The Role of Integrated Primary Care in Increasing Access to Effective Psychotherapies in the Veterans Health Administration

Veterans of the U.S. military often have considerable psychiatric needs. For a host of reasons—ranging from high rates of preenlistment mental illness (1) to combat experiences (2) and other ills of war (3), challenges of reintegration after military service (4), lower socioeconomic status (5), and worse physical health and health behaviors (6)—veterans have higher rates of psychiatric disorders (7–9) and suicide (10) compared with civilians. Accordingly, mental health care is an essential component of the extensive array of services offered by the Veterans Health Administration (VHA). Access to care, which refers to obtaining needed health care in a timely manner (11), has been “both a high priority and a persistent challenge” for VHA in recent years because demand for care has outpaced the system’s capacity to deliver it (12). The negative repercussions of poor access to care are illustrated by the association between unmet mental health care need and suicidal ideation among veterans (13).

VHA has undergone substantial organizational changes over the past three decades (14). Efforts to expand access to health care for veterans have included implementing the patient-centered medical-home model of primary care, expanding use of telehealth, opening VHA community-based outpatient clinics in rural areas, and increasing use of community-based (non-VHA) care (see 12, 14–16). More than a decade ago, VHA undertook a major initiative, termed Primary Care-Mental Health Integration, specifically designed to increase access to effective mental health treatment (14, 17, 18). The embedding of behavioral health providers in the primary care setting as part of the health care team is broadly referred to as “integrated primary care” (IPC) and has become increasingly popular over the past two decades (19).

What Is IPC?

IPC is a population-based model of care in that it seeks to serve the entire population of a primary care clinic, rather than a small subset of patients with the most significant mental health concerns (20, 21). To ensure continued access in the system for new patients, treatment is time limited, and stepped care is emphasized. That is, individuals who do not benefit sufficiently from brief treatment can be stepped up to a higher level of care, such as specialty mental health treatment.

IPC is also highly collaborative in that behavioral health providers are integrated as part of the care team to provide consultation to medical staff about how to manage behavioral health concerns and coordinate care for behavioral health and physical conditions (22). Ideally, IPC services are both measurement based and evidence based (14). Goals of IPC include improving detection of behavioral health problems, increasing access to and engagement in appropriate care, improving quality and coordination of care through collaboration between behavioral health and medical providers, and decreasing mental health stigma (18).

Although there are many IPC models, two specific models are most commonly used (19). The primary care behavioral health (PCBH) model is most widely implemented in clinical care, and the collaborative care (CC) model has been most widely researched. In PCBH, sometimes referred to as colocated collaborative care or the behavioral health consultant model, a licensed mental health clinician, such as a psychologist or clinical social worker, is embedded in the primary care team to provide brief assessment and intervention for a wide range of mental health, substance use, and health behavior concerns (23). The prototypic course of treatment ranges from one to six 15- to 30-minute appointments, spaced every two to four weeks.

CC, also referred to as care management and disease management, typically involves nurse care managers providing education and psychosocial support around the use of medications for specific behavioral health concerns (e.g., depression), in consultation with the primary care provider as well as a supervising psychiatrist (24, 25). The support provided in CC is protocol driven and guided by algorithms for stepping care up or down on the basis of the patient’s progress over time. In contrast, brief psychotherapy in PCBH is typically guided by the individual clinician’s judgment.

Many types of behavioral health providers deliver IPC, and they often play different roles on an integrated care team. IPC teams may include primary care providers; nurse care managers; licensed practical nurses; health technicians; nonprescribing behavioral health providers; and others, such as clinical pharmacists, registered dieticians, medical social workers, and peer support specialists (26). Psychiatrists serve several key roles in IPC care teams. In the CC model, consultant psychiatrists provide caseload review, curbside consultation, and education. This allows the psychiatrist’s expertise to be extended to a larger proportion of the patient population. For instance, primary care providers can be encouraged to do more psychiatric prescribing with ready access to psychiatrist expertise (27). Psychiatrists practicing in the CC model have been described as meeting the triple aim of health care: simultaneously improving access to care and care quality while keeping costs low (28). Psychiatrists are less frequently integrated into the PCBH model; psychologists or clinical social workers typically provide consultation to the primary care team as well brief psychotherapy directly to patients. However, in PCBH, psychiatrists may begin a patient on psychiatric medication with the intention that the primary care provider will continue prescribing once the patient is stabilized (29).

Some unique aspects of the U.S. Department of Veterans Affairs (VA) system enable IPC. VA has one shared electronic medical record that all providers use. This allows easier communication and coordination between behavioral health and medical providers (14). The medical record also includes a component in which patient status can be assessed with common psychological assessments, which enables IPC to be measurement driven.

In addition, the pay structure of VA allows easier access to IPC, because IPC is considered a primary care service, and typically special copays for mental health treatment do not apply. Also, veterans incur one copay per day of service, so “warm handoffs” do not result in additional copays. As described in the previous paragraph, VA also has a diverse behavioral health work force, ranging from licensed independent providers to dependent providers, such as health technicians and peer support specialists (26). VA also has well-developed systems to provide IPC to rural veterans, including community-based clinics and telehealth. Although other health care systems share some of these same features, it is likely that, combined, these features have made IPC successful in VA.

IPC Improves Access to Mental Health Services

It is clear from the existing literature that IPC in general, and CC and PCBH in particular, improve access to mental health services. Access occurs on a continuum from initial access to behavioral health services in primary care to initiation, engagement, and completion of specialty mental health care. Access outcomes can also include gaining access to psychiatric medication and receiving guideline-concordant care. In IPC, access to behavioral health care is not limited to direct in-person or phone contact with a behavioral health provider. Behavioral health providers, such as psychiatrists or psychologists, consult with members of the primary care team to enable them to deliver behavioral health care in the form of psychiatric medication or health behavior coaching.

The research evidence that CC improves access is very strong. A 2012 meta-analysis of 79 CC studies concluded that CC not only increases access to pharmacotherapy and care manager support but also improves depression, anxiety, and quality of life outcomes for primary care patients (30). Much less research has investigated whether CC improves access specifically to evidence-based psychotherapy.

CC models have been increasingly including cognitive-behavioral therapy (CBT) as part of stepped-care delivery models (31), thereby increasing access to CBT, an evidence-based treatment. For instance, the STEPS-UP model for depression and posttraumatic stress disorder (PTSD) among active-duty military members includes CBT delivered in a variety of modalities and has resulted in patients using more mental health care (21). However, data on engagement in specific evidence-based psychotherapies resulting from STEPS-UP have not been reported.

Another CC intervention that included CBT in a stepped-care model found that veterans with PTSD who were randomized to CC were 18 times more likely to receive cognitive processing therapy (CPT) than those receiving usual care (29). This CC intervention delivered CPT by telephone to rural veterans to improve access to care. This study demonstrates the potential for carefully designed CC models to increase engagement in evidence-based psychotherapy.

Relatively less research has investigated whether PCBH increases access to care. However, a recent systematic review of 23 PCBH studies found that PCBH increased initial and sustained access to behavioral health services in primary and specialty care (32). Effect sizes were small to medium, and most studies used naturalistic, uncontrolled designs. For instance, one large national VA study found that primary care patients who received same-day PCBH services were eight times more likely to initiate psychotherapy and two times more likely to initiate antidepressant treatment than patients receiving usual primary care (33). Another national VA study found that patients referred by PCBH providers were three times more likely to engage in specialty mental health treatment than patients referred by primary care providers (34). These studies demonstrate how PCBH increases access to the continuum of mental health services. However, the field has yet to study whether PCBH increases access specifically to evidence-based psychotherapies either within primary care or as part of stepped care.

What Components of IPC Drive Increased Access?

There are several ways that IPC likely improves access to mental health care. First, IPC enables better access by virtue of being a population-based model of care (35). In this approach, a larger number of patients, and thus a larger proportion of the population, use the service, compared with specialty care models. IPC services tend to involve briefer visits, fewer visits, and longer intervals between visits compared with higher intensity specialty care (20).

In addition to these structural differences, the approach to treatment differs, with population-based care in IPC employing more of a “consultant” rather than “treatment” orientation (20). The goal is to support the care team in helping the patient improve his or her functioning, rather than to treat to the point of complete remission. Care is therefore more focused, which ensures that individuals’ engagement is more episodic, in turn allowing frequent turnover in cases to facilitate access. Also, when mental health providers support other primary care team members in delivering behavioral health services, this further increases access to the primary care population. For instance, a psychiatrist can provide education and support to a primary care provider prescribing an antidepressant, or a psychologist may teach nurses motivation-interviewing strategies to encourage more health-promoting behaviors.

Another way IPC facilitates access to care is by addressing patient treatment preferences. Patients prefer to receive care for mental health concerns in the primary care setting rather than the specialty mental health setting (36–39). IPC leverages the patient’s relationship with the primary care provider to facilitate acceptance of referrals within the team (40). Treatment offered in the primary care setting as part of routine health care (20) may be more acceptable to patients because they are already familiar with the primary clinic, do not have to expend extra effort to seek services, and may perceive less stigma compared with the specialty mental health setting.

Patients also prefer services that are convenient. IPC facilitates access to care by increasing ease and convenience for patients, eliminating delays and hassles, and facilitating immediate action on problem recognition to capitalize on motivation and thereby optimize engagement. In general, IPC reduces barriers to treatment, particularly when IPC is executed in the ideal fashion with same-day access, in which the mental health appointment is completed immediately following the primary care appointment in which a treatment need is identified (20, 41). Same-day access eliminates many common practical treatment barriers, such as contacting the provider or clinic, setting up another appointment, finding child care or transportation, and taking time to return to the clinic (42).

IPC is also believed to improve the quality of primary care by expanding the breadth of expertise offered, because it uses an entire team of providers rather than one single provider. This interdisciplinary approach requires a high level of communication and collaboration (19, 43). Because of their specialized training, embedded behavioral health providers contribute expertise in psychopharmacology and behavioral treatments to the primary care team, which can complement the primary care provider’s and nurse’s ability to be credible sources of medical information (44).

Collaboration can take many forms in a primary care visit. At the most minimal level, warm handoffs occur (a tenet of the PCBH model), such that the primary care provider introduces and transfers patient care to the behavioral health provider after the primary care visit (45). Additional collaborative efforts in IPC can take the form of group medical visits or shared medical appointments involving multiple care team members interacting with patients during the same visit (46, 47).

Evidence-Based Psychotherapies Designed for IPC

Not only does IPC provide the opportunity to increase access to psychotherapy in specialty mental health settings, but it also allows for provision of brief evidence-based treatments to be delivered in primary care without increasing the burden on the primary care team. The evidence base supporting treatments that are feasible to deliver in primary care is limited yet growing. Here, we review research on interventions for the most common presenting concerns in primary care.

Depression

Depression is common in primary care (48). IPC enables behavioral treatments to be provided in primary care as a supplement or alternative to antidepressants. The evidence base for CC for depression is very strong. In a Cochrane review of 79 studies (30), greater improvement in depressive symptoms as well as medication adherence and quality of life was found among primary care patients who received CC for depression. CC for depression includes care being provided jointly by a primary care and a mental health professional, a structured depression management plan (typically including medication management and psychosocial support), scheduled follow-ups, and enhanced interprofessional communication (30).

With respect to the PCBH model, reviews have found that brief treatments for depression, including three to nine sessions of CBT or problem-solving therapy, have significant but small effects on depressive symptoms compared with primary care treatment as usual (49, 50). A recent meta-analysis of 29 randomized controlled trials (RCTs) examining multimodal CBT in primary care concluded that different delivery formats of CBT were effective for depression, such as face to face, online, guided self-help, and telephone (50). Recent research has begun to examine treatments that may be more consistent with the brevity of care provided within PCBH, ideally one to four sessions (51). This includes studies examining brief treatments, such as two to four sessions of behavioral activation (52) or e-health and mobile technology as a way to deliver intervention content to primary care patients outside of a clinic setting (53). For instance, in a recent open trial, a web and mobile-phone CBT intervention with a focus on positive coping strategies was found to decrease depressive symptoms (54). Overall, primary-care-delivered CBT for depression is effective, but more research is needed to test briefer models of treatment and determine how technology can more effectively deliver treatment content.

Anxiety

IPC offers a pathway to improving treatment for anxiety symptoms and disorders, which are highly prevalent and burdensome in primary care yet underrecognized and undertreated (55). Although the vast majority of CC trials have targeted depression (55), more research is now examining CC for anxiety, and evidence is emerging that CC is effective for anxiety disorders (56). A recent meta-analysis of seven RCTs (57) found that patients who received CC had greater reductions in anxiety compared with those in the usual primary care condition. Many CC interventions for anxiety use a stepped-care approach, typically beginning with guided self-help materials based on CBT principles.

For example, in a recent study of stepped CC for panic disorder and generalized anxiety disorder in community primary care (N=329), the lowest intensity level of care was completing a guided (by telephone) self-help workbook based on CBT principles, followed by pharmacotherapy or referral for specialty mental health services (58). CC was found to be more effective in reducing anxiety and improving quality of life than usual care. By offering self-management resources grounded in CBT as well as medication management, CC for anxiety increases access to interventions adapted from evidence-based psychotherapy and to evidence-based pharmacotherapy.

In another trial (N=704) investigating which technology resources can assist the delivery of CC for anxiety, it was found that CC that included computerized CBT was superior to usual care. Adding an Internet support group did not confer additional benefit, however (59). The value of telephone, text, Internet-based, and mobile application support as part of IPC care is an area ripe for additional research. given that early studies have found that certain types of technology can increase access to care.

Limited research has examined anxiety interventions in the context of PCBH. A meta-analysis of 12 RCTs found that psychological treatment for anxiety disorders in primary care (most often CBT for generalized anxiety disorder or panic disorder) was effective (60). However, a recent review of psychological treatment for anxiety in primary care found that few existing interventions were feasible for use in real-world PCBH practice because of either the high number of sessions (more than six), long duration of sessions (longer than 30 minutes), or narrow focus of intervention (single anxiety disorder) (61).

A few studies have examined brief interventions designed specifically for use in VA PCBH settings, including one trial (N=180) of brief CBT for anxiety and depression symptoms among primary care patients with heart failure or chronic obstructive pulmonary disease (62). Brief CBT increased access to care, with 84% of intervention participants versus only 5% of usual care participants receiving any psychotherapy, and significantly reduced anxiety and depression. This study highlights the unique ability of IPC to increase access to effective psychotherapy that addresses comorbid mental and physical health disorders.

Another RCT (N=223) found that an average of seven sessions (many completed by telephone) of CBT for older adults with generalized anxiety disorder yielded greater reductions in anxiety and worry compared with usual primary care, regardless of whether CBT was delivered by doctoral- or bachelor’s-level providers (63). This study illustrates the potential for PCBH to expand access to effective psychotherapy through avenues such as telephone-delivered care to reach patients in rural areas or with other transportation barriers. It also supports the benefit of lay providers (working under appropriate clinical supervision) to help address shortages in the mental health workforce.

PTSD

Like depression and anxiety, PTSD is very common among primary care patients, and individuals with PTSD are unlikely to seek mental health services outside of the primary care setting (64–66). The research base on IPC interventions targeting PTSD symptoms is small but consistently growing over time. CC interventions originally designed to treat depression have been adapted for PTSD, but some of the major trials have failed to demonstrate that CC decreases PTSD symptoms more than usual care.

One study (N=195) found that CC increased access to mental health visits and medication refills but did not improve PTSD symptoms or functioning among veterans with PTSD (67). Similarly, another study (N=404) found that CC did not improve PTSD symptoms more than usual care at Federally Qualified Healthcare Centers, although patients who engaged in CC did have more mental health visits and higher prescription rates (68). CC approaches that allow patients to opt into psychotherapy for PTSD have yielded better results than the traditional CC models focused on increasing access to pharmacotherapy.

Engel et al. (69) compared traditional CC with an enhanced version that included options for telephone, online, and in-person CBT. They found that the enhanced CC resulted in more care utilization and reductions in PTSD and depression among 666 military members. Fortney et al. (29) compared a type of CC that provided telephone-based CPT with usual care (N=133). They found that CC significantly increased access to evidence-based PTSD treatment for rural veterans and that patients who received CPT had significant reductions in PTSD symptoms. In conclusion, CC for PTSD seems most effective when it increases access to evidence-based psychotherapy.

Less research has been conducted investigating models of PCBH treatment for PTSD. However, existing studies uniformly have found preliminary evidence that PCBH interventions reduce PTSD symptoms, despite a wide variety of specific approaches being delivered to reduce PTSD symptoms. Cigrang and colleagues (70, 71) conducted a pilot study followed by a small RCT (N=67) among military personnel comparing four 30-minute sessions of prolonged exposure with usual care and found significant reductions in PTSD symptoms favoring prolonged exposure.

Another intervention similarly delivered four 30-minute sessions to veterans but focused on teaching PTSD symptom-management strategies. Participants also engaged in at-home symptom management with the PTSD Coach mobile app. This intervention was found to reduce PTSD symptoms and increase engagement in additional mental health treatment in a small RCT (72). Engel et al. (69) tested a unique adaptation of the PCBH model among 80 combat veterans and found that primary care patients who received nurse-assisted online CBT had larger reductions in PTSD than those who received usual care. Group mindfulness training also was found to reduce PTSD symptoms compared with usual care among 60 primary care veterans (73).

Practitioners also can deliver interventions in PCBH to increase patient engagement in specialty mental health services. A CBT intervention focused on restructuring negative treatment-seeking beliefs effectively increased engagement in psychotherapy for PTSD in specialty care (74). In review, practitioners can implement a large variety (e.g., one on one, group, e-health, therapist delivered, nurse delivered) of treatment modalities within PCBH to reduce PTSD symptoms, but more rigorously designed research is needed for the field to be confident in these results.

Hazardous Alcohol Use

Hazardous alcohol use is most commonly addressed in primary care according to a screening, brief intervention, and referral to treatment (SBIRT) model. Primary care practices are recommended to conduct regular screening for hazardous alcohol use, and in the VHA, 40% of recent returning veterans screen positive (75). Most individuals with hazardous alcohol use or alcohol use disorders are not engaged in treatment (76–78). IPC offers the primary care team assistance in the SBIRT process, providing access to embedded behavioral health providers with specialized skills, including the ability to provide brief interventions and facilitate referrals to more specialized addiction settings.

Research examining CC for hazardous alcohol use, including alcohol use disorder, has found that CC can increase engagement in specialty care and reduce drinking. For instance, a CC intervention combining evidence-based pharmacotherapy (i.e., naltrexone), measurement-based care, and psychosocial support was found to increase abstinence from heavy drinking compared with usual care among 163 veterans with alcohol use disorder (79). Other research has found CC to be effective in increasing engagement in services among patients endorsing hazardous drinking compared with enhanced referral processes (80). Research also supports CC as an effective method for providing telephone-delivered brief alcohol interventions (81).

Meta-analyses including 56 RCTs (82) and a recent Cochrane review (83) have demonstrated the efficacy of brief alcohol interventions (i.e., five or fewer sessions, less than 60 minutes total duration) in primary care for patients with hazardous drinking. However, there is little evidence that brief alcohol interventions alone improve outcomes for patients with alcohol use disorder, which makes them most appropriate as an early intervention in a stepped-care model. An advantage of brief alcohol interventions is that the average length and number of appointments matches the PCBH model (84), which helps to ease translation into practice. Research has begun to examine how to overcome other implementation issues by examining how to support primary care practices and result in sustainable programs (85) as well as using e-health to help support provider trainings (86) or direct patient support (87).

Tobacco Use

Tobacco use is another important presenting concern in primary care. Approximately 80% of smokers visit primary care each year, and it is recommended that treatment for tobacco use disorders is provided in primary care (88). In a stepped-care model, IPC providers can help provide treatment to those patients who do not respond to interventions delivered by the primary care provider (89). Previous research has found that the combined impact of behavioral treatments and medications increases the chance of quit success above either provided alone (90), which supports the idea that primary care is an ideal place for intervention.

A recent meta-analysis reviewed 36 studies examining brief IPC interventions for tobacco use and found that participants who received an intervention had greater odds of smoking cessation (91). Similar to brief alcohol interventions, these interventions matched the brief format necessary to fit into IPC and consisted of motivational interviewing and CBT strategies. Researchers are now examining electronic cigarettes (92) because of their increased popularity. In addition, research is focusing on the implementation of these evidence-based interventions, especially those that use e-health to decrease the burden on the primary care team and have the potential to reduce disparities in receipt of care (93–95).

Insomnia

Insomnia has received a fair amount of attention in primary care because of its prevalence (96), which can be as high as 90% among veterans (97). Because patients prefer behavioral treatments for insomnia to medications (98), a significant amount of research has examined CBT for insomnia (CBT-I), which is highly effective (99, 100). A recent literature review found strong evidence for brief (i.e., six or fewer sessions) behavioral treatments for insomnia, which can easily translate into PCBH settings (101).

A meta-analysis of 87 RCTs investigating CBT-I found that at least four sessions of face-to-face treatment outperformed self-help interventions, regardless of the presence or absence of medications (102). A recent systematic review reported promising results from mobile-phone-delivered sleep interventions, with 14 out of 16 studies reviewed improving sleep quality (103). In summary, brief CBT-I is a good fit for IPC, and early evidence indicates that e-health can assist in treating insomnia among primary care patients. Accordingly, VHA has nationally disseminated a six-session CBT-I protocol that can be delivered in primary care (101) as well as the CBT-I Coach mobile app to supplement this treatment (104).

Conclusions and Future Directions

IPC increases access to initial and continued mental health treatment using a wide variety of services supported by behavioral health providers located in the primary care setting. Although this review describes two widespread models of IPC (CC and PCBH), increasingly components of these models are being blended. Traditionally, CC has targeted specific prevalent disorders and focused on pharmacotherapy supplemented by patient education and psychosocial support. PCBH, in contrast, has sought to modify full-length psychotherapies to be delivered briefly for the full range of behavioral health concerns seen in primary care. Several innovative IPC interventions include psychiatric medication, evidence-based psychotherapies, and other components (e.g., self-management tools), often delivered in a stepped-care fashion. The research reviewed here demonstrates that these blended approaches are often superior to traditional CC models (e.g., 29, 58, 69) for depression, anxiety, and PTSD. A notable feature of blended approaches is that they require psychiatrists, psychotherapists, and nurse care managers each to play unique roles in care delivery while working collaboratively with the larger primary care team. The multiple integrated roles in a blended model can be complicated to implement, but the research demonstrates that the payoff of successful implementation is increased access to effective psychotherapies and improved patient well-being.

The review of PCBH services shows that brief psychotherapies in primary care have excellent potential to decrease mental health symptoms, hazardous alcohol and tobacco use, and insomnia. However, the field of PCBH research is at a different developmental stage for each of these presenting concerns. For depression and anxiety, most of the CBT interventions tested need to be further shortened so they are feasible for PCBH providers to deliver. When PCBH providers have treatments that are longer than they can feasibly deliver, they make idiosyncratic choices regarding what to deliver; consequently, they may leave out active components, reducing the effectiveness of the treatment. PCBH research focused on PTSD treatment has many promising early studies, but larger, full-scale RCTs are needed for researchers to be confident that specific PCBH interventions reduce PTSD symptoms. PCBH interventions for hazardous alcohol use, tobacco use, and insomnia are well developed but not implemented widely enough. More implementation-science studies are needed for researchers to understand how to overcome barriers and capitalize on facilitators for implementation.

Practitioners are using a variety of technological and e-health tools to assist in delivering behavioral health services to primary care patients. Research has demonstrated that telephone-delivered care, websites, and mobile apps that deliver CBT content are important components of IPC services (e.g., 50, 58, 69, 73, 102). More research is needed to understand the optimal way to incorporate e-health into IPC. It is important to understand that e-health consists of tools used to deliver an intervention and is not a stand-alone intervention.

The supportive accountability model offers a useful framework on how to incorporate technology into clinical services. It describes how adding professional support increases patient engagement in self-management materials (electronic or paper based) through accountability to a clinician who is seen as trustworthy, helpful, and experienced (105). More research is needed on how to strike the correct balance between clinician support and e-health materials in IPC services.

IPC improves access to mental health services by providing brief, focused treatment to many primary care patients with a wide variety of presenting concerns. More patients engage in care when it can be accessed immediately and is offered in a less-stigmatizing setting where many patients prefer to receive treatment. Our review describes how interventions that blend components of CC and PCBH, often in a stepped-care approach, have advantages over traditional CC models. Future PCBH research will benefit from focusing on brief psychotherapies that are feasible to deliver in primary care and from conducting implementation-science studies. E-health holds great promise for expanding the reach and depth of IPC services, and future research should focus on finding the best combination of e-health and professional support.