Abstract
Introduction
One in six U.S. adults is unable to access needed mental health care (Tikkanen et al., 2020), and access is particularly challenging in rural areas and other regions where the supply of mental health providers is limited or declining (Andrilla et al., 2018; Bishop et al., 2016). This inability to access mental health services is a critical barrier preventing patients from obtaining evidence-based care aligned with their mental health needs.
One strategy to improve access is to transition patients with stable or low-acuity mental health conditions out of specialty mental health services and into primary care. Such transitions are consistent with recovery oriented care, which posits that recovery exists on a continuum of improved health and wellness (Bishop et al., 2016; Fletcher et al., 2019; Sheedy & Whitter, 2009; Smith et al., 2019; Stengel et al., 2012). This continuum of recovery often helps define when to “step up” care for patients requiring more intensive mental health services (Harter et al., 2018; Heddaeus et al., 2018; van Straten et al., 2015), but less frequently defines when to “step down” care for patients who no longer need mental health specialty services (Fletcher et al., 2019; Smith et al., 2019). This “stepping down” is critical for improving access, because when stable patients transition to primary care, outpatient mental health providers have more capacity to meet the needs of new patients and those with high-acuity conditions (Fletcher et al., 2019; Smith et al., 2010, 2019).
Although there have been efforts to define best practices for transitioning patients from inpatient psychiatric care to outpatient mental health settings (Brodsky et al., 2018; Tyler et al., 2019), there are no widely established recommendations on transitioning patients from outpatient mental health settings to primary care. Despite this lack of best practices, some organizations have developed standardized workflows to support these transitions and have described their experiences in the research literature.
The purpose of this rapid literature review was to identify research articles that described efforts to transition patients from outpatient mental health services to primary care and to summarize the transition practices and implementation strategies deployed.
Methods
This rapid literature review was conducted as part of Kaiser Permanente Washington's Learning Health System program, which leverages research capabilities to accelerate evidence-based decision making. We conducted this review over a 3-month period to meet our timeline for developing recommendations on optimizing access to mental health services in Kaiser Permanente Washington. To meet this accelerated timeline, we deployed existing rapid review methods (Haby et al., 2016; Hartling et al., 2015) such as sequential (instead of independent) dual review and data abstraction.
Search protocol
We searched PsycINFO and CINAHL for empirical studies or reviews published between January 2000 and August 2019 that described practices for transitioning patients from outpatient mental health services to primary care. Search terms included terms and synonyms for (a) care transitions, (b) mental health, and (c) outpatient care providers. A list of search terms is in Table 1. After completing an initial round of abstract and full-text reviews, we searched PubMed using the same categories of search terms to ensure the comprehensiveness of our search strategy and identify additional articles of interest.
Database search terms.
Study selection
We included English-language articles of any study design that described processes for transitioning patients from outpatient mental health settings to primary care. We excluded studies focused exclusively on transitions out of inpatient mental health settings and studies in which patients were simply discharged and not transitioned to primary care. We limited our search to literature published in the year 2000 or later to prioritize findings most likely to be relevant to current practice. We did not assess risk of bias because of our accelerated timeline and because the purpose of this rapid review was to provide a narrative summary of existing transition practices, not to draw conclusions about the effectiveness of specific interventions.
We used Evidence for Policy and Practice Information-Reviewer (EPPI-Reviewer) version 4.0 (Thomas et al., 2010) to screen articles identified from the literature search. One reviewer (KM) screened all abstracts for eligibility, erring on the side of inclusion to ensure no potentially relevant articles were missed. All articles included in the abstract review phase were sent to a second reviewer (PB), who reviewed the full text of each article against our inclusion criteria. Questions and uncertainties were resolved through team discussion and consensus.
Data abstraction
One team member (PB) abstracted data from included articles and a second team member (KM) reviewed abstracted data for completeness and accuracy. Abstracted data included: study design, location, setting, population, conditions treated, sample size, provider type, and descriptions of transition practices and implementation strategies. We defined “transition practices” as processes and protocols that directly facilitated the transition of patients from mental health specialty services to primary care, and we defined “implementation strategies” as approaches and techniques that enhanced the adoption and sustainment of those transition practices (Kirchner et al., 2020).
The lead author (PB) reviewed abstracted data across articles and identified categories of transition practices and implementation strategies that appeared the most frequently, defined as those described in four or more articles. The lead author then drafted a definition for each of these categories of transition practices, along with examples from the included articles. Additional team members (KM, JR, KC, and EW) reviewed and provided feedback; the team further refined these definitions and categorizations through iterative discussions. For the implementation strategies, the team referred to categories of implementation strategies as described by Waltz et al. (Waltz et al., 2015).
Results
Rapid evidence search results
Literature searches resulted in a review of 3,761 abstracts and 29 full text articles (Figure 1). Of these, only 11 articles met our inclusion criteria (Berkowitz et al., 2018; Chang et al., 2014; Durbin et al., 2004, 2012; Fletcher et al., 2019; Hamilton-West et al., 2017; Jespersen et al., 2009; Koenig et al., 2013; Noseworthy et al., 2014; Röhricht et al., 2017; Smith et al., 2019). Characteristics of included articles are summarized in Table 2. The articles represent a range of study designs, including quality improvement (QI) projects or evaluations, qualitative studies, a literature review, a cross-sectional study, a nonrandomized difference-in-difference study, and a measure validation study. The projects took place in the United States, Canada, England, and Australia. Nearly all studies were based in some form of integrated health system, including four studies in the U.S. Department of Veterans Affairs (VA), two in the United Kingdom's National Health Service (NHS), two in Canadian provinces, one in an Australian state, and one within a U.S. community health partnership that later became an accountable care organization (ACO). Participants included patients, health care staff, and other stakeholders. Although we did not exclude child or adolescent populations, all included articles focused on adults. These patients had a range of mental health diagnoses, including depression, stress-related disorders, bipolar disorder, and schizophrenia. Despite this variety of diagnoses, most articles outlined specific criteria for identifying stable patients eligible for transition.

Literature flow diagram.
Characteristics of included studies.
Abbreviations: PC = primary care; PCP = primary care provider; OT = occupational therapist; QI = quality improvement; VA = U.S. Department of Veterans Affairs; NHS = National Health Service (United Kingdom); MFT = marriage and family therapist
Notes:
Patients were excluded if they: were taking antipsychotic medications, lithium, or valproic acid paired with a bipolar diagnosis; had started a new psychotropic medication in the prior 6 months; had a psychiatric emergency room visit and/or psychiatric inpatient hospitalization in the prior 12 months; or were deemed at high risk for suicide.
Not further specified.
Served as designated liaisons for patients transitioning from mental health services to primary care, as well as a single point of referral back to mental health services if needed. This role was staffed by community psychiatric nurses and occupational therapists.
Eligible patients were those who: had no acute psychiatric hospital admission in prior 12 months; posed no current risk to self or others; had well-established medication regimen and required little assistance with adherence; had settled accommodation and were able to meet their own basic living needs; and supported transition plan.
Psychiatric primary care liaison nurses had experience providing mental health care and served as designated liaisons to support patients transitioning from mental health services to primary care.
Peer support workers had personal experience transitioning from outpatient mental health services to primary care and provided emotional and practical support to patients undergoing transitions.
Eligible patients were those who: had no recent medication changes; had symptoms under control; had no recent change in social situation; were not experiencing a major life event; and supported transition plan
Liaison nurses worked with both mental health and primary care teams and served as designated liaisons to coordinate care for patients transitioning from mental health services to primary care.
Article does not report number of participants in included studies
Criteria revolved around stability of illness, treatment, and social circumstances; low risk to self and others; and belief among treating team that transition would benefit patient.
As a requisite for inclusion in this review, all articles described the transition of stable patients from outpatient mental health services to primary care. In reviewing these descriptions, we identified six commonly described transition practices and two commonly described implementation strategies, each of which appeared in four or more articles (Table 3). Descriptions and examples of these transition practices and implementation strategies are in Table 4 and summarized in the article text.
Categories of transition practices and implementation strategies described in included articles.
Summary and examples of categories of transition practices and implementation strategies.
Abbreviations: EHR = electronic health record.
Transition practices
All 11 articles included descriptions of at least two transition practices, defined as processes or protocols that directly facilitated patient transitions from mental health specialty services to primary care. Articles reported a median of four transition practices (range 2–6); identified categories of transition practices are summarized here.
Assessment of recovery and stability
Seven articles described the use of standardized processes and criteria to identify patients who would be most likely to benefit from transitioning to primary care. These criteria for stability included: no psychiatric hospital admission or emergency department visit in prior 12 months; no currently identified risks to self or others; no current prescription for antipsychotic medication; no recent medication changes; symptoms under control; stable housing situation; able to meet basic living needs; no recent major life events; and patient support for transition plan.
Patient engagement
The most commonly described transition practice was patient engagement, with 10 articles describing the use of patient-centered approaches to engage patients as active participants in their care. For example, six articles described how mental health providers partnered with patients and elicited their preferences in determining whether to move forward with a transition to primary care (Fletcher et al., 2019; Hamilton-West et al., 2017; Jespersen et al., 2009; Noseworthy et al., 2014; Röhricht et al., 2017; Smith et al., 2019). In addition, an NHS study in East London described a process of working with patients to help them identify their recovery goals and develop a recovery and transition plan based on those goals (Röhricht et al., 2017).
Shared treatment planning
The second-most described transition practice was shared treatment planning, with eight articles describing how primary care providers, mental health providers, patients, and families collaborated to create a single treatment plan. These shared treatment plans were tailored to each individual patient and might include mental health goals, physical health goals, family information, living arrangements, mental health triggers, and plans for preventing and responding to a relapse of mental health symptoms. Five articles described the use of multidisciplinary team meetings or joint case conferences to engage mental health and primary care providers in shared treatment planning (Berkowitz et al., 2018; Chang et al., 2014; Koenig et al., 2013; Noseworthy et al., 2014; Röhricht et al., 2017), and two articles recommended the use of overlapping visits in which the mental health provider accompanied patients to their first post-transition primary care visits (Koenig et al., 2013; Noseworthy et al., 2014).
Care coordination
Seven articles described a process of establishing clear roles and responsibilities to guide patients through the transition period and ensure continuity of care. This frequently involved the deployment of specific staff members (such as nurse care managers or community health workers) to assist patients with the transition process, coordinate care across multiple providers, address patients’ barriers to care, and facilitate connections to community resources. For example, an NHS project in Southeast England described the creation of a new role called “primary care mental health specialists” staffed by community psychiatric nurses and occupational therapists (Hamilton-West et al., 2017). Responsibilities of this role included facilitating communication across mental health and primary care providers, involving patients in care decisions, and serving as the patients’ primary point of contact when issues arose.
Follow-up and support
Seven articles described how mental health providers and/or primary care providers implemented formal follow-up procedures to monitor patients during and after the transition process. This often involved systematically checking in with patients about their transition experience and establishing clear protocols for how primary care and/or mental health providers would handle a reemergence of mental health symptoms. For example, a Baltimore community health partnership deployed nurses to conduct telephone follow-up with patients after their transition (Berkowitz et al., 2018), and the East London NHS project recruited peer support workers who had personal experience transitioning out of mental health services to provide patients with emotional and practical support during the transition process (Röhricht et al., 2017).
Medication management
Seven articles described the establishment of clear processes and responsibilities for the administration of psychiatric medications. For example, as part of the U.S. Department of Veterans Affairs (VA)'s FLOW initiative in South Texas, mental health providers ensured patients had at least a 6-month supply of psychiatric medications at the time of transition, and they instructed patients to contact their primary care provider at least 1 month prior to needing a refill (Fletcher et al., 2019; Smith et al., 2019). In the Southeast England NHS project, responsibilities of the newly created role of primary care mental health specialists included providing patient education on medication adherence (Hamilton-West et al., 2017).
Implementation strategies
Only six articles included descriptions of implementation strategies, defined as approaches or techniques that enhanced the adoption and sustainment of transition practices. Identified categories of implementation strategies are summarized below.
Train and educate stakeholders
In five articles, health systems provided mental health and primary care staff with opportunities to learn about the transition process and best practices for managing mental health conditions in primary care. Such educational opportunities included implementation strategies identified by Powell et al. (Powell et al., 2015), such as developing educational materials, distributing those materials, conducting educational meetings and providing ongoing training. For example, the East London NHS project provided psychopharmacology training for nurses and created a website with self-learning modules about managing mental health conditions in primary care (Röhricht et al., 2017).
Use evaluative and iterative strategies
Four articles mentioned efforts to involve staff and patients in continuously evaluating and improving the transition process. These process improvement activities included specific implementation strategies (Powell et al., 2015) such as auditing and providing feedback, conducting cyclical small tests of change, purposely reexamining the implementation, conducting local consensus discussions, capturing and sharing local knowledge, and staging implementation scale-up. For example, the VA's FLOW initiative in South Texas used internal and external facilitators to assess progress on the transition initiative, address barriers, and develop site-specific solutions to any challenges encountered (Smith et al., 2019).
Additional implementation strategies
The VA's FLOW initiative also described unique implementation strategies that were not described in any other included articles. These included an assessment of readiness to implement, the development of an implementation blueprint, the involvement of local opinion leaders, and the use of a blended facilitation model that combined external facilitators (who provided content and process expertise) and internal facilitators (who worked directly with staff and providers to address site-specific barriers) (Smith et al., 2019).
Reported outcomes
Six articles reported outcomes of their transition initiatives, which included health care spending (Berkowitz et al., 2018), provider and patient feedback (Hamilton-West et al., 2017; Noseworthy et al., 2014), providers’ caseload-related stress (Fletcher et al., 2019), and the percentage of patients who returned to specialty mental health care (which ranged from 2% [Fletcher et al., 2019; Smith et al., 2019] to 10% [Röhricht et al., 2017]). Two articles reported that 0.5% (Fletcher et al., 2019) or 1.0% (Röhricht et al., 2017) of transitioned patients returned to specialty mental health services after experiencing brief psychiatric crises or mental health-related hospitalizations. No other articles reported any post-transition patient safety events. Overall, the reported outcomes were favorable toward the transition initiative; however, our decision to forgo risk-of-bias assessment, as well as the descriptive nature of included studies, prevents us from drawing conclusions about the effectiveness of particular interventions.
Discussion
This rapid literature review summarizes common themes from 11 articles about practices for transitioning stable patients from outpatient mental health services to primary care. Most of these fell into six categories of transition practices, with patient engagement appearing the most frequently (10 articles), followed by shared treatment planning (eight articles), assessment of recovery and stability, care coordination, follow-up and support, and medication management (seven articles each). Less frequently, articles included descriptions of implementation strategies, with five articles describing efforts to train and educate stakeholders and four articles describing the use of evaluative and iterative strategies.
These findings suggest organizations are employing common practices to help patients transition from mental health specialty services to primary care, but there are opportunities for an increased focus on implementation strategies to enhance the adoption and sustainment of these transition practices. Only six articles included descriptions of implementation strategies, even though previous research has shown many of these strategies—including educational meetings, ongoing trainings, audit and feedback, and practice facilitation—have been associated with effective implementations of clinical program change (Goorts et al., 2021; Ivers et al., 2014; Singh et al., 2021; Thoele et al., 2021; Varsi et al., 2019).
Overall, we identified a profound need for more research on transitioning stable patients from outpatient mental health services to primary care. Only 11 articles described such transitions, and none of these articles evaluated the effectiveness of transition initiatives using experimental study designs or formal evaluations grounded in conceptual models (Glasgow et al., 1999; Nilsen, 2015; Proctor et al., 2011). Future studies could use frameworks such as Proctor et al.'s 2009 conceptual model of implementation research (Proctor et al., 2009) to explore the extent to which transition initiatives are associated with improvements in implementation outcomes such as uptake (e.g., number of patients who transitioned); acceptability (e.g., clinician and patient satisfaction with the transition process); or sustainability (e.g., how long the new transition process is maintained); as well as service outcomes such as safety (e.g., number of post-transition psychiatric hospitalizations) and client outcomes such as function (e.g., changes in mental health symptoms). Future studies also could provide insight on the effectiveness of transition initiatives in different types of health care systems and among patients with varying levels of mental health needs. Such research would advance the field and provide important insight for health care decision-makers seeking to implement effective mental health transition practices that fit their population and setting.
Limitations
This rapid review has several limitations. First, our review sought to summarize mental health transition practices and implementation strategies as described in the research literature. It was not designed to evaluate evidence of effectiveness; so, we are unable to draw conclusions about best practices or the relative importance of specific transition processes or strategies. Second, we are limited to summarizing the practices and strategies reported in our identified articles; therefore, we may be missing specific details or entire initiatives that have not been described in the research literature. Third, it is possible that we missed some relevant articles due to our rapid timeline for literature searching and study selection. Fourth, nearly all included studies took place in integrated health systems, meaning that many of the transition initiatives benefited from a unified leadership structure and shared electronic health record system that facilitated communication between mental health and primary care. Similar transition initiatives might be more difficult to implement in private mental health specialty practices with limited infrastructure and incentives to support these transitions. Moreover, some patients may wish to continue seeing their mental health providers even after symptoms improve. Achieving equitable, patient-centered care may require balancing the needs and preferences of existing patients with those of underserved populations seeking to access limited mental health specialty resources.
Conclusions and implications
This rapid literature review identified and described six categories of transition practices and two categories of implementation strategies that could support efforts to transition stable patients from outpatient mental health services to primary care. More research is needed to understand how specific interventions might affect patient outcomes, as well as the feasibility of implementing these interventions in heterogeneous health care settings. As of this writing, the coronavirus disease 2019 (COVID-19) pandemic is contributing to an increase in symptoms of anxiety, depression, and other mental health conditions across the U.S. (Czeisler et al., 2020), placing additional stress on an already overburdened mental health care system (Auerbach & Miller, 2020). Amid these challenging circumstances, we have an opportunity to pursue systems solutions to improve access to mental health care. Doing so will require thinking not just how to “step up” care for patients with more intensive needs, but also how to “step down” care to support patients on their journey toward mental health recovery and empowerment.
Plain language summary
Many patients face difficulty accessing mental health care when they need it. One way to improve access is to transition stable patients out of mental health specialty services and into primary care, which opens more availability in mental health specialty services. Several organizations are trying new practices and implementation strategies to support these transitions, and we conducted a rapid literature review to learn from their experiences. We searched for research articles describing practices for transitioning stable patients from outpatient mental health services to primary care, as well as implementation strategies that could improve the adoption and continued use of those practices. In 11 included articles, we identified six commonly described transition practices, with patient engagement appearing in 10 articles, followed by shared treatment planning (eight articles), assessment of recovery and stability, care coordination, follow-up and support, and medication management (seven articles each). Fewer articles included descriptions of implementation strategies, with five articles describing efforts to train and educate stakeholders and four articles describing the use of evaluative and iterative strategies. These findings suggest organizations are using similar practices to help patients transition from mental health specialty services to primary care, but there are opportunities to focus more on implementation strategies that could improve the adoption and continued use of these transition practices. More research is needed to better understand whether certain transition efforts are effective, as well as whether specific implementation strategies could improve the success of those transition efforts in different health care settings.
