Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Rehabilitation in Later Life: Multimorbidity, Autonomy, and Participation

Abstract. Rehabilitation in later life is the coordinated modification of health-related functional limitations, environmental constraints, behavioral strategies, and care relationships to sustain activities and participation that older people value. This fundamental critical narrative research monograph examines how multimorbidity, frailty, changing intrinsic capacity, acute illness, rehabilitation services, autonomy, and community participation interact over time. It traces the development of ecological gerontology, functional assessment, the International Classification of Functioning, Disability and Health, person-centered geriatric assessment, and contemporary rehabilitation trials. Original randomized studies of home adaptation, structured physical activity, and multidisciplinary care are evaluated alongside large epidemiological investigations, systematic reviews, and the 2026 geriatric rehabilitation process synthesis. Evidence supports the modifiability of particular functional outcomes and the value of coordinated care, while the effects of treatment on participation, durability, equity, and personally selected goals remain less consistently measured. The monograph integrates psychological accounts of agency, self-efficacy, fear, adaptation, and decision-making with clinical and environmental evidence, then defines priorities for longitudinal, participatory, and implementation research. It is an evidence-informed critical narrative synthesis, not a new randomized trial, a protocol-based systematic review, or a meta-analysis.

Keywords: geriatric rehabilitation; later life; multimorbidity; functional ability; autonomy; participation; frailty; comprehensive geriatric assessment; rehabilitation psychology; healthy aging; intrinsic capacity; person-centered care; activities of daily living; shared decision-making; home modification; community inclusion; functional recovery; rehabilitation outcomes; polypharmacy; rehabilitation services

Scientific Overview and Research Questions

Rehabilitation in later life is best studied as a changing relationship among bodily capacities, psychological processes, daily activities, interpersonal support, and the opportunity to participate in society. A person may regain strength after hospitalization and still be unable to visit friends because transportation, inaccessible housing, or fear of falling prevents community mobility. Conversely, mobility aids, environmental adaptations, and shared planning can preserve valued roles despite persistent impairment. The WHO International Classification of Functioning, Disability and Health (2001) (https://www.who.int/classifications/international-classification-of-functioning-disability-and-health) supplies a vocabulary for studying these relationships. The WHO healthy aging framework (https://www.who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability) emphasizes what people can be and do in the environments available to them. Source 23 Source 26

The questions organizing this monograph are causal, conceptual, and normative. How does the co-occurrence of chronic conditions shape rehabilitation capacity and treatment burden? Which elements of rehabilitation reliably change activities and real-world participation? How should autonomy be recognized when assistance is required? Which psychological mechanisms determine whether a person enters, persists in, or withdraws from rehabilitation? How can researchers distinguish meaningful benefit from changes in a convenient performance score? How do community resources, caregivers, financing, and service organization affect the durability of gains? The discussion compares original interventions with evidence syntheses and separates measured effects from mechanisms proposed to explain them.

This monograph uses a structured, question-led critical narrative review of identifiable primary studies, historical conceptual scholarship, professional guidance, and high-quality syntheses. The evidence base includes randomized trials, large observational analyses, a Cochrane review, and a 2026 review of rehabilitation processes. Sources were selected for direct relevance to later-life functioning, multimorbidity, autonomy, environmental adaptation, and participation; this is not an exhaustive protocol-driven search with duplicate screening, risk-of-bias adjudication for every eligible study, or pooled quantitative estimation. Causal conclusions are restricted to the populations, comparisons, and outcomes actually studied.

Definitions and Boundaries of the Research Object

Later-life rehabilitation refers here to multidisciplinary or targeted actions intended to optimize functioning and enable valued activities among older adults experiencing illness, injury, disability, or reduced capacity. It encompasses prevention of additional functional decline, recovery after a discrete event, restoration of skills, modification of routines and settings, and sustained access to support. It may occur in hospitals, rehabilitation facilities, primary care, homes, or communities. Age itself is an imperfect eligibility rule: an older person with high capacity, a younger person living with extensive disability, and an older person after a hip fracture present distinct questions.

Multimorbidity denotes the coexistence of two or more long-term conditions, although specific research definitions and condition lists vary. Frailty describes heightened vulnerability to stressors and is measured through distinct operational frameworks. Functional ability concerns opportunities to perform meaningful activities through the interaction of intrinsic capacity and environment. Autonomy is the capacity to participate in determining goals and governing choices, including when supported by other people or technology. Participation concerns involvement in life situations; attendance, performance, subjective significance, and freedom of choice are analytically different dimensions. These terms must remain distinct when outcomes are measured. The ICF (https://www.who.int/classifications/international-classification-of-functioning-disability-and-health) and WHO healthy aging account (https://www.who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability) ground this distinction. Source 23 Source 26

The research object is the later-life rehabilitation trajectory through multimorbidity, autonomy, and participation. Generic models of cognitive aging, the entire evidence base of every chronic disease, and condition-specific neurological therapies are adjacent subjects. The present synthesis draws on those literatures where they explain rehabilitation organization, functional transfer, or older adults' choices, while retaining a dedicated focus on late-life transitions. A functional change measured in a clinic and a return to a personally meaningful community role therefore occupy different positions in the explanation.

History of Geriatric Rehabilitation and Functional Thinking

The modern history of this field includes successive changes in what counts as success. Disease-oriented practice prioritized survival, symptom reduction, and biomedical recovery. Functional research made basic activities, walking, transfers, and the consequences of hospitalization visible as outcomes in their own right. Geriatric assessment joined medical, psychological, social, and functional information in a single care process; subsequent experimental designs assessed whether organized geriatric services improved the probability of remaining at home. The Ellis et al. (2017) Cochrane review (https://doi.org/10.1002/14651858.CD006211.pub3) documents a mature body of trials on comprehensive geriatric assessment, while Bachmann et al. (2010) (https://doi.org/10.1136/bmj.c1718) investigated rehabilitation programs specifically designed for geriatric inpatients. Source 7 Source 2

The development of standardized functional measures allowed investigators to compare risk and change across cohorts. Guralnik et al. (1994) (https://doi.org/10.1093/geronj/49.2.m85) investigated the Short Physical Performance Battery (SPPB), relating lower-extremity performance to self-reported disability and subsequent outcomes. The WHO formally endorsed the ICF in 2001, creating an international classification that places activities, participation, and contextual factors alongside body structures and functions. These developments shifted rehabilitation toward the measured conditions under which people live, while retaining the need for rigorous clinical evaluation. Source 11

A further transition has been the movement from standardized impairment reduction to person-valued endpoints and co-produced goals. WHO Rehabilitation 2030 (https://www.who.int/initiatives/rehabilitation-2030/) calls for rehabilitation to be integrated across health services and the life course. The 2025 second-edition WHO ICOPE guidance (https://www.who.int/publications/i/item/9789240103726) describes pathways for identifying declines in intrinsic capacity, assessing care needs, and constructing personalized plans. These initiatives are policy and practice frameworks; their adoption does not itself establish the effectiveness of every component in every setting. Source 25 Source 27

Ecological Gerontology and Person–Environment Fit

Ecological gerontology explains why the same physiological capacity can produce very different levels of daily functioning. Lawton and Nahemow (1973) (https://doi.org/10.1037/10044-020) conceptualized aging through transactions between personal competence and environmental demands. Stairs, complex transportation, limited lighting, confusing service interfaces, and fragmented care can impose demands that exceed current capacity. Handrails, accessible routes, predictable information, and reliable assistance may change performance without changing the underlying disease. Environment is therefore an active determinant of functioning and a legitimate intervention target. Source 12

This approach is particularly useful in multimorbidity because reserve and task demands fluctuate. A person managing pain, breathlessness, fatigue, and sensory impairment may perform a task successfully in a quiet home but encounter disabling barriers in a clinic or crowded street. Clinical measures that ignore environmental press risk treating a particular setting as a fixed property of the person. Environmental assessment should document actual living conditions, physical and social barriers, local service access, and the support needed to perform chosen activities.

Ecological reasoning also cautions against equating an accessible environment with a low-demand environment. Accessible environments can support challenging and meaningful activity; an intervention that eliminates all risk by eliminating the person's opportunities for participation may undermine the very outcome rehabilitation seeks. The scientific question is whether accommodations permit a person to pursue priorities with an acceptable distribution of benefits and risks. This makes person–environment fit a testable interaction, not a decorative metaphor.

Multimorbidity as a System of Interacting Demands

Multimorbidity affects more than the sum of disease severities. Barnett et al. (2012) (https://doi.org/10.1016/S0140-6736(12)60240-2) examined 40 conditions among more than 1.7 million people registered in Scottish primary care, showing the importance of age and socioeconomic deprivation in patterns of multiple conditions. A cross-sectional design establishes population distribution and associations; it cannot demonstrate that multimorbidity causes a specific rehabilitation failure. Yet it provides a strong reason to question the external validity of single-condition models when applied to older adults with competing care requirements. Source 4

Conditions may interact in several ways. Joint pain may constrain activity prescribed for cardiovascular health; sensory loss can complicate instruction; depressed mood may reduce engagement; medication adverse effects can alter balance or attention. These relationships require condition-specific assessment rather than a generic assumption that two diagnoses automatically imply additive disability. Treatment burden is also an exposure: appointments, travel, medication organization, administrative work, and competing recommendations consume finite time, energy, money, and attention.

Boyd et al. (2005) (https://doi.org/10.1001/jama.294.6.716) analyzed the difficulties of applying disease-specific clinical guidelines to older patients with multiple conditions. The American Geriatrics Society Expert Panel (2012) (https://doi.org/10.1111/j.1532-5415.2012.04188.x) emphasized patient preferences, evidence interpretation, prognosis, treatment complexity, and care planning. NICE NG56 (https://www.nice.org.uk/guidance/NG56) similarly foregrounds shared priorities and the reduction of treatment burden. In rehabilitation research, intervention intensity should therefore be considered alongside competing disease management work, access costs, fatigue, and the possibility that additional therapy time displaces valued life activities. Source 6 Source 1 Source 15

Frailty, Physiological Reserve, and Heterogeneous Aging

Frailty predicts vulnerability in many care contexts, but its major measures capture different constructs. Fried et al. (2001) (https://doi.org/10.1093/gerona/56.3.m146) proposed a physical phenotype grounded in criteria including weight loss, exhaustion, weakness, slowness, and low activity. Rockwood and Mitnitski (2007) (https://doi.org/10.1093/gerona/62.7.722) described frailty in relation to accumulated deficits. The first emphasizes a specified physical syndrome; the second integrates a broader inventory of health problems. Their scores can correlate without being interchangeable, and intervention selection should identify which definition was used. Source 8 Source 17

Frailty is closely related to but conceptually different from disability and multimorbidity. A person can have several stable diagnoses and retain substantial functional capacity. Another may have low reserve and become dependent after a relatively minor illness. A frailty score may predict a transition but does not by itself establish its cause or justify excluding a person from rehabilitation. Prognostic tools should inform the magnitude and kind of assistance offered, with results interpreted against the person's goals and actual setting.

The variability of older adults is a fundamental research fact. The WHO healthy aging framework (https://www.who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability) specifically recognizes diversity in capacity at a given age. Heterogeneity should shape trial inclusion, outcome stratification, accessibility design, and interpretation of average effects. A result concentrated among participants with moderate frailty may have limited transportability to those living with advanced cognitive impairment, severe instability, or limited community support. Source 26

Intrinsic Capacity and the Functional Ability Framework

Intrinsic capacity describes the combined physical and mental capacities available to a person. Functional ability describes what those capacities permit in conjunction with a particular environment. The WHO healthy aging framework (https://www.who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability) includes meeting needs, making decisions, moving, relating to others, and contributing to society among valued functions. The distinction matters because rehabilitation can operate through bodily improvement, environmental redesign, altered task organization, assistive technologies, and interpersonal support. Source 26

The WHO ICOPE handbook, second edition (2025) (https://www.who.int/publications/i/item/9789240103726) presents person-centered assessment and care pathways that consider declines in intrinsic capacity and social support needs. Its value for a research synthesis is architectural: it specifies where screening, individualized planning, referral, and longitudinal coordination belong in a care pathway. The empirical validity of a particular screening rule or service configuration still depends on studies of sensitivity, appropriateness, outcomes, implementation, and cost. Source 27

Longitudinal rehabilitation can be described as a series of transitions between capacity, capability in context, and actual participation. These transitions are affected by setbacks and adaptation. Recovering a physical skill may enable an activity but may require transportation, confidence, and a supportive companion before it becomes a sustained social practice. Studies should observe all three levels rather than assume that improvements transfer automatically.

Autonomy, Agency, and Supported Decision-Making

Autonomy in later-life rehabilitation is the person's effective authority over personally important choices, exercised through information, communication, preference formation, and accessible alternatives. Dependence in a physical task can coexist with high decision-making autonomy. A participant who chooses where to live, identifies preferred activities, and accepts selected assistance can retain substantial control over life organization despite needing help with dressing or mobility. Clinical practice must therefore distinguish assistance with execution from substitution of another person's preferences.

Ryan and Deci's (2000) (https://doi.org/10.1037/0003-066X.55.1.68) self-determination framework links autonomy, competence, and relatedness to human motivation and well-being. It provides a psychological hypothesis about conditions that may support sustained rehabilitation engagement. It does not establish that autonomy support alone improves every geriatric endpoint. Mechanistic trials would need to measure actual decision opportunity, perceived autonomy, engagement, and relevant outcomes over time. Source 18

Multimorbidity creates hard decisions between competing benefits. A rigorous shared-decision process clarifies the person's preferred trade-offs, the likely time course and magnitude of benefits, risks, treatment burden, and available supports. NICE NG56 (https://www.nice.org.uk/guidance/NG56) provides guidance for tailoring multimorbidity care around what matters to the person. Supported communication, accessible formats, advocacy, and additional deliberation time can make participation possible when hearing, cognition, language, fatigue, or institutional routines would otherwise narrow choice. Decisions should be revisited as capacity, prognosis, and priorities change. Source 15

Goal Selection, Motivation, and Adaptation

Rehabilitation goals connect a desired future activity to actionable steps in treatment. A goal may involve walking a specified distance, using a kitchen safely, returning to a community group, or participating in family decisions. A clinically plausible goal becomes psychologically relevant when the person understands why it matters and can identify how intermediate exercises relate to it. Goal selection itself may change when illness and resources shift. Baltes and Baltes (1990) (https://doi.org/10.1017/CBO9780511665684.003) described selection, optimization, and compensation as an adaptive framework in adult development and aging. Source 3

Goal setting is also an intervention studied empirically. Levack et al. (2015) (https://doi.org/10.1002/14651858.CD009727.pub2) synthesized rehabilitation trials concerning goals and strategies to enhance goal pursuit; a 2016 summary (https://pubmed.ncbi.nlm.nih.gov/26771917/) emphasized heterogeneity and very low-certainty evidence for several comparisons. The existence of some favorable effects does not identify a universal superior goal-setting procedure. Studies differ in whether they compare any goal setting with none, structured procedures with usual practice, or explicit strategies for pursuing agreed goals. Source 13

Mechanisms worth testing include clarity of action plans, expectations of success, experience of competence, reminders and environmental cues, social support, and the match between personally valued roles and prescribed tasks. Outcome measurement should record who selected the goal, whether the goal changed, its attainment, and whether attainment altered real-world participation. A treatment that improves a laboratory performance score while exhausting the person or displacing cherished activities may provide a narrower benefit than aggregate outcomes imply.

Depression, Fear, Pain, and Behavioral Engagement

Mood, fear of falling, pain, fatigue, and uncertainty shape the behavioral opportunity to rehabilitate. These experiences can be consequences of illness, antecedents of avoidance, or responses to specific environments. Their temporal and causal positions vary across people. A person who avoids stairs after falling may protect against immediate danger while progressively losing conditioning and access to social activities; another may avoid stairs because they remain objectively unsafe. Assessment must distinguish accurate hazard appraisal from overgeneralized fear and identify modifiable environmental conditions.

A randomized home intervention by Gitlin et al. (2006) (https://doi.org/10.1111/j.1532-5415.2006.00703.x) combined occupational and physical therapy, environmental adaptation, and coping strategies in 319 older adults with functional difficulty. The intervention group reported improvements in daily activities, self-efficacy, and fear of falling at measured follow-up points. This complex design supports the efficacy of the assigned program in its studied population, while limiting attribution of the result to a single psychological component. Source 10

Depression and anxiety screening can illuminate barriers to engagement, but results require contextual and clinical interpretation. Fatigue may reflect cardiopulmonary disease, sleep disruption, medication effects, mood, or multiple interacting causes. Rehabilitation teams must address symptom burden and accessible pacing while assessing safety, changing health needs, and informed participation. Psychological support is integrated with medical assessment, rather than assumed to substitute for investigation of new clinical problems.

Cognition, Communication, and Cognitive Accessibility

Cognitive functioning influences learning new routines, organizing complex treatment, understanding consent, remembering instructions, and navigating services. However, a single cognitive screen does not measure the whole capacity to make a particular decision. Hearing and vision loss, language differences, educational background, acute illness, pain, unfamiliar settings, and fatigue can change test performance or apparent participation. Research and practice should examine how information is delivered and what supports enable the person to express a stable preference.

In rehabilitation, cognitive accessibility can involve simplified task sequencing, written and visual reminders, consistent routines, conversational pacing, a reliable support person, and environmental cues. These accommodations are plausible means of improving task performance, but their individual efficacy is context-specific. Severe cognitive impairment is frequently excluded from efficacy trials; Skoumal et al. (2026) (https://doi.org/10.1007/s40520-026-03450-z) explicitly noted recurrent exclusion of participants with moderate to severe dementia in geriatric rehabilitation studies. This exclusion limits claims about applicability to a substantial group of potential service users. Source 19

Cognitive assessment should therefore be understood as both a measure of abilities and an examination of barriers to meaningful involvement. Trials can distinguish cognitive change, procedural learning, communication success, decision participation, and functional outcomes rather than using one undifferentiated cognitive endpoint. Research including participants with cognitive impairment requires appropriate consent procedures, accessible measures, and transparent reporting of proxy versus self-report data.

Functional Measurement and Construct Validity

Measurement is central to the scientific identity of later-life rehabilitation. The ICF separates body functions, activities, participation, and environmental factors. The WHO Disability Assessment Schedule 2.0 (https://www.who.int/standards/classifications/international-classification-of-functioning-disability-and-health/who-disability-assessment-schedule) evaluates domains including cognition, mobility, self-care, interpersonal relations, life activities, and participation. The SPPB developed by Guralnik et al. (1994) (https://doi.org/10.1093/geronj/49.2.m85) emphasizes timed lower-extremity performance. Each instrument offers a different inferential window. Source 24 Source 11

Basic activities of daily living assess practical tasks such as bathing, dressing, transferring, and toileting; instrumental activities include more complex tasks such as shopping, transportation, and managing household demands. A standardized mobility test captures performance under specified conditions; a report of frequency of leaving home captures behavior within a social and physical environment. A participation outcome asks whether the person can assume a chosen social role with desired frequency and acceptable effort. Correlations across measures do not establish that they are equivalent constructs.

A defensible measurement battery balances clinical feasibility, responsiveness to change, comparability, and participant relevance. It must address ceiling and floor effects; assistive device use; cognitive, linguistic, and sensory accessibility; proxy-report discrepancies; and whether respondents interpret an item consistently over time. Patient-selected goal attainment adds relevance but can complicate comparison when goals differ. The solution is complementary measurement with explicit estimands, rather than a single omnibus score described as independence.

Longitudinal Change and Causal Inference

Rehabilitation trajectories involve recovery, stabilization, recurrence, and adaptation. Baseline function is shaped by preexisting disability, acute precipitating events, access to treatment, prognosis, and previous support. A cohort observed only at discharge may mistake natural recovery for treatment efficacy, while a study ending at six weeks may miss environmental barriers that reemerge when the person returns home. Repeated observations across clinical and community contexts help identify durable functional transfer.

Randomized trials are strongest for estimating the effect of assigning an eligible population to a specified intervention versus a specified comparator, provided missing data, adherence, outcome selection, and other biases are addressed. They may not identify which component of a multicomponent program caused the result. Observational cohorts capture real-life heterogeneity and longer-term patterns but face confounding by severity, referral, socioeconomic status, and family support. Qualitative studies can clarify experienced autonomy, implementation barriers, and the meanings of participation without producing a population average causal effect.

Analysts must attend to competing outcomes such as death, institutionalization, severe illness, and loss to follow-up. An analysis restricted to survivors able to complete walking tests may present an overly favorable picture of functional outcomes. Outcomes such as living at home require careful interpretation because they also depend on housing, financing, availability of caregiving, and individual preference. Design, denominators, subgroup definition, and attrition should be reported near numerical results.

Comprehensive Geriatric Assessment and Coordinated Care

Comprehensive geriatric assessment (CGA) combines multidimensional assessment with multidisciplinary planning and coordinated follow-up. Its domains encompass health conditions, medicines, cognition, mood, mobility, nutrition, social conditions, and functional needs, with team arrangements varying by setting. It is a care process, not merely a diagnostic questionnaire. The trial evidence addresses particular configurations of geriatric assessment embedded in hospital systems.

The Ellis et al. (2017) Cochrane review (https://doi.org/10.1002/14651858.CD006211.pub3) included 29 trials involving 13,766 participants in nine predominantly high-income countries. Compared with usual care, CGA increased the probability of being alive and living at home at three to twelve months (RR 1.06, 95% CI 1.01–1.10; 16 trials, 6,799 participants). It reduced nursing-home admission (RR 0.80, 95% CI 0.72–0.89; 14 trials, 6,285 participants). The review found little or no difference in mortality (RR 1.00, 95% CI 0.93–1.07; 21 trials) or dependence (RR 0.97, 95% CI 0.89–1.04; 14 trials). Certainty differed by outcome, and setting-specific costs remained uncertain. Source 7

These findings require precise interpretation. A higher probability of living at home is meaningful, but it cannot alone establish that autonomy, quality of life, or preferred residence improved for each individual. A nursing-home transition may reflect poor access to home supports or a freely chosen arrangement. Future trials should measure preferred residence, perceived control, caregiver strain, and feasible alternatives alongside residence and survival, thereby connecting service-level effects to the ethical and psychological constructs at issue.

Specialist Inpatient Rehabilitation Evidence

Inpatient geriatric rehabilitation involves intensive interdisciplinary care during a specific episode of functional risk or decline. Bachmann et al. (2010) (https://doi.org/10.1136/bmj.c1718) reviewed 17 randomized trials involving 4,780 participants in general and orthopedic geriatric rehabilitation programs. At discharge, pooled estimates favored rehabilitation for functional improvement (OR 1.75, 95% CI 1.31–2.35), nursing-home admission (RR 0.64, 95% CI 0.51–0.81), and mortality (RR 0.72, 95% CI 0.55–0.95). Estimates were more modest at later follow-up, with differences in populations, programs, and settings. Source 2

The review compared specialized programs with usual care as defined in their respective trials. These interventions often entailed assessment, assigned therapies, interdisciplinary meetings, and reevaluation, so no single element can claim the entire pooled effect. Differences between general geriatric and orthopedic programs limit indiscriminate pooling of functional outcomes. The original BMJ correction (2010) (https://doi.org/10.1136/bmj.c3484) amended descriptions of two programs while stating that analyses used correctly abstracted data; source evaluation should retain this provenance.

Inpatient rehabilitation should be judged by what happens after discharge as well as inside the institution. A therapy gain may be lost if the home remains inaccessible, medication regimens conflict, necessary support is unavailable, or activity practice ceases. Functional measures at discharge are necessary intermediate indicators, not a full account of the later-life rehabilitation trajectory.

Physical Activity, Mobility, and Disability Prevention

Mobility is both a physiological capacity and a prerequisite for many social activities. The LIFE randomized trial by Pahor et al. (2014) (https://doi.org/10.1001/jama.2014.5616) enrolled 1,635 sedentary adults aged 70 to 89 with physical limitations who could initially walk 400 meters. A structured moderate-intensity physical activity program was compared with health education over a mean follow-up of 2.6 years. Major mobility disability occurred in 30.1% of the activity group and 35.5% of controls (HR 0.82, 95% CI 0.69–0.98). Persistent mobility disability was also less frequent in the activity group (HR 0.72, 95% CI 0.57–0.91). Source 16

The primary endpoint was inability to complete a 400-meter walk, measured under specified criteria. It is a meaningful functional indicator, but its improvement does not demonstrate corresponding gains in friendship, civic activity, decision-making autonomy, or satisfaction with daily life. Participants were selected, sedentary, and still capable of walking 400 meters at entry; the intervention's efficacy should not simply be projected onto people who are bedbound, have markedly different risks, or lack access to similar supervised programs.

The trial also emphasizes adherence and competing demands. Physical activity requires available time, transport, safe settings, motivation, and ongoing accommodations for pain or illness. An effective rehabilitation program includes the conditions that make sustained participation feasible, while maintaining medical assessment of adverse events and individual limitations. The result supports a particular behavioral intervention and outcome, not a universal prescription of the same training dose for every older adult.

Frailty-Targeted Multicomponent Interventions

The SPRINTT study by Bernabei et al. (2022) (https://doi.org/10.1136/bmj-2021-068788) extended prevention research to older adults with physical frailty and sarcopenia across European sites. The randomized study enrolled 1,519 community-dwelling adults aged at least 70, identified through specified physical performance and muscle-mass criteria, and compared a multicomponent physical activity, nutritional counseling, and technological support strategy with healthy-aging education. In the prespecified primary analysis subgroup with SPPB scores of 3–7 (1,205 participants), mobility disability occurred in 283 of 605 intervention participants (46.8%) and 316 of 600 controls (52.7%), with a hazard ratio of 0.78 (95% CI 0.67–0.92). Reported falls and intervention-related events also require assessment of the benefit–harm balance. This study's rationale and subgroup analysis recognize that mobility risk reflects several interacting dimensions of vulnerability. Source 5

The interpretation of SPRINTT requires attention to subgroup, intervention fidelity, competing outcomes, and primary endpoint definition. Frailty and sarcopenia were defined by particular operational criteria, so other definitions could yield different eligible populations. Nutritional support, technology, and physical activity were combined; the design estimates the program package rather than each element's independent contribution. Clinical research should distinguish prevention of incident major mobility disability, changes in performance tests, and the wider ability to participate in society.

A key scientific challenge is deciding which patients benefit from which intensity and configuration of intervention. Clinical fragility may increase the value of support while simultaneously raising intervention burden and attrition. Predictive stratification should be examined prospectively rather than inferred solely from retrospective differences in responders. Aging is heterogeneous, and average trial effects provide only one part of the evidence for individualized rehabilitation planning.

Home-Based Rehabilitation and Environmental Modification

Home programs directly address the setting in which activities are performed. In a trial of 188 community-living older adults with physical frailty, Gill et al. (2002) (https://doi.org/10.1056/NEJMoa020423) compared a six-month home-based program emphasizing physical abilities with an educational control program. Findings supported prevention of functional decline in the studied population, with important variation between moderate and severe frailty. This design shows why intervention setting and baseline vulnerability matter to the interpretation of functional benefit. Source 9

Gitlin et al. (2006) (https://doi.org/10.1111/j.1532-5415.2006.00703.x) tested a program incorporating occupational therapy, physical therapy, home changes, and compensatory strategies. The outcomes included daily activities, perceived confidence, and fear of falling. Szanton et al. (2019) (https://doi.org/10.1001/jamainternmed.2018.6026) studied the person-directed CAPABLE intervention among 300 low-income older adults in Baltimore. At five months, the adjusted activities-of-daily-living disability score ratio was 0.70 (95% CI 0.54–0.93) relative to attention control; corresponding between-group effects were not sustained at twelve months. The latter limitation is critical for evaluating durability. Source 10 Source 20

Home modification has a particular explanatory advantage: it changes task demands while allowing people to remain within familiar routines. Yet program effectiveness is sensitive to housing tenure, landlord permission, local availability of workers, device costs, and participant choice. A modification that improves bathing safely may increase confidence; whether it changes participation outside the home needs separate measurement. The trial literature supports person–environment interventions while inviting stronger tests of their long-term and distributional effects.

Discharge, Reablement, and Continuity Across Settings

Transitions from hospital to rehabilitation unit, home, or community service can reorganize an older person's daily life. A discharge destination is not the same as the quality of a transition. Information transfer, medication reconciliation, accessible transport, continuity of equipment, caregiver preparation, and follow-up services may determine whether gains persist. Fragmentation increases the burden of care work and creates opportunities for missed problems or avoidable inactivity.

Reablement commonly denotes time-limited, goal-oriented support to recover or maintain the ability to perform daily activities, especially at home. Programs vary substantially in duration, eligibility, provider mix, and outcome specification; findings should therefore be attributed to defined models rather than to the label alone. The WHO ICOPE pathway (https://www.who.int/publications/i/item/9789240103726) and Rehabilitation 2030 initiative (https://www.who.int/initiatives/rehabilitation-2030/) provide frameworks for coordinating assessment, intervention, and longer-term care, although implementation effectiveness must be evaluated locally. Source 27 Source 25

Research should follow the transition as an exposure rather than treat settings as unrelated episodes. Important indicators include time from referral to care, delayed equipment, failed appointments, changes in caregiver burden, unplanned readmissions, actual activity in the home, and whether the person's goals remain visible to successive teams. Process measures are particularly valuable when clinical outcomes improve but the causal contribution of handoffs and coordination remains unclear.

Family, Caregiving, and Relational Autonomy

Rehabilitation in later life commonly unfolds within networks of relatives, friends, paid caregivers, volunteers, and clinicians. These relationships can make decisions accessible, provide encouragement, and permit activity that would otherwise be difficult. They can also alter power. A caregiver may appropriately help with a transfer while unnecessarily deciding which social activities the person should pursue. Relational autonomy analyzes how effective self-direction depends on available relationships without equating the receipt of help with the absence of agency.

Caregiving is not an unlimited or cost-free rehabilitation resource. Time, financial strain, physical demands, and emotional distress can change support reliability and affect patient outcomes. Team discussions should distinguish care that the person has chosen, duties that a caregiver has agreed to, and assumptions about what relatives will supply. This distinction is especially important when discharge decisions presume stable unpaid support that has not been verified.

Dyadic studies can examine concordance between patient and caregiver goals, conflicts over acceptable risks, and changes in mutual expectations. A patient may prioritize a social outing while a family member prioritizes preventing falls; neither report should be treated as a full representation of the relationship. Longitudinal methods that observe both participants and their environments are necessary to understand how practical assistance, control, and participation coevolve.

Assistive Technology and Accessible Environments

Assistive devices, accessible housing, communication systems, transportation, and digital interfaces can change the relationship between impairment and activity. Walking aids, hearing supports, visual accessibility, home adaptations, and task reminders can preserve options. Device provision alone is an incomplete intervention: fitting, instruction, maintenance, affordability, user preference, and real-world acceptance shape whether the device becomes useful.

The United Nations 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) places rehabilitation and assistive technology within the wider aim of independence and participation. The WHO Rehabilitation 2030 program (https://www.who.int/initiatives/rehabilitation-2030/) similarly treats assistive products and equitable service delivery as systems questions. Normative principles clarify legitimate aims; empirical studies must still measure actual uptake, performance, injury risk, satisfaction, and access inequalities. Source 22 Source 25

Technology trials require comparison of supported versus unsupported implementation. A sophisticated tool may perform well in a laboratory yet fail when instructions are inaccessible or connectivity is unstable. Outcomes must include whether the system permits user control, whether assistance can be refused or modified, and whether the person retains a workable alternative when a device fails. Dependability and governance become part of functional safety.

Community Participation and Social Roles

Community participation is a central endpoint of later-life rehabilitation because daily functioning matters within a life of relationships, identity, and commitments. Participation can include visiting friends, volunteering, attending cultural or religious events, shopping, caring for others, voting, and using public space. These activities carry different personal meanings. Counting out-of-home trips may provide a useful behavioral measure but cannot substitute for determining whether the activities were freely selected and valued.

The WHO healthy aging framework (https://www.who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability) recognizes relationships, social contribution, mobility, and decision-making as elements of functional ability. NICE NG32 (https://www.nice.org.uk/guidance/NG32) addresses independence and mental well-being through community and interpersonal opportunities for people aged 65 or older. The guideline's recommendations are a practice and policy framework; their existence should not be treated as independent proof of causal benefit for every community program. Source 26 Source 14

Participation outcomes can be conceptualized across access, actual involvement, frequency, chosen intensity, satisfaction, and perceived belonging. Studies should identify barriers such as transport, unsafe walkways, age discrimination, service hours, and insufficient income. A rehabilitation program may increase capacity while producing little participation benefit if these barriers persist. Conversely, social policy and accessibility modifications can improve participation without changing clinical impairment scores. Both patterns are evidence relevant to the research object.

Equity, Cultural Context, and Access to Rehabilitation

Age, income, neighborhood, rural residence, disability status, language, gender, migration experience, and health-system organization can affect entry to care and the durability of rehabilitation. Such differences may influence diagnosis, referral, physical access, transportation, informal support, and follow-up. Average treatment effects are consequently insufficient to establish equity. Selection into trials and selective outcome completion may systematically exclude those facing the greatest barriers.

The Scottish epidemiological work of Barnett et al. (2012) (https://doi.org/10.1016/S0140-6736(12)60240-2) highlighted socioeconomic patterning of multimorbidity. CAPABLE, studied by Szanton et al. (2019) (https://doi.org/10.1001/jamainternmed.2018.6026), directly investigated low-income adults in a particular U.S. city; the study increases knowledge about a frequently underserved population while still requiring transportability checks before broader application. Trials from well-resourced settings must be interpreted against the availability of professionals, home modifications, and continuing community services elsewhere. Source 4 Source 20

Culture also shapes what counts as valued independence. Some participants prefer interdependent living arrangements, strong family decision involvement, or shared daily responsibilities. Researchers should ask about desired choices rather than assume one household pattern represents a universally superior outcome. Language accessibility, meaningful consent, and culturally relevant participation measures are methodological requirements for comparing rehabilitation outcomes across groups.

The Contemporary Evidence on Rehabilitation Processes

An important 2026 contribution is Skoumal et al.'s systematic review (https://doi.org/10.1007/s40520-026-03450-z) of core components of geriatric rehabilitation. The authors searched major bibliographic databases through January 31, 2025 and included 36 studies—31 randomized trials and five controlled prospective cohorts—with 10,647 patients. They reported extensive heterogeneity in rehabilitation terminology, team arrangements, workflows, context, and assessment, including more than 90 instruments. Comprehensive geriatric assessment was infrequently documented; patients with moderate to severe dementia were often excluded; and no included study fully described the entire rehabilitation process. Source 19

These are findings about reporting, organizational definition, and the existing evidence base, not evidence that every service is ineffective. Rehabilitation packages can improve outcomes even while the mechanisms and sequence of their components remain difficult to isolate. The review highlights an obstacle to accumulation: studies labeled 'geriatric rehabilitation' may operationalize significantly different processes, making pooled estimates and cross-country transfer difficult to interpret.

The review also identifies a knowledge gap between effectiveness and process science. Better descriptions of goal negotiation, team decisions, assessment feedback, care transitions, caregiver involvement, and participation outcomes would permit replication of successful models and explanations of failed implementation. Future studies can link process indicators to functional outcomes while preserving patient priorities and contextual variation. A complete account requires evidence about what an intervention delivers, to whom, how consistently, and under which environmental conditions.

Critical Comparison of Interventions and Outcomes

The principal interventions in the examined literature have different causal targets. Exercise programs primarily test whether activity changes mobility and disability risk; home adaptations test changes in daily-task demands and behavioral compensation; comprehensive geriatric assessment tests care coordination; inpatient programs test specialized interdisciplinary packages; and goal-setting interventions examine mechanisms of direction and engagement. Collapsing them under a single estimated effect of 'rehabilitation' would erase treatment content and differences in comparison conditions.

The evidence is strongest for some narrowly defined endpoints. LIFE (https://doi.org/10.1001/jama.2014.5616) supports reduced major mobility disability in its eligible population; Ellis et al. (https://doi.org/10.1002/14651858.CD006211.pub3) support improved likelihood of living at home after inpatient CGA; Bachmann et al. (https://doi.org/10.1136/bmj.c1718) support benefits of specialized inpatient programs in the trials reviewed; and Szanton et al. (https://doi.org/10.1001/jamainternmed.2018.6026) support short-term reduction in ADL disability in a person-directed home intervention. These outcomes represent related but nonidentical constructs, and their effects cannot be combined without attention to program, population, follow-up, and estimand. Source 16 Source 7 Source 2 Source 20

Autonomy and participation are less consistently represented as primary endpoints. Goals may be documented without measuring who controlled the decision, and community involvement may be inferred from mobility or residence rather than observed directly. Claims that any particular clinical program improves 'independence' should specify whether the term means physical ability, reduced assistance, residence, perceived control, or freely chosen participation. Precision protects both scientific interpretation and the relevance of research to older adults.

Clinical and Research Ethics

The ethical center of later-life rehabilitation is the person's right to informed and meaningful participation in choices that affect daily life. United Nations CRPD Article 26 (https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities/article-26-habilitation-and-rehabilitation.html) calls for comprehensive, voluntary and community-oriented rehabilitation, while Article 19 (https://www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities/article-19-living-independently-and-being-included-in-the-community.html) concerns choice of residence and inclusion. Research ethics includes informed consent, accessibility, burdens of assessments, privacy, fair inclusion, and attention to changes in decisional capacity or acute illness. Source 22 Source 21

A program that increases physical performance yet diminishes patient control through coercive routines has an ethical problem even if one measured outcome improves. Conversely, accepting a person's preference for more assistance or a safer form of participation may be a successful outcome when consistent with informed choice. Trial endpoints should include experienced control and satisfaction with decisions, not just clinical judgments of suitable living arrangements.

Clinical research in multimorbid older adults requires monitoring for falls, adverse events, acute decompensation, medication effects, exhaustion, and other intervention-related burdens appropriate to the setting. This monograph describes evidence and principles rather than prescribing exercise doses, medication changes, or clinical decisions for individual readers. High-stakes implementation should follow current professional guidance and qualified clinical assessment; external clinical peer review of this authored synthesis remains a separate editorial requirement.

Artificial Era and Digitally Mediated Rehabilitation

Digitally mediated rehabilitation is an emerging setting in which wearable sensing, teleconsultation, automated reminders, accessible communication systems, and artificial decision-support tools may influence access and continuity. Artificial systems can help coordinate appointments, summarize goals, or support rehearsal of tasks. Their value must be defined through observed effects on functioning and choice rather than inferred from technical sophistication. A system that offers many recommendations but increases care burden or obscures responsibility may worsen the rehabilitation experience.

Research in the Artificial Era should distinguish human clinical judgment, algorithmic recommendation, environmental assistance, and artificial action. Questions include whether patients understand and can contest recommendations; whether interfaces remain accessible with sensory or cognitive impairments; whether reminders support or override preferences; whether data surveillance changes perceived autonomy; and whether automated triage reproduces disparities in rehabilitation access. These are testable hypotheses requiring prospective and participatory evaluation, not empirical conclusions about currently deployed systems.

A clinically sound design preserves clinician accountability for diagnosis and treatment decisions, transparent documentation of provenance, privacy safeguards, fallback when technology fails, and meaningful user control. Human–Artificial partnerships may become part of the ecology of rehabilitation, but their effectiveness should be tested on patient-selected outcomes, community participation, safety, and distributional access alongside conventional functional measures.

Priorities for Research and Scientific Synthesis

The next generation of geriatric rehabilitation studies should register primary estimands that separate impairment, activity, participation, and autonomy. A trial can simultaneously report objective mobility performance, daily activities, preferred participation goals, experiences of decision control, caregiver effects, resource use, and adverse events. Prespecified follow-up after discharge should identify both immediate improvement and maintenance under ordinary environmental conditions. Transparent reporting of attrition, mortality, institutionalization, and incomplete outcomes is essential.

Research on multimorbidity should model interacting health problems and treatment burden while avoiding simplistic diagnosis counts as sufficient explanation. Trials need transparent inclusion of people with sensory or cognitive impairment and those with severe frailty, along with supported-consent protocols. Comparative effectiveness designs may test combinations and sequencing of interventions, using factorial or pragmatic designs where feasible. Longitudinal observational and qualitative work can expose mechanisms missed by standardized trials and support more valid measure development.

Implementation research should study pathway fidelity, referral delays, team coordination, equipment provision, caregiver training, and environmental barriers across low- and high-resource contexts. Equity-focused analyses require meaningful subgroup representation and assessment of access before randomization. International collaboration can improve transportability while acknowledging local differences in housing, transport, family structures, and care systems. The Skoumal et al. (2026) (https://doi.org/10.1007/s40520-026-03450-z) findings make clear that coherent process specification is a prerequisite for learning which forms of rehabilitation are reproducible and for whom. Source 19

Conclusion

Rehabilitation in later life concerns the preservation and transformation of functional ability within the real conditions of older people's lives. Multimorbidity, frailty, and acute events create changing demands, while goals, motivation, family relations, clinical organization, and environmental access determine the translation of capacities into daily action. Psychological autonomy and participation are central research outcomes because a person's ability to direct life and remain involved in valued relationships and activities may change independently of impairment.

The scientific literature establishes meaningful effects of defined programs on mobility disability, some daily activities, and the probability of remaining at home. It also shows heterogeneity in eligibility, intervention components, outcome measures, durability, and reporting of entire rehabilitation pathways. The most productive synthesis therefore examines the relationship among mechanisms and endpoints rather than using 'independence' as an undifferentiated label. Further progress requires studies of what matters to older adults, which supports permit that participation, and how the benefits of rehabilitation survive transitions between settings.

References