based approaches in primary care and in providing hands-on recommendations for providers in these settings.