Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Rehabilitation Service Transitions, Access, and Implementation

Abstract. Rehabilitation service transitions are organizational and experiential processes through which individuals move among acute care, inpatient rehabilitation, outpatient care, home-based support, community services, education, employment, and long-term participation. This research monograph examines how continuity, service accessibility, implementation conditions, and institutional accountability shape the opportunity to achieve and sustain functional goals. It integrates the International Classification of Functioning, Disability and Health (ICF), multidimensional access theory, continuity-of-care research, early supported discharge trials, transitional care experiments, health equity research, and implementation science. The central synthesis is that a successful transition requires a functioning pathway: a clinically appropriate plan, intelligible transfer of information, reliable receiving services, accessible material and social environments, and continuing revision of goals with the person. Hospital readmission, length of stay, functional status, participation, patient experience, equity, implementation fidelity, and sustainability represent different outcomes and require distinct measurements. Evidence is strongest for selected coordinated stroke early supported discharge services and more modest for general discharge planning; broad extrapolation to every impairment or service setting remains unwarranted. The review develops a methodological agenda for longitudinal, participatory, and equity-sensitive evaluation of rehabilitation pathways. It is a critical narrative research synthesis, not a prospectively registered systematic review or meta-analysis.

Keywords: rehabilitation psychology; transitions of care; rehabilitation services; continuity of care; early supported discharge; access to health care; disability equity; integrated care; implementation science; ICF; community participation; rehabilitation outcomes; service delivery; patient-centered care.

Introduction: The Rehabilitation Pathway as a Scientific Object

A rehabilitation intervention can demonstrate efficacy in a clinic while the service system fails to deliver it at the point where a person needs it. The gap arises through referral delays, discontinuous therapy, poor communication, transportation barriers, affordability, inaccessible environments, and the absence of a receiving team. These conditions alter what an individual can do after leaving an institution, how rehabilitation goals are maintained, and whether therapeutic gains become meaningful participation in everyday life. The scientific subject of this monograph is the rehabilitation service pathway: the ordered yet revisable set of organizations, relationships, resources, decisions, and actions through which a person gains access to rehabilitative support across settings and time.

The World Health Organization's Rehabilitation 2030 initiative (https://www.who.int/initiatives/rehabilitation-2030/) situates rehabilitation within health systems rather than at the margins of specialty services. Its priorities encompass service integration, universal health coverage, the multidisciplinary workforce, information systems, and equitable community access. These priorities are especially consequential when a person moves from an intensive inpatient environment to settings in which medical supervision, family labor, housing, transport, and local public services are differently organized. The transition is therefore a change in the entire opportunity structure of recovery. Source 22

Rehabilitation psychology contributes a distinctive question: how do service conditions interact with cognition, affect, self-efficacy, identity, environmental fit, and social participation? A transfer plan may be technically accurate yet psychologically inaccessible when instructions cannot be understood, an individual's priorities were excluded, or a caregiver cannot assume the work silently assigned to them. Conversely, a respectful and comprehensible service encounter cannot compensate for absent therapy, unavailable equipment, or unaffordable transportation. Explanation must accommodate interaction across personal, organizational, and structural levels.

This monograph treats patients and disabled persons as participants with variable preferences, capabilities, and contexts. It does not assume that every person must recover an earlier level of functioning, that discharge is a uniform endpoint, or that independence means the absence of assistance. Its organizing question is whether the transition preserves meaningful access to supports chosen in light of the person's goals and functioning.

Scope, Research Questions, and Review Method

The review focuses on service transitions among acute care, rehabilitation hospitals, inpatient units, outpatient departments, home rehabilitation, primary care, social care, and community participation services. Psychological treatment, neurological recovery, individual goal-setting, and assistive technology appear where they illuminate service access or implementation, while their intervention-specific mechanisms remain the province of neighboring research dossiers. Mental health service ecology, generic digital health, and general health service implementation are treated as intersecting literatures rather than substitutes for the rehabilitation-specific object.

Five questions guide the synthesis. What components make a rehabilitation transition continuous from the person's perspective? Which theories distinguish access from simple service use? What does comparative evidence establish about coordinated discharge and follow-up? Why do evidence-based service models fail to become routine practice? Which measurements permit evaluation of functioning, participation, equity, implementation, and unintended effects over time?

The method is a critical narrative synthesis using original randomized studies, systematic reviews, conceptual and measurement papers, professional guidelines, and international policy documents. Priorities in source selection are original authorship, identifiable DOI or official permanent URL, explicit design, transparent population, and relevance to at least one research question. Evidence from general hospital transitions is presented as context rather than automatically attributed to rehabilitation populations. WHO, NICE, Cochrane, implementation science, and rehabilitation-specific stroke evidence are read in relation to each other. Sources were checked through the bibliographic records or openly accessible text identified in References; information available only in an abstract is interpreted at that level.

This is not a systematic review conducted under a prospective protocol. It does not assert exhaustive database coverage, independent duplicate screening, formal study-by-study risk-of-bias scoring, or pooled effect estimation. The distinction matters because a transparent synthesis can identify mechanisms and research questions without claiming the inferential completeness of a meta-analysis. Numerical estimates are limited to outcomes verifiable in original abstracts or published review summaries, with the population and comparator named. The MRC framework for complex interventions (https://doi.org/10.1136/bmj.n2061) and StaRI reporting guidance (https://doi.org/10.1136/bmj.i6795) inform the methodological analysis.

Historical Formation of Rehabilitation Services

The history of rehabilitation services involves changes in the meaning of disability, the organization of specialized medicine, the relationship between hospital and community, and the status of the person receiving care. Institutional traditions concentrated professional resources in specialized facilities; community-oriented approaches extended attention to the conditions in which people live, work, travel, and maintain relationships. Their coexistence remains visible in contemporary systems, where some interventions are delivered by highly specialized teams and others depend on primary care, social services, educational settings, and family support.

A formal turning point was the WHO International Classification of Functioning, Disability and Health (https://www.who.int/standards/classifications/international-classification-of-functioning-disability-and-health), endorsed by the World Health Assembly in May 2001. The ICF organizes health-related functioning through body functions and structures, activities, participation, and environmental factors. It provides an important common language across professions while leaving the individual's own evaluation of meaningful life outcomes to appropriately chosen measures and shared deliberation. The ICF has made it more difficult to equate a completed hospital episode with a completed rehabilitation process. Source 21

Health service research developed complementary concepts. Penchansky and Thomas (1981) (https://doi.org/10.1097/00005650-198102000-00001) conceptualized access as fit between clients and health services along five dimensions. Andersen (1995) (https://pubmed.ncbi.nlm.nih.gov/7738325/) revisited a behavioral model emphasizing the relationship among predisposing, enabling, and need factors and service use. Haggerty and colleagues (2003) (https://doi.org/10.1136/bmj.327.7425.1219) integrated multidisciplinary traditions on continuity. Together these developments moved inquiry beyond the quantity of services supplied toward the processes by which people reach, understand, receive, and sustain care. Source 1

The transition from concept to tested intervention became visible in randomized discharge and care-coordination trials in the 1990s and 2000s, including Naylor et al. (1999) (https://doi.org/10.1001/jama.281.7.613) and Coleman et al. (2006) (https://doi.org/10.1001/archinte.166.17.1822). These trials addressed service-level organization rather than simply testing an additional clinical treatment. Stroke early supported discharge research subsequently established an especially relevant rehabilitation-specific literature because it compared organized home-based continuation with conventional inpatient pathways.

WHO's 2019 guide for action (https://www.who.int/publications-detail-redirect/9789241515986), its 2023 Package of Interventions for Rehabilitation (https://www.who.int/publications/i/item/9789240067097), and the 2022 global disability health-equity report (https://www.who.int/publications/i/item/9789240063600) reflect a further shift: rehabilitation is evaluated as a public health system capacity with distributive, workforce, and quality implications. These documents establish planning directions, while controlled trials and observational designs remain necessary to determine the effects of particular delivery models. Source 23 Source 25 Source 24

Functioning, Participation, and the Ecological Unit of Analysis

The ICF is essential because a service pathway can influence outcomes without directly changing pathology. A functioning limitation can arise from an impairment interacting with inaccessible transport, housing, communication, policy, or service organization. A person whose motor capacity improves during inpatient treatment may still be unable to participate in work or community life when outpatient services are too distant or the environment remains inaccessible. Consequently, service evaluation requires both measures of clinical change and measures of performance and participation in ordinary settings.

The unit of analysis can be the person, the episode of care, the transition between providers, the multidisciplinary team, the service network, or the jurisdiction. Each unit answers a different causal question. Individual outcome studies assess change in function or experience. Process evaluations examine referral completion, data transfer, goal continuity, and availability of responsible staff. Organizational studies evaluate staffing, adoption, fidelity, and cooperation. Population studies assess the distribution of need and access. Confusing these units invites ecological fallacies and misplaced claims of efficacy.

Functional measurement should also distinguish capacity, performance, and perceived participation. Standardized tests conducted under controlled conditions show what a person can do in that setting. Community performance depends on opportunities and environmental demands. Patient-reported outcome measures reveal dimensions that clinical performance measures may miss. The WHO framework supplies domains for these comparisons, while no single classification provides a complete assessment of autonomy, belonging, or personal priorities.

The UN Convention on the Rights of Persons with Disabilities, Article 26 (https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities/article-26-habilitation-and-rehabilitation.html), explicitly links habilitation and rehabilitation to autonomy, full participation, multidisciplinary assessment, and local service availability. In research design this normative commitment encourages attention to who defines outcomes, whose environments are made usable, and whose needs are excluded by conventional eligibility rules. Rights commitments and efficacy estimates answer distinct questions; both matter to evidence-informed system design. Source 20

What Counts as a Service Transition?

A service transition is a change in the organization, setting, provider responsibility, or practical accessibility of rehabilitation support. It may accompany discharge from acute hospital care to inpatient rehabilitation; transfer from inpatient therapy to home and outpatient services; movement from childhood to adult programs; transition from publicly funded to privately financed services; or a change from time-limited treatment to community assistance. The most visible event is a transfer date, but the relevant process extends through preparation, arrival, engagement, and follow-up.

Transitions differ in complexity. A direct transfer within a unified organization may retain records and staff relationships. A transfer across institutions or sectors often requires new eligibility assessments, financing authorization, transport, scheduling, language accommodation, consent, and communication systems. Long-term transitions may require renewed goals as social roles and circumstances change. Analyses that treat all transfers as equivalent hide the mechanisms that cause failure.

The handoff has multiple components: identification of current functioning and risks; communication of rehabilitation goals and equipment needs; designation of a receiving professional; confirmation that a referral was accepted; coordination with the person and caregivers; and verification that the first relevant contact occurred. A completed referral is an administrative event. Effective access additionally requires an available and acceptable service that the person can use and that can address the reason for referral. This is consistent with the process-sensitive account proposed by Levesque, Harris, and Russell (2013) (https://doi.org/10.1186/1475-9276-12-18).

A transition can also fail after initial engagement. Therapy may end prematurely, goals may be revised without the person's participation, transportation may become unavailable, or follow-up may fail to identify deterioration. Accordingly, pathway research should define a time horizon and distinguish initiation, sustained access, treatment intensity, functional response, and meaningful participation. These are separately observable outcomes rather than interchangeable names for successful discharge.

Continuity: Informational, Management, and Relational

Haggerty et al. (2003) (https://doi.org/10.1136/bmj.327.7425.1219) described continuity of care across informational, management, and relational dimensions. Informational continuity means that clinically relevant history and personal circumstances are available and used over time. Management continuity concerns coherent care plans and coordinated responses to changing needs. Relational continuity concerns an ongoing therapeutic relationship or an intelligible succession of responsible professionals. Rehabilitation frequently requires all three, especially where treatment spans disciplines and social settings.

Information transfer alone cannot establish clinical continuity. A discharge summary may reach a clinic while its recommendations conflict with available services, the referral is rejected, or the new team lacks staff. Likewise, a single familiar provider may be valuable but unable to coordinate occupational therapy, psychology, social work, nursing, equipment provision, and community services. Management continuity involves explicit interprofessional responsibilities and an actionable plan. Relational continuity adds trust, the ability to disclose difficulties, and confidence that someone remains accountable when the plan changes.

Continuity should be measured from both system and recipient perspectives. System measures include record transfer, referrals accepted, time to follow-up, missed appointments, and responsible coordinator assignment. Person-reported measures examine whether the individual understands whom to contact, experiences consistent goal-setting, can participate in decisions, and feels that important information was carried forward. A service may demonstrate excellent data transmission while failing these experiential criteria.

Such distinctions also refine intervention evaluation. A navigation program could improve appointment completion without improving function; a multidisciplinary pathway could improve function despite a modest effect on readmission; and a well-received relationship could remain constrained by finance or geography. The appropriate inference requires measurements aligned with the mechanism being tested.

Access as a Multidimensional Process

Access is an evolving relationship between service conditions and people's abilities and resources. The classic taxonomy developed by Penchansky and Thomas (1981) (https://doi.org/10.1097/00005650-198102000-00001) differentiated availability, geographic accessibility, accommodation, affordability, and acceptability. The framework directs attention to dimensions that are particularly acute for disabled people: a geographically nearby clinic can be inaccessible because the building has steps; an available appointment can be unusable without accessible transportation; and a nominally affordable intervention can impose substantial caregiver or equipment costs.

Levesque, Harris, and Russell (2013) (https://doi.org/10.1186/1475-9276-12-18) integrated the service side with users' abilities to perceive needs, seek care, reach providers, pay, and engage. Their corresponding service dimensions are approachability, acceptability, availability and accommodation, affordability, and appropriateness. This framework is valuable for rehabilitation because access failures arise at successive points. A person may lack information that a service exists, encounter burdensome entry criteria, travel successfully yet find no appropriate therapist, or disengage because the intervention is incomprehensible or conflicts with priorities.

The Andersen (1995) model (https://pubmed.ncbi.nlm.nih.gov/7738325/) places utilization within individual and contextual factors. Combining these perspectives identifies the need for both demand-side and supply-side measures. Clinical need is an important determinant of use, but observed use also reflects systems of eligibility, transportation, communication, financing, stigma, trust, and prior experiences. A low-use group may have low need, substantial unmet need, or a mixture; utilization alone cannot decide which explanation holds. Source 1

Access analysis should therefore use meaningful denominators. The number of people receiving rehabilitation has little interpretive value without estimating who has rehabilitation needs, who is referred, who is eligible, who can attend, and who continues treatment. Stratification by disability characteristics, geography, age, income, language, race or ethnicity where ethically appropriate, and service coverage helps expose differences that aggregate rates obscure. Access and equity require attention to the conditions that create the observed distributions, not merely their description.

The Demand for Rehabilitation and the Distribution of Services

The scale of the rehabilitation need is substantial, but estimates of potential benefit are not equivalent to observed demand or service availability. Cieza et al. (2020) (https://doi.org/10.1016/S0140-6736(20)32340-0), using Global Burden of Disease 2019 estimates, reported approximately 2.41 billion individuals worldwide living with health conditions that could benefit from rehabilitation, with a 95% uncertainty interval of 2.34–2.50 billion. This measure identifies modeled potential need for selected conditions, not the number of people referred, receiving care, waiting for services, or likely to benefit from a particular program. It is nevertheless powerful evidence against treating rehabilitation as a rare specialist requirement.

The WHO Rehabilitation 2030 initiative (https://www.who.int/initiatives/rehabilitation-2030/) and 2019 health-systems guide (https://www.who.int/publications-detail-redirect/9789241515986) connect this need to systems that must establish workforce capacity, accessible services, financing, information, and accountability. Their emphasis on strengthening rehabilitation across health care is important because service capacity is unequally distributed among hospital centers, smaller municipalities, rural communities, and low-resource systems. A highly specialized service in a metropolitan hospital may serve people who can reach it while remaining inaccessible to others with equally substantial functional needs. Source 22 Source 23

Estimating effective service coverage requires several stages: population need, identification, referral, appointment availability, initiation, retention, adequate intensity, and achieved outcomes. Each stage can introduce selection. Waiting lists represent people already detected by the system and may exclude those never identified. Recorded treatment completion can systematically exclude people experiencing barriers to attendance. Evaluations should therefore distinguish the reach of a service among all potentially eligible people from outcomes among those who completed treatment.

The WHO Package of Interventions for Rehabilitation (2023) (https://www.who.int/publications/i/item/9789240067097) describes essential interventions and associated personnel, assistive products, and equipment requirements across conditions. Its planning utility depends on matching identified interventions to local capacity. The existence of an evidence-based intervention in a guidance document does not by itself establish staff training, financing, community transportation, referral integration, or service accountability. Rehabilitation need becomes an implementation problem precisely at that point. Source 25

Handoffs, Information Transfer, and Discharge Communication

Information transfer is one of the best documented weaknesses of transitions between care settings. The review by Kripalani et al. (2007) (https://doi.org/10.1001/jama.297.8.831) examined 55 observational studies of discharge communication and 18 controlled studies of interventions. It reported low availability of discharge summaries at the first postdischarge visit and substantial variability in content and timeliness. These findings concern hospital-to-primary-care communication broadly; their specific proportions should not be treated as contemporary rehabilitation-system estimates. They nevertheless establish a mechanism plausibly relevant to multidisciplinary rehabilitation.

Rehabilitation handoffs are unusually information-intensive because treatment decisions depend on more than a diagnosis and medication list. Relevant materials include current functioning, activity limitations, cognitive and communication abilities, psychological needs, rehabilitation goals, adaptations, assistive devices, caregiver capacities, environmental risks, and plans for ongoing therapy. The 2023 NICE stroke rehabilitation guideline (https://www.nice.org.uk/guidance/ng236/chapter/Recommendations) specifies a rich transfer-information set covering clinical, functional, psychological, social, and follow-up requirements. It also recommends procedures for timely communication between providers.

A standardized template can improve completeness, but its utility depends on whether the recipient can act on it. A transfer packet that recommends unavailable services is informationally complete yet operationally incomplete. A discharge record copied through an electronic system may still contain outdated goals or ambiguous responsibility. The receiving team should be identifiable, the plan should specify immediate actions, and the person and supporters should have accessible information about the next contact and what to do when needs change.

Research should move beyond binary measures such as whether a discharge letter was sent. More informative outcomes include receipt before the first visit, accuracy of therapy goals, agreement between sending and receiving teams, patient comprehension, documented acceptance of the referral, appropriate equipment availability, and time to first delivered intervention. Record audits and interviews can be combined to detect discrepancies between formally closed and practically unresolved transitions.

The Psychological Work of Leaving an Institution

Discharge is a psychological transition as well as a change in service location. Institutional environments provide routines, predictable access to specialized staff, and externally organized schedules. Community settings offer opportunities to resume valued roles while transferring many coordination tasks to individuals and their supporters. A person's expectations, uncertainty, perceived control, fatigue, cognitive capacity, and confidence in communicating needs can shape how that change unfolds. These features must be studied without presuming a uniform emotional response to discharge.

Rehabilitation plans frequently require activities that demand memory, executive function, communication, motivation, and physical endurance. After neurological injury or other disabling conditions, some of those capacities may themselves be under rehabilitation. A system that relies on independent appointment management, reading complex written instructions, arranging travel, and recognizing clinical deterioration can impose demands beyond the person's present functioning. This mismatch is a service-design characteristic and should be measured directly, rather than being assigned solely to patient adherence.

Continuity can support perceived agency when people understand the next stage, participate in decisions, have reliable contacts, and see their personal priorities preserved across providers. The relationship between such conditions and outcomes is mediated by service resources, impairment-specific needs, family circumstances, and wider access barriers. Psychological engagement is one mechanism in a larger pathway, not an explanation for every variation in rehabilitation success. The conceptual distinction follows the functioning framework of the WHO ICF (https://www.who.int/standards/classifications/international-classification-of-functioning-disability-and-health) and the user-side dimensions of access described by Levesque et al. (2013) (https://doi.org/10.1186/1475-9276-12-18). Source 21

Patient interviews can identify transition burdens that routine administrative data omit. Relevant questions include whether goals were negotiated, whether instructions were understandable, whether the service respected communication preferences, whether another professional could be reached, and whether opportunities for valued activity actually expanded. Longitudinal qualitative inquiry is especially useful because the meaning of recovery and acceptable assistance can change over months and years.

Family, Caregivers, and the Distribution of Responsibility

Family members and informal supporters often participate in transport, medication management, exercises, equipment setup, communication, personal assistance, emotional support, and coordination between services. Their participation can be valuable, but a plan that assumes unlimited caregiver availability risks translating organizational shortages into unpaid household labor. Capacity varies according to employment, economic resources, health, relationships, skill, and willingness. Reliable rehabilitation planning therefore evaluates what support is available, which tasks it will involve, and whether training, respite, or professional services are needed.

Caregiver involvement also requires respect for the person's preferences and legal autonomy. Some individuals want strong family involvement; others prefer a different support network or privacy. Supported decision-making, accessible communication, and explicit consent can prevent caregivers from becoming default substitutes for the person's voice. The Convention on the Rights of Persons with Disabilities, Article 26 (https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities/article-26-habilitation-and-rehabilitation.html) links rehabilitation to independence and participation, both compatible with chosen forms of interdependence and support. Source 20

The NICE stroke rehabilitation guidance (https://www.nice.org.uk/guidance/ng236/chapter/Recommendations) calls for information, training where family members and carers are willing and able to help, and recognition that their support needs may change. This is a condition-specific clinical guideline rather than a universal protocol for all rehabilitation populations. Its design principle is broadly relevant: assigning a task to a caregiver should involve evaluation of capability, consent, practical conditions, and continuing support.

Evidence about caregiver outcomes requires independent measurement. Patient functional improvement does not prove that caregiver burden fell. A program may shorten inpatient stay yet increase the number of unpaid care hours, household expenditures, or stress. Research should collect caregiver strain, opportunity costs, satisfaction, training adequacy, perceived safety, and the distribution of work, with attention to selection bias when caregivers who are most burdened are least able to participate in follow-up surveys.

Multidisciplinary Coordination and Service Integration

Rehabilitation commonly involves physicians, nurses, rehabilitation psychologists, neuropsychologists, physical therapists, occupational therapists, speech-language pathologists, social workers, rehabilitation assistants, and professionals providing equipment and community support. The appropriate composition varies across health conditions and settings. Multidisciplinary practice becomes integrated when these contributors coordinate goals, share actionable information, and adjust responsibilities as needs evolve. Merely listing several professions on a discharge summary does not establish integrated care.

The scientific distinction is between co-location, communication, and coordinated action. Co-location may improve opportunities for exchange; communication transmits information; coordinated action requires agreement about priorities, timing, resource responsibility, and escalation when a service fails. A transfer of care can contain all the expected professional names while lacking a person who verifies whether the next stage actually begins. A useful organizational measure is therefore closed-loop referral completion, including confirmation by the receiving provider and the individual.

Coordination operates across formal and informal systems. Housing adaptations, disability benefits, transport, workplace accommodations, school supports, and social care affect the feasibility of clinical recommendations. Rehabilitation managers may influence these systems indirectly, but no single clinical team ordinarily controls them. Network governance, case management, negotiated interagency arrangements, and service contracts can become more important than the skill of an individual provider. The WHO guidance on integrating rehabilitation into health systems (https://www.who.int/health-topics/rehabilitation/integrating-rehabilitation-into-health-systems) highlights governance, financing, service delivery, workforce, assistive technology, and information as mutually implicated system components. Source 26

Theoretical models should specify a mechanism before assigning an outcome to coordination. Shared records might reduce duplication; a named coordinator might reduce unresolved referrals; joint goal-setting might improve perceived alignment; and rapid community therapy might preserve functional gains. A study that simply compares organizations labeled integrated and usual care may combine all these features, making it difficult to identify which component caused an effect. Process measures and implementation fidelity are necessary to interpret composite service interventions.

Early Supported Discharge After Stroke

Early supported discharge (ESD) is a central rehabilitation-specific experimental model. It offers appropriately selected people the possibility of leaving inpatient care earlier while continuing coordinated rehabilitation with a specialist multidisciplinary team in the community. Its evaluative importance is that the intervention changes the location and organization of care while seeking to maintain effective rehabilitation. Earlier departure alone is not equivalent to ESD: the continuation, expertise, coordination, and intensity of services are integral to the model.

The Cochrane review by Langhorne, Baylan, and the Early Supported Discharge Trialists (2017) (https://doi.org/10.1002/14651858.CD000443.pub4) included 17 trials involving 2,422 participants, generally selected stroke survivors with moderate disability. Hospital stay was shortened by a mean of 5.5 days according to the published summary (reported 95% CI approximately 3–8 fewer days). For death or dependency at scheduled follow-up, the odds ratio was 0.80 (95% CI 0.67–0.95), based on 16 trials and 2,359 participants. The review characterized this evidence as moderate certainty. Its subgroup analyses suggested that coordinated multidisciplinary teams were especially important; weaker evidence supported service models lacking such organization.

The review also found no clear difference in death alone, patient mood, or health status, and low-certainty evidence did not establish a difference in readmission. Thus ESD evidence supports a particular pattern of outcomes, not universal superiority on every measure. The participating populations and clinical selection criteria restrict generalization to people with severe disability, unstable medical conditions, minimal home support, or different rehabilitation diagnoses. Cross-country variation in usual care may further alter the incremental benefit of providing a specialist community team.

The 2023 NICE guideline NG236 (https://www.nice.org.uk/guidance/ng236/chapter/Recommendations) recommends appropriately assessed ESD with continuing multidisciplinary support and clear transfer planning. The 2025 Canadian Stroke Best Practice Recommendations (https://pmc.ncbi.nlm.nih.gov/articles/PMC12871428/) similarly emphasize coordinated specialist community teams, continued therapy, and timely commencement. Guidelines specify implementation conditions and clinical judgments, whereas the RCT synthesis estimates effects under studied conditions. Their combined implication is that service transition quality depends on what replaces inpatient care, not simply on the speed of discharge.

Discharge Planning and Transitional Care Interventions

Discharge planning is a broader class of intervention than rehabilitation-specific ESD. It includes individualized identification of care needs, preparation for transfer, and arrangement of postdischarge support. In its 2022 Cochrane review (https://doi.org/10.1002/14651858.CD000313.pub6), Gonçalves-Bradley and colleagues included 33 trials of individualized discharge planning versus usual discharge care. The published synthesis concluded that structured plans probably produce small reductions in hospital length of stay and readmissions for relevant hospital populations, while evidence about health status and cost remained limited or uncertain.

The difference between a modest aggregate effect and a specific successful program matters. Naylor et al. (1999) (https://doi.org/10.1001/jama.281.7.613) conducted an RCT of comprehensive discharge planning and nurse-led home follow-up among 363 older patients at risk of readmission. At 24 weeks, at least one readmission occurred in 20.3% of intervention participants and 37.1% of controls, according to the published trial abstract. The same abstract reported no statistically significant differences in functional status, depression, or satisfaction. The enrolled population was older medical and surgical patients, so these findings illustrate transitional care organization without constituting direct evidence of effectiveness for every specialized rehabilitation condition.

Coleman et al. (2006) (https://doi.org/10.1001/archinte.166.17.1822) tested a care transitions intervention that emphasized active involvement of patients and caregivers, medication management, records, follow-up, and recognition of warning signs. Its randomized design was influential in treating transitions as a focused intervention target. Again, the mechanism and target population must be preserved when extrapolating to rehabilitation: care coordination is relevant, but rehabilitation also requires discipline-specific intensity, assistive support, and activity-oriented goals.

These studies indicate why a general discharge checklist may be insufficient for complex rehabilitation trajectories. A plan is only one component of a pathway. Effects on readmission and hospital stay do not establish effects on community participation, confidence, return to work, or environmental accessibility. Trials should prospectively distinguish these outcomes, specify the standard-care comparator, and quantify intervention uptake and resource inputs.

Clinical Guidelines and Conditions for Safe Transitions

Clinical guidance can identify minimum service conditions when the evidence has already been translated into formal recommendations. NICE NG236 (2023) (https://www.nice.org.uk/guidance/ng236/chapter/Recommendations) calls for specialist stroke rehabilitation across hospital and community services and continuity for as long as support remains useful for agreed goals. Its recommendations on discharge address safe and enabling home environments, appropriate equipment, shared health and social care plans, communication of psychological and cognitive needs, caregiver circumstances, and access to follow-up.

The guideline also recommends follow-up within 72 hours after hospital-to-community transfer by a specialist stroke rehabilitation team for people who continue to have rehabilitation needs. The Canadian Stroke Best Practice Recommendations (2025) (https://pmc.ncbi.nlm.nih.gov/articles/PMC12871428/) set out their own service-timing recommendations for early supported discharge and emphasize local specialist team readiness. Exact timing is guideline-, condition-, and jurisdiction-specific; it must not be reinterpreted as a universal target for all rehabilitation services.

Eligibility criteria require attention to both individual and environmental factors. A person may be medically ready for transfer but lack appropriate equipment, transport, accessible accommodation, or community staffing. Conversely, using restrictive eligibility criteria to exclude those with complex needs can make a program appear efficient while widening inequities. Measurement should examine the characteristics of people judged unsuitable for the pathway and whether alternate services are provided.

Because clinical service decisions depend on the person's diagnosis, stability, goals, support, and local system, this monograph is an analysis of research and delivery frameworks. It does not prescribe a discharge date, therapy intensity, or individualized care regimen. Implementation of condition-specific clinical recommendations belongs to qualified clinical teams operating within their local responsibilities and guidelines.

Equity, Disability Rights, and Unequal Access

Rehabilitation service systems distribute opportunities across populations. That distribution is shaped by who is recognized as needing care, which conditions are covered, where providers are located, who can pay, whose communication needs are met, and which outcomes institutions are rewarded for achieving. Equity analysis studies the mechanisms generating those differences. Simply reporting different service-use rates does not establish whether groups have comparable needs or whether a disparity represents unequal access, but it offers a starting point for testing those questions.

The WHO Global Report on Health Equity for Persons with Disabilities (2022) (https://www.who.int/publications/i/item/9789240063600) documents structural conditions associated with poorer health and barriers within health systems. The UN Convention on the Rights of Persons with Disabilities, Article 26 (https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities/article-26-habilitation-and-rehabilitation.html) articulates an obligation to organize and strengthen rehabilitation, including community availability, individualized assessment, participation, and voluntary involvement. Rights-based commitments identify what systems should make possible. Effectiveness research assesses whether particular policies and delivery arrangements achieve those goals under identifiable conditions. Source 24 Source 20

Many rehabilitation transitions involve compounded disadvantage. A person with cognitive or communication impairment may face a referral process requiring inaccessible forms and unsupported telephone communication. Someone living in a rural area may receive nominal coverage but have no accessible transportation to the approved provider. An individual with complex psychosocial needs can fall between systems whose eligibility rules address only medical impairment. Such examples articulate plausible causal pathways and must be examined with local evidence rather than presented as uniformly established prevalence estimates.

Equity-oriented evaluations should report care need, referral, eligibility, initiation, time to service, retention, treatment dose, participant-selected outcomes, and reasons for discontinuation by relevant subgroups. They should also evaluate adverse consequences of policy changes. Moving care from a hospital to the community may increase geographic reach for some people while shifting transportation or caregiver costs onto others. Methodological care is required to avoid attributing effects of unequal service systems to stable characteristics of the people experiencing those systems.

Financing, Transportation, Geography, and Workforce

Rehabilitation delivery depends on material resources. Workforce availability, professional distribution, reimbursement policy, travel time, transport accessibility, scheduling, communication technologies, and equipment supply condition what can be offered and to whom. An intervention with strong clinical efficacy may have limited impact when clinics lack trained personnel, clients cannot attend consistently, or funding ends before functional and social goals can be addressed.

Geography should be analyzed as effective travel burden rather than distance alone. The relevant quantity may include accessible transit availability, travel cost, physical transfers, caregiver accompaniment, appointment frequency, and the temporal burden imposed by recovery-related fatigue. Rural service provision can be supported by outreach, itinerant specialists, home visits, and digital modes when appropriate, but no delivery mode automatically resolves the human and environmental constraints of access. Availability, affordability, acceptability, and appropriateness remain separate domains under the Levesque et al. framework (2013) (https://doi.org/10.1186/1475-9276-12-18).

Financing influences both service supply and selection. Episode-based payment may reward short stays; professional fee schedules may reward discrete visits; bundled or population-based arrangements can create different incentives for coordination. These are institutional mechanisms requiring empirical evaluation rather than universally beneficial or harmful models. An economic comparison must identify who pays and who benefits. Reduced inpatient expenditure can coexist with increased caregiver time, equipment costs, or unmet needs. Measuring total societal and household costs is particularly relevant for a transition framed as efficient.

Workforce planning is not reducible to clinician headcount. Teams need skills across conditions, accessible communication, cultural responsiveness, interdisciplinary work, and continuity protocols. The WHO rehabilitation health-systems guide (2019) (https://www.who.int/publications-detail-redirect/9789241515986) and 2023 Package of Interventions (https://www.who.int/publications/i/item/9789240067097) connect evidence-based interventions to resource requirements. Implementation studies should record staffing ratios, turnover, caseload, training, supervision, and treatment capacity so that an observed service effect can be understood and potentially reproduced. Source 23 Source 25

Implementation Science: Determinants and Mechanisms

Implementation science studies how evidence-based interventions become embedded in routine systems, under what conditions they are adopted, and whether their delivery remains effective and equitable. In rehabilitation, the object may be a clinical intervention, coordinated early supported discharge, closed-loop referral, integrated documentation, or a community multidisciplinary pathway. Scientific evaluation must distinguish the intervention delivered to recipients from the strategies used to implement it.

The original Consolidated Framework for Implementation Research (CFIR) presented by Damschroder et al. (2009) (https://doi.org/10.1186/1748-5908-4-50) synthesized determinants of implementation across domains such as intervention characteristics, inner and outer settings, individuals, and process. The updated CFIR (Damschroder, Reardon, Widerquist, & Lowery, 2022) (https://doi.org/10.1186/s13012-022-01245-0) revised that architecture using practitioner feedback. A determinant framework is useful for asking why one clinic establishes a multidisciplinary discharge pathway while another cannot: staff workload, leadership, professional roles, local financing, referral relationships, compatibility with existing documentation, and patients' circumstances may all influence implementation.

Behavioral mechanisms can be examined with the Theoretical Domains Framework validated by Cane, O'Connor, and Michie (2012) (https://doi.org/10.1186/1748-5908-7-37). Knowledge, skills, professional identity, beliefs about capability, environmental resources, and social influences are relevant when professionals must change established referral or documentation practices. This framework helps specify potential determinants of action. It does not demonstrate which determinant is causally responsible in a particular setting without appropriate research.

The RE-AIM framework of Glasgow, Vogt, and Boles (1999) (https://doi.org/10.2105/ajph.89.9.1322) evaluates reach, effectiveness, adoption, implementation, and maintenance. For a rehabilitation pathway, reach asks what proportion of eligible individuals enter it; adoption asks how many suitable organizations and providers implement it; implementation examines fidelity and resources; maintenance asks whether effects and organizational use persist. A program can show excellent outcomes among enrolled participants while reaching few of those in greatest need.

Implementation strategies should be named concretely. Examples include audit and feedback on missed transitions, supervised team training, protected coordination time, interoperable referral templates, practical patient navigation, redesign of appointment scheduling, and negotiated service agreements. The strategies require separate assessment of feasibility and costs. Labeling a complex program as integrated care without describing its components prevents meaningful replication and weakens causal explanation.

Implementation Outcomes and Their Measurement

Proctor et al. (2011) (https://doi.org/10.1007/s10488-010-0319-7) distinguished implementation outcomes from clinical and service outcomes and specified eight constructs: acceptability, adoption, appropriateness, feasibility, fidelity, implementation cost, penetration, and sustainability. This distinction is especially important for rehabilitation transitions because a new pathway may be operationally feasible yet clinically ineffective, or clinically promising yet unsustainable with available staffing and funding.

Acceptability concerns perceptions among participants and providers. Adoption concerns initial uptake by organizations or professionals. Appropriateness concerns perceived fit for needs and circumstances; feasibility concerns practical deliverability. Fidelity measures the degree to which essential intervention components were delivered. Penetration concerns integration within a service setting or system, while sustainability concerns continued practice over time. Implementation cost includes resources required to establish and maintain the new delivery mode. None of these outcomes should be substituted for measures of activity, participation, quality of life, or clinical safety.

Operational definitions must specify their denominators and timing. Adoption measured as the number of trained clinics differs from adoption measured as the number actively delivering a pathway. Fidelity measured by the presence of a discharge checklist differs from fidelity measured by actual continuity of therapy. A clinic can pass an administrative audit while participants encounter unmet needs after leaving. Careful measurement therefore combines documentation, provider observation, service tracking, and recipient reports.

Mixed-method designs can reveal why outcomes diverge. Quantitative data may identify delays or differential referral completion; interviews and observation can show how staff negotiate roles, why documents are misinterpreted, how families experience transitions, and which aspects of the intervention were adapted. The StaRI statement (Pinnock et al., 2017) (https://doi.org/10.1136/bmj.i6795) encourages transparent reporting of both the implementation strategy and the intervention being implemented, improving the interpretability of claims across settings.

Research Designs and Causal Inference

Causal questions about rehabilitation services differ from questions about individual therapies. The intervention is often delivered by teams or institutions, affects communication networks, and evolves during implementation. Individual randomization may be feasible for some transitions, as in the classic discharge trials, but cluster randomization may be more appropriate when staff behavior or organizational procedures cannot be isolated across participants. A stepped-wedge design can introduce a program sequentially across sites while collecting repeated observations, but secular trends and treatment-time interactions require explicit modeling.

Controlled interrupted time series, difference-in-differences designs, and natural experiments may be useful when systemwide reform rules out random allocation. Their validity depends on credible assumptions about counterfactual trends, timing, co-occurring reforms, patient case mix, and changes in measurement. A before-and-after reduction in readmission does not independently establish that a new rehabilitation program caused the reduction. Staffing expansion, evolving admission practices, and broader epidemiological changes may provide competing explanations.

The MRC framework for complex interventions (Skivington et al., 2021) (https://doi.org/10.1136/bmj.n2061) emphasizes context, program theory, stakeholder engagement, and the interplay among development, feasibility, evaluation, and implementation. In rehabilitation, a well-specified program theory can link named components to intermediate mechanisms: referral tracking reduces uncompleted handoffs; shared goals improve coordination; home-based therapy changes accessibility; and consistent follow-up may support engagement. Each link should be testable and distinguished from assumptions about eventual outcomes.

A strong evaluation should report eligibility criteria, allocation and contamination, local comparison services, implementation dose, staffing, outcome definitions, follow-up duration, missing data, adverse events, and cost perspectives. Patient-reported measures may be informative yet vulnerable to expectation and response effects; administrative data may be more complete yet insensitive to valued participation. Studies that combine methods and predefine primary estimands are better placed to separate genuine benefits from documentation artifacts.

Economic and equity effects deserve prospective inclusion. Programs serving individuals with fewer resources may appear less successful when outcomes are measured without adjusting for structural constraints; adjusting away those constraints may obscure precisely the inequality that policy should address. Multilevel causal models and qualitative analyses should therefore make explicit which variables are confounders, mediators, moderators, or policy targets rather than mechanically controlling for every available covariate.

Functional Outcomes, Participation, and Patient-Reported Experience

A discharge pathway can be judged through many outcomes: time to community therapy, readmission, mortality, complications, functional independence, psychological adjustment, return to work, participation, quality of life, and the experience of service continuity. Their relation is contingent. Shorter hospital stays may be valuable when safe community support exists; the same metric can become misleading if discharge merely transfers burden elsewhere. A low readmission rate does not prove improved participation, and a meaningful gain in social roles may occur without a major change on a narrow impairment scale.

The WHO ICF (https://www.who.int/standards/classifications/international-classification-of-functioning-disability-and-health) organizes domains of body functions, activities, participation, and environment. In studies of rehabilitation pathways, pre-specified functioning measures should capture clinically relevant abilities and real-world performance. Patient-reported participation can incorporate valued social roles, autonomy, and personally meaningful goals. A standardized outcome set aids comparison, while individualized goal measures make visible outcomes important to specific people. Both approaches require attention to reliability, responsiveness, accessibility, and interpretability. Source 21

The 2017 Cochrane early supported discharge review (https://doi.org/10.1002/14651858.CD000443.pub4) illustrates why endpoint distinctions matter. It identified a difference in death or dependency and hospital stay for selected people with stroke, but did not establish uniform benefits in mood or subjective health. The 2022 discharge-planning review (https://doi.org/10.1002/14651858.CD000313.pub6) similarly found clearer evidence for some organizational outcomes than for broad patient health measures. A scientific conclusion must preserve those differences rather than use the general term improved outcomes.

Measurement also requires appropriate timing. Outcomes at discharge capture a particular episode, while six-month or annual follow-up can reveal sustainability, delayed access failures, and role recovery. Evaluation should consider people who die, enter institutions, move jurisdictions, or cannot complete standard assessments. Excluding such individuals without transparent analysis can bias estimates toward successful transitions. Accessible data collection methods and supported communication are part of research validity.

Data Systems, Referral Closure, and Longitudinal Accountability

Service records commonly document referrals, appointments, diagnoses, and encounters but can underrepresent ongoing access and meaningful outcomes. A rehabilitation system can record a referral as sent even when the receiving clinic never accepts it; it may record therapy as delivered without documenting whether the treatment addresses current goals. The solution is to treat transition information as a sequence of verifiable events linked to functional needs and responsibilities.

A robust information model may include a person-selected goal profile, functioning and communication needs, intended service, referral date, receiving organization, acceptance or rejection, first attended contact, delays, delivered intervention, revised goals, and reasons for discontinuation. Such variables should have clear definitions and privacy safeguards. Data exchange must support the person as a participant in their own care rather than reduce them to a passive object transferred across databases.

Accountability should distinguish the professionals responsible for an event. Sending clinicians are positioned to document needs and initiate referral; receiving teams are positioned to confirm appropriateness and service availability; coordinators can identify uncompleted transfers; managers can audit delays; and individuals can report whether access actually occurred. Institutional governance must allocate responsibility for unresolved cases, especially those crossing public and private, medical and social, or pediatric and adult service boundaries.

The Kripalani et al. (2007) (https://doi.org/10.1001/jama.297.8.831) evidence on information deficits is historically important, while contemporary implementation studies should test the quality of digital interoperability and actual workflow effects. Electronic transfer can reduce the time needed to deliver information while increasing documentation burden or propagating poor-quality records. Audits should include completeness, timeliness, usability, error correction, recipient comprehension, and consequences for functioning.

Longitudinal tracking enables scientific evaluation of service episodes and cumulative barriers. However, linkage across institutions can raise risks involving consent, access, cybersecurity, and secondary use. Quality improvement programs should implement data minimization, access controls, and meaningful governance alongside measures of transition success. Better measurement is an institutional obligation as well as a statistical opportunity.

Digital Access, Remote Care, and Assistive Support

Remote rehabilitation, electronic coordination, and assistive technologies can change service access by reducing travel requirements, allowing specialists to support geographically distant people, and improving communication among providers. Their relevance to rehabilitation service transitions is organizational: they can facilitate continuity across settings when designed around actual clinical needs, available technology, and user capabilities. The efficacy of particular remote rehabilitation procedures is a separate intervention-specific question.

Digital access comprises several layers. Network connectivity and a usable device are necessary for many platforms, but accessibility also depends on sensory, motor, cognitive, and communication fit; affordability; privacy; caregiver or technician support; and the presence of appropriate clinicians. Someone may have broadband access yet be unable to use a video interface with the required controls or may lack a private setting for psychological care. A referral that substitutes remote attendance for in-person contact can improve convenience for one group and reduce meaningful access for another.

WHO's integration guidance (https://www.who.int/health-topics/rehabilitation/integrating-rehabilitation-into-health-systems) emphasizes the larger system conditions of workforce, assistive products, service delivery, and financing. The NICE stroke rehabilitation guideline (2023) (https://www.nice.org.uk/guidance/ng236/chapter/Recommendations) includes telerehabilitation among condition-specific options and specifies attention to individual suitability. Research should compare remote, in-person, and hybrid pathways using appropriately selected populations, actual service intensity, attrition, accessibility supports, and function and participation outcomes. Source 26

Electronic records create further opportunities for shared plans and referral closure, but their effects should be assessed in terms of workflow and outcomes. More digital communication is not inherently more coordinated care. A system that sends multiple automated notifications while no clinician assumes responsibility for an uncompleted handoff may increase message volume without continuity. Effective digital transition support requires legible assignments, usable information, error recovery, and recipient confirmation.

In the wider Artificial Era, adaptive computational systems may support scheduling, accessible explanations, translation, referral anomaly detection, and longitudinal review. These are research possibilities whose effectiveness and safety must be demonstrated against transparent comparators. Algorithms trained on historical service use can reproduce under-identification of need; automated referral prioritization can create unexplained inequity. The appropriate research questions concern accuracy, human oversight, accessibility, privacy, procedural fairness, and observed effects on care and participation. Automation should be evaluated as part of the service system in which it operates.

Comparative Populations and Generalizability

Stroke research provides substantial evidence for coordinated early supported discharge, but rehabilitation pathways differ across traumatic injury, spinal cord injury, chronic pain, musculoskeletal disorders, neurodevelopmental conditions, sensory disability, and progressive illness. These populations vary in clinical course, functional priorities, caregiver needs, equipment requirements, and available services. A transition program may succeed when a specialist team can maintain therapy intensity close to home yet perform differently when needs are unpredictable or community services are scarce.

Generalizability requires more than a shared diagnostic label. Differences in severity, baseline function, age, multimorbidity, cognitive demands, socioeconomic conditions, health coverage, geography, and institutional culture can modify the effectiveness of the same nominal intervention. The Langhorne and Baylan (2017) (https://doi.org/10.1002/14651858.CD000443.pub4) review explicitly concentrated on selected stroke populations and noted variation associated with team coordination. Treating its overall effect as universal for all rehabilitation populations would ignore the studied intervention and selection conditions.

Some transitions occur across the life course rather than across buildings. People moving from pediatric to adult disability programs may encounter altered eligibility and care relationships; others move from active treatment to long-term community supports, return to education or employment, or change service systems after relocation. These changes can alter responsibilities and access even without an acute clinical event. Research on such pathways should identify the specific services and legal context involved instead of assuming that hospital discharge models transfer unchanged.

Studies should recruit and retain people across the range of communication and cognitive abilities, supported by accessible consent and outcome collection. Studies excluding severe impairment or complex social circumstances may produce valid estimates for selected participants while leaving the groups with greatest service needs poorly represented. A transportability analysis should specify what must remain true for an intervention to work elsewhere: specialist staffing, adequate home conditions, receiving-service availability, financing, and adherence to active treatment components.

Comparative inquiry can be productive when it distinguishes invariant mechanisms from context-dependent inputs. The need for reliable responsibility transfer may be widespread; the practical procedure for achieving it differs across a national health service, fragmented insurance system, rural public program, and humanitarian emergency. A scientifically mature rehabilitation service theory identifies both the common mechanism and the resource conditions supporting it.

Economic Evaluation and Unintended Consequences

Economic evaluation of rehabilitation transitions asks whether additional resources used to coordinate services yield benefits worth their opportunity costs. The evaluation perspective matters. A hospital may save bed-days while households acquire transportation expenses and unpaid care responsibilities. A community program may require more direct rehabilitation spending while reducing later dependence or supporting valued activity. Neither scenario is adequately evaluated by comparing hospital expenditure alone.

The Cochrane review of early supported discharge (2017) (https://www.cochrane.org/evidence/CD000443_services-reducing-duration-hospital-care-people-acute-stroke) reported variation in cost estimates across the included trials. Its published summary described estimates from individual trials ranging from lower to higher costs compared with conventional care. The 2022 Cochrane review of discharge planning (https://www.cochrane.org/evidence/CD000313_discharge-planning-hospital) found uncertain effects on costs and on several patient health outcomes. These findings counsel against treating reductions in length of stay as proof of overall cost savings.

Cost-effectiveness research should define the outcome of value and its time horizon. Functional gains, quality-adjusted life-years, participation, avoidance of institutionalization, caregiver burden, and patient-selected goal achievement represent different benefits and cannot be collapsed without an explicit valuation method. Cost data should include program development, clinician time, equipment, transport, administrative coordination, maintenance, and relevant indirect consequences. Where possible, sensitivity analyses should show how results depend on staffing assumptions and the availability of existing community infrastructure.

Unintended consequences deserve their own measures. An early discharge program may increase caregiver demands; an automated triage system may disadvantage people with complex needs; a highly standardized pathway may improve uniformity while reducing responsiveness; and a service consolidation may strengthen clinical expertise while lengthening travel for rural populations. An intervention may also improve measured hospital efficiency while delaying return to valued roles. These possibilities are hypotheses for empirical investigation, not findings attributed to every real-world program.

Economic and social evaluation should therefore be integrated into trial design rather than appended after the intervention has been implemented. Programs that appear efficient within one institution can become inequitable across an entire service network. The relevant object is the sustained rehabilitation pathway and its effects on people, households, and the organizations that support them.

Critical Discussion: What the Evidence Establishes

Several findings can be stated with differentiated confidence. First, conceptual and international normative work establishes a coherent rationale for examining functioning, access, and participation across settings. The ICF (https://www.who.int/standards/classifications/international-classification-of-functioning-disability-and-health) provides the language of person–environment functioning; Levesque et al. (2013) (https://doi.org/10.1186/1475-9276-12-18) specify access as a process of interacting service features and user abilities; Haggerty et al. (2003) (https://doi.org/10.1136/bmj.327.7425.1219) distinguish dimensions of continuity. These are influential explanatory and classificatory frameworks rather than direct experimental demonstrations that any particular service redesign improves health. Source 21

Second, randomized evidence supports some specific transition interventions. Naylor et al. (1999) (https://doi.org/10.1001/jama.281.7.613) demonstrated reductions in readmissions for a nurse-led intervention among selected older hospitalized patients. The Cochrane discharge-planning synthesis (2022) (https://doi.org/10.1002/14651858.CD000313.pub6) found probable but generally small improvements in length of stay and readmission in relevant populations. The Cochrane stroke ESD synthesis (2017) (https://doi.org/10.1002/14651858.CD000443.pub4) provides more directly rehabilitation-specific evidence for coordinated specialist services among selected stroke survivors. Differences in design, case mix, intervention content, and comparison services matter to interpretation.

Third, process and implementation research explains why an intervention cannot be judged only by its formal name. Proctor et al. (2011) (https://doi.org/10.1007/s10488-010-0319-7) identify distinct implementation outcomes, while CFIR (https://doi.org/10.1186/s13012-022-01245-0) and RE-AIM (https://doi.org/10.2105/ajph.89.9.1322) guide analysis of contexts, reach, adoption, and maintenance. These frameworks make barriers and implementation processes measurable. Their application must be linked to direct observations and comparative data before attributing effects to hypothesized mechanisms.

Important limitations remain. The ESD evidence centers on stroke and selected patient groups. General discharge evidence spans heterogeneous conditions and interventions. Service outcomes are often measured more consistently than person-defined participation. Implementation models have broad descriptive reach, but causal tests of particular mechanisms may be sparse. Contemporary health systems differ substantially from those in some classic trials. Administrative and financial outcomes may be measured more reliably than autonomy, belonging, and long-term opportunity. Unequal inclusion in research can create uncertainty precisely where policy implications are greatest.

This synthesis therefore supports a conditional conclusion: appropriately designed, coordinated, accessible, and adequately resourced transitions can improve particular service and functional outcomes, while the magnitude and distribution of benefit must be established in each population and context. The evidence favors describing interventions precisely, measuring several outcome domains, and examining who is reached. Scientific strength lies in making those conditions explicit.

Future Research Priorities

Future studies should investigate whole-pathway access rather than only the moment of discharge. Cohorts linked across acute care, rehabilitation, community therapy, social services, and participant-reported outcomes can identify when unmet needs emerge and whether they accumulate. Studies should predefine entry points, denominators for reach, competing events such as death or institutionalization, and meaningful follow-up periods. Reporting should specify whether access was offered, initiated, delivered at appropriate intensity, and sustained.

Comparative trials should isolate the active components of coordinated service models. A factorial design could examine the contribution of a named coordinator, confirmed referral acceptance, shared goal records, and rapid therapy initiation, provided the design is feasible and ethically appropriate. Cluster-randomized or well-justified quasi-experimental studies can test implementation at organizational scale. Independent replication is needed across geography, clinical severity, service financing, and cultural contexts.

Measurement science requires development and validation of accessible indicators for transition continuity and person-valued participation. Standardized functioning measures should be combined with individual goals and experiential reports, and instruments should be tested for responsiveness and measurement equivalence where appropriate. Missing data mechanisms require explicit investigation because people with communication, cognitive, mobility, or socioeconomic barriers may be less likely to complete conventional follow-up.

Equity-oriented implementation studies should test practical strategies that reduce barriers without reallocating burdens invisibly. Examples include transport assistance, accessible referral communications, interpreter services, supported navigation, community outreach, and flexible home-based delivery. Programs should be evaluated for differential uptake, costs to households, participant autonomy, and unintended exclusion. Decisions about which outcomes matter should include people who rely on services and their chosen supporters.

Finally, digital and Artificial Era systems create opportunities to track uncompleted referrals and support coordination while introducing questions of privacy, automation bias, explanation, accountability, and discriminatory access. Rigorous research should compare them with well-defined nonautomated procedures and measure actual effects on functioning and participation. Technical success is a separate outcome from equitable clinical and psychological benefit.

Conclusion

Rehabilitation service transitions are the point at which clinical rehabilitation, health-system organization, psychological adaptation, and social environments meet. A successful transition preserves access to appropriate services while enabling goals to be revised with the person as needs and circumstances change. Its achievement depends on the fit among clinically meaningful plans, receiving teams, resources, communication, responsibility, and ordinary life conditions.

The strongest directly relevant controlled evidence concerns coordinated early supported discharge for selected people after stroke. Broader discharge planning and transitional-care research provides complementary evidence but shows more modest or heterogeneous effects and cannot be transferred without attention to population and intervention differences. Models of access, continuity, functioning, and implementation explain why these results vary and identify what researchers must measure in order to understand them.

A mature science of rehabilitation service pathways evaluates clinical and functional outcomes alongside reach, continuity, participation, caregiver demands, equity, implementation quality, cost, and sustainability. These domains jointly define the empirical field. The central practical implication is to make each transition accountable to the person's continued opportunity to receive useful support and to participate meaningfully in life beyond institutional care.

References