Characteristics of Included Sources
After duplicates were removed, the database searches yielded 593 unique sources of evidence. 530 studies were identified as irrelevant based on screening of their title and abstract, and the 63 remaining full texts were assessed for eligibility. Hand-searching of reference lists of all sources that met eligibility criteria and relevant review papers resulted in an additional 23 sources to screen. Thus, in total 86 full texts were assessed for eligibility. 66 sources were excluded at this stage. Exclusion reasons were wrong outcomes (n = 32), wrong patient population (n = 9), wrong study design (n = 8), wrong setting (n = 7), conference abstract (n = 7), and paediatric population (n = 3). Data charting was completed for the remaining 20 sources that met the eligibility criteria. See Fig. 1 for the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow diagram.
Of the 20 included sources, the majority were quantitative research studies (n = 11) or mixed methods studies (n = 6). Additionally, two sources were qualitative studies and one source described a program without reporting on research findings. For sources reporting on research (n = 19), most studies included participants from multiple stakeholder groups (for example, patients, PCPs, clinicians from SCSs, and others; n = 8). Many studies used patients’ charts as data sources (n = 8) and others collected data only from patients directly (n = 3). The countries of origin of the research studies and programs included in this review were Australia (n = 7), the United Kingdom (n = 6), New Zealand (n = 3), the Netherlands (n = 2), Ireland (n = 1), and Canada (n = 1). Due to the heterogeneity in healthcare systems in different countries and differences in the clarity of reporting settings across the sources, it was difficult to group and quantify the SCSs reported. SCSs included Community Mental Health Centres, Community Mental Health Teams, Community Mental Health Services, Public Mental Health Services, specialist mental health services, public psychiatric services, early intervention for psychosis services, and other settings not well defined. Although addiction settings were considered in this scoping review, no sources focused on addiction settings met all the eligibility criteria for inclusion.
Barriers and Facilitators to Discharge
Study extraction data were coded within the broader categories of barriers and facilitators to discharge. Individual codes were then subcategorized into patient-related, primary care-related, and process/systems-related barriers or facilitators to discharge (see Table I). See Appendix D for the full coding table for this research question. Some data associated with discharge to primary care from the included evidence sources was not categorized as barriers or facilitators to discharge because of contradictory information across sources. For example, Filia and colleagues (24) reported that those with longer illness duration and longer time taking clozapine were more likely to be discharged to primary care than to private psychiatry. However, Jespersen and colleagues (25) reported that less chronicity was associated with discharge, and Ramanuj and colleagues (26) reported less time spent with secondary care was associated with discharge. Furthermore, Filia et al. (24) and Jespersen et al. (25) both found that those with fewer contacts with SCSs were more likely to be discharged. Thus, it was decided that within the context of this scoping review, which aims to map evidence rather than interpret it, duration of SMI and time spent within a SCS could not be categorized as barriers or facilitators. Additional miscellaneous factors associated with discharge (female sex, diagnoses of less high prevalence disorders, and fewer family contacts (25)) were also not categorized.
Patient-Related Barriers
Regarding barriers, patient-related codes included patient stability, care needs, socioeconomic status, engagement with treatment, and readiness for discharge. Patient stability was noted in three sources as recent onset of symptoms (27), a history of high-risk events (28), and recent use of acute crisis care (26). Care needs were reported in five sources and included a high need for SCSs (29, 30), high-risk symptoms (27), high symptom load (24), psychosocial impairment (24), high substance use (24), and need for medications (26). Socioeconomic status factors, reported in three sources, included having a limited support network (27), experiencing homelessness (31), and difficulties paying for primary care services in private or semi-private models of care (32). Issues related to engagement with treatment posing barriers to discharge were reported in four sources as low motivation to engage in treatment (27), low medication and/or treatment compliance (24, 29, 30), and having a Community Treatment Order (CTO; (24)). Finally, factors related to readiness for discharge were reported in two sources as having an unexpected or abrupt discharge (33), feeling passed on by the SCS (33), and concerns about losing contact with a psychiatrist (11, 33).
Primary Care-Related Barriers
-
Primary care-related codes for barriers were accessibility and care context-related factors and PCP ability to meet patient needs. Accessibility and care context-related factors were reported in four sources and included factors such as patients having a low level of personal organizational skills (29), primary care having less patient accountability (29), patient fears about unfamiliarity and stressors in the physical environment of primary care (11), patient and PCP concerns about time constraints in primary care (11, 30), and patient preference to remain in secondary care (32). Factors related to PCP ability to meet patient needs were reported in four sources and included the factors of patient concerns about quality of psychiatric care from PCPs (11, 32), managing medication complexity (27, 29), and the need to establish a therapeutic relationship (29).
Process/Systems-Related Barriers
Process/systems-related codes included quality of communication and support across care settings and work and time required to facilitate discharge. Communication and support factors, reported in three studies, included poor communication between secondary and primary care (30, 33), lack of support from and/or between PCPs and SCSs (29, 30), and a lack of information about the transition process (29). The work and time required to facilitate discharge was noted in one source (29) as time required for the transition process and the amount of paperwork required.
Patient-Related Facilitators
Codes for patient-related facilitators included stability, strengths, and readiness for discharge. Stability factors as facilitators were reported in six sources as general stability (27, 28), functional remission of SMI or high overall functioning (25–27, 34), having employment (25), having fewer medical conditions (25), having fewer medication needs (26), having less psychosocial stress (25), having an absence of substance abuse (29), having less time spent in or fewer encounters with acute care (25, 26), and not having or had a CTO (25). Patient strengths were reported in four sources as having a strong support system (27, 34), high motivation (27, 34), skills (27, 34), medication compliance (29), good cognitive function (28), insight into their SMI (28, 34), and the ability to attend appointments and blood tests independently (29). Readiness for discharge factors, reported in five sources, included feeling prepared for discharge (33), being aware of and expecting discharge (11, 33), approving of the discharge (29, 34), having faith in the transfer of care (34), viewing primary care as less stigmatizing (32), and not feeling pressured to be discharged (11).
Primary Care-Related Facilitators
Primary care-related codes for facilitators were accessibility and care context related-factors and PCP ability to meet mental health care needs. Accessibility and care context related-factors, reported in three sources, included patient preference for primary care over SCSs (32), the accessibility and convenience of primary care for patients (11, 29, 32), patients’ physical access to a pharmacy (29), and patient ability to financially access primary care services in semi-private or private healthcare systems (32). Factors related to PCP ability to meet mental healthcare needs were reported in four sources as patients’ belief in PCPs’ ability to meet their needs (32), PCPs’ ability to recognize when additional support is needed (33), the ability for patients to receive coaching, mental health monitoring, and consistent contact within primary care (35), having an established strong and trusting relationship between the patient and PCP (33), and the primary care office having a welcoming environment (11).
Process/Systems-Related Facilitators
Codes for process/systems-related facilitators were the discharge planning process and communication and support across services. Four sources reported factors related to the discharge planning process, including having a collaborative and planned process involving an interdisciplinary team (8, 9), including PCPs in the discharge planning process (9), including patients in discharge planning and care plan development (11, 33), having a personalized and flexible discharge process (33), having transparency in the discharge process (11), and recognizing patients’ self-management ability when planning discharge (33). Seven sources reported factors related to communication and support across services including having ongoing communication and support between primary care, SCSs, and patients (7, 9, 30, 33, 35), facilitated re-entry or access to SCSs when needed (29, 33), and primary care and SCS healthcare providers having faith in transfers of care (34).
Table I Barriers and Facilitators of Discharge
Barrier
|
Sources (N = 20)
|
Patient-related
|
|
Patient stability
|
3
|
Care needs
|
5
|
Socioeconomic status
|
3
|
Engagement with treatment
|
4
|
Readiness for discharge
|
2
|
Primary care-related
|
|
Accessibility and care context related-factors
|
4
|
PCP ability to meet patient needs
|
4
|
Process/systems-related
|
|
Quality of communication and support across care settings
|
3
|
Work and time required to facilitate discharge
|
1
|
Facilitators
|
Sources (N = 20)
|
Patient-related
|
|
Stability
|
6
|
Patient strengths
|
4
|
Readiness for discharge
|
5
|
Primary care-related
|
|
Accessibility and care context-related factors
|
3
|
PCP ability to meet mental health care needs
|
4
|
Process/systems-related
|
|
Discharge planning process
|
4
|
Communication and support across services
|
7
|
Facilitated Discharge Programs
11 facilitated discharge programs were reported across 12 of the evidence sources: the Transition into Primary Care Psychiatry (TIPP) clinical model (36), the Consultation and Liaison in Primary Care Psychiatry (CLIPP) model (25, 37), the Recovery and Enablement Track (10), the Enhanced Primary Care (EPC) Pathway (38), a primary care liaison service (25), the Wellington Mental Health Liaison Service (11), a modified shared care protocol (9), the Primary Care Mental Health Specialist (PCMHS) Service (7), a shared care model for clozapine (24, 29), the PARTNERS (develoPing integrAted primaRy care for paTieNts with sERiouS mental illness) program (35), and a planned discharge process (8). Of the programs, five were developed and employed in Australia (8, 9, 24, 25, 29, 37), four in the United Kingdom (7, 10, 35, 38), one in Canada (36), and one in New Zealand (11).
For this research question, the major components of discharge programs were ascribed codes (see Appendix E for coding table). Figure 2 demonstrates the frequency of common components among the 11 discharge programs. A new professional role supporting SCSs, PCPs, and patients through the transition of patient care from SCSs to PCPs was identified in seven programs across nine evidence sources (8, 9, 24, 25, 29, 35–38). The development and provision of a detailed care plan, including relapse signatures and relapse interventions plans, was integrated into eight programs (7–9, 11, 25, 35–38). Likewise, provision of mental health supports distinct from PCPs was identified as a component in eight programs (7, 8, 10, 24, 25, 29, 35, 36, 38). Consultations with psychiatrists were available within seven programs (9, 24, 25, 29, 35–38). Facilitated re-entry into a SCS was a component of six programs (7, 9, 10, 24, 25, 29, 38). Six programs also included discharge planning meetings with various relevant stakeholders (8, 9, 11, 25, 36, 37). Lastly, two programs provided formal education/training programs to PCPs (11, 38). Primarily, transition programs were delivered across the boundaries of secondary and primary care services; however, one program was delivered in the context of a SCS, focusing on patient readiness for transfer of care (10), and another was delivered in the context of a SCS to improve processes related to patient transfer (8).