In this issue of International Psychogeriatrics, Maria Lapid and colleagues present a review of the literature on end-of-life (EOL) care for individuals with schizophrenia (Lapid etal., Reference Lapid, Baruth, Ho and Mohammad2020a). In “End-of-life care in schizophrenia: a systematic review,” Lapid etal. delineate the ethically complex and inherently multidisciplinary nature of caring for individuals with serious mental illness (SMI) and serious medical illness at the EOL. The authors frame barriers to effective EOL care for individuals with schizophrenia through a variety of lenses including healthcare disparities, bioethics, and clinical palliative care needs. Lapid etal. provide a compelling review of an issue of public health importance; as they identify in their review, individuals with SMI experience excess mortality from serious medical illnesses, differential utilization of and access to health care services, and unique socioeconomic challenges at the EOL.

The disparities in care and clinical challenges identified by Lapid etal. result from longstanding siloing of physical and mental health delivery (Horvitz-Lennon etal., Reference Horvitz-Lennon, Kilbourne and Pincus2006). Such siloing has been demonstrated to affect health care almost universally. However, the impact is disproportionately significant among individuals with serious illnesses – both mental and physical. Because patients with serious medical illness experience such a high degree of behavioral health (BH) comorbidity (Garrido etal., Reference Garrido, Penrod and Prigerson2014; Schuyler, Reference Schuyler2010), siloing of mental and physical health in serious illness care (SIC) is especially problematic. This is particularly true with respect to the care of patients with SMI or dementia and comorbid life-limiting medical illness as such patients are likely to require intensive mental health and SIC longitudinally (Lloyd-Williams etal., Reference Lloyd-Williams, Abba and Crowther2014; Shalev etal., Reference Shalev, Fields and Shapiro2020a).

Given the deeply ingrained divisions between physical and mental health care, there is a need for models of BH and SIC. This is particularly true in palliative care – a framework of holistic care for individuals with chronic medical illnesses which is delivered by an interprofessional team across care settings and addresses multiple aspects of SIC including symptom management, serious illness communication, and caregiver support (Tatum and Mills, Reference Tatum and Mills2020).

Over the past several decades, integrated BH and physical health has been robustly modeled and implemented across a range of healthcare settings including primary care, geriatrics, and oncology (Ell etal., Reference Ell, Xie, Quon, Quinn, Dwight-Johnson and Lee2008; Franx etal., Reference Franx, Dixon, Wensing and Pincus2013; Horvitz-Lennon etal., Reference Horvitz-Lennon, Kilbourne and Pincus2006; Unutzer etal., Reference Unutzer2002). However, there is a significant gap in BH integration into palliative and EOL care. The dearth of such models is complicated by deficiencies in accountability and payment structures to fuel codelivery of BH and palliative care and by lack of workforce pipelines for clinicians and administrators able to work at the BH–palliative care interface. Many of the challenges addressed by Lapid etal. demonstrate the need for integrated BH–palliative care models and workforce.

Models of integrated BH and SIC would improve care for a multitude of patients. Lapid etal. demonstrate the need for such models vis-à-vis individuals with SMI, but geriatric patients with dementia would also benefit significantly from such models (Ingravallo etal., Reference Ingravallo, Mignani, Mariani, Ottoboni, Melon and Chattat2018). There are disparities in advance care planning for individuals with dementia, particularly among racial and ethnic minorities (Pettigrew etal., Reference Pettigrew2020). Such differences are fueled by misassumptions and lack of clinicians trained to provide palliative care for this population. For instance, clinicians may have a poor understanding of capacity assessment for advance care planning among older and/or cognitively impaired individuals. As with individuals with SMI, there has not been an operationalization of the role of capacity assessment in the advance care planning process for individuals at risk for dementia, and this may be another reflection of the BH–palliative care gap (Kiriaev etal., Reference Kiriaev, Chacko, Jurgens, Ramages, Malpas and Cheung2018).

Although there are not yet widely disseminated models of BH–palliative care codelivery, frameworks for such models exist. Cheung etal. present such a conceptual framework of BH–SIC integration generated through literature review, survey data, and semistructured interviews with program leaders. The Cheung model focuses on BH integration into five aspects of SIC: (1) person- and family-centered care, (2) clinical functions, (3) workforce, (4) clinical support structures, and (5) policies to enhance and incentivize effective integrated care (Cheung etal., Reference Cheung, Spaeth-Rublee, Shalev, Li, Docherty, Levenson and Pincus2019). Cheung etal. identify key elements of each of these aspects, as well as barriers to implementation and policy recommendations to promote integrated BH–SIC integration. Policy-oriented recommendations include alternative payment models that promote, rather than undermine, integration, and quality measures that integrate BH and SIC. Although the Cheung model provides a rich conceptual framework for understanding how a fully integrated model of care might function, it has yet to be applied in vivo to program development or evaluation.

Applied models of BH–SIC codelivery exist predominantly in the oncologic setting. Kelly Irwin and colleagues created The Bridge, a model grounded in collaborative care to facilitate codelivery of oncology and BH care for patients with both SMI and cancer (Irwin etal., Reference Irwin2019). Patients with SMI and new cancer diagnoses within one system were flagged and proactively enrolled in a 12-week program providing them with access to a cancer center psychiatrist and social work case manager who helped optimize engagement with treatment. The intervention team helped patients assess goals of care to inform person-centered treatment planning. The intervention team additionally communicated with the patient’s outpatient community mental health team and their oncology team to promote interdisciplinary collaboration and increase access to expert psycho-oncology. The social work case manager served as the primary contact for study participants, and the psychiatrist provided 24-hour pager coverage. After the 12-week intervention, patients were then transitioned to community care as indicated. The intervention team continued tracking patients for 12 additional weeks and would continue clinical follow-up if indicated. Qualitative results demonstrated good feasibility; most patients completed assessments at all timepoints, and both patients and oncologists found the intervention to be helpful. However, as with the existing conceptual and applied models, The Bridge depends on highly specialized psycho-oncologic workforce with appropriate training in both BH and SIC.

The creation and implementation of models of SIC that integrate BH require workforce that can navigate both domains effectively. Unfortunately, there are significant limitations in BH workforce that pose a challenge to the development of integrated models. Despite the need, only a few modest models of workforce development exist in the literature. The IMhPaCT project in Australia (Integrated Mental health and Palliative Care Task project) was a year-long project in Adelaide and rural South Australia in which palliative care and mental health teams caring for SMI patients did joint in-service education on mental health within palliative care (Taylor etal., Reference Taylor, Swetenham, Myhill, Picot, Glaetzer and van Loon2012). In addition, mental health clinicians completed home visits with palliative care teams to establish collaborative working relationships and teach mental health evaluation. A mental health learning package was produced for hospice nurses, and mental health-palliative care champions were established to advocate for increased collaboration. Unfortunately, the initiative was limited: no data were gathered, the team involved was small (36 palliative care professionals in the mental health workshops, none of whom were physicians), and the project was terminated prematurely because of funding deficits. Another model in a geriatric palliative care setting employed interprofessional geriatric psychiatry education models to help address perceived gaps in knowledge among unit staff and was well received (Grief etal., Reference Grief, Lowe and Campbell2013).

Promising data exist on the potential for e-learning and other novel educational modalities to disseminate BH–SIC knowledge. For instance, an international group of geriatricians and psychiatrists used a Delphi approach to create an adaptive e-learning module to teach international health care providers capacity, the global capacity education e-tool (Peisah etal., Reference Peisah, Lerman, Herrmann, Rezmovitz and Shulman2019). The tool was piloted with an international cohort of health care professionals and showed significant improvements in pre- to post-module test scores on capacity. Given that capacity is a significant topic at the BH–SIC intersection, the global capacity education e-tool may serve as a model for creating BH–SIC cross-training that can be easily disseminated and utilized by health care providers. E-learning is a particularly attractive paradigm for furthering cross-training given the dearth of experts at this intersection, the need for dissemination to a wide multidisciplinary audience, and the potential need for just-in-time training models.

On the physician level, psychiatrists receive very limited exposure to palliative care, despite significant interest among trainees (Forster etal., Reference Forster, Proskurin, Kelly, Lovell, Ilchef and Clayton2017; Irwin etal., Reference Irwin, Montross, Bhat, Nelesen and von Gunten2011). Even psychiatrists subspecializing in consultation–liaison (CL) psychiatry, which has significant overlap with palliative care, receive uneven training in palliative care (data on geriatric psychiatry training programs is forthcoming) (Shalev etal., Reference Shalev, Nash, Levenson and Shapiro2020b). Few psychiatrists pursue specialty training in palliative medicine, despite the opportunity to do so (“Workforce Statistics | AAHPM,” n.d.). Among SIC clinicians, physicians in specialties involved in SIC including hospice and palliative medicine, oncology, and cardiology have no BH training requirements (Shalev etal., Reference Shalev2019). Unfortunately, nonphysician professionals involved in serious illness and BH have similar deficits in cross-training opportunities. These training siloes not only preclude the development of clinicians with both BH and SIC skills but also restrict the opportunities to learn about the collaboration “across the aisle” between SIC and BH clinicians.

At this time, no BH–SIC cross-training has been widely disseminated or integrated into requirements for program accreditation or clinician certification. Unfortunately, siloed training leads to siloed care. To address the challenges and redress the disparities in EOL care for individuals with SMI as discussed by Lapid etal., new training paradigms must be developed. Psychiatrists need to be exposed to serious illness and palliative care. Although not all psychiatrists will become palliative care clinicians, all should have comfort working fluently with palliative care specialists and understand the basic framework for SIC delivery. Geriatric and CL psychiatrists have expertise working across medical and mental health systems and caring for patients with comorbid medical issues. To that end, CL and geriatric psychiatrists should assume a leadership role in making psychiatry more palliative care fluent and in providing opportunities for mental health cross-training of our colleagues practicing SIC.

Evidence shows that despite the lack of models and training paradigms, there is interest in further expansion of the psychiatry–palliative care interface. Scholars in both the psychiatric and the palliative care literature have called for increased interaction between these fields (Balon etal., Reference Balon, Motlova, Beresin, Coverdale, Louie and Roberts2015; Meier and Beresford, Reference Meier and Beresford2010). The urgency of these calls has increased markedly in the era of coronavirus disease 2019 (COVID-19), which has challenged both BH and palliative care in unprecedented ways (Lapid etal., Reference Lapid, Baruth, Ho and Mohammad2020b). In their recent commentary in International Psychogeriatrics on geriatric EOL care in COVID-19, Maria Lapid etal. identify a multitude of complex challenges: social stigma and ageism, an unprepared workforce, and a dearth of advance care planning among others. They lay out recommendations to address these challenges including innovative care delivery models and workforce development. Many of the challenges and recommendations identified strongly correlate with those among individuals with SMI.

While the mental health gap in SIC clearly affects patients with SMI, it reflects deficits that extend more broadly to geriatric patients, individuals with dementia, and individuals affected by COVID-19. This gap can be bridged to the benefit of both clinicians and patients. Through the development of national curricula, the adjustment of training program accreditation requires SIC exposure for mental health trainees and viceversa, and the creation of opportunities for collaborative learning between disciplines, the care of patients with any mental health needs at the EOL can be improved and the disparities identified by Lapid etal. can be redressed.