RESEARCH DOSSIER
Stroke Rehabilitation: Mood, Agency, and Community Reintegration
Abstract. Stroke rehabilitation is a multidimensional psychological and social process in which depression, anxiety, apathy, emotional expression, fatigue, perceived agency, and available community opportunities influence life after cerebrovascular injury. This fundamental research monograph investigates conceptual foundations, historical development, measurement validity, original intervention trials, longitudinal and qualitative research, systematic reviews, and contemporary stroke rehabilitation guidance. A structured critical narrative synthesis distinguishes post-stroke emotional syndromes from one another, differentiates self-efficacy from effective choice and environmental access, and evaluates rehabilitation outcomes across home, relationships, work, transport, and leisure. Large randomized fluoxetine trials do not support routine use for enhancing overall functional recovery, despite a positive earlier small motor trial; evidence for treatment of established post-stroke depression requires separate evaluation. Participation reviews show heterogeneous intervention effects and illuminate differences between personally chosen goal attainment and broad social integration. The synthesis identifies priorities for accessible measurement, long-term care continuity, equitable inclusion, causal mechanism testing, and independent replication. It presents neither new participant data nor a registered systematic review, independently pooled effect estimate, or individualized clinical treatment advice.
Keywords: stroke rehabilitation; post-stroke depression; post-stroke anxiety; apathy; post-stroke fatigue; emotionalism; stroke self-efficacy; sense of agency; rehabilitation psychology; community reintegration; social participation; patient-centered rehabilitation; supported goal setting; stroke caregiver; early supported discharge; stroke intervention trials; ICF; participation outcomes; Artificial Era
The Scientific Object of Stroke Psychosocial Rehabilitation
Stroke psychosocial rehabilitation examines how a cerebrovascular event reorganizes emotional experience, motivation, practical decision-making, family relationships, and participation in social life. Its outcome is the person's ability to pursue meaningful activities and roles within a changed bodily and social situation. Clinical recovery can include motor improvement, yet effective recovery also involves rebuilding routines, restoring access to relationships, and gaining influence over decisions. The World Health Organization's International Classification of Functioning, Disability and Health (ICF) locates participation within interactions among health conditions, body functions, activities, environmental factors, and personal circumstances. This provides a common vocabulary for the psychological study of post-stroke life. WHO (2001)
The present research object joins three consequential domains: mood, agency, and community reintegration. Mood includes depression, anxiety, emotional lability, and subjective emotional adjustment; agency concerns perceived capability, goal choice, communication of preferences, and the ability to initiate and sustain activity. Reintegration concerns roles in home, work, friendship, neighborhood, culture, and recreation. These domains influence one another through multiple pathways. A person may achieve good mobility yet remain isolated; another may live with enduring impairment while securing meaningful roles through adaptation and social support. Neither example permits outcome to be reduced to impairment scores alone. Duncan et al. (1999) Winstein et al. (2016)
This monograph addresses the evidence for these relationships through a structured critical narrative synthesis of primary randomized studies, observational cohorts, psychometric validations, systematic reviews, and contemporary professional guidance. It considers which conclusions follow from controlled trials and which remain hypotheses arising from longitudinal or qualitative research. The work is an integrative scholarly interpretation of published evidence, with no original participant-level dataset, registered exhaustive review protocol, independent meta-analysis, or claim of new clinical treatment effects. Stroke-specific psychosocial rehabilitation is its focus; detailed vascular pathophysiology, general motor-recovery science, and rehabilitation of traumatic brain injury are treated in their own disciplines. Yao et al. (2026) NICE (2023)
History of the Research Tradition
The scientific history of stroke rehabilitation began with the measurement of survival and bodily disability and broadened through studies of activities of daily living, cognitive effects, depressive symptoms, and social participation. Earlier clinical practice generally defined recovery through impairment reduction and independence in basic tasks. By the late twentieth century, rehabilitation researchers were developing measures that captured patients' reports of communication, memory, emotion, and participation. The Stroke Impact Scale validation by Duncan and colleagues exemplifies this transition: its eight domains placed the survivor's own account beside clinical observations and exposed dimensions that conventional disability scales could miss. Duncan et al. (1999)
The ICF, endorsed by the World Health Assembly in 2001, institutionalized a conceptual separation among impairment, activity limitation, and participation restriction while treating the environment as a contributor to functioning. The expansion of early supported discharge services subsequently tested whether organized rehabilitation could move across hospital and home settings. Langhorne and Baylan's Cochrane synthesis evaluated a service model rather than one isolated physical therapy technique, making continuity of rehabilitation and the conditions of community living central research questions. These changes created the intellectual and institutional foundation for examining mood and agency as integral elements of participation. WHO (2001) Langhorne et al. (2017)
In parallel, post-stroke depression research matured from descriptive case series into prospective observational synthesis and controlled interventions. Hackett and colleagues summarized observational studies in 2005; Hackett and Pickles updated the synthesis in 2014. Larger randomized investigations later challenged the proposition that a particular antidepressant routinely enhances motor recovery. Contemporary guidelines increasingly integrate screening for mood, assessment of fatigue, personalized goals, family support, and the return to vocational and social roles. The historical trajectory is therefore visible in changing outcomes, measurement instruments, and service models, each carrying different assumptions about what a recovered life means. Hackett et al. (2005) Hackett & Pickles (2014) FOCUS Trial Collaboration (2019) Yao et al. (2026)
The ICF and an Ecological Account of Participation
The ICF makes rehabilitation outcome relational and contextual. Body-function changes after stroke include weakness, sensory deficits, pain, aphasia, attention impairment, and changes in emotional regulation. Activity describes the execution of tasks, while participation concerns involvement in life situations. A usable wheelchair-accessible transit route, supportive employer, and communication partner are environmental conditions that can change participation without changing the underlying brain lesion. Conversely, inaccessible transport or public stigma can generate restriction even when clinical measures suggest adequate capacity. The ICF gives researchers an analytic framework for organizing these influences, though it does not itself establish their causal weights. WHO (2001)
For psychology, an ecological formulation requires attention to how environmental opportunities become perceived and enacted. Confidence in walking alone may depend on fear of falling, exposure to safe practice, accessible routes, assistance when needed, and successful prior experiences. Self-efficacy is a belief about capability in a task context; actual participation also depends on resources, permissions, social relationships, and bodily endurance. Interventions may therefore target skills, beliefs, and environmental barriers simultaneously. Contemporary Canadian recommendations address driving, vocational roles, mood, cognition, sleep, fatigue, and social engagement within one community-participation module, illustrating the breadth of what must be coordinated. Jones et al. (2008) Yao et al. (2026)
This framework guards against a common interpretive error: a statistically significant improvement in a laboratory motor test cannot automatically be translated into increased work participation, social satisfaction, or independent choice. Outcome correspondence requires demonstrated transfer across context. The construct of reintegration is also plural: resumption of a previous role, modification of a role, and creation of a new role may all represent valued participation. Assessments should record what a survivor values and whether a particular environment makes those goals attainable, with repeated measurements as needs, supports, and opportunities change. Duncan et al. (1999) Egan et al. (2026)
Heterogeneity of Stroke and Recovery Time
Stroke survivors differ in lesion location, type, severity, premorbid health, cognitive and communication function, age, social position, and access to services. The psychological consequences of a small focal event may include profound aphasia or fatigue, while a larger motor deficit may coexist with relatively preserved mood and decisional abilities. Cohort averages consequently describe a distribution rather than a typical person. Depression prevalence estimates depend on the timing of assessment, recruitment setting, exclusion criteria, and the measure used. Reviews pooling heterogeneous studies must be read as summaries of selected populations and instruments. Hackett & Pickles (2014) Zhan et al. (2023)
Time is an active variable in psychosocial rehabilitation. The acute period may involve disorientation, medical uncertainty, delirium, and abrupt loss of familiar control. Hospital discharge changes the practical demands placed on survivors and their supporters; longer-term life reveals persistent difficulty with work, driving, intimacy, household management, and leisure. Emotional problems may emerge or recur after apparently successful early recovery. NICE guidance accordingly recognizes evolving psychological needs and recommends attention to new or persisting mood problems at later reviews. Follow-up schedules focused exclusively on the first hospital stay will systematically miss part of the recovery trajectory. NICE (2023) Yao et al. (2026)
Individual trajectories also depend on what outcomes are selected. Early change on a motor impairment scale and later change in meaningful participation may follow different curves, with delayed improvements through adaptation or new environmental support. Repeated measures must therefore specify a time origin, follow-up intervals, competing events such as recurrent stroke, and whether participants with aphasia, cognitive impairment, or institutional care are represented. Longitudinal observational estimates are vulnerable to attrition and survival selection. Multisite trials strengthen causal inference for their assigned interventions, while still requiring careful interpretation of who was eligible and how outcomes were measured. Winstein et al. (2016) Langhorne et al. (2017)
Depressive Symptoms and Post-Stroke Depression
Depression after stroke encompasses depressive symptom burden as well as clinically evaluated depressive disorders. The distinction matters because symptom scales differ in thresholds and include somatic items overlapping with neurological illness. In a 2014 meta-analysis of 61 observational studies involving 25,488 people, Hackett and Pickles estimated a pooled proportional frequency of 31% (95% confidence interval 28%–35%). The pooled figure summarizes markedly variable samples and time windows; it is neither a prediction for every survivor nor a diagnosis based solely on having experienced stroke. This prevalence estimate remains an important historical benchmark against which newer clinical practice can be evaluated. Hackett & Pickles (2014)
The practical impact of depression extends across motivation, sleep, adherence to activity, perceived capability, and relationships. However, observed links between depression and functional outcomes are open to reciprocal and confounded explanations. More severe neurological impairment can increase both distress and activity limitation, while reduced participation may reinforce loss of pleasure and hopelessness. Lesion-related neurobiological changes and prior mental-health vulnerability may add other pathways. Robust research must distinguish depressive illness from fatigue, apathy, grief, emotionalism, and communication-related difficulty in expressing distress. Robinson and Jorge reviewed neuropsychiatric mechanisms and the clinical implications of this complex post-stroke condition. Robinson & Jorge (2016) Allida et al. (2023)
Screening is a gateway to assessment, not a substitute for it. Canadian recommendations call for validated screening by trained professionals when medically appropriate and for timely evaluation when a screen indicates risk. Choice of instrument and format matters for people who cannot read, speak, or comprehend standard questionnaires; a negative score obtained through inaccessible administration is not reassuring. Diagnostic reasoning should examine chronology, functional impairment, pre-stroke mental health, medication effects, pain, sleep, and the possibility of suicidality. Treatment and safety decisions belong to qualified clinicians who can integrate these findings with neurological and medical care. Yao et al. (2026) NICE (2023)
Neurobehavioral, Personal, and Social Pathways to Depression
Biological explanations of post-stroke depression address neural injury, altered networks, inflammatory processes, and changes in monoaminergic or stress-related signaling. Psychological explanations concern loss of valued activity, threatened identity, reduced predictability, and appraisal of future capacity. Social explanations concern isolation, economic insecurity, unmet care needs, and shifts in close relationships. The pathways can coexist and change over time. A neurological association by itself does not prove that a particular lesion or neurotransmitter change causes an individual episode of depression, and social associations may partly reflect prior socioeconomic disadvantage. Robinson & Jorge (2016) Hackett & Pickles (2014)
A defensible clinical model follows the sequence from precipitating event through current disability, personal interpretation, social opportunities, and feedback from everyday experiences. For example, difficulties in communication can make social encounters effortful; withdrawal can decrease sources of positive reinforcement; reduced engagement can then amplify low mood. Similar behavioral feedback processes are central to psychosocial-behavioral treatment approaches, including the Living Well with Stroke trial. The model supplies potentially testable mediators, but their independent contribution requires repeated measurement and designs capable of separating earlier causes from later consequences. Mitchell et al. (2009) Barrows & Thomas (2018)
Studies often find that depressive symptoms and participation correlate. The inference from a correlation to a single directional mechanism is weak unless temporal ordering, baseline severity, common causes, and measurement error are considered. Randomization of mood intervention may establish an effect on a mood scale without establishing that mood change caused functional improvement. Future mechanistic trials should measure candidate mediators before final participation outcomes, evaluate both direct and indirect effects with appropriate assumptions, and include environments where participation can actually occur. The distinction between explanatory plausibility and experimentally supported treatment effect is central to this monograph. Allida et al. (2023) Zhang et al. (2026)
Anxiety, Fear, and Threat Appraisal
Anxiety following stroke may involve worry about recurrence, falls, safety outside the home, loss of employment, or anticipated failure in public tasks. Formal anxiety disorders and elevated anxiety scores are different outcomes. Campbell Burton and colleagues synthesized observational studies reporting anxiety after stroke; the diversity of measures and settings cautions against treating any pooled figure as universal. In clinical and social life, avoidance behaviors may narrow the opportunities needed for rebuilding competence. Some avoidance is proportionate to genuine hazards, so rehabilitation assessment must distinguish warranted precautions from restrictions sustained primarily by threat appraisal. Campbell Burton et al. (2013) NICE (2023)
The relationship between anxiety and participation is especially clear in complex environments. Shopping, transportation, public speaking, or meeting strangers after aphasia demand coordination of attention, movement, and communication. The survivor may have sufficient component skills in a quiet therapy room yet anticipate difficulty managing the combined demands outside it. Graded practice, accessible assistance, and realistic safety planning can be reasoned about as ways to test feared predictions and support confidence; the extent to which each method improves long-term post-stroke participation is an empirical question rather than a guaranteed effect. Jones et al. (2008) Winstein et al. (2016)
Anxiety is not interchangeable with depression, apathy, or fatigue. A survivor who wants to go out but fears falling has a different treatment target from a survivor who has low initiation or lacks transport. Assessments should identify the person's own explanation of restricted participation, relevant medical risk, emotional triggers, and available supports. NICE advises reviewing mental-health and emotional functioning alongside cognitive and behavioral changes, while Canadian guidance integrates sleep and fatigue with psychosocial evaluation. This multidomain process supports proportional interventions and reduces the risk of attributing every difficulty to a single psychological construct. NICE (2023) Yao et al. (2026)
Apathy, Initiation, and Motivation
Apathy after stroke concerns diminished goal-directed behavior, initiative, or emotional engagement. It may share features with depression but does not require sadness, guilt, or despair. Zhang and colleagues synthesized 39 observational studies involving 5,168 stroke survivors and estimated pooled apathy prevalence of 33.0% (95% CI 27.6%–38.4%). Results varied by country, stroke subtype, measurement instrument, and study quality. The estimate should therefore prompt careful recognition of a common syndrome rather than be interpreted as a stable biological prevalence across populations. Zhang et al. (2023)
Apathy has special importance for the study of agency because the capacity to express a preference can remain present while initiation is reduced. A person may value returning to social life yet fail to begin planning, leave home, or complete multi-step activities without prompts. Executive dysfunction, fatigue, depression, and environmental monotony can produce superficially similar behavior. Observing one missed appointment or one uninitiated task cannot establish a syndrome. Assessment requires patterns across settings, collateral observations interpreted alongside the person's account, and the separation of inability, inaccessibility, and reduced spontaneous initiation. Zhang et al. (2023) Winstein et al. (2016)
Rehabilitation implications follow from the functional analysis rather than from a single label. External structure, reminders, rewarding activities, communication support, and shared goal review may address some initiation barriers, while medical and neuropsychological assessment may uncover treatable contributors. Clinical trials that measure only depression may overlook whether an intervention affects apathy. Conversely, an increase in task completion under intensive prompting should be distinguished from a durable increase in self-directed action. This distinction is essential when evaluating agency: observable activity, autonomy of choice, and externally maintained performance are related yet separate research outcomes. Jones et al. (2008) Ris et al. (2026)
Emotionalism and Disorders of Emotional Expression
Post-stroke emotionalism involves episodes of crying or laughing that may be unexpected, exaggerated, or difficult to control. It can occur independently of the individual's prevailing emotional state and may create distress or social embarrassment. Broomfield and colleagues' 2024 review describes the diagnostic, pathophysiological, and treatment questions surrounding the condition. Confusing emotionalism with depression may lead both researchers and family members to misread the meaning of behavior. The person's report of subjective mood and the observed pattern of expressive episodes must be considered as distinct kinds of evidence. Broomfield et al. (2024)
Emotional expression changes can carry direct participation costs. A survivor anticipating involuntary crying during conversation may decline invitations or avoid speaking in public even while wishing for connection. Others may interpret episodes as sadness, inability to cope, or lack of decisional capacity. Communication partners therefore influence the practical consequences of the symptom. Education about the condition, careful assessment, and respectful strategies for managing difficult situations can reduce avoidable misunderstanding, although the evidence for any specific intervention must be evaluated separately and clinical treatment needs individualized specialist judgment. Broomfield et al. (2024) NICE (2023)
The condition also reveals an important measurement distinction: emotion as experienced, emotion as expressed, and emotion as attributed by observers need not coincide. A scale that treats frequency of crying as a direct index of depressive mood may misclassify participants. Researchers studying social reintegration should ask whether perceived embarrassment, observer reactions, and opportunities for disclosure mediate participation outcomes. Longitudinal comparisons should specify whether emotionalism changed while depressive symptoms remained stable, and should identify the proportion of survivors excluded because communication or cognitive deficits made standard assessment difficult. Broomfield et al. (2024) Barrows & Thomas (2018)
Post-Stroke Fatigue and the Management of Activity
Fatigue after stroke is often persistent, disabling, and disproportionate to straightforward measures of physical exertion. Zhan and colleagues analyzed 66 studies with 11,697 participants and reported a pooled prevalence of 46.79% (95% CI 43.41%–50.18%); the estimate varied with the fatigue instrument and characteristics of the included studies. Fatigue is not a proxy for unwillingness or poor motivation. Its possible contributors include sleep disorders, depression, pain, deconditioning, medical conditions, and medication effects, as well as mechanisms linked to stroke itself. Treating the observed symptom as a multidimensional target is more precise than inferring a single cause. Zhan et al. (2023) Cumming et al. (2018)
Energy limitations shape agency because they constrain the timing, duration, and predictability of meaningful tasks. A person may be able to accomplish a goal under optimal conditions but not sustain it across work, household, transport, and social demands on the same day. Self-management strategies therefore address planning, prioritization, rest, communication of energy needs, and the relation between activity and recovery. Canadian guidance recommends education about energy conservation, assessment of sleep and comorbidities, and gradual activity progression while acknowledging low-certainty evidence for several specific treatments. The success of a plan should be evaluated against valued roles rather than only fatigue scores. Yao et al. (2026) NICE (2023)
Research on fatigue requires care with overlaps among symptom scales and interventions. Fatigue items may appear in depression inventories; activity limitation may appear in both fatigue and disability measures; shared-method variance can inflate associations. An intervention that increases physical conditioning may improve tolerance but fail to address cognitive exhaustion during conversation. Trials should specify primary endpoints, intervention dose, adherence, and longer-term transfer, while carefully monitoring adverse experiences. Pharmacological questions belong to stroke specialists: the existence of experimental drug studies does not establish routine treatment effectiveness, and depression medication must not be presented as treatment for fatigue without another indication. Cumming et al. (2018) Yao et al. (2026)
Cognition, Communication, and Accessible Assessment
Executive control, attention, memory, spatial awareness, and language can influence how a survivor understands choices and implements plans. Aphasia can restrict spoken or written expression while leaving many intentions and preferences accessible through adapted methods. It is particularly dangerous to interpret difficulty answering a standard mood questionnaire as an absence of psychological experience. Barrows and Thomas evaluated Dynamic Visual Analogue Mood Scales in 46 stroke survivors, including participants with aphasia, and found promising convergent performance; that sample and cross-sectional design do not establish validity across all severity levels or acute clinical settings. Barrows & Thomas (2018)
Psychometric accessibility must be considered a property of the interaction between instrument and participant. Reading requirements, response speed, complex syntax, motor demands, visual impairments, and fatigue can all distort scores. Alternative response modalities may include pictorial scales, supported conversation, shorter sessions, and trained observation, but instruments require validation for the intended construct and context. A caregiver-proxy score may provide valuable collateral information while still differing systematically from a survivor's own interpretation of mood, autonomy, or social satisfaction. In studies of agency, such proxy discrepancies are potentially meaningful data rather than mere measurement noise. Barrows & Thomas (2018) Duncan et al. (1999)
Accessible assessment changes who becomes visible in evidence. If trials routinely exclude people with severe aphasia, cognitive impairment, or communication support needs, published findings will overrepresent those easiest to enroll. This selection may also affect estimates of therapy acceptability and mood-disorder prevalence. NICE specifically recommends adjustments for communication and cognitive needs; Canadian guidance emphasizes inclusive assessment and coordinated rehabilitation. Research protocols should document inclusion criteria, supported consent procedures, mode of measurement, reasons for missing responses, and the feasibility of repeating assessment after fatigue or fluctuating attention. NICE (2023) Yao et al. (2026)
Agency, Self-Efficacy, and the Experience of Control
Agency in stroke rehabilitation is a set of related capacities and opportunities: initiating action, choosing goals, anticipating consequences, evaluating results, and influencing surrounding conditions. Perceived self-efficacy is one important component, referring to beliefs about capability for particular behaviors. Jones, Partridge, and Reid developed the Stroke Self-Efficacy Questionnaire to capture confidence relevant to functional performance. Confidence should be studied as an appraisal rather than mistaken for actual competence, and competence should be distinguished from environmental opportunity. A survivor can have strong confidence but encounter inaccessible streets; another may possess a skill yet expect failure after a frightening experience. Jones et al. (2008) WHO (2001)
Agency is dynamic. Repeated successful activity can provide information that supports confidence, while setbacks, unpredictable fatigue, and repeated decisions made without the person can undermine it. In treatment settings, meaningful choice includes the ability to select the goal, negotiate the pace, reject an unwanted activity, and revise priorities as circumstances change. Agency is also relational: therapists, families, employers, and service institutions control opportunities to practice and to make consequential choices. The association between self-efficacy and functioning therefore requires interpretation in the context of assistance and structural barriers. Jones et al. (2008) Rosewilliam et al. (2011)
A rigorous investigation of agency would measure both subjective confidence and observable participation over time, while recording whether choices are genuinely person-directed. Changes on self-efficacy scales may be proximal outcomes of rehabilitation without proving independent causal effects on long-term work or relationships. The study design should accommodate communication impairments and distinguish personal valuation from professional preference. When confidence is high but activity remains low, environmental access, fatigue, anxiety, and transportation should be examined before labeling the person as unmotivated. This framework makes clinical goal setting a testable psychological process rather than an administrative formality. Jones et al. (2008) Ris et al. (2026)
Identity Reconstruction and Biographical Continuity
Stroke may interrupt a person's practical narrative about who they are: a worker, partner, parent, athlete, neighbor, or independent traveler. Identity change involves memories of prior roles, present bodily experience, others' reactions, and expectations for the future. The loss of one role does not have a uniform psychological meaning. Some people seek its restoration; others pursue an adapted form or a new activity that expresses enduring values. A research program centered only on whether the pre-stroke role resumed can overlook a successful transition chosen by the person. Participation frameworks offer a way to record these diverse trajectories without turning biography into a single score. WHO (2001) Duncan et al. (1999)
Psychosocial adjustment also takes place through ordinary experiences of competence and reciprocity. The ability to contribute to a household task, help another person, or participate in a group may matter independently of a clinician's description of motor ability. Qualitative research examining priorities beyond the first post-stroke year has identified mental adaptation, practical accommodations, financial safety, work, and opportunities to give back as relevant aspirations. Such findings illuminate mechanisms and unmet questions but do not establish that every participant shares an identical ordering of priorities. Lo et al. (2023)
Identity reconstruction is methodologically difficult because personal narratives vary across culture, time, relationships, and interviewer context. Repeated interviews can reveal changes in meaning that standardized scales may miss, while quantitative measures permit comparisons across treatment groups. The two forms of evidence answer different questions. When investigating identity and agency, researchers should avoid assigning a fixed moral value to independence or dependency; receiving assistance may allow the person to pursue goals more freely. The scientific outcome is whether practical life accords with the person's own priorities and accessible opportunities. WHO (2001) Ris et al. (2026)
Participation as a Measurable Outcome
Participation refers to involvement in life situations, including personal relationships, community events, domestic responsibilities, and valued work. It cannot be inferred reliably from a single measure of arm strength or walking speed. Duncan and colleagues' Stroke Impact Scale provided patient-reported domains including emotion and participation and helped expand the interpretation of outcomes. Yet a questionnaire's participation domain remains a particular operationalization: it can capture perceived restriction, frequency of activity, or satisfaction depending on its items. These are related constructs, and their scores should be described with their actual meanings. Duncan et al. (1999) WHO (2001)
Participation measurement faces a tension between standardized comparability and individualized value. A broad scale can show average change across randomized groups, but may undervalue a particular personally meaningful activity; a person-specific measure can detect targeted goal attainment while limiting cross-person comparison. Egan and colleagues' 2026 meta-analysis separated person-specific participation goals, broad participation, and instrumental activities of daily living and found different pooled estimates by outcome category. The distinction demonstrates why it is misleading to describe all these endpoints as one interchangeable measure of social reintegration. Egan et al. (2026)
Good research should specify whether the intended construct is amount of participation, diversity of roles, social connectedness, autonomy of decision, satisfaction, or environmental accessibility. Measurement occasions must be long enough to reveal opportunities outside a short course of treatment. Researchers should also describe floor and ceiling effects, response modes, proxy use, and clinically relevant change thresholds where established. A favorable shift on an impairment scale with no meaningful change in social participation is a scientifically important result, just as increased access through an environmental adaptation may occur without any change in neurological impairment. Duncan et al. (1999) Zhang et al. (2026)
Domestic Activity and the Environment of Everyday Choice
Home is a primary setting where recovery becomes concrete. Preparing meals, managing medication with professional oversight, attending to personal care, using appliances, and maintaining routines require combinations of motor, cognitive, emotional, and social capacity. Activity scales are useful, but the same numerical score can correspond to different lives depending on household layout and available assistance. NICE recommends assessing the home environment before discharge, including equipment, adaptations, and support for independent living. This requirement recognizes that environmental conditions can determine whether a person can exercise an otherwise available skill. NICE (2023)
Domestic rehabilitation can also make personal control visible. A survivor may complete a task with modified equipment, ask for specific assistance, or decide to share rather than perform a household responsibility. None of these outcomes can be adequately interpreted without asking whether the arrangement reflects personal preference and safety. For families, the transition from professional services to home may involve abrupt growth in caregiving labor and negotiation of roles. Assessment therefore must include the capacity and well-being of people asked to provide support, as well as the stroke survivor's priorities. Yao et al. (2026) Bakhtiari-Dovvombaygi et al. (2025)
Research on home interventions should measure functional transfer in real environments instead of assuming that clinic-based mastery will generalize. Observation across tasks can reveal how fatigue, interruption, competing demands, and inaccessible spaces limit performance. Intervention studies should distinguish effects of skills training from those of equipment provision, environmental redesign, or additional caregiver time. Economic evaluation can clarify whether a service offers durable participation benefits rather than moving uncompensated costs to families. This ecological logic links the household to broader participation and prevents rehabilitation from being defined exclusively by self-care independence. WHO (2001) Langhorne et al. (2017)
Vocational Roles and Return to Work
Work participation connects economic security, personal identity, daily structure, social relationships, and practical agency. Return to work is usually recorded as an observable status, yet it can conceal substantial differences: hours, workload, task quality, accommodations, earnings, job security, and whether the role was chosen. Following stroke, fatigue and cognitive changes can remain consequential even when physical abilities appear adequate. Rehabilitation research should therefore distinguish initial work resumption from sustained, health-compatible employment and from valued nonemployment roles such as caregiving or volunteering. The person's goals and circumstances determine which outcome is relevant. Li et al. (2024) Yao et al. (2026)
Li and colleagues' 2024 systematic review and meta-analysis examined interventions designed to support return to work after stroke. Syntheses in this field face intervention heterogeneity because work-focused support may combine vocational counseling, occupational therapy, employer negotiation, fatigue management, and graded adjustments. A pooled finding cannot by itself identify which component is causally active or transferable across labor markets. Work outcomes also depend on insurance rules, legal protections, job flexibility, and local transportation. The same functional condition can therefore produce different employment results in different environments. Li et al. (2024) WHO (2001)
For study design, employment endpoints should be defined prospectively, distinguishing any paid work from sustained work and measuring the quality of accommodation. Controls should account for premorbid occupation, economic conditions, stroke severity, and employer resources. Survivors who were already retired or outside formal employment require other meaningful participation measures. Policy evaluations should include perspectives of workers and employers while maintaining the survivor's decision authority. Employment after stroke is consequently a psychosocial and institutional outcome, and success requires more than a favorable isolated neurological score. Li et al. (2024) Ris et al. (2026)
Transportation, Driving, and Access to Public Space
Travel is often the mechanism linking restored capacity to social participation. Reaching a clinic, market, workplace, park, or friend's home may require walking endurance, spatial judgment, vision, communication, and knowledge of an accessible transport system. A person unable to drive may remain highly active where public transit is inclusive, while an individual with adequate mobility may be restricted in a car-dependent setting. The ICF situates these differences among environmental influences on participation. Canadian guidance includes driving and community participation precisely because access to destinations is part of rehabilitation rather than an incidental travel concern. WHO (2001) Yao et al. (2026)
Driving after stroke introduces a specific safety and legal domain. It requires relevant physical, visual, cognitive, and medical assessment in accordance with local regulations. Psychological readiness and confidence cannot substitute for fitness-to-drive determination, and prior driving experience alone cannot establish current safety. Rehabilitation practitioners can explore alternative mobility options and the consequences of losing driving access, particularly for work, relationships, and medical follow-up. Such discussions are clinically and socially important because transport restriction may produce isolation independently of the direct neurological effects of stroke. NICE (2023) Winstein et al. (2016)
Experimental studies of mobility and participation should measure actual journeys and destinations where feasible, rather than treating walking speed as the endpoint of public accessibility. Environmental audits, travel diaries, interviews, and participation scales can reveal different aspects of the same process. Researchers must account for route safety, seasonal conditions, availability of caregivers, financial cost, and urban versus rural infrastructure. Transport interventions may benefit people by changing services or the built environment without modifying a person's impairment; these are legitimate rehabilitation outcomes if the causal pathway and participant-valued benefits are documented. WHO (2001) Yao et al. (2026)
Family Systems and Caregiver-Survivor Relationships
Family members and informal caregivers often participate in medication organization, appointments, daily activities, emotional support, and the negotiation of new roles. Their work can enable the survivor's goals, yet caregiver time and emotional capacity are finite resources. Stroke changes intimate relationships as well as functional routines, and some people have no caregiver available. Contemporary Canadian guidelines recommend assessment of both survivor and family support needs, and provision of individualized psychosocial assistance, training, and access to community services. Family-centered practice must remain compatible with the survivor's authority over personal decisions. Yao et al. (2026) Bakhtiari-Dovvombaygi et al. (2025)
Caregiver distress, burden, and quality of life constitute outcomes in their own right. A 2025 systematic review by Bakhtiari-Dovvombaygi and colleagues examined interventions directed to stroke survivors and caregivers; a 2026 synthesis by Lin and colleagues evaluated randomized support interventions with outcomes including caregiver burden and quality of life. These literatures differ in populations, intervention composition, and measurement. They support taking caregiver needs seriously while cautioning against attributing a particular effect to education, psychological support, or practical training without an appropriate comparison. Bakhtiari-Dovvombaygi et al. (2025) Lin et al. (2026)
Dyadic studies have an analytic advantage when they collect separate accounts from survivor and supporter. Disagreement about goals may reveal communication barriers, risk concerns, unequal control, or ordinary differences in values. A model that substitutes the caregiver's preference for the survivor's voice can inadvertently erase autonomy, especially when aphasia is present. Trials should record the caregiver's baseline capacity, social support, and health, and monitor whether interventions shift work from professional services to unpaid households. The outcome of successful reintegration is better understood through sustained, mutually workable relationships and accessible external resources than through a simple count of caregiver hours. Barrows & Thomas (2018) Yao et al. (2026)
Psychological Therapies and Behavioral Rehabilitation
Psychological rehabilitation may address depressive symptoms, anxiety, coping, adjustment, and patterns of activity avoidance. Interventions discussed in modern stroke guidance include cognitive behavioral approaches, problem solving, interpersonal approaches, and other structured forms of psychological support. Their effects cannot be generalized from nonstroke populations without considering aphasia, cognitive impairment, fatigue, and treatment accessibility. The 2023 Cochrane review of treatment for post-stroke depression concluded that evidence for several pharmacological and psychological approaches remains of very low certainty, emphasizing the need for more reliable trials. Clinical recommendations and evidence certainty must be reported separately. Allida et al. (2023) Yao et al. (2026)
The Living Well with Stroke randomized trial by Mitchell and colleagues enrolled 101 participants with clinically diagnosed depression after ischemic stroke. An eight-week psychosocial-behavioral intervention combined with antidepressant treatment was compared with usual care including antidepressant treatment. The trial reported reductions in depressive symptoms and greater remission at treatment end and at 12 months under its specified measures. It provides direct evidence for one intervention package and a defined population. Because the design combined behavioral intervention with usual pharmacological care, it cannot by itself isolate every behavioral component or establish equal effectiveness in people excluded from the trial. Mitchell et al. (2009)
An intervention is clinically meaningful when it addresses a problem the survivor recognizes and can use in daily life. Treatment research should specify dose, delivery mode, adaptations for communication and attention, adverse experiences, therapist training, and care setting. Mood improvement does not automatically prove enhanced community participation. A stronger evaluation includes both symptom severity and person-valued outcomes after treatment, assessing whether restored engagement survives changes in environmental support. Psychosocial services should also provide pathways to assessment when symptoms indicate potential danger or complex psychiatric comorbidity; such decisions require qualified clinical oversight. NICE (2023) Allida et al. (2023)
Antidepressant Trials and the Distinction Between Mood and Motor Recovery
The history of fluoxetine research provides a clear example of how treatment hypotheses change with trial scale. The 2011 FLAME trial randomized 118 people after ischemic stroke and observed greater mean motor-scale improvement among analyzed participants receiving fluoxetine than placebo over three months. This result prompted interest in serotonergic enhancement of neuroplastic recovery. Its clinical interpretation depended on a modest sample, selected patients, and a particular motor endpoint, so replication in larger and more representative trials was necessary before claiming a general rehabilitation benefit. Chollet et al. (2011)
Three large trials supplied that test. FOCUS randomized 3,127 participants in the United Kingdom and found no improvement in the distribution of modified Rankin Scale outcomes with six months of fluoxetine, although recorded new depression episodes were fewer and fractures more frequent. AFFINITY enrolled 1,280 participants in Australia, New Zealand, and Vietnam and likewise found no functional benefit, with increased falls, fractures, and seizures. EFFECTS enrolled 1,500 participants in Sweden and reported no functional benefit, alongside increased fractures and hyponatremia. These primary comparisons directly weaken the proposition that routine fluoxetine improves overall post-stroke function. FOCUS Trial Collaboration (2019) AFFINITY Trial Collaboration (2020) EFFECTS Trial Collaboration (2020)
A reduced incidence of depression in a trial cannot simply be reinterpreted as a favorable overall balance for preventive medication, because adverse effects and the original functional endpoint also matter. Conversely, results against routine use to enhance recovery do not answer every question about treatment of a diagnosed depressive disorder. The 2023 Cochrane treatment synthesis judged portions of the depression-treatment evidence to be of very low certainty, and contemporary guidance emphasizes individualized clinical assessment. This monograph reports evidence and tradeoffs without recommending any drug or dose for an individual survivor. Prescribing, contraindication evaluation, and monitoring belong to clinicians considering the complete medical history. Allida et al. (2023) Yao et al. (2026)
Exercise, Behavioral Activation, and Meaningful Activity
Physical activity contributes to cardiovascular health and may provide experiences of mastery, social connection, and improved perceived capability. The psychological argument for incorporating activity into stroke rehabilitation is therefore multidimensional. Activity can provide access to positive reinforcement and a setting for graded confidence building, while also having physical and medical demands that must be assessed individually. A study demonstrating improvement in gait, endurance, or strength does not automatically establish relief of a depressive disorder. The relevant scientific question is whether an activity-based program produces an independently measured improvement in mood, participation, or both, compared with an appropriate control. Winstein et al. (2016) Yao et al. (2026)
Behavioral activation principles suggest structuring achievable activities around personally valued goals, monitoring their emotional consequences, and gradually expanding opportunity. After stroke, these principles require adaptation for weakness, unpredictable fatigue, language difficulty, cognitive load, and access to safe environments. An apparent lack of engagement can reflect insufficient transportation, pain, or low initiation rather than rejection of a goal. The intervention should specify who chose an activity and how it was modified, how progress was negotiated, and whether an activity was rewarding to the survivor. Evidence for a named psychotherapy in stroke populations must be distinguished from an untested clinical inference that its general principles will transfer. Mitchell et al. (2009) Allida et al. (2023)
Trials of exercise or combined rehabilitation interventions ought to separate psychological endpoints from cardiopulmonary or motor endpoints and include sufficient follow-up to assess durability. Relevant comparisons may contrast structured group activity, individually supported activity, and usual services, while considering social exposure and therapist contact as possible active components. Measurement should include intervention attendance, actual activity outside supervised sessions, adverse events, and changes in participation. The care context matters because the same exercise prescription may have different implications for a person returning to an accessible community center and a person living far from any adapted facility. WHO (2001) Egan et al. (2026)
Adaptive Interventions and the Evidence for Social Participation
Participation interventions after stroke include occupational adaptation, social-skills and communication support, community-based practice, rehabilitation groups, family education, and multidimensional service packages. They differ in the social-ecological level at which they act: the individual, the interpersonal relationship, or the surrounding institution. Egan and colleagues' systematic review and meta-analysis distinguished changes in personally chosen participation goals from changes in broader participation and instrumental activities. The review found more favorable evidence for person-specific goals than for broad participation measures, an important warning against equating a targeted improvement with generalized community reintegration. Egan et al. (2026)
Network meta-analysis adds an indirect comparative layer when many interventions have not been tested head-to-head. Zhang and colleagues evaluated social participation interventions through a social-ecological framework and identified differences among intervention categories. Zhou and colleagues examined nonpharmacological intervention comparisons in a substantially larger evidence network. Such reviews can be informative, but ranking estimates depend on transitivity, consistency, intervention definitions, outcome comparability, and the distribution of study quality. Network rank should be interpreted alongside effect magnitude, credible or confidence intervals, and certainty judgments rather than treated as a direct clinical hierarchy. Zhang et al. (2026) Zhou et al. (2026)
A recurring design challenge is that participation depends on opportunities available after therapy ends. An intervention can improve a person's confidence and technical performance, yet achieve limited generalization when transport is inaccessible or employment accommodations are unavailable. Future studies should therefore record environmental exposures and social opportunity as explanatory variables. Researchers should examine who benefits, whether benefits extend to people with aphasia, and whether independent replication reproduces a meaningful change. The principal comparative inference is that a tailored individual goal may be easier to improve than a complex societal outcome, while both remain worth studying as separate endpoints. Egan et al. (2026) WHO (2001)
Early Supported Discharge and Rehabilitation Across Settings
Early supported discharge is a coordinated service arrangement in which selected people with stroke leave hospital earlier and receive organized rehabilitation at home. Its significance extends beyond location: professional coordination, availability of equipment, family capacity, and access to timely therapy determine whether hospital gains can be sustained in daily life. Langhorne, Baylan, and the Early Supported Discharge Trialists synthesized 17 trials with 2,422 participants and found evidence that early supported discharge can improve some outcomes in appropriately selected stroke populations. Selection, multidisciplinary organization, and local service capacity are intrinsic to the intervention being evaluated. Langhorne et al. (2017)
The psychological transition home may reveal unmet needs that were difficult to anticipate in hospital. At home, goals become concrete through meals, transport, errands, family relationships, and the presence or absence of friends. Symptoms of depression or anxiety may become apparent only when old roles are attempted, while a person's priorities can change rapidly. Discharge planning is therefore a continuity problem involving follow-up, communication, and supported decision-making. The Canadian recommendations on planning rehabilitation explicitly treat service coordination and individualized pathways as elements of appropriate stroke care rather than as optional administrative details. Nelson et al. (2026) NICE (2023)
Evidence about early supported discharge should not be generalized to all medically unstable patients or to settings without the professional infrastructure provided in studied services. The causal contrast is an organized service package compared with the control services actually available; it does not show that merely leaving a facility earlier improves mood or agency. Economic evaluation needs to account for costs transferred to households, caregiver time, and later unmet care. Psychosocial outcomes are particularly sensitive to how services follow patients after the most intensive phase has ended, making longitudinal assessment indispensable to the study of sustainable reintegration. Langhorne et al. (2017) Bakhtiari-Dovvombaygi et al. (2025)
Life After Discharge and the Persistence of Unmet Needs
The period beyond the first post-stroke year is scientifically important because rehabilitation services often become less intensive while the demands of work, relationships, and community life persist. Long-term needs can involve chronic fatigue, difficulties in concentration, changes in identity, economic insecurity, and reduced leisure. Lo and colleagues used qualitative inquiry and observations to investigate priorities after the first year and documented concerns that extend beyond basic physical recovery. These findings support the inclusion of survivor-defined priorities in long-term studies, although qualitative accounts from particular settings cannot establish population prevalence. Lo et al. (2023)
Follow-up must detect problems that emerge after the acute crisis. A survivor who initially reports little distress may later experience depression upon failing to return to a valued role; another may regain confidence through successful adaptation despite persistent physical impairment. Long-term rehabilitation is therefore concerned with changing goals and contexts. Canadian recommendations address regular health follow-up and the assessment of mood, cognition, sleep, fatigue, sexuality, and participation, while NICE outlines personalized review and access to ongoing support. Implementing such recommendations requires service availability and clear responsibility rather than a single discharge conversation. Yao et al. (2026) NICE (2023)
Scientific cohorts should distinguish trajectories of improvement, stability, delayed deterioration, and adaptation, with repeated measures of symptoms, function, environment, and participation. Selective loss to follow-up can produce overly optimistic averages when people with communication impairment or the greatest need are difficult to recontact. Researchers should document institutional residence, recurrent stroke, mortality, and changes in caregiver availability as competing or modifying events. A clinically meaningful follow-up study also asks whether rehabilitation goals still match the survivor's priorities, rather than treating the baseline goal list as permanently valid. WHO (2001) Ris et al. (2026)
Digital Support, Telerehabilitation, and Accessibility
Digital services can support remote consultations, home exercise guidance, appointment reminders, symptom reporting, education, and contact with clinicians or peers. Their possible value in post-stroke psychosocial recovery lies in improving continuity, reducing some travel burdens, and creating alternative communication routes. Those benefits depend on the device, interface, network access, privacy protections, and whether the chosen task is clinically appropriate for remote delivery. A person with aphasia, visual field loss, or executive dysfunction may face substantial barriers to interfaces designed for a general population. The availability of an application alone is therefore not evidence of accessible rehabilitation. NICE (2023) Yao et al. (2026)
Technology-based rehabilitation should be assessed through the complete service arrangement. Does a person receive useful guidance and timely human follow-up? Can fatigue-limited or communication-impaired users complete the procedure? Are caregivers expected to supply unpaid technical labor? Is the intervention safe and responsive when mood symptoms worsen? These implementation questions complement efficacy trials. In particular, a digital program intended to increase self-management can reduce effective agency when the user has little control over notifications, interpretation, or the consequences of recorded data. Evaluation should distinguish genuine user choice from automated workflow completion. WHO (2001) Nelson et al. (2026)
Theoretical potential must be separated from verified clinical effect. For stroke rehabilitation, the evidence base differs across physical telerehabilitation, remote psychological treatment, and social connection platforms; results from one application category should not be transferred to another. Outcome studies should preregister appropriate measures of accessibility, patient-reported acceptability, mood, and sustained participation, and report technical failures and dropout. Remote tools may also complement in-person services rather than replacing them. Their long-term contribution is contingent on clinical integration and equitable access to the resources required for meaningful use. Winstein et al. (2016) Yao et al. (2026)
Inequality, Culture, and the Distribution of Rehabilitation Opportunities
Stroke rehabilitation takes place within unequal distributions of income, housing, transport, language access, health insurance, educational opportunity, and informal care. These conditions influence who can attend treatment, purchase assistive equipment, resume paid work, and maintain social relationships. The ICF recognizes environmental determinants of participation, but empirical research must still measure their distribution and effects. Two survivors with equivalent neurological impairment may experience substantially different community participation because their local environments provide different affordances. Structural disadvantage can also precede stroke and confound associations between post-stroke mood and later functioning. WHO (2001) Nelson et al. (2026)
Cultural variation matters for the meaning of independence, obligations within households, public expressions of distress, and desired participation. Researchers should avoid assuming that employment, living alone, or a specific leisure activity has the same significance across societies. A person's account of a valued role is necessary to interpret its loss or modification. Instruments translated into another language require more than literal translation: conceptual equivalence, accessible communication, and measurement performance should be investigated. A single pooled prevalence estimate or intervention effect across diverse settings can conceal substantial variation in baseline resources and access. Hackett & Pickles (2014) Duncan et al. (1999)
Equity-sensitive trials should report sampling frames, inclusion of aphasia and cognitive impairment, recruitment location, socioeconomic distribution, and who declined or could not access participation. Community-based studies can combine standardized scales with structured environmental observations and qualitative interviews. Evaluation must also consider whether an intervention reduces a burden or merely shifts it to households with limited resources. The scientific purpose is to establish how opportunities modify the relationship between symptoms and participation, including whether interventions proven effective in well-resourced rehabilitation centers remain useful in under-resourced settings. WHO (2001) Langhorne et al. (2017)
Measurement, Psychometrics, and Construct Overlap
The main constructs in this field have multiple operational definitions. Depression may be identified by diagnostic interview or threshold on a symptom scale; apathy may be classified using informant or clinical ratings; fatigue may be self-reported; agency may be inferred from task-specific confidence; participation may be measured by perceived restriction, frequency of action, or goal attainment. Such methods do not yield interchangeable numbers. A well-designed study states which construct each instrument targets, what its validation population was, and how aphasia, motor disability, or fatigue may interfere with administration. Hackett & Pickles (2014) Zhang et al. (2023) Duncan et al. (1999)
Overlap is substantial. A lack of initiation might increase depression or apathy scores; tiredness may appear in fatigue and depression questionnaires; reduced activity may occur because of bodily impairment, avoidance, or inaccessible environments. Correlated measures may partly reflect shared items or common reporting method. Discriminant validity can be evaluated through factor structures, clinical interviews, cross-method measures, and responsiveness to different interventions. The Dynamic Visual Analogue Mood Scales were evaluated with stroke participants including people with aphasia, illustrating the importance of testing accessible response formats rather than simply excluding users who cannot complete ordinary verbal scales. Barrows & Thomas (2018) Jones et al. (2008)
For comparative studies, the minimal measurement design includes baseline severity, a clearly specified primary endpoint, documentation of missingness, appropriate follow-up, and transparent reporting of effect estimates with uncertainty. Participant-valued outcomes may require individualized targets supplemented by standardized measures. Researchers should report whether questionnaire changes exceed established measurement error and whether observed shifts correspond to meaningful daily life. The goal is to establish what changed and for whom, while avoiding claims about a broad construct when the instrument captured only one component. Egan et al. (2026) Duncan et al. (1999)
Causal Inference, Trial Design, and Evidence Synthesis
Causal claims about psychological rehabilitation require an intervention contrast that can support them. Randomized trials reduce confounding in the treatment assignment but remain subject to differential attrition, unblinded outcome assessment, adherence problems, imprecise estimates, and selective reporting. Observational cohorts are valuable for describing symptom trajectories and plausible predictors, while requiring explicit control of prior disability, baseline mood, age, social resources, and treatment selection. Cross-sectional studies of motivation and community participation cannot establish which variable caused the other. These differences in design explain why findings must be interpreted at the level of the question each study can answer. Hackett & Pickles (2014) Mitchell et al. (2009)
The fluoxetine program is a model case in cumulative evidence. An encouraging small trial of motor recovery was followed by three substantially larger randomized trials in different health systems; the later results did not support routine fluoxetine to improve broad functional outcomes and identified adverse events. This does not erase the original experiment or prove that all pharmacological treatment of clinically diagnosed depression is ineffective. It identifies the particular hypothesis that failed a stronger empirical test, while preserving separate treatment questions for which evidence certainty differs. Chollet et al. (2011) FOCUS Trial Collaboration (2019) AFFINITY Trial Collaboration (2020) EFFECTS Trial Collaboration (2020) Allida et al. (2023)
Evidence syntheses introduce additional assumptions. Conventional meta-analysis requires comparable effect definitions, appropriately handled dependence, and assessment of heterogeneity and publication bias. Network meta-analysis further requires valid indirect comparisons. A systematic review's inclusion criteria and risk-of-bias judgments are part of its conclusions, not merely a preface. The present monograph is a structured critical narrative synthesis based on published studies; it does not claim exhaustive screening, preregistration, pooled participant analysis, or new quantitative effect estimates. The underlying studies retain their own populations and limitations, with discrepancies treated as substantive information for future research. Egan et al. (2026) Zhang et al. (2026) Allida et al. (2023)
Safety, Clinical Judgment, and Individualized Care
Post-stroke mood disorders may coexist with suicidal ideation, cognitive impairment, medication interactions, recurrent vascular risk, and the challenges of recovery from a major medical event. Safe evaluation requires a qualified clinical team able to distinguish affective symptoms from neurological syndromes, assess urgent risk, and coordinate psychological with medical care. The availability of a screening score does not establish diagnosis, and general evidence from randomized trials does not support individualized prescribing without a complete assessment. Contemporary NICE and Canadian guidance emphasize multidisciplinary planning and adapted communication for survivors with varied needs. NICE (2023) Yao et al. (2026)
A treatment balance includes possible harms, not only statistically favorable outcomes. In the FOCUS, AFFINITY, and EFFECTS fluoxetine trials, the absence of a functional benefit coincided with important adverse events, varying by study. These findings are relevant to routine preventive prescribing intended to enhance stroke recovery. They do not justify discontinuation of a medicine without medical supervision or an assumption that individual depression care is unnecessary. Psychological therapies likewise require monitoring of distress, suitability, burden, and access barriers, particularly when an intervention increases demands on fatigued participants. FOCUS Trial Collaboration (2019) AFFINITY Trial Collaboration (2020) EFFECTS Trial Collaboration (2020) Allida et al. (2023)
Person-centered safety includes respecting the survivor's preferences through supported communication and decision-making. Families may provide necessary practical assistance while still holding views different from the survivor's. Clinical teams should distinguish informed refusal from communication difficulty, and task-specific capacity from global judgments based on aphasia or movement impairment. Referral processes for worsening depression, new neuropsychiatric symptoms, or safety concerns should be explicit. The scientific literature supports coordinated assessment and service provision while leaving individual diagnosis, monitoring, and treatment decisions to accountable professionals. Barrows & Thomas (2018) Rosewilliam et al. (2011) NICE (2023)
Human–Artificial Support and Research in the Artificial Era
The Artificial Era adds a new class of tools and collaborators to the practical ecology of human rehabilitation. Language-based systems, assistive interfaces, and adaptive prompts could be studied as sources of reminders, accessible information, communication scaffolding, or support for planning. Their presence does not establish that an artificial system has clinical judgment or that a conversational interaction is equivalent to qualified rehabilitation care. The psychologically relevant questions concern a stroke survivor's effective agency, understanding, control over data, accessibility, and continuing contact with human services. A broad ecological view permits such systems to be investigated without making unsupported assertions about their therapeutic effect. WHO (2001) Yao et al. (2026)
Automated assistance may improve execution of a task while changing who selects the task, interprets progress, or decides when it has been completed. Such redistribution of control should be made observable in research protocols. Studies could compare patient-directed prompting with fixed automated instruction, measure participation in familiar environments, and record whether a survivor can correct or disregard an artificial recommendation. For aphasia and cognitive impairment, usability and comprehension need independent evaluation; a visually attractive interface is not evidence that instructions can be used safely. These are hypotheses and prospective study designs rather than findings already demonstrated for post-stroke mood or community reintegration. Barrows & Thomas (2018) Jones et al. (2008)
The central scientific requirement is that claims about artificial rehabilitation supports be tied to precise task definitions, validated measures, and appropriately controlled outcomes. Data provenance, consent, privacy, accessibility, and clinical escalation pathways are essential parts of the intervention. An artificial companion can be studied in relation to loneliness or motivation, but any claimed improvement requires prospective evidence comparing relevant alternatives and monitoring potential dependency or withdrawal from human relationships. This inquiry belongs within psychological rehabilitation and human–artificial interaction as a research frontier, with explicit separation between established stroke science and emerging technological possibilities. WHO (2001) Allida et al. (2023)
Research Priorities and a Cumulative Scientific Agenda
A priority for stroke psychosocial rehabilitation is the inclusion of people often excluded from controlled studies: those with aphasia, cognitive impairment, substantial disability, multiple chronic conditions, and limited digital access. Studies need accessible measures and supported consent procedures that preserve the participant's own preferences. Representative sampling will improve estimates of mood, apathy, and fatigue and clarify whether known interventions generalize to diverse community contexts. Long-term follow-up should include life transitions after rehabilitation services end, distinguishing durable gains from immediate completion of a treatment protocol. Barrows & Thomas (2018) Hackett & Pickles (2014) Yao et al. (2026)
Intervention development should target mechanisms clearly enough to permit testing. Behavioral activation might be compared with other depression treatments while measuring actual valued activity; agency-focused goal setting might be assessed through both subjective choice and observed control over decisions; environmental accommodations might be randomized or evaluated through strong natural experiments where feasible. Trials should prespecify a small set of meaningful primary outcomes and report harms, intervention fidelity, economic burden, and individual differences. Mixed-methods work can then explain why outcomes differ across households, professions, cultures, and service organizations. Mitchell et al. (2009) Ris et al. (2026) Egan et al. (2026)
Cumulative science further requires independent replication, transparent publication of null results, and careful treatment of overlapping reviews. The most scientifically useful question is often not which package ranks first in a broad indirect network, but which component works for which outcome, in which population and environment, with what uncertainty. Research should test durable transfer from therapy settings to work, leisure, transportation, and relationships, treating survivor-defined goals and validated standard measures as complementary. This agenda connects the clinical literature on mood to a reproducible science of agency and community participation. Zhang et al. (2026) Zhou et al. (2026) WHO (2001)
Conclusion
Stroke rehabilitation is an extended psychological and social process in which emotional symptoms, capacity for self-directed action, and opportunities for participation jointly shape the course of life after cerebrovascular injury. Post-stroke depression, anxiety, apathy, fatigue, emotionalism, and communication difficulty have distinct meanings and can coexist. Their assessment demands validated methods that remain accessible to people with changing cognitive and expressive abilities. Stroke-specific psychological care gains scientific precision when these phenomena are considered alongside the practical settings in which people attempt to resume valued relationships, household activity, work, travel, and leisure. Hackett & Pickles (2014) Zhang et al. (2023) WHO (2001)
The strongest evidence supports a differentiated interpretation of interventions. Some psychosocial-behavioral treatment has shown promising benefits for depression in defined trial populations, while overall certainty across treatments remains limited in key systematic reviews. Large placebo-controlled fluoxetine trials failed to demonstrate improved functional recovery and revealed important adverse events, correcting the optimism generated by an earlier smaller trial. Adaptive participation interventions demonstrate why personally selected goals and broad community outcomes should be measured separately. Early supported discharge findings underscore the importance of organized services and continuity across care settings, with effects dependent on appropriate selection and delivery. Mitchell et al. (2009) Allida et al. (2023) FOCUS Trial Collaboration (2019) AFFINITY Trial Collaboration (2020) EFFECTS Trial Collaboration (2020) Egan et al. (2026) Langhorne et al. (2017)
A comprehensive scientific account of stroke recovery therefore requires patient-defined goals, high-quality causal designs, long-term measurement, and explicit evaluation of environmental access and service equity. Agency depends on effective influence over decisions as well as bodily capacity; reintegration depends on actual opportunities as well as personal motivation. Future investigations should connect mood, functional performance, personal choice, and participation with measurement tools and designs capable of distinguishing those pathways. The enduring research problem is how coordinated psychological and social rehabilitation supports people in building lives they themselves recognize as meaningful after stroke. Ris et al. (2026) Yao et al. (2026) Duncan et al. (1999)
References
- AFFINITY Trial Collaboration. (2020). Safety and efficacy of fluoxetine on functional outcome after acute stroke (AFFINITY): A randomised, double-blind, placebo-controlled trial. The Lancet Neurology, 19(8), 651–660. https://doi.org/10.1016/S1474-4422(20)30207-6
- Allida, S. M., Hsieh, C.-F., Cox, K. L., Patel, K., Rouncefield-Swales, A., Lightbody, C. E., House, A., & Hackett, M. L. (2023). Pharmacological, non-invasive brain stimulation and psychological interventions, and their combination, for treating depression after stroke. Cochrane Database of Systematic Reviews, 2023(7), CD003437. https://doi.org/10.1002/14651858.CD003437.pub5
- Bakhtiari-Dovvombaygi, H., Zare-Kaseb, A., Nazari, A. M., Rezazadeh, Y., & Bahramnezhad, F. (2025). The effect of interventions on quality of life, depression, and the burden of care of stroke patients and their caregivers: A systematic review. Journal of Neuroscience Nursing, 57(1), 44–50. https://doi.org/10.1097/JNN.0000000000000803
- Barrows, P. D., & Thomas, S. A. (2018). Assessment of mood in aphasia following stroke: Validation of the Dynamic Visual Analogue Mood Scales (D-VAMS). Clinical Rehabilitation, 32(1), 94–102. https://doi.org/10.1177/0269215517714590
- Broomfield, N. M., Blake, J., Gracey, F., & Steverson, T. (2024). Post-stroke emotionalism: Diagnosis, pathophysiology, and treatment. International Journal of Stroke, 19(8), 857–866. https://doi.org/10.1177/17474930241242952
- Campbell Burton, C. A., Murray, J., Holmes, J., Astin, F., Greenwood, D., & Knapp, P. (2013). Frequency of anxiety after stroke: A systematic review and meta-analysis of observational studies. International Journal of Stroke, 8(7), 545–559. https://doi.org/10.1111/j.1747-4949.2012.00906.x
- Chollet, F., Tardy, J., Albucher, J.-F., Thalamas, C., Berard, E., Lamy, C., Bejot, Y., Deltour, S., Jaillard, A., Niclot, P., Guillon, B., Moulin, T., Marque, P., Pariente, J., Arnaud, C., & Loubinoux, I. (2011). Fluoxetine for motor recovery after acute ischaemic stroke (FLAME): A randomised placebo-controlled trial. The Lancet Neurology, 10(2), 123–130. https://doi.org/10.1016/S1474-4422(10)70314-8
- Cumming, T. B., Yeo, A. B., Marquez, J., Churilov, L., Annoni, J.-M., Badaru, U., Ghotbi, N., Harbison, J., Kwakkel, G., Lerdal, A., Mills, R., Naess, H., Nyland, H., Schmid, A., Tang, W. K., Tseng, B., van de Port, I., Mead, G., & English, C. (2018). Investigating post-stroke fatigue: An individual participant data meta-analysis. Journal of Psychosomatic Research, 113, 107–112. https://doi.org/10.1016/j.jpsychores.2018.08.006
- Duncan, P. W., Wallace, D., Lai, S. M., Johnson, D., Embretson, S., & Laster, L. J. (1999). The stroke impact scale version 2.0: Evaluation of reliability, validity, and sensitivity to change. Stroke, 30(10), 2131–2140. https://doi.org/10.1161/01.STR.30.10.2131
- EFFECTS Trial Collaboration. (2020). Safety and efficacy of fluoxetine on functional recovery after acute stroke (EFFECTS): A randomised, double-blind, placebo-controlled trial. The Lancet Neurology, 19(8), 661–669. https://doi.org/10.1016/S1474-4422(20)30219-2
- Egan, M., Kessler, D., Duong, P., Duquette, K., Gurgel-Juarez, N., Sikora, L., & Linkewich, E. (2026). Adaptive interventions for enhancing participation poststroke: A systematic review and meta-analysis. Canadian Journal of Occupational Therapy, 93(1), 104–118. https://doi.org/10.1177/00084174251352305
- FOCUS Trial Collaboration. (2019). Effects of fluoxetine on functional outcomes after acute stroke (FOCUS): A pragmatic, double-blind, randomised, controlled trial. The Lancet, 393(10168), 265–274. https://doi.org/10.1016/S0140-6736(18)32823-X
- Hackett, M. L., Yapa, C., Parag, V., & Anderson, C. S. (2005). Frequency of depression after stroke: A systematic review of observational studies. Stroke, 36(6), 1330–1340. https://doi.org/10.1161/01.STR.0000165928.19135.35
- Hackett, M. L., & Pickles, K. (2014). Part I: Frequency of depression after stroke: An updated systematic review and meta-analysis of observational studies. International Journal of Stroke, 9(8), 1017–1025. https://doi.org/10.1111/ijs.12357
- Jones, F., Partridge, C., & Reid, F. (2008). The Stroke Self-Efficacy Questionnaire: Measuring individual confidence in functional performance after stroke. Journal of Clinical Nursing, 17(7B), 244–252. https://doi.org/10.1111/j.1365-2702.2008.02333.x
- Langhorne, P., Baylan, S., & Early Supported Discharge Trialists. (2017). Early supported discharge services for people with acute stroke. Cochrane Database of Systematic Reviews, 2017(7), CD000443. https://doi.org/10.1002/14651858.CD000443.pub4
- Li, J., Pan, X., Wang, Z., Zhong, W., Yao, L., & Xu, L. (2024). Interventions to support the return to work for individuals with stroke: A systematic review and meta-analysis. Journal of Occupational Rehabilitation, 34(4), 740–755. https://doi.org/10.1007/s10926-024-10178-y
- Lin, M., Xie, S., Wang, B., & Siu, A. K.-Y. (2026). Effectiveness of interventions for caregiver burden and quality of life of informal caregivers of stroke survivors: A systematic review and meta-analysis. Frontiers in Public Health, 14, 1932478. https://doi.org/10.3389/fpubh.2026.1932478
- Lo, S. H. S., Chau, J. P. C., Lam, S. K. Y., & Saran, R. (2023). Understanding the priorities in life beyond the first year after stroke: Qualitative findings and non-participant observations of stroke survivors and service providers. Neuropsychological Rehabilitation, 33(5), 794–820. https://doi.org/10.1080/09602011.2022.2049827
- Mitchell, P. H., Veith, R. C., Becker, K. J., Buzaitis, A., Cain, K. C., Fruin, M., Tirschwell, D. L., & Teri, L. (2009). Brief psychosocial-behavioral intervention with antidepressant reduces poststroke depression significantly more than usual care with antidepressant: Living well with stroke: Randomized, controlled trial. Stroke, 40(9), 3073–3078. https://doi.org/10.1161/STROKEAHA.109.549808
- National Institute for Health and Care Excellence. (2023). Stroke rehabilitation in adults (NICE Guideline NG236). NICE. https://www.nice.org.uk/guidance/ng236/chapter/Recommendations
- Nelson, M. L. A., Shi, J., Lindsay, M. P., Salbach, N. M., Yao, J. K., Timpson, D., Ritsma, B. R., Auger, L.-P., Beaumont, J., Bowes, R., Chibane, I. S., Courtice, S. J., Delgado, R., Dunlop, M., Foley, N., Ghavami, K., Guolla, T., Kean, D., MacFayden, S., ... Mountain, A. (2026). Canadian Stroke Best Practice Recommendations: Rehabilitation, recovery and community participation following stroke. Part one: Stroke rehabilitation planning for optimal care delivery, 7th edition update 2025. American Journal of Physical Medicine & Rehabilitation, 105(1), 59–75. https://doi.org/10.1097/PHM.0000000000002802
- Ris, I., Marwaa, M. N., Nielsen, R. W., Larsen, P., & Kristensen, H. K. (2026). Patient-centered goal-setting in stroke rehabilitation: A scoping review. Frontiers in Rehabilitation Sciences, 7, 1744900. https://doi.org/10.3389/fresc.2026.1744900
- Robinson, R. G., & Jorge, R. E. (2016). Post-stroke depression: A review. American Journal of Psychiatry, 173(3), 221–231. https://doi.org/10.1176/appi.ajp.2015.15030363
- Rosewilliam, S., Roskell, C. A., & Pandyan, A. D. (2011). A systematic review and synthesis of the quantitative and qualitative evidence behind patient-centred goal setting in stroke rehabilitation. Clinical Rehabilitation, 25(6), 501–514. https://doi.org/10.1177/0269215510394467
- Winstein, C. J., Stein, J., Arena, R., Bates, B., Cherney, L. R., Cramer, S. C., Deruyter, F., Eng, J. J., Fisher, B., Harvey, R. L., Lang, C. E., MacKay-Lyons, M., Ottenbacher, K. J., Pugh, S., Reeves, M. J., Richards, L. G., Stiers, W., & Zorowitz, R. D. (2016). Guidelines for adult stroke rehabilitation and recovery: A guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke, 47(6), e98–e169. https://doi.org/10.1161/STR.0000000000000098
- World Health Organization. (2001). International classification of functioning, disability and health (ICF). World Health Organization. https://www.who.int/standards/classifications/international-classification-of-functioning-disability-and-health
- Yao, J. K., Salbach, N. M., Lindsay, M. P., Nelson, M. L. A., Shi, J., O'Connell, C., Barclay, R., Bastasi, D., Boulos, M. I., Boyce, J., Claveau, G., Flowers, H. L., Foley, N., Gopaul, U., Kim, E. S., Lo, A., McDonald, A. M., McIntyre, A., O'Connor, C., ... Mountain, A. (2026). Canadian Stroke Best Practice Recommendations rehabilitation, recovery, and community participation following stroke, part three: Optimizing activity and community participation following stroke, 7th edition update, 2025. American Journal of Physical Medicine & Rehabilitation, 105(3), 238–252. https://doi.org/10.1097/PHM.0000000000002845
- Zhan, J., Zhang, P., Wen, H., Wang, Y., Yan, X., Zhan, L., Chen, H., Xu, N., & Lu, L. (2023). Global prevalence estimates of poststroke fatigue: A systematic review and meta-analysis. International Journal of Stroke, 18(9), 1040–1050. https://doi.org/10.1177/17474930221138701
- Zhang, H., Feng, Y., Lv, H., Tang, S., & Peng, Y. (2023). The prevalence of apathy in stroke patients: A systematic review and meta-analysis. Journal of Psychosomatic Research, 173, 111478. https://doi.org/10.1016/j.jpsychores.2023.111478
- Zhang, H., Yang, Y., Li, F., Ma, L., Abudureyimu, A., Cheng, L. J., Li, K., & Wu, X. V. (2026). Comparative effects of social participation interventions for stroke survivors: A network meta-analysis using a social ecological model. Annals of Physical and Rehabilitation Medicine, 69(1), 102029. https://doi.org/10.1016/j.rehab.2025.102029
- Zhou, X., Song, M., Qin, H., Liu, R., Wang, Y., & Zhou, L. (2026). Effects of non-pharmacological interventions for social participation in stroke survivors: A systematic review and network meta-analysis. Annals of Physical and Rehabilitation Medicine, 69(4), 102123. https://doi.org/10.1016/j.rehab.2026.102123
Social Connection, Friendship, and Leisure
Social reintegration includes reciprocity, companionship, conversation, cultural activity, and access to collective spaces. Frequency of contact is one indicator, but perceived belonging and the ability to choose relationships are additional dimensions. Stroke may alter communication, mobility, and endurance and change how others respond to the survivor. The social consequences depend on accessible venues, transportation, invitations, employment schedules, and the willingness of communities to accommodate difference. Psychological research must therefore examine interpersonal experiences and practical conditions together. The Canadian 2025 community-participation recommendations treat leisure and social roles as core domains of long-term recovery. Yao et al. (2026) WHO (2001)
Reduced engagement can have several origins. Depression can diminish pleasure; anxiety can increase avoidance; apathy can inhibit initiation; fatigue can make attendance costly; aphasia can require more patient conversational partners. These mechanisms imply different evaluation and support questions despite a similar outward appearance of isolation. A person who wants social contact but encounters inaccessible transportation faces a structural problem, while another may benefit from targeted mood treatment. Peer-support arrangements and communication adaptations are plausible supports, but claims about lasting benefit require controlled comparisons and measures meaningful to participants. Zhang et al. (2023) Barrows & Thomas (2018) Yao et al. (2026)
Interventions that target social participation show promise alongside considerable heterogeneity. The 2025-to-2026 literature includes network meta-analyses that compare multidimensional and psychological interventions, but their network assumptions and risk-of-bias assessments are crucial. A high rank in a network is not proof of a durable benefit for a particular survivor. Evaluations should measure social participation after services end, whether new relationships persist, and whether improvements are accessible to participants with substantial aphasia, mobility restrictions, or limited income. Zhang et al. (2026) Zhou et al. (2026)