Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Mental Health Service Systems and Clinical Care Pathways

Abstract. Mental health service systems shape the movement from recognized psychological need to assessment, treatment, crisis response, and sustained clinical support. This fundamental research monograph investigates transitions among primary care, specialist outpatient services, emergency departments, inpatient psychiatric settings, and community programs. Its central questions concern access, the reliability of referrals, informational and relational continuity, the allocation of clinical responsibility, patient safety after discharge, and the conditions under which organized care improves outcomes. A critical narrative synthesis integrates foundational access and quality frameworks, cross-cultural pathways research, randomized trials of collaborative depression care and early psychosis services, cohort studies and meta-analyses of postdischarge risk, implementation theory, and World Health Organization and NICE guidance. The analysis distinguishes causal evidence about coordinated intervention packages from observational associations and process measures; it evaluates the limits of generalization across age, diagnosis, health systems, and socioeconomic circumstances. The resulting account treats mental health pathways as time-dependent, multidimensional arrangements whose effectiveness depends on clinical appropriateness, reliable handoffs, meaningful patient participation, institutional accountability, and equitable access. The monograph also identifies methodological priorities for registry linkage, comparative causal evaluation, patient-reported continuity, and digitally mediated or human–artificial clinical coordination. This is an evidence-informed narrative monograph; it reports no new empirical data or independent systematic review.

Keywords: mental health service systems; clinical care pathways; psychiatric services; transitions of care; continuity of care; primary care; collaborative care; specialty mental health care; emergency psychiatry; psychiatric discharge; postdischarge follow-up; suicide prevention systems; early psychosis; service equity; care coordination; patient-centered access; patient safety; implementation science; health services research; human–artificial clinical systems

Introduction: The Clinical Pathway as a Research Object

Mental health service systems organize how psychological distress is recognized, assessed, treated, monitored, and supported across institutions and over time. Their central unit of analysis is the care pathway: the empirically observable sequence of contacts, decisions, transfers, interruptions, and relationships through which a person encounters mental health care. A pathway can begin in primary care, an emergency department, a school, a community organization, a specialist clinic, or an inpatient unit; subsequent movement depends on referral procedures, clinical need, availability, eligibility, geography, and the person's preferences. System organization can therefore alter the consequences of treatment even when the content of an individual clinical intervention is held approximately constant. This monograph examines those organizational transitions as a distinct subject within clinical psychology and psychopathology, informed by public health, health services research, implementation science, and patient-safety research. World Health Organization (2025) Levesque et al. (2013)

The primary research question is how a service system converts a recognized mental health need into sustained, appropriate, coordinated care. Answering it requires separate inquiry into the architecture of services, the transitions joining them, the temporal continuity of clinical work, the transfer of clinical information, and outcomes experienced by patients and families. A consultation, a referral, a completed appointment, and a durable therapeutic relationship are different events. Likewise, a discharge document is a record of transition preparation, whereas effective follow-through depends on whether receiving services actually accept responsibility and establish contact. Research on psychiatric pathways must therefore reconstruct both the formal organizational design and the realized trajectory of each person. Administrative records, patient-reported experience, clinical outcomes, and qualitative observation capture different parts of that trajectory. Haggerty et al. (2003) Donabedian (1988)

The evidence considered here includes foundational models of quality and access; comparative pathways research; randomized trials of collaborative and coordinated care; observational studies of outpatient follow-up after hospitalization; syntheses of discharge risk; and contemporary organizational guidance. The method is a critical narrative synthesis organized around explanatory mechanisms and research designs. It is neither a newly conducted clinical experiment nor a protocol-registered systematic review. Particular weight is assigned to studies that specify their comparator, outcome, timing, and population. Where an association is reported from a service registry, causal interpretation is explicitly limited by confounding, selection, and incomplete observation. Where a randomized intervention bundles several service components, the trial can establish the effect of the package more securely than the contribution of any single component. Chung et al. (2017) Stanley et al. (2018) Archer et al. (2012)

The subject has practical significance for suicide prevention, treatment engagement, early psychosis, depressive disorders, serious mental illness, pediatric transitions, physical-mental comorbidity, and recovery after crisis. Yet no universal pathway can be derived from diagnosis alone. Capacity, workforce, laws, cultural models of distress, social resources, insurance rules, and the distribution of specialist expertise condition how an intervention can be delivered. A global analysis must avoid treating high-income service infrastructure as the default form. The World Health Organization's 2024 Mental Health Atlas, published in 2025, offers a comparative institutional picture drawing on reports from 144 countries; its country-level descriptive indicators are indispensable for context but cannot by themselves determine which local pathway causes better clinical outcomes. World Health Organization (2025) World Health Organization (2022)

Defining Systems, Episodes, Pathways, and Transitions

A mental health service system is a set of organizations, professionals, financing mechanisms, referral rules, records, community supports, and governance structures that jointly make care available. A clinical care pathway is the ordered but revisable trajectory through those arrangements associated with an identified need or episode. The term pathway is used descriptively when researchers reconstruct actual movements and prescriptively when organizations design a desired sequence of assessments, decisions, and contacts. Confusing these meanings causes a familiar measurement problem: documenting that a pathway protocol exists says little about whether an individual traversed it. The distinction also matters when services are fragmented among independent agencies that share no records, referral agreements, or accountability for handoffs. Donabedian (1988) Levesque et al. (2013)

The pathway can be represented as a time-indexed sequence of states: first concern, recognition, help-seeking, initial contact, assessment, referral, receiving-service acceptance, intervention, reassessment, escalation or step-down, and sustained follow-up. Each transition has a waiting time, a probability of completion, a potential for information loss, and a distribution of responsibility. A patient may cycle repeatedly through emergency assessment, inpatient stabilization, and community contacts, or remain entirely within primary care. A pathway can also terminate involuntarily through loss of insurance, geographical displacement, stigma, refusal based on eligibility, or disengagement following coercive encounters. The analytically important event is thus often a missing link: a specialist appointment never scheduled, a safety plan never transferred, or a new clinician unable to access the prior assessment. Gater et al. (1991) Kripalani et al. (2007)

Several neighboring objects require clear boundaries. The psychological efficacy of a therapy belongs primarily to treatment research; service systems research addresses whether and how that therapy reaches patients, is coordinated with other care, and continues when responsibility changes. Community mental health ecology emphasizes neighborhoods, informal supports, housing, participation, and social institutions; the present dossier concentrates on formal and hybrid clinical transfers that connect hospital, emergency, specialist, and primary-care services. General implementation science explains how innovations are adopted, while a pathway inquiry examines the clinically consequential arrangement of those innovations in space and time. These fields overlap empirically and must exchange findings without losing their separate explanatory questions. Aarons et al. (2011) World Health Organization (2021)

An important distinction separates the availability of a service from its accessibility, continuity, clinical appropriateness, acceptability, and experienced benefit. A district can have a psychiatric clinic yet lack an effective referral route from primary care. A discharge plan can specify follow-up yet fail because no appointment capacity exists. A person can attend appointments without receiving coordinated medication monitoring or crisis support. Conversely, a person may obtain substantial support through trusted nonspecialist settings and require specialist escalation only episodically. Evaluating system performance therefore requires multidimensional indicators and an explicit account of the patient population and baseline need. Levesque et al. (2013) World Health Organization (2025)

Historical Development of Mental Health Service Organization

The institutional history of mental health care involved transformations in the sites and authority through which distress was interpreted. Hospital-centered psychiatry, asylum systems, general-practice recognition, community mental health programs, and multidisciplinary networks emerged at different moments in different countries. These forms often coexisted, and their histories cannot be reduced to a single transition from institutional treatment to community care. The material consequences of deinstitutionalization depended on the availability of housing, outpatient clinicians, social welfare provision, and mechanisms for sustained support. International guidance now favors community-based, person-centered services that respect rights and support personal agency, while acknowledging differences in existing infrastructure and resources. World Health Organization (2021) World Health Organization (2022)

Comparative empirical work made the pathway itself visible. Gater and colleagues' multicenter study of 1,554 newly referred patients in 11 countries examined how people reached psychiatric services and what was associated with delays. In better-resourced centers general practitioners frequently mediated entry; other centers displayed more varied routes, including traditional healers. Somatic complaints appeared commonly and could shape referral. The study demonstrated that psychiatric access is structured by social and institutional intermediaries as well as by symptoms. It was an observational comparison, not a randomized test of cultural or professional referral models, so the reported cross-center differences cannot be read as causal effects of one entry system. Gater et al. (1991)

The modern services-research tradition also borrowed principles from quality improvement and general medical systems. Donabedian's distinction among structure, process, and outcome provides a durable analytic vocabulary: workforce and clinic capacity constitute structure; referral decisions, communication, and visits constitute process; patient function, harm, symptom change, and experience constitute outcomes. Subsequent continuity research clarified that receiving care over an extended period involves the persistence of relationships, coherent management, and useful information across contacts. These foundations expanded the meaning of quality beyond facility count or number of appointments and made discontinuity experimentally and administratively investigable. Donabedian (1988) Haggerty et al. (2003)

The development of collaborative care in primary care supplied a particularly influential experimental response to fragmentation. The IMPACT trial evaluated a structured depression care program among older primary-care patients, with trained care management and psychiatric supervision integrated into ordinary settings. It illustrated how organizational changes could deliver evidence-based treatment through nonpsychiatric entry points. Later international trials examined whether nonspecialist counselors could extend mental health care in resource-constrained settings. These traditions belong to the same history of service-system redesign, though their staffing, financing, diagnoses, and outcome measures differ enough to resist a simplistic global model. Unützer et al. (2002) Patel et al. (2010) World Health Organization (2025)

Help-Seeking, Recognition, and Entry into Care

The first transition begins before a medical encounter. People interpret distress through family relationships, social norms, religion, cultural idioms, bodily symptoms, workplace conditions, and prior experiences of health services. Recognition of a possible mental health problem does not automatically produce a request for specialized treatment; personal preference, treatment expectations, stigma, cost, and geographic accessibility can all intervene. Comparative pathway studies demonstrate that referrals are mediated through different institutions, including general medical care and community actors. An adequate access model must therefore follow the person from perceived need through the ability to seek, reach, pay for, and engage with care, rather than treating the referral as the natural beginning. Gater et al. (1991) Levesque et al. (2013)

Levesque, Harris, and Russell's patient-centered access framework locates access at the interface between characteristics of health systems and abilities of populations. Approachability, acceptability, availability and accommodation, affordability, and appropriateness interact with the abilities to perceive, seek, reach, pay, and engage. For mental health research this framework changes the interpretation of waiting-list data. A short recorded wait can coexist with high unmet need when people never enter the queue; an apparent failure to engage can reflect transport cost, language exclusion, an inaccessible booking system, or a previous coercive clinical encounter. Pathway evaluation should document these mechanisms instead of explaining every missed visit as an individual motivational deficit. Levesque et al. (2013)

Large international survey programs establish that contact with treatment varies considerably across disorders and countries. The World Mental Health Surveys analyzed services for anxiety, mood, and substance disorders in 17 countries, reporting substantial variation in service receipt and sector of contact. Such surveys typically capture recalled use and self-reported symptoms rather than a complete dated sequence of medical events, and they may exclude people outside household sampling frames. Their strength lies in showing disparities between estimated need and treatment contact; linked clinical records are better suited to testing the timing and reliability of specific referrals after an index encounter. The two evidence types answer complementary questions and should not be combined as though they share a denominator. Wang et al. (2007) World Health Organization (2025)

Front-door design includes mental health recognition within primary care, general hospitals, emergency departments, school and youth services, and specialist access points. Each can improve reach for one population while imposing different barriers on another. A primary-care route can normalize discussion of distress but may rely on brief consultations and local referral capacity. An emergency department can provide urgent stabilization but has limited capacity to guarantee subsequent relational continuity. Specialist self-referral can shorten some routes while disadvantaging people unfamiliar with diagnostic language or referral procedures. Effective design makes the route intelligible, defines escalation thresholds and receiving responsibility, and preserves options for people whose initial point of contact proves unsuitable. World Health Organization (2021) NICE (2016)

Structure, Process, Outcome, and the Logic of Evaluation

Donabedian's structure-process-outcome model separates prerequisites, care activities, and consequences. In a mental health system, structure includes the density and distribution of professionals, community teams, crisis units, interoperable records, transport availability, and funding arrangements. Process includes assessment, communication, medication reconciliation, appointment booking, joint review, and follow-up contact. Outcomes include changes in symptoms, quality of life, functional participation, adverse events, trust, and preventable service disruption. The causal ordering from structure to process to outcome is a hypothesis that must be examined, not a guarantee: a well-staffed service may perform poorly, and favorable outcomes may reflect baseline case mix or community resources rather than service quality. Donabedian (1988) Proctor et al. (2011)

A reliable pathway has both a clinical logic and a coordination logic. Clinical logic specifies which problems warrant particular assessments or interventions, by whom, with what monitoring. Coordination logic specifies who receives a referral, who can see the assessment, who responds if an appointment is missed, and who owns responsibility during the interval between organizations. A referral request without confirmation of acceptance represents an open handoff; a discharge letter without a reachable receiving practitioner represents incomplete transfer. Organizational evaluation must make these handoff states visible while respecting confidentiality and the person's choices. Guidance on inpatient-to-community transition emphasizes early planning, shared decisions, and timely communication among inpatient teams and community providers. NICE (2016)

The performance of a pathway cannot be summarized responsibly by a single utilization measure. Hospital admissions may be beneficial when appropriately used during a severe crisis; lower admission rates do not inherently imply better care. Similarly, reduced emergency contacts may indicate successful support or, alternatively, barriers to urgent access. Frequent outpatient visits may represent high-quality continuity or ineffective repetitive assessment. Readmission is influenced by illness severity, housing, family resources, and local bed policy. Evaluations should prespecify patient-centered endpoints, record adverse events, and examine utilization as one component of a broader outcome profile. World Health Organization (2022) Chung et al. (2017)

The temporal dimension creates another problem: a system may improve contact at seven days while failing to sustain support at three or twelve months. Survival analysis, episode-based registries, repeated patient-reported outcome measures, and qualitative follow-up reveal different aspects of durability. Studies also need to examine whether pathway benefits are concentrated among people already easy to reach, widening disparities for people with unstable housing, limited language access, or complex comorbidity. Implementation outcomes such as acceptability, adoption, fidelity, and sustainability should be studied alongside clinical results, yet the two domains must remain distinct in analysis. Proctor et al. (2011) Aarons et al. (2011)

Relational, Informational, and Management Continuity

Continuity describes the coherence of care as it is experienced across encounters. Haggerty and colleagues differentiated informational continuity, management continuity, and relational continuity. Informational continuity uses prior history and circumstances to make present decisions appropriately informed. Management continuity requires timely, complementary services arranged around changing needs. Relational continuity concerns an ongoing therapeutic relationship and the accumulated understanding between clinician and patient. The dimensions can be separated empirically: electronic transfer may improve information yet leave a person without a clinician they trust; a trusted clinician may provide relational security while lacking access to specialized services or up-to-date discharge records. Haggerty et al. (2003)

A pathway may preserve continuity through a stable named clinician, multidisciplinary team, care manager, shared plan, or explicit cross-agency agreement. The appropriate arrangement depends on clinical intensity, geography, and the person's preferred relationships. Team-based models can protect access when one practitioner is unavailable, but shared responsibility must not erase accountability. Conversely, an exclusive reliance on a single clinician may create vulnerability during leave, turnover, or relocation. The service-system challenge is to specify how the relationship, information, and coordinated plan persist when a particular provider changes. Models of collaborative care make continuity an organized sequence of measurement, review, and stepped intervention rather than merely a repeated office visit. Haggerty et al. (2003) Unützer et al. (2002)

Information continuity introduces privacy and epistemic issues. Clinical records may hold diagnoses, medication histories, observed behavior, and risk assessments generated under different circumstances; those entries require contextual interpretation rather than uncritical copying. A label recorded during acute distress can influence later encounters and potentially narrow a clinician's understanding of recovery. Successful transfer requires enough accurate information for safe decisions, correction mechanisms when records are wrong, and intelligible explanations of what will be shared with whom. General hospital studies identify substantial gaps in communication between inpatient and primary-care physicians, but their conclusions must be applied to psychiatric services with attention to different laws and clinical practices. Kripalani et al. (2007) NICE (2016)

Management continuity involves an active commitment to re-evaluate the plan as needs change. Step-up from primary care to specialist treatment requires clear eligibility criteria, a receiving service, and feedback to the referring clinician. Step-down after crisis demands a community response that can manage medication, psychosocial needs, and renewed risk without forcing the person to restart assessment from the beginning. The practical endpoint of a handoff is not delivery of a message; it is confirmation that the next responsible care process has begun. Research should identify the precise event counted as a completed transition and incorporate patient testimony where the administrative record falsely implies closure. Levesque et al. (2013) NICE (2016)

Primary Care, Collaborative Care, and Stepped Treatment

Primary care occupies an unusual position in mental health systems because it is simultaneously a point of first recognition, a provider of continuing physical health care, and a gateway to specialist consultation. Collaborative care responds to this conjunction through designated care managers, systematic symptom monitoring, clinical supervision, and planned treatment adjustments. Its defining contribution is organizational: access to specialist expertise can be distributed across a primary-care panel without requiring each patient to obtain every intervention through a conventional psychiatric clinic. Such arrangements depend on clear referral thresholds, defined roles, a registry or equivalent tracking method, and a mechanism for identifying nonresponse over time. An office that simply refers patients elsewhere lacks the longitudinal feedback architecture central to collaborative care. Unützer et al. (2002) Archer et al. (2012)

The IMPACT randomized trial recruited 1,801 older adults with depression from primary-care practices and compared collaborative depression care with usual care. At twelve months, 45% of the intervention group achieved a reduction in depressive symptoms of at least 50%, compared with 19% receiving usual care. This trial provides causal evidence for a structured treatment package within the studied population and health-care setting, including sustained follow-up and psychiatric consultation. It does not isolate which component accounts for the effect, nor does it imply that every diagnosis benefits equally from the same staffing model. Subsequent Cochrane evidence examining collaborative care for depression and anxiety supports the broader importance of organized multiprofessional treatment, with variations in study quality and implementation. Unützer et al. (2002) Archer et al. (2012)

A related trial studied patients with depression and chronic medical illnesses using a coordinated intervention that jointly addressed depressive symptoms and medical risk factors. It illustrates why segmentation between mental and physical health services may fail when conditions interact and the same patient must navigate incompatible follow-up schedules. Yet effectiveness in a selected research cohort does not automatically establish feasibility in rural, uninsured, or extremely understaffed contexts. International research includes models built around trained nonspecialist counselors and supervision; the MANAS cluster-randomized trial in Goa demonstrated the feasibility of evaluating such arrangements within primary-care systems. A pathway must be designed with adequate training, accountability, and escalation to professional expertise, rather than assuming task delegation alone ensures competent treatment. Katon et al. (2010) Patel et al. (2010)

Stepped care introduces a temporal decision rule: begin with an intervention matched to current need, monitor progress, and change intensity when the observed response or risk warrants. Its service-system value depends on the speed and reliability of the step, the existence of a higher-intensity receiving service, and the patient's consent and preferences. Measurement may improve detection of nonresponse without enabling action if specialist capacity is absent. Likewise, repeated questionnaires do not replace clinical judgment when safety, unusual symptoms, or comorbid substance use change the meaning of a score. The central evaluative question is whether decision thresholds, actual availability, and clinically meaningful follow-through operate together for patients with diverse needs. Unützer et al. (2002) Levesque et al. (2013)

Specialist Referral, Triage, and Shared Responsibility

Specialist services may be organized by diagnosis, age, intensity, professional discipline, geographic catchment, or crisis function. These divisions can concentrate expertise while multiplying transitions. A referral is a request for review by another unit; acceptance, assessment, and an actionable plan are separate events that may occur weeks apart. The reliability of a referral system depends on transparent criteria, consistent communication, and a named process for responding when the first destination is inappropriate or full. Patients who are repeatedly redirected between services may experience long periods in which clinical responsibility is ambiguous. A complete pathway study measures this interval instead of categorizing it merely as administrative delay. Levesque et al. (2013) Haggerty et al. (2003)

Triage typically prioritizes acuity, functional impairment, complexity, anticipated benefit, and safety considerations, but its rules can produce systematic inequities when expressed through narrow diagnostic categories or incomplete assessments. An individual may present with somatic complaints, fluctuating symptoms, or a crisis related to housing insecurity, making a single referral label insufficient. The comparative research on routes to psychiatric care provides early evidence that the service an individual reaches reflects local professional networks and modes of symptom interpretation. Triage evaluation therefore requires assessment of false rejection, delayed recognition, diagnostic reclassification, and the trajectory of people declined by a specialist clinic. A well-documented accepted referral is only one part of the system's performance distribution. Gater et al. (1991) Wang et al. (2007)

Shared-care arrangements divide tasks among a referring general practitioner, mental health clinician, pharmacist, and sometimes community support personnel. Such arrangements can preserve familiar relationships and specialist access if they specify who monitors symptoms, medicines, physical health, and renewed risk. Failure arises when each professional assumes that another will contact the person or interpret a changed clinical observation. Clear documentation should therefore identify responsibilities and next decision points, with authorized information exchange and routes for the patient to raise concerns. These are organizational principles rather than an individual therapeutic prescription, and their local enactment depends on professional scope, confidentiality rules, and resources. NICE (2016) Kripalani et al. (2007)

Research designs can distinguish referral completion from referral appropriateness. First, a linked-record cohort identifies patients at an index decision and follows them to their actual next contact. Second, chart review assesses whether the stated clinical problem and the receiving assessment were concordant. Third, interviews elicit how the patient understood the referral and any practical barriers. Fourth, stratified analysis estimates whether waiting and refusal differ by language, deprivation, age, or comorbidity. Comparison across services needs careful adjustment for clinical case mix, and data missing because a person sought support outside the network must be treated as potentially informative rather than as proof of disengagement. Levesque et al. (2013) Proctor et al. (2011)

Emergency Departments, Crisis Services, and Acute Assessment

Emergency and crisis pathways begin at a moment when distress, perceived danger, impaired functioning, or family concern has exceeded the capacity of existing support. Their function includes urgent assessment, stabilization, treatment of co-occurring physical problems, consideration of hospitalization when clinically necessary, and creation of a credible next step. An emergency visit is usually brief relative to the course of an illness or psychosocial crisis, so its value depends on the continuity it creates beyond the department. The care transition is particularly vulnerable when emergency records, psychiatric specialists, inpatient facilities, and community providers are administered separately. A documented disposition is not equivalent to verified access to follow-up support. NICE (2016) World Health Organization (2021)

Clinical safety depends on multiple processes: appropriate assessment of current difficulties, review of known history and available supports, attention to recent medication or substance changes, discussion of patient preferences, and practical arrangements for follow-up. Policies aimed at making emergency visits shorter can alter flow without ensuring that receiving services have capacity. Conversely, long waits may increase distress and lead to repeated assessment, while a clinically warranted admission may be unavailable because of bed constraints. A systems-level analysis therefore examines patient flow alongside appropriateness and safety, and interprets length of stay or admission rates in relation to case mix, availability, and outcomes rather than as unconditional markers of efficiency. Donabedian (1988) World Health Organization (2022)

The Safety Planning Intervention with structured follow-up has been examined among patients presenting with suicide-related concerns in US Veterans Health Administration emergency departments. Stanley and colleagues compared an intervention cohort with usual care and reported lower observed suicidal behavior over six months, alongside greater outpatient engagement. The evaluation was a nonrandomized cohort comparison. Its results are consistent with the potential value of bridging immediate crisis contact to ongoing services, but confounding and site-specific implementation prevent attribution of the entire association to one discrete element. An intervention can also affect documentation and subsequent contact ascertainment, making outcome definitions important. This study is evidence about an integrated service package in its setting, not a universal discharge algorithm. Stanley et al. (2018)

Care transitions after crisis must account for patients who refuse a proposed intervention, cannot use scheduled appointments, or have previously experienced coercion. A pathway that depends exclusively on compliance may lose the people with greatest barriers. Engagement-centered design includes understandable information, accessible contact arrangements, options for community support, and transparent expectations about confidentiality. Rights-based WHO guidance emphasizes person-centered care and the role of community services in avoiding unnecessary institutional restriction. Evaluating these arrangements requires patient-reported trust, perceived coercion, meaningful choice, and unmet need as well as clinical utilization. The question is whether the system remains usable under circumstances of high emotional and practical strain. World Health Organization (2021) NICE (2016)

Inpatient Psychiatry and Hospital–Community Interfaces

Inpatient psychiatric care may provide intensive assessment, containment during severe crisis, coordinated treatment initiation, and support when community options are temporarily insufficient. Hospitalization also changes a person's living arrangements, social contacts, occupational routines, and relationship to health professionals. A service-system analysis therefore examines admission criteria, geographic placement, continuity with preexisting clinicians, planning from the beginning of the stay, and the capacity of community services to receive the person later. Treating discharge as the administrative end of an episode obscures the change of responsibility that follows; the person remains in a sequence of clinical and social transitions. NICE (2016) World Health Organization (2021)

The NICE guideline on transitions between inpatient mental health settings and community or care-home settings offers an explicit organizational model in the United Kingdom. It calls for involvement of patients and relevant supporters, connection to community life, coordinated discharge plans, and transfer of key information. Its recommendations encompass both planned and unplanned admissions, with adjustments for the feasibility of early planning. This guideline is a jurisdiction-specific synthesis of evidence and expert guidance; it is not a randomized trial and its exact procedural duties should not be presumed identical under other legal or financing systems. For research purposes it provides measurable process variables such as whether discharge planning started early and whether the receiving professionals obtained an actionable plan. NICE (2016)

Information transfer is a recurrent point of failure. In general hospital settings, Kripalani and colleagues reviewed evidence of communication gaps between hospital-based and primary-care clinicians. Although the study did not exclusively concern psychiatric patients, it identifies a general mechanism relevant to mental health transitions: information may arrive late, omit pending investigations, or fail to identify responsibilities. Psychiatric records add specific complexities concerning fluctuating risk, interpretation of behavior under crisis, sensitive disclosures, and consent. A useful transition document must identify what changed during hospitalization, what remains uncertain, the next planned contact, and who holds responsibility for medication and risk follow-up; however, the content and disclosure rules require adaptation to local law and individual circumstances. Kripalani et al. (2007) NICE (2016)

A hospital may be located far from a person's home because local beds or specialist facilities are unavailable. Such placements can disrupt family contact, community clinicians, and continuity of housing or employment. The guideline draws particular attention to maintaining social links and support arrangements during such transitions. Researchers should evaluate geographic distance, travel burden, housing security, and the time until community providers resume involvement. Patients who lack stable accommodation or have intersecting physical health needs may require several coordinated services whose responsibilities overlap. The conceptual unit is the transition network, not merely the discharge destination: a successful hospital episode requires a credible social and clinical environment into which treatment can continue. NICE (2016) World Health Organization (2022)

Postdischarge Risk, Follow-Up, and the Interpretation of Evidence

The period after psychiatric hospital discharge is a central test of system continuity because clinical responsibilities change at a time when some patients remain vulnerable. A systematic review and meta-analysis by Chung and colleagues assembled 100 studies with 183 patient samples and estimated a pooled suicide rate of 484 deaths per 100,000 person-years after discharge, with a 95% confidence interval of 422 to 555. The rate was highest in studies examining the first three months, at 1,132 per 100,000 person-years. Between-study heterogeneity was very high, with I-squared of 98% for the overall estimate. These are population-level rates from observational datasets, not a prediction of an individual's future risk and not proof that discharge itself causes the observed deaths. Chung et al. (2017)

Clinical service guidance addresses this vulnerable transition through organized follow-up. NICE NG53 recommends follow-up within seven days of discharge and within 48 hours when a risk of suicide has been identified; it also specifies early transfer of discharge information and care plans. These times belong to a particular UK guideline and should be interpreted in its regulatory context. A researcher must distinguish compliance with a process measure from evidence that a particular interval independently causes benefit. Adherence may be correlated with resources, staff availability, illness course, patient engagement, and multiple other aspects of quality. Therefore, studies should measure the appropriateness and content of contact as well as the time interval. NICE (2016) Haggerty et al. (2003)

Fontanella and colleagues analyzed Medicaid data linked with mortality records for 139,694 youths aged 10 to 18 years who had been hospitalized for psychiatric reasons. A follow-up mental health visit within seven days was associated with a lower risk of death by suicide during days eight to 180 after discharge, with an adjusted relative risk of 0.44 and a 95% confidence interval of 0.23 to 0.83. The observational design, exclusion criteria, and small number of suicide deaths impose substantial limits on causal interpretation and broad generalization. Nevertheless, the study illustrates the value of linking service timing with a consequential outcome rather than studying appointment attendance alone. Fontanella et al. (2020)

Postdischarge research should identify a clear index discharge, classify discharge destination and clinical condition, and define whether follow-up means a booked appointment, completed encounter, direct outreach, or ongoing treatment. Time-varying exposure creates additional methodological problems: patients who survive long enough to receive a follow-up visit differ from those who do not, and measured effects can be distorted without careful temporal design. A robust assessment should consider competing events, such as readmission, geographical relocation, and discontinuation of insurance coverage. Ideally, observational findings would be triangulated with experiments or quasi-experiments altering service processes while recording both benefits and unintended consequences. Fontanella et al. (2020) Chung et al. (2017)

Child, Adolescent, and Life-Course Transitions

Age transitions introduce a second form of discontinuity: a person may remain clinically vulnerable while organizational eligibility changes. Child and adolescent services often differ from adult services in staffing, assessment practices, family involvement, school links, and consent arrangements. A fixed administrative age threshold can therefore create a gap even when symptoms and functional needs are continuous. The pathway should be studied in terms of preparation, joint meetings, records, receiving-service acceptance, and the young person's participation in decisions. Research also needs to identify youths who never reach adult services, because datasets limited to adult clinics omit precisely the cases most vulnerable to loss at the transition boundary. NICE (2016) Levesque et al. (2013)

The relative role of families and caregivers varies with age, developmental capacity, legal context, and the young person's preferences. During childhood, parents may arrange appointments and hold extensive information about learning, daily routines, and behavior. During adolescence and young adulthood, confidentiality expectations and personal agency may change. A service designed around parental coordination can become ineffective when the young adult must independently navigate insurance, booking procedures, and several clinics. Conversely, excluding supportive relatives without discussion can leave practical needs unmet. The research problem is to operationalize supported autonomy and continuity while respecting the person's rights and specific circumstances. World Health Organization (2021) NICE (2016)

Later-life transitions include interactions among depression, cognitive change, chronic physical illness, social isolation, and care-home placement. A person may move between general medical care, psychiatric consultation, rehabilitation, residential support, and specialist dementia services. Communication requirements expand when cognitive impairment limits the ability to convey prior treatment and when family involvement is changing. NICE explicitly addresses communication with services supporting people with dementia, learning disabilities, or autism. These recommendations can guide a study of whether receiving teams were informed and how they incorporated individual communication needs; their effectiveness still requires outcome evaluation in the relevant populations. NICE (2016) Katon et al. (2010)

The life-course perspective cautions against imposing one adult outpatient model on children, older adults, or people with lifelong disabilities. Measures of functional recovery should reflect age-appropriate education, employment, relationships, independence, and participation rather than rely solely on symptom scores. Service outcomes should be stratified by age and accessibility needs, while investigators assess whether differences are caused by clinical complexity or by administrative design. Transitions at school leaving, insurance change, relocation, retirement, and entry to long-term care provide quasi-natural opportunities for research when policy variations are clearly documented and comparable groups can be constructed. Levesque et al. (2013) World Health Organization (2022)

Long-Term Coordination in Serious Mental Illness

People living with severe and persistent mental health difficulties may require coordinated clinical and practical support over long periods. Frequent changes among acute wards, emergency departments, specialist teams, and primary care can produce repeated assessments without cumulative understanding. The organizational challenge is to sustain care while responding to changing levels of need. Intensive case management and assertive community approaches attempt to address this by combining outreach, care coordination, and access to multidisciplinary expertise. Their evaluation requires distinguishing actual service intensity, team structure, fidelity, and local baseline service provision, because an intervention called case management can differ substantially across locations. Dieterich et al. (2017) World Health Organization (2021)

The Cochrane review by Dieterich and colleagues examines intensive case management for people with severe mental illness and synthesizes trials with varying comparators and health systems. Its importance is conceptual as well as empirical: intervention effects depend on what ordinary care already provides and whether the intervention resembles assertive community treatment. An additional contact or case manager does not automatically create a clinically coherent pathway. The receiving team's capacity, breadth of services, responsiveness during crisis, and continuity through staff change all matter. Reviews that pool heterogeneous programs require careful interpretation, especially when hospital use is influenced by local admission policies and resource constraints. Dieterich et al. (2017) Donabedian (1988)

Coordination must extend beyond symptom treatment when housing, employment, or family circumstances threaten a person's ability to attend care. Mental health services frequently depend on social-sector partners whose funding and eligibility rules follow different institutional logics. A discharged patient may have medication and outpatient appointments arranged but no secure place to live; a person may need vocational support that the psychiatrist's team does not control. Rights-based community mental health models emphasize participation and social inclusion, making these issues part of the effectiveness of service delivery. This dossier investigates their clinical interface and accountability, while a separate community-service ecology inquiry addresses the broader system of social supports. World Health Organization (2021) World Health Organization (2022)

A rigorous longitudinal study would reconstruct time between contacts, clinician continuity, outreach attempts, physical-health follow-up, crisis escalation, patient-reported recovery, and participation outcomes. An observed decline in hospitalization could represent better preventive support or a loss of access, so it must be interpreted with symptoms, unmet need, and adverse events. Assignment to intensive programs may favor the most clinically complex people, creating confounding by indication in observational comparisons. Trials with explicitly specified control services provide more secure evidence for bundled programs, but transport to different jurisdictions requires independent examination of staffing, service availability, and access rules. Dieterich et al. (2017) Proctor et al. (2011)

Early Psychosis and Coordinated Specialty Care

Early intervention for psychosis is a crucial test of integration because initial treatment often involves psychiatric assessment, family engagement, psychological therapies, medication decisions, education, employment support, and ongoing review. Fragmentation at this stage may delay effective intervention and disrupt developmental opportunities. Coordinated specialty care packages attempt to bring these elements into a team operating around the person's goals. Their research question differs from a trial of one antipsychotic or one psychotherapy: it concerns whether a defined arrangement of interventions and relationships changes engagement, symptoms, hospitalization, and functioning compared with services normally available. Kane et al. (2016) Correll et al. (2018)

The NIMH RAISE Early Treatment Program compared a coordinated specialty-care model with usual community care for first-episode psychosis. Its two-year outcomes demonstrated the feasibility and clinical relevance of testing organization of care across US community settings. The result must be situated in the comparator's actual resources and in the observed treatment period. It should not be represented as evidence that any arbitrary team labeled coordinated care will reproduce the same findings. Fidelity, supervision, team communication, family involvement, and the capacity to support school and employment are plausible active components, but isolating their individual contributions requires designs beyond a simple package-versus-usual-care contrast. Kane et al. (2016)

Correll and colleagues' systematic review and meta-analysis combined ten randomized trials involving 2,176 participants with early-phase psychosis. At the end of treatment, early intervention services showed lower all-cause discontinuation, with a pooled risk ratio of 0.70 (95% confidence interval 0.61–0.80), and fewer participants experienced psychiatric hospitalization, with a risk ratio of 0.74 (95% confidence interval 0.61–0.90). School or work involvement also favored early intervention. These estimates concern the research interventions, studied populations, and time windows; follow-up after special services end may reveal different patterns. Heterogeneity in programs and outcomes limits claims about one universal service component. Correll et al. (2018)

The evidence links a service-system outcome, treatment engagement, to a developmental outcome, continued participation in ordinary life. Yet keeping someone enrolled in a program is valuable only insofar as care remains useful, acceptable, and responsive. A patient may receive many contacts but have little influence over treatment goals; another may pursue education successfully with intermittent clinical support. Comparative effectiveness research should therefore measure personal recovery and quality of life alongside service retention and hospitalization. Investigators should also examine transitions out of early-intervention programs, because a temporary period of intensive care can generate a new boundary when funding or time-limited eligibility ends. Kane et al. (2016) World Health Organization (2021)

Task Sharing and Mental Health Pathways in Resource-Constrained Systems

Many health systems have limited specialist mental health personnel and substantial gaps between need and contact with care. Task sharing attempts to extend access by training nonspecialist providers to deliver defined interventions with supervision, referral routes, and safeguards. Its success depends on more than the technical skill of individual providers. A useful pathway must identify who recognizes complicated presentations, when a specialist becomes involved, how clinical records are retained, and how service users return for reassessment. Workforce expansion without reliable escalation creates a risk that complex or deteriorating cases will remain in settings without sufficient clinical support. World Health Organization (2025) Patel et al. (2010)

The MANAS trial in Goa used a cluster-randomized design to evaluate an intervention led by lay health counselors for common mental disorders in primary care. Cluster allocation was appropriate to organizational interventions delivered at facility level, though it also creates analytic demands concerning clustering, baseline differences, and the small number of service units. The trial provided evidence for a locally designed care pathway, including structured support and professional input. It did not establish that an identical staffing model will be appropriate where regulations, language, supervision capacity, or availability of medicines differ. The scientific generalization is that integrated task-sharing systems can be tested rigorously, not that workforce substitution by itself is sufficient. Patel et al. (2010)

The COPSI randomized controlled trial in India examined community-based intervention for people with schizophrenia and their caregivers, adding evidence about sustained support outside hospital-centered structures. Such trials expand the empirical horizon from brief detection and referral to continuing clinical and family engagement. Their interventions are complex packages with culturally and infrastructurally specific components. Outcomes may depend on family involvement, travel distances, medication access, household economics, and the availability of trained providers. Research comparing these systems should specify which features of care were delivered, how frequently, by whom, and with what independent clinical oversight. Chatterjee et al. (2014) World Health Organization (2021)

A 2024 systematic review and meta-analysis by Faisal and colleagues evaluated collaborative care for common mental disorders in low- and middle-income countries. Its relevance lies in examining whether evidence from high-resource primary-care settings generalizes to different institutional conditions. Review findings must be read in light of intervention heterogeneity, different diagnostic populations, and variation in usual care. Country-level reports such as the WHO Mental Health Atlas establish background capacity and policy conditions but do not substitute for patient-level effectiveness studies. Together, they support research designs that combine contextual system description with credible comparisons of patient outcomes and implementation integrity. Faisal et al. (2024) World Health Organization (2025)

Inequality, Social Determinants, and Geographic Access

Service pathways are stratified by income, education, disability, language, migration status, place of residence, and institutional trust. These factors influence the ability to notice symptoms, seek assistance, travel to appointments, communicate preferences, and navigate eligibility rules. The access framework of Levesque and colleagues is useful precisely because it combines service properties with the capacities and resources of the population. A system can appear equitable when it allocates the same appointment procedure to everyone while producing unequal effective access. For example, identical online booking may disadvantage people without digital access, and identical travel requirements may disproportionately burden rural residents. Levesque et al. (2013)

Social determinants also affect the clinical course for which a pathway is being designed. Housing precarity, unemployment, discrimination, and exposure to violence can shape both distress and the feasibility of keeping an appointment or maintaining medication supplies. The systematic review of reviews by Lund and colleagues documents the importance of social determinants for mental disorders and the wider development agenda. For service-system analysis, these findings demand distinction between two causal questions: whether a social circumstance influences illness risk and whether a service reform reduces the access gap created by that circumstance. These are related but require different outcome measures, comparison groups, and strategies for controlling confounding. Lund et al. (2018) World Health Organization (2022)

Geographic barriers operate at multiple scales. A remote region may have no specialist clinic, a city neighborhood may have specialized care but prohibitively expensive transport, and a well-connected hospital may discharge patients across administrative boundaries whose records are incompatible. Telehealth can reduce some distance-related barriers while creating others through connectivity, privacy, or device requirements. The WHO Atlas monitors national indicators including telehealth and community care, but local equity requires more granular spatial, socioeconomic, and patient-level data. Aggregating rural and urban regions can conceal service deserts, while comparing crude appointment rates may miss differential need and diagnosis severity. World Health Organization (2025) Levesque et al. (2013)

Equity-sensitive pathway research should stratify the likelihood of passing each major transition, then consider whether observed disparities persist after appropriate adjustment for need. It should also examine who is excluded by the data system: people without insurance, informal migrants, those using nonmedical services, and individuals who stop care before receiving a formal diagnosis. A reduction in average waiting time can coexist with worsening outcomes among a smaller high-need group. Equity analysis must therefore treat distribution as a substantive outcome of service organization rather than a secondary demographic description. Changes in access and experience can be evaluated with mixed methods, provided the causal interpretation is calibrated to the design. Wang et al. (2007) Levesque et al. (2013)

Physical Health, Comorbidity, and Integrated Clinical Accountability

Mental health care is frequently delivered alongside care for cardiovascular disease, diabetes, chronic pain, disability, and substance-related problems. Distinct services may use different records, medication lists, assessment schedules, and treatment priorities. The pathway may fail when mental health teams assume that general medical needs are being monitored elsewhere, while primary-care clinicians assume that psychiatric services are addressing all urgent concerns. Integrated care should therefore specify responsibilities for physical monitoring, review medication changes, and reconcile recommendations when several clinicians are involved. The relevant evidence concerns the organization of collaborative clinical action rather than a claim that every physical and mental health condition has the same biological mechanism. Katon et al. (2010) NICE (2016)

The TEAMcare trial, reported by Katon and colleagues, tested collaborative care for patients with depression and chronic medical conditions. Its design demonstrates that integration can be operationalized through coordinated medical and mental health targets, team-based review, and adjustment of care. The intervention was a composite, and its findings are most secure for patients resembling those enrolled in the study and health systems able to supply comparable resources. The trial does not prove that simple colocation of psychiatric and medical offices will achieve the same outcomes. Research must measure the actual sequence of decisions and whether practitioners act on each other's assessments. Katon et al. (2010)

Co-occurring substance use can complicate assessment, crisis management, referral eligibility, and sustained engagement. A system that routes a person to an addiction service while deferring mental health treatment until a separate process is completed may create mutually reinforcing access barriers. Similarly, medication interactions or withdrawal-related symptoms may be difficult to interpret when care is fragmented. The organizational principle is a credible cross-specialty assessment and transfer pathway with explicit ownership of clinical decisions. Questions of specific medications, diagnostic criteria, and individual treatment are governed by the relevant disorder and pharmacotherapy research; the service-system inquiry studies where and why the clinical handoffs work or fail. World Health Organization (2022) NICE (2016)

Measurement must capture whether integrated care produces improved symptoms, physical risk indicators, functioning, service experience, and reduction in adverse events. Analyses should account for illness severity and initial access, as people with multiple conditions are more likely to receive intensive services and experience poor outcomes regardless of system design. Randomization at practice or team level can compare pathway programs, but intervention spillover and clustered outcomes need explicit statistical handling. Longitudinal registry designs can examine real-world scale while facing missing information across insurance providers and care systems. These complementary methods are needed before concluding that an integration policy is effective at population level. Katon et al. (2010) Proctor et al. (2011)

Community Interfaces, Rights, and Recovery-Oriented Service Design

Clinical treatment trajectories enter the larger social world at every transition. Home and community environments influence whether a person can rest, travel, work, obtain food, maintain relationships, and recognize changing distress. A pathway focused only on the next medical appointment may overlook the practical conditions required for follow-through. The WHO's guidance on community mental health services emphasizes rights-based, person-centered approaches and alternatives that can support participation and autonomy. The question for a clinical-pathway study is how these community services are connected to formal assessment, crisis response, specialist treatment, and continuing medical responsibility, including when a person does not wish to engage with a particular institution. World Health Organization (2021) World Health Organization (2022)

The balanced care model proposed by Thornicroft and Tansella articulates the need for an appropriate relationship between hospital and community components, with service configuration varying by local resources. It supplies a comparative framework rather than a single universal service blueprint. A low-resource district may prioritize a dependable route from primary care to specialist assessment and back; a highly resourced city may need to reduce duplication among specialized teams. The balance must be assessed through access, outcomes, safety, patient experience, and institutional feasibility. Simply increasing the share of care labeled community-based is inadequate if support networks cannot manage crises or deliver necessary clinical interventions. Thornicroft & Tansella (2013)

Recovery-oriented care expands outcome evaluation to include identity, autonomy, relationships, and participation. Clinical symptom changes remain important but do not exhaust the experiences by which people judge services. A person may have residual symptoms yet return to work and maintain valued relationships; another may experience symptom relief without stable housing or trustworthy care. This distinction matters when hospitals or community teams are rewarded for throughput, because administrative completion may obscure unmet social and clinical needs. A comprehensive pathway assessment should solicit the person's priorities, record actual opportunities for decision-making, and examine whether transitions preserve those priorities across service providers. World Health Organization (2021) Haggerty et al. (2003)

Institutional coercion creates a particularly important tension. Emergency and inpatient systems sometimes exercise legally authorized restrictions during crises, and such experiences may influence subsequent trust and willingness to seek care. Rights-based service design calls for transparent explanation, proportionality, access to advocacy, and ongoing involvement of the person in future planning, subject to jurisdiction-specific law. The effectiveness question extends beyond whether an admission resolved the immediate crisis to whether the subsequent pathway remains clinically safe, respectful, and acceptable. Studying these outcomes requires qualitative inquiry and patient-reported measures alongside clinical records, because the institutional account of a transition may differ from the person's experience. NICE (2016) World Health Organization (2021)

Mapping Care Pathways and Measuring Transitions

A research pathway must be reconstructable from observations rather than inferred from an organizational chart. The minimum event structure includes a date, setting, type of contact, clinical purpose, referral origin and destination, referral acceptance, the person or team responsible, and a meaningful disposition. Repeated events allow researchers to identify cycles between emergency care and hospitalization, transitions from primary care to specialty consultation, and periods without recorded support. Yet registry data often encode administrative transactions rather than actual therapeutic relationships, making corroboration necessary. Electronic records, insurer claims, appointment systems, discharge summaries, and patient narratives differ in completeness and may omit out-of-network services or informal support. Haggerty et al. (2003) Kripalani et al. (2007)

Several operational definitions are especially important. Referral completion can mean acceptance by a receiving service or the first completed visit; these should be reported separately. Waiting time can be measured from clinician referral, patient request, eligibility approval, or booking, each producing different conclusions. Continuity can refer to the same named clinician, a team, preserved records, or an agreed management plan. Follow-up may be in person, by telephone, or another clinically appropriate method. Attrition may represent an unwanted loss of care, informed refusal, a deliberate step-down, or relocation. A study that collapses all these events into attendance rates loses the mechanisms it purports to explain. Haggerty et al. (2003) Levesque et al. (2013)

A useful analytic representation treats services as nodes and clinically meaningful transfers as directed edges. The weight of an edge can reflect transition probability, delay, unmet need, risk, or patient burden. Such graphs can reveal bottlenecks and loops, but network centrality is not itself a clinical outcome. A hospital may appear central because all admissions must pass through it; that fact neither proves effective coordination nor establishes that more hospital care is desirable. Process mining and sequence analysis can identify recurrent trajectories, provided the data are sufficiently complete and the temporal order is valid. This application is a methodological synthesis derived from general quality and continuity frameworks, not an assertion that a single graph model has already been universally validated. Donabedian (1988) Proctor et al. (2011)

Data quality audits should examine duplicate patient identities, incompatible coding, delayed data exchange, missing discharge destinations, and unobserved transfers between institutions. Linkage error may be associated with migration, unstable housing, or name changes, biasing the apparent pathway of disadvantaged populations. Administrative categories can also reflect reimbursement rules rather than clinically meaningful distinctions. Patient interviews can identify referrals that were nominally closed without an appointment, and clinician interviews can expose unofficial workarounds. Triangulation is especially valuable when the research question concerns responsibility and communication, variables that rarely have one reliable numerical proxy. Levesque et al. (2013) Aarons et al. (2011)

Experimental, Quasi-Experimental, and Longitudinal Designs

Randomized trials offer an important design for comparing service models when the intervention can be allocated ethically and consistently. Because care coordination changes the work of a clinic, ward, or team, cluster randomization may be more appropriate than individual assignment and can reduce contamination between arms. Trials must report how clusters were chosen, the number of clusters, baseline balance, clustering adjustments, intervention fidelity, and how the control condition was organized. The IMPACT and MANAS studies exemplify the possibility of experimentally evaluating coordinated mental health programs, but they also demonstrate that such programs contain several interacting components whose mechanisms require supplementary investigation. Unützer et al. (2002) Patel et al. (2010)

Observational linked-record cohorts are well suited to rare and consequential outcomes following hospitalization, especially when long follow-up or very large samples are needed. However, engagement with follow-up care is affected by severity, housing, insurance, caregiver support, prior adherence, and the perceived usefulness of services. These same factors can influence later hospitalization and mortality, yielding confounding. Analyses must align time zero with eligibility, classify exposure without introducing immortal-time bias, report missingness, and examine sensitivity to unmeasured confounding. Fontanella and colleagues' youth study supports an association between timely follow-up and lower suicide mortality but remains vulnerable to these constraints. Fontanella et al. (2020)

When a randomized trial is impractical, interrupted time-series designs can examine changes following a clearly timed reform, using repeated observations before and after implementation. Strong studies assess secular trends, seasonality, preexisting changes, concurrent policies, and whether coding or patient composition changed at the intervention boundary. Difference-in-differences comparisons may be useful when jurisdictions adopt care models at different times and credible parallel trends can be argued. These designs can provide evidence of service-system effects under transparent assumptions, but a simple before-and-after difference without a plausible counterfactual is weak causal evidence. Researchers should specify exactly which policy changed the pathway and whether implementation was immediate or gradual. Donabedian (1988) Proctor et al. (2011)

Longitudinal causal questions must also distinguish intermediate events from outcomes. A follow-up visit may mediate an intervention's effect, but conditioning on it can induce bias if it is influenced by postassignment clinical deterioration. Readmission can be an adverse outcome, a necessary protective event, or a competing event for outpatient participation. Time-to-event methods should consider multiple transitions and competing risks, while longitudinal outcome models should account for repeated observations. Randomized evidence about whole packages, quasi-experimental evidence about policy changes, and qualitative evidence about mechanisms can be synthesized together without pretending that all designs estimate the same causal contrast. Chung et al. (2017) Fontanella et al. (2020)

Evidence Hierarchies and the Limits of Comparative Effectiveness

A service-system claim has at least three evidentiary components: the pathway was implemented, the pathway changed the process it targeted, and patient-centered outcomes improved because of the changed process. These propositions may rest on different data. A registry can show a shorter mean time to follow-up without proving that patients felt safer or experienced fewer symptoms. A trial can show improved depression response from a collaborative-care package while leaving uncertain whether the case manager, psychiatric supervision, or monitoring interval was indispensable. Critical appraisal must preserve these distinctions instead of converting process success into global clinical effectiveness. Unützer et al. (2002) Proctor et al. (2011)

Meta-analysis is informative when studies examine sufficiently comparable populations, programs, and outcomes. The Cochrane synthesis of collaborative care for depression and anxiety and the meta-analysis of early psychosis services support identifiable benefits of particular organizational packages. Yet effect estimates describe combinations of staffing, duration, training, and usual-care comparison. They are not universal coefficients that can be applied to any country or facility. In early psychosis research, outcomes at the end of a specialized program may differ from outcomes after the program terminates. The appropriate policy interpretation is to test whether essential conditions of the evaluated service can be reproduced, then reassess outcomes under local implementation. Archer et al. (2012) Correll et al. (2018)

Evidence for discharge transitions is particularly heterogeneous. The pooled suicide rate described by Chung and colleagues comes from studies with widely differing national contexts, sample selection rules, diagnostic composition, observation periods, and time trends; the very high I-squared signals that a single pooled number inadequately characterizes every setting. Similarly, a large observational association between timely follow-up and mortality cannot isolate the effect of contacting a patient from the contribution of the broader services that make such contact possible. Quantitative estimates should therefore be accompanied by denominators, follow-up intervals, uncertainty, and an account of what causal conclusion the design permits. Chung et al. (2017) Fontanella et al. (2020)

Negative, mixed, and null findings deserve equal methodological attention. A coordination program may fail because the mechanism is weak, because it was implemented without fidelity, because services already provide similar support, or because the selected outcome lacks sensitivity. It may improve appointment attendance while leaving symptoms unchanged, or improve patient experience while increasing short-term utilization. A review that reports only dramatic positive trials will systematically overstate confidence. Publication bias, selective reporting, inadequate controls, short follow-up, and intervention heterogeneity should be assessed before drawing policy conclusions. These considerations also argue for transparent trial registration, accessible protocols, and independent replication in varied health-system contexts. Archer et al. (2012) Dieterich et al. (2017)

Measurement of Quality, Safety, and Patient-Experienced Continuity

Clinical pathway measurement requires a balanced outcome framework. Process indicators may include the fraction of accepted referrals, median time to first contact, availability of a named clinician, transfer of a complete care plan, and continuity after discharge. Clinical outcomes may include validated symptom measures, functioning, crisis recurrence, treatment-related harm, and mortality. Patient-centered outcomes include whether the plan was understood, whether preferences were respected, and whether services remained accessible during changing circumstances. A service quality dashboard must prevent a strong process score from concealing poor outcome or equity performance. Donabedian (1988) Haggerty et al. (2003)

Continuity measures must be construct-valid. Counts of visits with one clinician capture an aspect of relational continuity but do not measure whether the relationship was therapeutic. A copied discharge summary demonstrates data transfer but does not prove that new clinicians read or applied its contents. A care manager's name on a form suggests assigned responsibility, but effective management continuity also requires response to change. Researchers should therefore combine administrative proxies with chart review, patient-reported continuity, and direct examination of handoff processes. Test-retest reliability, missing data, and measurement invariance across languages and populations become important when comparing services. Haggerty et al. (2003) Levesque et al. (2013)

Safety indicators require particular care. Suicide mortality is rare and clinically important, but event rates may be unstable in small local samples and can be misclassified. Self-harm, crisis presentations, adverse medication events, and readmissions offer additional information yet are not interchangeable outcomes. A short-term rise in recorded safety events after implementing better monitoring could reflect improved detection rather than increased harm. A study should document the ascertainment method, time window, and opportunity for observation, then use multiple indicators to assess plausible interpretations. The postdischarge literature demonstrates both the urgency of studying risk and the limitations of using a single aggregated outcome. Chung et al. (2017) Stanley et al. (2018)

Patient experience supplies information unavailable in administrative measures. An appointment may be marked completed while the person found it inaccessible, misunderstood the next steps, or experienced it as coercive. Qualitative interviews, accessible surveys, and participatory study design can identify specific features of trust, dignity, and comprehension. Sampling must include people who discontinued care or were refused entry, otherwise service evaluations selectively represent those who successfully navigated the system. Rights-based WHO guidance supports the substantive importance of such experience measures, while implementation frameworks can guide the assessment of acceptability and feasibility at organizational level. World Health Organization (2021) Proctor et al. (2011)

Implementation, Workforce, Financing, and System Resilience

Evidence-based service models are implemented through organizations that differ in staffing, leadership, information technology, workforce stability, and reimbursement. Aarons and colleagues' model of evidence-based practice implementation in public service sectors highlights the institutional conditions influencing adoption and sustained use. The Consolidated Framework for Implementation Research, updated by Damschroder and colleagues, organizes contextual determinants that can shape implementation. These frameworks provide explanatory categories, not a guarantee that identifying barriers will change clinical outcomes. Their utility for pathways lies in explaining why a model that worked in a controlled study may falter when shifted to routine services. Aarons et al. (2011) Damschroder et al. (2022)

Implementation outcomes should be measured explicitly. Proctor and colleagues distinguish acceptability, adoption, appropriateness, feasibility, fidelity, implementation cost, penetration, and sustainability. A hospital may adopt a discharge checklist and achieve high fidelity while community appointment capacity remains inadequate. A primary-care clinic may accept collaborative care yet lack psychiatric consultation time for nonresponding patients. Penetration measures who actually receives the service among the eligible population, while sustainability asks whether the program remains operational when initial research funding ends. These outcomes clarify the chain from policy announcement to delivered care and can be compared with clinical endpoints. Proctor et al. (2011)

Workforce composition determines what transitions are possible. Specialists may be concentrated in cities, crisis teams may operate only during restricted hours, and nursing or social-work capacity may be insufficient for outreach. The WHO Mental Health Atlas compares national workforce and service indicators and documents major disparities, but country averages conceal local service geography and the distribution of clinical expertise. A scientifically grounded redesign should estimate the number and characteristics of people needing the pathway, the intensity and duration of required contacts, and the realistic capacity of the receiving teams. Reassigning tasks can improve reach only when supervision and escalation remain workable. World Health Organization (2025) Patel et al. (2010)

Service resilience concerns the ability to maintain clinically necessary connections during staff turnover, disasters, funding changes, digital outages, or population displacement. A pathway dependent on one external software service may fail during connectivity loss; one dependent on an irreplaceable specialist may fail during absence. Resilience can be studied through simulations, interrupted time-series observation during shocks, and examination of whether high-risk patients retain contact under constrained conditions. Organizational redundancy, appropriate documentation, and clearly assigned fallback responsibilities are plausible safeguards, but their actual effectiveness requires evaluation against clinical and patient-experience outcomes. System resilience thus becomes an empirical property of the pathway rather than a general promise of institutional preparedness. World Health Organization (2025) Proctor et al. (2011)

Digital Records, Telehealth, and Human–Artificial Clinical Systems

Digital systems can alter clinical pathways by making referrals traceable, enabling appointment reminders, supporting remote consultations, and sharing relevant records across institutions. Their contribution should be evaluated at the level of the completed clinical transition. A system that sends an electronic discharge summary may improve information transmission but still leave patients without timely care, and a telehealth service may expand geographic reach while excluding individuals with limited connectivity or privacy. The WHO Mental Health Atlas includes indicators on digital service capacity, providing policy context rather than causal proof of benefit. Research must examine actual access, clinical safety, user experience, and whether digital infrastructure creates new points of failure. World Health Organization (2025) Levesque et al. (2013)

Automated triage, decision support, and language interfaces may organize incoming information or assist with administrative navigation. They also create risks involving misclassification, unequal performance across populations, privacy leakage, automation bias, and ambiguous responsibility when an error occurs. A model trained on historical utilization may reproduce earlier service inequities by interpreting unmet need as low predicted demand. Clinical deployment requires external validation, appropriately calibrated uncertainty, human oversight, and monitoring for shifts in case mix. These are research and governance requirements derived from the epistemic structure of triage and patient-safety evaluation, rather than claims that a particular artificial system has demonstrated clinical effectiveness. Donabedian (1988) Proctor et al. (2011)

In the horizon of the Artificial Era, human–artificial systems become an additional interface within clinical service organization. The scientific question is what information and decisions artificial components can support, how their performance is measured, and who retains responsibility for clinically consequential action. Artificial cognition and agency may be studied through observable decision policies and functional effects without inferring subjective experience from fluent language. A pathway study would test whether an artificial interface changes referral completion, error rates, equitable access, clinician workload, patient comprehension, and subsequent health outcomes. Experimental assignment and carefully monitored rollouts are needed before effects can be attributed to the interface itself. World Health Organization (2022) Proctor et al. (2011)

A research program should compare human-only, digitally assisted, and human–artificial pathway configurations while ensuring comparability of populations and clinical settings. Outcome assessment must include potentially hidden errors: unsupported reassurance, inappropriate escalation, missed referrals, or disclosure of sensitive information. It should also investigate whether automated services redistribute work onto patients or clinicians rather than removing it. Provenance of records, auditability of recommendations, reversibility of erroneous routing, and service continuity during outages are essential operational variables. The horizon From Homo to Artificial broadens the study of service mediation while preserving the empirical requirement to validate every claimed clinical benefit and safety property. World Health Organization (2021) Damschroder et al. (2022)

Governance, Accountability, and Cross-Sector Clinical Responsibility

Mental health systems are governed through financing, professional regulation, hospital administration, insurance, public health authority, community organizations, and, in some settings, social welfare agencies. A person may move across several of these systems during one illness episode, with different definitions of eligibility and responsibility. Governance therefore enters the research pathway as a causal candidate: it determines whether a referral can be funded, a record transferred, or a community team permitted to act. Cross-sector coordination depends on incentives and obligations that persist beyond an individual clinician's goodwill. Observational comparisons should document governance differences explicitly rather than attributing all variation to the motivation or skill of frontline workers. World Health Organization (2025) Aarons et al. (2011)

Accountability at a transition can be investigated by asking which actor has authority to accept the person, arrange the next contact, respond to deterioration, and revise the plan. A single named coordinator may improve clarity without having sufficient resources to obtain an appointment or housing support. Conversely, a multidisciplinary network can operate effectively if responsibilities and escalation rules are explicit. The organizational design should preserve patient participation, lawful information sharing, and the ability to contest inaccurate records. The NICE transition guideline illustrates one regulatory arrangement involving hospitals, primary care, community teams, and carers, but the relevant legal and financing details must be established independently for each jurisdiction. NICE (2016) Haggerty et al. (2003)

Financing arrangements can reward visits, episodes, contracts, or population outcomes, and each design may shape clinical behavior. Incentives focused narrowly on reducing admissions might discourage appropriate inpatient treatment; incentives focused solely on appointment volume might favor easy-to-reach cases and short contacts over difficult coordination. Pathway quality research should therefore evaluate the entire bundle of outcomes and the distribution of benefits. Implementation studies can measure whether new contractual responsibilities were adopted and sustained, while comparative effectiveness designs must test whether altered incentives caused clinically meaningful changes. Transparent accounting of resources is necessary when reforms claim to provide equivalent outcomes at lower cost. Donabedian (1988) Proctor et al. (2011)

International comparison is valuable because different systems solve similar problems through different institutional structures. Nevertheless, countries differ in population health, data coverage, funding, workforce, legal protections, and the availability of family and community support. The WHO Atlas documents such institutional variation but should not be used to rank individual clinical pathways by causality without stronger patient-level evaluation. A balanced comparative program combines system description, qualitative process reconstruction, matched or quasi-experimental analyses, and repeated measurement after reforms. The central question is whether institutional responsibility survives the boundary between one provider and the next in a form meaningful to patients. World Health Organization (2025) Thornicroft & Tansella (2013)

Critical Scientific Disputes and Unresolved Tensions

One recurring debate concerns standardization versus clinical adaptation. Pathway protocols can reduce omissions and clarify responsibility, yet rigid sequences can misfit fluctuating symptoms, cultural interpretations, neurodevelopmental differences, or complex comorbidity. A protocol is most useful when it specifies reliable safety and communication functions while allowing the clinician and patient to adapt treatment content and timing to the actual circumstances. Evaluations should distinguish deviation caused by poor implementation from warranted adaptation. Implementation frameworks can make that distinction visible through fidelity assessment, but the desired degree of uniformity remains a substantive scientific and ethical question. Damschroder et al. (2022) Proctor et al. (2011)

A second dispute concerns specialization versus integration. Specialist teams can maintain deep expertise and concentrated resources, yet additional eligibility boundaries create new opportunities for discontinuity. Generalist settings can improve reach and continuity but may lack the knowledge or time to manage severe or complex difficulties. The collaborative-care and early-psychosis literature offers evidence that coordinated, specialized expertise can be delivered within integrated programs. Still, these findings do not define one optimal number of services or one professional mix for every region. The relevant comparison is among complete, feasible pathways under stated resource conditions, not among abstract labels such as specialist or integrated. Unützer et al. (2002) Correll et al. (2018)

A third dispute concerns utilization as a proxy for success. Some systems seek fewer hospitalizations, lower emergency attendance, and faster discharge. Such changes can reflect beneficial preventive care, but also reduced access or displacement of burden onto families and community services. Conversely, more contact may indicate appropriate detection of unmet need. Outcomes must be interpreted in the context of baseline need and preferences. The postdischarge literature demonstrates that minimizing institutional contact without securing continuing support can leave consequential vulnerabilities unaddressed. This does not establish that hospitalization is always beneficial; it establishes that pathway evaluation requires the clinical and social meaning of each event. Chung et al. (2017) NICE (2016)

A fourth debate concerns individual responsibility and system responsibility. Service engagement is sometimes characterized as patient adherence, yet the capacity to attend depends on transport, language, trust, affordability, and compatible scheduling. Patient preference also matters: a person may make a reasoned choice to change or discontinue a proposed service. The challenge is to distinguish informed choice from a loss produced by inaccessible arrangements. Rights-based care and multidimensional access theory support inquiry into these mechanisms. Research should identify where the institution could act to make participation possible while avoiding a model that treats every missed appointment as evidence of a defective patient. Levesque et al. (2013) World Health Organization (2021)

Priorities for the Next Generation of Pathway Research

Future studies should compare pathways as time-dependent organizational systems, using a common event vocabulary while preserving clinical and cultural specificity. Shared definitions of referral, acceptance, first contact, discharge, follow-up, treatment interruption, and completed transfer would improve comparison across settings. Such standardization requires a companion account of who remains outside administrative observation. Linked national records can support large-sample follow-up, but locally conducted interviews remain necessary to detect invisible barriers, misunderstanding, and the practical meaning of continuity. Researchers should publish transition definitions and linkage rules alongside their clinical outcomes. Haggerty et al. (2003) Levesque et al. (2013)

The strongest causal programs will use randomized or quasi-experimental reforms that manipulate one or more clearly defined coordination processes and assess effects across clinical, social, and safety outcomes. Factorial or adaptive designs may help separate the value of active referral, appointment preparation, clinician continuity, and outreach intensity when ethically and practically feasible. Long-term follow-up is essential to determine whether early program gains endure after specialized funding or eligibility ends. Trials should transparently document ordinary care, implementation fidelity, clustering, attrition, and differential exposure. This approach builds on evidence from collaborative care and coordinated specialty care while testing their component mechanisms more directly. Archer et al. (2012) Kane et al. (2016)

An equity-centered agenda will examine transition success among people with unstable housing, rural residence, language barriers, disabilities, multiple diagnoses, and limited digital access. Average improvements can conceal deterioration in small but high-need populations. Research should therefore report both population means and the distribution of waiting times, completion, functioning, and harm. Particular care is needed where insurance data omit people who have no coverage, or where mortality-linked registries underrepresent migration and other forms of loss to follow-up. The WHO Atlas provides a macro-level baseline for international comparison but requires supplementation with individual-level and place-based measures. World Health Organization (2025) Lund et al. (2018)

Digital and artificial interfaces require independently replicated clinical pathway evaluation before routine expansion is justified by claims of efficiency. Research should test error detection, inappropriate triage, privacy, model drift, clinician workload, patient trust, and responsibility during failure. Governance studies should also examine who can correct a machine-generated record and whether patients can reach a human professional when automated navigation fails. An effective human–artificial system is empirically defined by measured performance within the service network, including adverse consequences and distributional effects. The broad research horizon of the Artificial Era makes these questions increasingly important without changing the evidentiary requirements for clinical safety. Damschroder et al. (2022) World Health Organization (2021)

Integrative Scientific Synthesis

The evidence supports a multi-level understanding of mental health care pathways. At the population level, access depends on the relationship between people's capacities and the approachability, acceptability, availability, affordability, and appropriateness of services. At the organizational level, staff, funding, records, and clinical governance determine which transitions can occur. At the encounter level, patient–clinician relationships, assessment, shared decisions, and communication shape the quality of care. Across time, continuity depends on whether these elements remain coherent as needs and settings change. The distinctive research object is the organization of the links among levels rather than the properties of any single provider. Levesque et al. (2013) Haggerty et al. (2003)

Several findings have comparatively secure empirical foundations. Structured collaborative depression care can improve outcomes relative to usual care under conditions demonstrated in randomized trials; coordinated early psychosis programs can improve engagement and reduce some hospital use over the trial periods; and the postdischarge interval carries substantial population-level risks in observational evidence. However, the causal contribution of each service component often remains unresolved, and the translation of effect sizes between countries or populations requires fresh evaluation. These differences in certainty should remain visible in any policymaking based on the synthesis. Unützer et al. (2002) Correll et al. (2018) Chung et al. (2017)

The synthesis also shows that safety and equity cannot be appended to an otherwise complete pathway evaluation. A system that improves average follow-up but excludes particular populations has altered the distribution of opportunity and vulnerability. A system that sends complete electronic records while losing relational continuity has solved only one of several coordination problems. Effective transfer requires information, receiving responsibility, appropriate resources, and the person's meaningful participation; each is measurable and potentially modifiable. Organizational accountability extends across the interval between services, when formal institutional borders can otherwise conceal an ongoing clinical need. NICE (2016) World Health Organization (2021)

A scientifically credible program of improvement follows the complete sequence from identified need to observed follow-through and sustained benefit. It specifies what the pathway is meant to achieve, maps where people currently encounter barriers, chooses interventions with evidence suited to the local context, and evaluates outcomes using appropriate counterfactual designs. It also measures the long-term capacity of organizations to sustain the model and invites patient experience into the definition of success. This program is a synthesis of established access, continuity, quality, and implementation research, with empirical hypotheses remaining open wherever mechanisms or generalizability have not yet been tested. Donabedian (1988) Proctor et al. (2011)

Conclusion

Mental health service systems and clinical care pathways constitute a fundamental research domain because clinical outcomes depend in part on how people move between settings and how responsibility persists across those movements. The pathway begins with perceived need and includes recognition, entry, assessment, referral, treatment, crisis response, discharge, and continuing contact. Each transition can sustain or interrupt care. The analytical task is to connect organizational structure and specific coordination processes to patient-centered outcomes, while distinguishing administrative completion from a functioning clinical relationship. Levesque et al. (2013) Haggerty et al. (2003)

The strongest evidence comes from carefully evaluated service packages, including collaborative care in primary-care depression and coordinated early psychosis services. Large observational cohorts and systematic reviews identify important vulnerabilities around psychiatric discharge, although their findings must be read with attention to confounding and heterogeneity. Global system surveys establish differences in workforce and infrastructure, whereas comparative pathway studies show that the route to psychiatric care varies with local institutions and social context. These evidence streams are complementary and retain distinct causal meanings. Unützer et al. (2002) Correll et al. (2018) Fontanella et al. (2020) World Health Organization (2025)

A clinically coherent pathway combines appropriate assessment, an intelligible plan, reliable information transfer, confirmed receiving responsibility, access to needed services, and opportunities for reassessment. Equity, recovery, patient rights, and practical continuity form part of its quality because they condition whether care can be used and sustained. Researchers must examine failures in the intervals between organizations, including people absent from service records, and evaluate adverse consequences alongside intended benefits. NICE (2016) World Health Organization (2021)

The continuing agenda is to build causal and longitudinal evidence about service transitions across health systems, ages, diagnoses, and social circumstances. Better event data, stronger comparative designs, explicit process measures, and patient participation can clarify which arrangements improve outcomes and under what conditions. Digital and human–artificial mediation introduces new research questions about delegation, auditability, and clinical responsibility. Mental health pathway science will advance when it measures the movement of care as carefully as it measures the interventions delivered within each service. Proctor et al. (2011) Damschroder et al. (2022)

References

  • Aarons, G. A., Hurlburt, M., & Horwitz, S. M. (2011). Advancing a conceptual model of evidence-based practice implementation in public service sectors. Administration and Policy in Mental Health and Mental Health Services Research, 38(1), 4–23. https://doi.org/10.1007/s10488-010-0327-7
  • Archer, J., Bower, P., Gilbody, S., Lovell, K., Richards, D., Gask, L., Dickens, C., & Coventry, P. (2012). Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews, 2012(10), CD006525. https://doi.org/10.1002/14651858.CD006525.pub2
  • Chatterjee, S., Naik, S., John, S., Dabholkar, H., Balaji, M., Koschorke, M., Varghese, M., Thara, R., Weiss, H. A., Williams, P., McCrone, P., Patel, V., & Thornicroft, G. (2014). Effectiveness of a community-based intervention for people with schizophrenia and their caregivers in India (COPSI): A randomised controlled trial. The Lancet, 383(9926), 1385–1394. https://doi.org/10.1016/S0140-6736(13)62629-X
  • Chung, D. T., Ryan, C. J., Hadzi-Pavlovic, D., Singh, S. P., Stanton, C., & Large, M. M. (2017). Suicide rates after discharge from psychiatric facilities: A systematic review and meta-analysis. JAMA Psychiatry, 74(7), 694–702. https://doi.org/10.1001/jamapsychiatry.2017.1044
  • Correll, C. U., Galling, B., Pawar, A., Krivko, A., Bonetto, C., Ruggeri, M., Craig, T. J., Nordentoft, M., Srihari, V. H., Guloksuz, S., Hui, C. L. M., Chen, E. Y. H., Valencia, M., Juarez, F., Robinson, D. G., Schooler, N. R., Brunette, M. F., Mueser, K. T., Rosenheck, R. A., . . . Kane, J. M. (2018). Comparison of early intervention services vs treatment as usual for early-phase psychosis: A systematic review, meta-analysis, and meta-regression. JAMA Psychiatry, 75(6), 555–565. https://doi.org/10.1001/jamapsychiatry.2018.0623
  • Damschroder, L. J., Reardon, C. M., Widerquist, M. A. O., & Lowery, J. (2022). The updated Consolidated Framework for Implementation Research based on user feedback. Implementation Science, 17, Article 75. https://doi.org/10.1186/s13012-022-01245-0
  • Dieterich, M., Irving, C. B., Bergman, H., Khokhar, M. A., Park, B., & Marshall, M. (2017). Intensive case management for severe mental illness. Cochrane Database of Systematic Reviews, 2017(1), CD007906. https://doi.org/10.1002/14651858.CD007906.pub3
  • Donabedian, A. (1988). The quality of care: How can it be assessed? JAMA, 260(12), 1743–1748. https://doi.org/10.1001/jama.1988.03410120089033
  • Faisal, M. R., Salam, F. T., Vidyasagaran, A. L., Carswell, C., Naseri, M. W., Shinwari, Z., Fulbright, H., Zavala, G. A., Gilbody, S., & Siddiqi, N. (2024). Collaborative care for common mental disorders in low- and middle-income countries: A systematic review and meta-analysis. Journal of Affective Disorders, 363, 595–608. https://doi.org/10.1016/j.jad.2024.07.086
  • Fontanella, C. A., Warner, L. A., Steelesmith, D. L., Brock, G., Bridge, J. A., & Campo, J. V. (2020). Association of timely outpatient mental health services for youths after psychiatric hospitalization with risk of death by suicide. JAMA Network Open, 3(8), e2012887. https://doi.org/10.1001/jamanetworkopen.2020.12887
  • Gater, R., de Almeida e Sousa, B., Barrientos, G., Caraveo, J., Chandrashekar, C. R., Dhadphale, M., Goldberg, D., al Kathiri, A. H., Mubbashar, M., Silhan, K., Thong, D., Torres-Gonzales, F., & Sartorius, N. (1991). The pathways to psychiatric care: A cross-cultural study. Psychological Medicine, 21(3), 761–774. https://doi.org/10.1017/S003329170002239X
  • Haggerty, J. L., Reid, R. J., Freeman, G. K., Starfield, B. H., Adair, C. E., & McKendry, R. (2003). Continuity of care: A multidisciplinary review. BMJ, 327(7425), 1219–1221. https://doi.org/10.1136/bmj.327.7425.1219
  • Kane, J. M., Robinson, D. G., Schooler, N. R., Mueser, K. T., Penn, D. L., Rosenheck, R. A., Addington, J., Brunette, M. F., Correll, C. U., Estroff, S. E., Marcy, P., Robinson, J., Meyer-Kalos, P. S., Gottlieb, J. D., Glynn, S. M., Lynde, D. W., Pipes, R., Kurian, B. T., Miller, A. L., . . . Heinssen, R. K. (2016). Comprehensive versus usual community care for first-episode psychosis: 2-year outcomes from the NIMH RAISE Early Treatment Program. American Journal of Psychiatry, 173(4), 362–372. https://doi.org/10.1176/appi.ajp.2015.15050632
  • Katon, W. J., Lin, E. H. B., Von Korff, M., Ciechanowski, P., Ludman, E. J., Young, B., Peterson, D., Rutter, C. M., McGregor, M., & McCulloch, D. (2010). Collaborative care for patients with depression and chronic illnesses. New England Journal of Medicine, 363(27), 2611–2620. https://doi.org/10.1056/NEJMoa1003955
  • Kripalani, S., LeFevre, F., Phillips, C. O., Williams, M. V., Basaviah, P., & Baker, D. W. (2007). Deficits in communication and information transfer between hospital-based and primary care physicians: Implications for patient safety and continuity of care. JAMA, 297(8), 831–841. https://doi.org/10.1001/jama.297.8.831
  • Levesque, J.-F., Harris, M. F., & Russell, G. (2013). Patient-centred access to health care: Conceptualising access at the interface of health systems and populations. International Journal for Equity in Health, 12, Article 18. https://doi.org/10.1186/1475-9276-12-18
  • Lund, C., Brooke-Sumner, C., Baingana, F., Baron, E. C., Breuer, E., Chandra, P., Haushofer, J., Herrman, H., Jordans, M., Kieling, C., Medina-Mora, M. E., Morgan, E., Omigbodun, O., Tol, W., Patel, V., & Saxena, S. (2018). Social determinants of mental disorders and the Sustainable Development Goals: A systematic review of reviews. The Lancet Psychiatry, 5(4), 357–369. https://doi.org/10.1016/S2215-0366(18)30060-9
  • National Institute for Health and Care Excellence. (2016). Transition between inpatient mental health settings and community or care home settings (NICE Guideline NG53). https://www.nice.org.uk/guidance/NG53/chapter/recommendations
  • Patel, V., Weiss, H. A., Chowdhary, N., Naik, S., Pednekar, S., Chatterjee, S., De Silva, M. J., Bhat, B., Araya, R., King, M., Simon, G., Verdeli, H., & Kirkwood, B. R. (2010). Effectiveness of an intervention led by lay health counsellors for depressive and anxiety disorders in primary care in Goa, India (MANAS): A cluster randomised controlled trial. The Lancet, 376(9758), 2086–2095. https://doi.org/10.1016/S0140-6736(10)61508-5
  • Proctor, E., Silmere, H., Raghavan, R., Hovmand, P., Aarons, G., Bunger, A., Griffey, R., & Hensley, M. (2011). Outcomes for implementation research: Conceptual distinctions, measurement challenges, and research agenda. Administration and Policy in Mental Health and Mental Health Services Research, 38(2), 65–76. https://doi.org/10.1007/s10488-010-0319-7
  • Stanley, B., Brown, G. K., Brenner, L. A., Galfalvy, H. C., Currier, G. W., Knox, K. L., Chaudhury, S. R., Bush, A. L., & Green, K. L. (2018). Comparison of the Safety Planning Intervention with follow-up vs usual care of suicidal patients treated in the emergency department. JAMA Psychiatry, 75(9), 894–900. https://doi.org/10.1001/jamapsychiatry.2018.1776
  • Thornicroft, G., & Tansella, M. (2013). The balanced care model for global mental health. Psychological Medicine, 43(4), 849–863. https://doi.org/10.1017/S0033291712001420
  • Unützer, J., Katon, W., Callahan, C. M., Williams, J. W., Jr., Hunkeler, E., Harpole, L., Hoffing, M., Della Penna, R. D., Noël, P. H., Lin, E. H. B., Areán, P. A., Hegel, M. T., Tang, L., Belin, T. R., Oishi, S., & Langston, C. (2002). Collaborative care management of late-life depression in the primary care setting: A randomized controlled trial. JAMA, 288(22), 2836–2845. https://doi.org/10.1001/jama.288.22.2836
  • Wang, P. S., Aguilar-Gaxiola, S., Alonso, J., Angermeyer, M. C., Borges, G., Bromet, E. J., Bruffaerts, R., de Girolamo, G., de Graaf, R., Gureje, O., Haro, J. M., Karam, E. G., Kessler, R. C., Kovess, V., Lane, M. C., Lee, S., Levinson, D., Ono, Y., Petukhova, M., . . . Wells, J. E. (2007). Use of mental health services for anxiety, mood, and substance disorders in 17 countries in the WHO World Mental Health Surveys. The Lancet, 370(9590), 841–850. https://doi.org/10.1016/S0140-6736(07)61414-7
  • World Health Organization. (2021). Guidance on community mental health services: Promoting person-centred and rights-based approaches. World Health Organization. https://www.who.int/publications/i/item/9789240025707
  • World Health Organization. (2022). World mental health report: Transforming mental health for all. World Health Organization. https://www.who.int/publications/i/item/9789240049338
  • World Health Organization. (2025). Mental health atlas 2024. World Health Organization. https://www.who.int/publications/i/item/9789240114487