Disease-nonspecific Activity Outcome Measures for Adult Rehabilitation Populations: A COSMIN-based Systematic Review of Psychometric Properties Authors: Shoji Kinoshita, Shigehito Shiota, Shunsuke Taito, Ryo Momosaki, Yukio Mikami, Masahiko Mukaino, Naoshi Ogata, Yohei Otaka, Masahiro Abo Published: 2026 ABSTRACT ABSTRACTObjectives:This study identified and characterized disease-nonspecific, clinician-administered, or performance-based activity outcome measures used in adult rehabilitation and evaluated their measurement properties using the COSMIN framework.Methods:We systematically searched MEDLINE, CINAHL, PsycINFO, Web of Science, the Cochrane Library, and Ichushi-Web. We included peer-reviewed original studies in adult rehabilitation samples that evaluated at least one measurement property of disease-nonspecific, clinician-administered, or performance-based activity measures, in which at least 50% of items assessed basic movements, mobility, or self-care. We excluded patient-reported or disease-specific instruments and pediatric samples. Two reviewers independently screened records, extracted instrument characteristics and measurement properties, and assessed methodological quality using the COSMIN Risk of Bias checklist. The certainty of evidence for each measurement property was graded using the COSMIN-modified GRADE approach.Results:We included 83 studies reporting on 20 measures. The Barthel Index (BI), Modified Barthel Index (MBI), Functional Independence Measure (FIM), and de Morton Mobility Index (DEMMI) accounted for most psychometric evaluations. Internal consistency and inter-rater or test–retest reliability were generally acceptable, whereas evidence for structural validity, construct validity, responsiveness, and measurement error was limited and heterogeneous. Content mapping indicated substantial variation in coverage of basic movements, mobility, and self-care, with only a subset of instruments demonstrating relatively broad activity coverage.Conclusions:Only a limited set of generic activity measures has, at best, moderate-certainty evidence for internal consistency and reliability in adult rehabilitation. These findings provide a practical basis for selecting measures by care phase and highlight priority targets for future disease-nonspecific activity measures. INTRODUCTION In rehabilitation medicine, restoration and maintenance of activity—encompassing basic movements, mobility, and self-care—is a central goal and key outcome of treatment. Standardized activity measures are used to support clinical decision-making and goal-setting and to evaluate intervention effects in trials and routine practice. As populations age and many patients live with multiple coexisting conditions, the trajectories of activity often extend across acute hospital care, inpatient rehabilitation, and community-based services, making it increasingly important to monitor functional status using measures that are comparable across diagnoses, settings, and phases of care.1) In this context, the choice of appropriate activity outcome measures is critical not only for individual patient management but also for coordinating care along the rehabilitation continuum and generating interpretable data at the system level. Therefore, activity measurement in rehabilitation should support continuity of assessment across phases of care, rather than relying on disconnected instruments at different stages of recovery. In the present systematic review, we use “activity” as a broad operational term for practical domains of functioning targeted for improvement, maintenance, and monitoring in rehabilitation medicine, with a particular focus on basic movements, mobility, and self-care. In adult rehabilitation, two generic activity measures have been particularly influential: the Barthel Index (BI) and Functional Independence Measure (FIM). In this context, activities of daily living (ADL) generally refer to conventional basic daily tasks, such as feeding, dressing, grooming, bathing, toileting, continence, transfers, and locomotion, although their content varies across instruments. The BI has been widely adopted as a global indicator of dependence in daily activities.2,3) However, psychometric studies have documented substantial floor effects in very dependent patients and ceiling effects in higher-functioning individuals, limiting its ability to detect meaningful changes across the full spectrum of functional recovery.4,5) The FIM was later introduced as a more detailed 18-item measure of functional independence (motor and cognitive domains) and has become a standard outcome in inpatient rehabilitation programs.6) However, FIM scoring procedures and training are governed by proprietary licensing agreements, which restrict their unrestricted clinical and research use and complicate broader implementation outside specific healthcare systems. Together, these considerations suggest that although the BI and FIM have played a central role in activity assessment, they may be suboptimal as freely accessible, cross-setting instruments for tracking activity from very low to high functional levels across the rehabilitation continuum. To guide the selection and standardization of activity outcome measures, it is important to systematically summarize the evidence on their psychometric properties. Several systematic reviews have addressed ADL and instrumental activities of daily living (IADL) instruments in specific populations and contexts. Prior work has mainly focused on self-reported or proxy-reported questionnaires for community-dwelling older adults, people with dementia, or patients after stroke or brain injury, showing that multiple tools are available and that a limited subset has acceptable reliability and some evidence of validity, whereas data on structural validity, responsiveness, measurement error, and interpretability remain incomplete.7,8,9,10,11) However, these reviews have largely been anchored in particular diagnoses, age groups, or care settings. They have not provided an integrated overview of disease-nonspecific, clinician-administered, or performance-based activity measures used across adult rehabilitation, nor have they detailed how their item content differentially covers basic movements, mobility, and self-care, underscoring the need for the present COSMIN-based systematic review. Therefore, we undertook a systematic review to identify and characterize disease-nonspecific, clinician-administered, or performance-based activity outcome measures used in adult rehabilitation populations and to evaluate their measurement properties using the Consensus-based Standards for the selection of health Measurement Instruments (COSMIN) framework. The COSMIN framework provides standards for assessing the methodological quality of studies on measurement properties, such as reliability, validity, responsiveness, and measurement error, and for grading the certainty of evidence for each property. Specifically, our study had the following aims: (1) to collate scales in which activity items—covering basic movements, mobility, and self-care—constitute the majority of the content; (2) to describe the characteristics and item content of these outcome measures and summarize instrument-level feasibility and cultural adaptability; and (3) to synthesize the available evidence on their psychometric properties, including internal consistency, reliability, structural/construct validity, measurement error, and responsiveness, in adult rehabilitation settings. By providing an integrated overview of the content and measurement properties of these activity outcome measures, this systematic review seeks to inform the selection of suitable instruments for clinical practice and research and provide a reference point for the future development of more comprehensive activity indices in rehabilitation. MATERIALS AND METHODS Study Design and Reporting This systematic review was conducted according to the COSMIN methodology for systematic reviews of health measurement instruments12,13) and is reported in line with the PRISMA 2020 statement.14) The full protocol, including the detailed search strategy and operational definitions, is publicly available at protocols.io.15) No amendments were made to the protocol after its publication. No individual patient-level data were collected for this study. Given that only data from published articles were used, ethical approval and informed consent were not required. Eligibility Criteria Studies and outcome measures were considered eligible according to the following inclusion criteria: peer-reviewed original research articles reporting empirical data; subjects were adults (aged ≥18 years) in rehabilitation or closely related clinical settings, regardless of diagnosis, care setting, or phase of rehabilitation; studies reporting data on at least one psychometric (measurement) property, such as internal consistency, reliability, validity, responsiveness, or measurement error; use of standardized, multi-item activity outcome measures; included measures in which the majority of items represented the performance of practical activities, and at least 50% of all items had to assess basic movements, mobility, or self-care; used clinician-administered or performance-based assessments. Studies were excluded according to the following criteria: systematic reviews, narrative reviews, case reports, conference abstracts, theses without corresponding journal publications, and non-peer-reviewed publications; studies using pediatric samples; studies using purely patient-reported outcome measures (PROMs) completed solely by patients; studies using disease-specific scales; studies using single-task performance tests without a multi-item scale structure. PROMs were excluded because this review was designed to inform the selection and development of clinician-administered or performance-based activity measures that can be applied even when self-report is not feasible, such as in patients with cognitive or communication impairment. Information Sources and Search Strategy The electronic search strategy was developed with methodological input from a medical librarian at Jikei University School of Medicine. We searched major biomedical and allied health databases (MEDLINE, CINAHL, PsycINFO, Web of Science Core Collection, Cochrane Library, and Ichushi-Web) from inception to 2 September 2025 using combinations of controlled vocabulary terms and free-text keywords. Search strategies were adapted to the syntax of each database. The full search strings are reported in a publicly available protocol.15) No language restrictions were applied during the search and screening stages. To identify additional relevant studies, we manually screened the reference lists of all the included articles and related systematic reviews. We did not search PEDro or OTseeker because they did not index relevant validation studies for this review question. Study Selection All records retrieved from the database searches were imported into Rayyan (web-based screening software),16) which was used both for screening and for automatic detection and removal of duplicates. Two reviewers (SK and SS) independently screened the titles and abstracts against the pre-specified eligibility criteria. The same reviewers then independently assessed the full texts to decide on final inclusion. The reasons for exclusion at the full-text stage were recorded in a structured list. Disagreements at either screening stage were resolved through discussion; when a consensus could not be reached, a third reviewer (ST) acted as an adjudicator. The study selection process, including the number of records at each stage and the reasons for exclusion, is summarized in a PRISMA flow diagram (Fig. 1). PRISMA flow diagram of study selection. Data Extraction Data extraction was performed independently by two reviewers using a standardized form that was piloted before use. For each included study, we recorded the bibliographic details, country, clinical setting, study design, study population, and basic demographic information. For each outcome measure, we summarized key characteristics, including the intended construct and target population, mode of administration (clinician-rated or performance-based), number of items, response format and scoring method, and any reported information on administration time, feasibility, and known floor or ceiling effects. To describe the content of each outcome measure, we reviewed the item-level information and mapped the types of activities. Based on this mapping, we summarized the main content areas covered by each scale and identified the overlapping and unique aspects of the activities assessed across the measures. In addition, we linked individual items to categories of the International Classification of Functioning, Disability, and Health (ICF) according to published ICF linking rules.17) For the purpose of content mapping, mobility was defined as movement-related activities, including changing body position, transfers, walking, and stair climbing. For each measurement property examined in a given study, we extracted details of the psychometric evaluation, including the property assessed (e.g., internal consistency, test–retest or inter-rater reliability, construct validity, criterion validity, and responsiveness). These results included Cronbach’s alpha coefficients, intraclass correlation coefficients, weighted kappa statistics, standard error of measurement, minimal detectable change, correlation coefficients, effect sizes, and indices of responsiveness when reported. Any discrepancies in the extracted data between the reviewers were resolved by discussion and, when necessary, by consulting a third reviewer. Data extraction forms, extracted data, and analytic decisions are available from the corresponding author upon reasonable request. Assessment of Methodological Quality The methodological quality of each study with respect to each reported measurement property was assessed using the COSMIN Risk of Bias checklist, following the guidelines for studies of health measurement instruments. Two reviewers independently applied the relevant COSMIN boxes and rated the quality of the design and statistical methods for each property as very good, adequate, doubtful, or inadequate. The “worst score counts” principle was used for each box. Disagreements were resolved through discussion, with a third reviewer available to arbitrate unresolved cases. Data Synthesis For each instrument and measurement property, we qualitatively summarized the direction and magnitude of the results across studies and graded the overall quality of evidence using the COSMIN-modified GRADE approach.18) In this process, we considered the COSMIN Risk of Bias ratings for individual studies, the consistency of findings across studies, the total sample size, and the number of studies (as indicators of imprecision), and we considered the applicability of the study populations and settings to adult rehabilitation practice (indirectness). Based on these considerations, we categorized the overall quality of evidence for each instrument–property combination as high, moderate, low, or very low. Quantitative pooling of coefficients or effect sizes was not attempted because of substantial clinical and methodological heterogeneity in patient populations, settings, instrument versions, and statistical methods. RESULTS Study Selection The database searches yielded 2973 records, and 17 additional records were identified through other sources. After the removal of duplicates, 2318 unique records remained for title and abstract screening. Of these, 2095 records were excluded as not relevant to activity outcome measures or measurement property studies. A total of 223 full-text articles were assessed for eligibility. Finally, 83 studies were included in this review. The study selection process and reasons for exclusion at each stage are summarized in the PRISMA flow diagram (Fig. 1). Characteristics of Included Studies and Outcome Measures The included studies were conducted in various countries and care settings, including acute hospitals, post-acute and inpatient rehabilitation units, outpatient clinics, and community or long-term care facilities. Sample sizes varied widely, from small validation studies in single centers to large observational studies in mixed rehabilitation cohorts. Across studies, participants represented various rehabilitation populations, most commonly older adults with neurological or musculoskeletal conditions; however, the outcome measures themselves were not restricted to a specific disease group. Rehabilitation settings ranged from acute hospital wards and early post-acute units to inpatient rehabilitation hospitals, geriatric wards, nursing homes, and community-based services, with post-acute and inpatient rehabilitation settings being the most common. Table 1 summarizes the main characteristics of the disease-nonspecific activity outcome measures identified in this review. Across the included studies, a range of clinician-rated or performance-based scales were used, most of which focused on basic self-care and mobility tasks and were designed to be applicable across diagnostic groups in adult rehabilitation. Table 1. Characteristics of included activity outcome measuresMeasureTarget populationMode of administrationSub/scales, number of itemsRange of scores/scoringOriginal languageAvailable translationsCopyrightCost of useAbility for basic movement scale (ABMS II)Adults after stroke in acute–subacute inpatient rehabilitation; measures basic movement capacityClinician-administered bedside observation of five fundamental movementsSingle scale; five items (turn over, sit up, maintain sitting, stand up, maintain standing)Each item graded 0–5 (6 levels in ABMS II incl. cueing); total 0–25 (higher=better basic mobility)Japanese (published in English)Limited/unknown; primarily English materials reportedAuthors (Tanaka, Hashimoto et al.); non-commercial research/clinical use permittedFree (no licensing fees; follow published procedures)Activity Measure for Post-Acute Care (AM-PAC)Adults across acute and post-acute care; diagnosis non-specificComputer adaptive test or fixed short forms; clinician report or patient reportThree domains: basic mobility, daily activity, applied cognitive; “6-Clicks” inpatient short forms have 6 items/domainItems rated 1–4; short-form raw 6–24; CAT and conversion to T-scores (mean 50, SD 10) availableEnglishSpanish, Mandarin Chinese, Brazilian Portuguese; additional versions vary by form (6-Clicks/CAT)Trustees of Boston University/distributed via Pearson/CREcare®Licensed; permission/fees may applyBarthel Index (Modified Barthel Index, PET-MBI)Adults with disability in personal care/mobility (stroke and other conditions); inpatient and communityClinician observation or interview; patient/caregiver report when neededSingle scale; ten ADL items (feeding, bathing, grooming, dressing, bowels, bladder, toilet use, transfers, ambulation/wheelchair, stairs)Commonly 0–100 weighted total (higher=more independent); 0–20 modified version also usedEnglish (US)Extensive translations worldwide (e.g., Spanish, Chinese, Persian, and many others)© MedChi (Maryland State Medical Society); widely treated as public domain for noncommercial useFree for clinical/research use; permission needed for modifications/commercial usede Morton Mobility Index (DEMMI)Older adults (acute, subacute, community); diagnosis non-specificClinician performance-based observationMobility; 15 itemsRaw 0–19 converted to interval 0–100 (higher=better mobility)EnglishMultiple validated versions (Dutch, German, Danish, Turkish, Slovenian, Brazilian Portuguese, others)© de Morton et al.; free for clinical/research use with attributionFree (with attribution)Functional Independence Measure (FIM)Adults in inpatient rehabilitation across diagnoses (e.g., stroke, TBI, SCI, MS); WeeFIM for younger childrenClinician-rated interdisciplinary observation/interview using standardized form by trained ratersMotor 13 + cognitive 5=18 items (eating, grooming, bathing, dressing, toileting, bladder/bowel, transfers, locomotion; communication, social cognition)Seven levels per item (1=total assist → 7=complete independence); total 18–126 (higher=more independent)English (US)Multiple official/published versions (e.g., German, Italian, Spanish, Swedish, Finnish, Portuguese, Turkish, French, Persian)FIM™ © UDSMR® (Uniform Data System for Medical Rehabilitation); licensed proprietary instrumentLicensed use only; institutional fees and rater training/certification typically requiredFunctional Status Score for the ICU (FSS-ICU)Critically ill adult ICU patients, including mechanically ventilated or recently extubatedClinician-rated, performance-based bedside assessmentSingle total score; five items (rolling; supine-to-sit; sitting at edge of bed; sit-to-stand; walking)0–7 per item; total 0–35 (higher=better)EnglishBrazilian Portuguese; Korean; Chinese; SpanishFree for clinical/research use with citation of original authors; local translation permissions may applyFreeKatz IndexOlder adults; chronic illness; general medicalClinician/patient interview; observationBasic ADL; six items (bathing, dressing, toileting, transferring, continence, feeding)0–6 or A–G categories (dependent→independent)EnglishWidely translated and used globally (multiple languages, including Japanese)Public domainFreeKlein–Bell ADL ScaleAdults in rehabilitation across diagnoses (e.g., neurologic, incl. SCI, stroke, orthopedics); designed for general physical disability (also used in children, but SR focuses on adults)Observer-rated, performance-based assessment by a clinician (direct observation against standardized behavioral criteria). Typical administration time 60–180 minSix domains: dressing; elimination; mobility; bathing/hygiene; eating; emergency telephone use. Total 170 items; 26 sub-dimensions described in linking workItem-level pass/fail. Tasks have weights (1–3) summed to a total independence score 0–313 (0–100%); higher=greater independenceEnglishSwedish version reported/used (linking and SCI studies); other official translations not identifiedManual: Klein RM, Bell BJ. The Klein–Bell ADL Scale Manual. Seattle: University of Washington, Division of Occupational Therapy, 1979. Current availability limitedManual currently unavailable; no formal fee schedule located. Likely permission required to reproduce items/scoringLucerne ICF-Based Multidisciplinary Observation Scale (LIMOS)Adults with stroke in inpatient neurorehabilitation (multidisciplinary setting); used for ADL across motor, cognition, communicationObserver-rated (multidisciplinary: physician, nurse, PT, OT, ST, neuropsychologist). Each discipline rates its own partICF-based structure. Original validation: Seven ICF chapters comprising 45 domains (items). Short-LIMOS and UL-LIMOS variants existFive-point performance scale per domain (1–5: dependent → independent incl. aids). Total and subscale scores; higher=more independentEnglish (published instrument); developed in Switzerland (German-speaking clinical setting)Portuguese (Brazil) Short-LIMOS (English). UL-LIMOS (Upper Limb, English)Authors (Ottiger et al., 2015) under CC BY for articleNo formal fee identified; instrument description available open-accessMelville-Nelson Self-Care Assessment (SCA)Skilled nursing/subacute; adults with disability; diagnosis non-specificClinician performance-based observation (OT-focused)Seven domains (bed mobility, transfers, dressing, eating, toileting, personal hygiene, bathing); multiple sub-activities per domainSelf-performance domain scored by sub-activities; total approx. 0–140 (higher=more limitation). Separate Support scale (0–3)EnglishLimited/unknown; primarily English materials reportedAuthors (Melville, Nelson)Permission from authors/institution; no standard fee reportedHierarchical Assessment of Balance and Mobility (HABAM)Hospitalized/frail older adults; acute geriatric inpatientsClinician performance-based observationThree domains (balance, transfers, mobility); hierarchical levels rather than fixed itemsDomain levels combined to yield total score; higher=better (hierarchical scale)EnglishNone official (English materials available)© HABAM developers (MacKnight et al.)Free for clinical/research (check source)Performance Assessment of Self-Care SkillsAdults incl. cognitive impairment; community and clinicClinician performance-based observation in naturalistic tasks26 tasks: 5 functional mobility, 3 basic ADL, 14 IADL (cognitive), 4 IADL (physical)Each task rated on three scales (independence, safety, adequacy) typically 0–3; criterion-referencedEnglishMultiple (e.g., Arabic, Portuguese, Spanish; others reported)PASS authors/universities (see manual)Nominal (copying/kit). Manual may require permissionPhysical and Cognitive Performance Test for Assisted Living Facilities (PCPT ALF)Assisted Living Facilities (ALF) residents; adults incl. with cognitive impairment; diagnosis non-specificPerformance-based assessment of ADL/IADL tasks with simultaneous rating of physical ability and cognitive support needsTwo subscales: physical ability (performance of ADL/IADL tasks) and cognitive support (type/level of cueing or assistance); item set developed from ALF-relevant tasksTotal score range approx. 8–51; physical ability 0–27; cognitive support 8–24; higher scores indicate greater ability/less support requiredEnglishNot reported (developed in English for ALF residents)Authors (Bowen et al., 2017)Free for research/clinical use per publicationRivermead Mobility IndexNeurological/rehab populations incl. stroke; diagnosis non-specific mobility scaleClinician interview/observation; yes/no performanceMobility; 15 items (0/1 each)0–15 (higher=better mobility)EnglishMultiple (incl. Japanese, Portuguese, Danish, Dutch, Chinese)Authors (Collen et al.); non-commercial clinical use generally permittedFree for clinical/research (check permissions)Ronchi Brief Evaluation Battery (RBEB)Older adults in subacute/long-term care; patients in occupational therapy programs (e.g., dementia or general geriatric rehab)Observer-rated, performance-based tests focused on self-sufficiency (washing, dressing, eating). Administered by OT/rehab staffBrief battery targeting ADL performance (washing, dressing, eating). Exact item count/sections not standardized in the abstract; center-developed batteryTask performance judged by completion/accuracy and, in some versions, time; composite/sectional scores described in authors’ reports (details vary)Italian (developed in residential nursing care home, Modena)Used primarily in Italian settingsAuthored by Baldelli et al.; no commercial manual identifiedNo fee schedule identified; contact authors/institution for permissionsSection GG, CARE item setMedicare/CMS post-acute settings (SNF, IRF, LTCH, HHA); adultsClinician coding based on usual performance during assessment windowSelf-care (GG0130: ~7 items), mobility (GG0170: ~17 items)Six-level performance code (06 independent → 01 dependent) + “activity not attempted” codesEnglishN/A (US CMS instrument; official language is English)CMSFree (public CMS instrument)SOFMER Activity Score (SAS)Rehabilitation inpatients/outpatients (mixed diagnoses); adultsClinician rating (team-based)v2: eight items (four motor, four cognitive)Five-level scale per item; higher=better functionFrenchLimited; primarily French-language resourcesSOFMER (French Society of Physical and Rehabilitation Medicine)Free for clinical/research (per publications)Sunnaas Index of ADLAdults in rehabilitation (incl. SCI, stroke); community independence focusClinician observation/interview with manualADL; 12 activitiesItems scored 0–3 (0=dependent, 3=independent) with SCI-specific adjustments for some itemsNorwegianNordic languages and English manual availableSunnaas Rehabilitation HospitalFree (with manual)Timed Movement Battery (TMB)Older adults (community-dwelling) with varying mobilityClinician-administered timed performance testsBattery of timed mobility tasks (e.g., gait, sit-to-stand, object manipulation)Task times recorded; composite/summary scores used (lower time=better)EnglishNot reportedDevelopers (Creel et al.)FreeTraffic Light System-BasicADL (TLS-BasicADL)Older adults in acute hospital settingsClinician observation/interview; color-coded reportingBasic ADL; ~13 items (independence in basic mobility and self-care)Color codes (green=independent, yellow=supervision, red=dependent) with corresponding ordinal levelsEnglish (Sweden)Limited (research use; English/Swedish)Developers (Asplin et al.)Free (research use)SD, standard deviation; PET, performance evaluation tool; TBI, traumatic brain injury; SCI, spinal cord injury; MS, multiple sclerosis; WeeFIM, Functional Independence Measure for Children; PT, physical therapist; OT, occupational therapist; ST, speech therapist; CC BY, creative commons attribution; CMS, Center for Medicare and Medicaid Services; SNF, skilled nursing facility; IRF, inpatient rehabilitation facility; LTCH, long-term care hospital; HHA, home health agency. Characteristics of included activity outcome measures SD, standard deviation; PET, performance evaluation tool; TBI, traumatic brain injury; SCI, spinal cord injury; MS, multiple sclerosis; WeeFIM, Functional Independence Measure for Children; PT, physical therapist; OT, occupational therapist; ST, speech therapist; CC BY, creative commons attribution; CMS, Center for Medicare and Medicaid Services; SNF, skilled nursing facility; IRF, inpatient rehabilitation facility; LTCH, long-term care hospital; HHA, home health agency. Content Mapping of Activity Outcome Measures Tables 2 and 3 present the content mapping of the included activity outcome measures. When the BI and FIM are used as reference points, both scales cover a core set of basic activities of daily living in self-care and mobility. However, several fundamental movement tasks, such as rolling and sitting or standing balance, which are explicitly assessed by instruments such as the Abilities of Basic Movement Scale II (ABMS-II)19) and the de Morton Mobility Index (DEMMI),20) are absent from the item sets of the BI and FIM. Only a few measures provided broad coverage from basic movement tasks through ambulation and stair climbing to more complex self-care activities, and most instruments tended to emphasize either mobility or self-care rather than both. The Section GG item set, introduced within the United States post-acute care quality reporting system, provides relatively broad coverage of basic movement, mobility, and self-care activities within a single framework, although only two studies reported its measurement properties in adult rehabilitation populations.21,22) Moreover, items related to domestic life, such as simple household tasks or light indoor chores, were rarely included and, when present, tended to appear only in a small subset of scales. In addition, items related to continence were included in the BI and FIM but were less frequently represented in other measures, contributing to differences in how autonomic functions were reflected in overall activity scores. Table 2. Content mapping of items across included activity outcome measures: mobilityMeasureMobility activityLying to sittingTurning overSit to standSittingStandingTransferCarryingWalkingMoving with wheelchairStairsUse of public transportPicking up an object from floorICF coded4100d4108d4104d4153d4154d420d430d450d465d4551d470d4105+d4400Ability for basic movement scale+++++Activity Measure for Post-Acute Care++++++Barthel Index (Modified Barthel Index, PET-MBI)++++de Morton Mobility Index+++++++Functional Independence Measure (FIM)++++Functional Status Score for the ICU+++++Katz Index+Klein–Bell ADL Scale+++Lucerne ICF-Based Multidisciplinary Observation Scale+++++++++Melville-Nelson Self-Care Assessment+++Hierarchical Assessment of Balance and Mobility+++++Performance Assessment of Self-Care Skills+++Physical and Cognitive Performance Test for Assisted Living Facilities++++Rivermead Mobility Index+++++++Ronchi Brief Evaluation BatterySection GG, CARE item set++++++++SOFMER Activity Score+++Sunnaas Index of ADL+++Timed Movement Battery+++++Traffic Light System-BasicADL++ Content mapping of items across included activity outcome measures: mobility Table 3. Content mapping of items across included activity outcome measures: self-care, domestic life, and other activitiesMeasureSelf-careDomestic lifeOtherBathingGroomingToiletingDressingFeedingMedication/health managementPreparing mealsHouseworkCommunicationSocial interactionMoney managementCognitionBowel/bladderICF coded510d520d530d540d550-560d570d630d620, d640-650d310-d399d710-d799d860 / d865b140-b189, d110-d299b525, b620Ability for basic movement scaleActivity Measure for Post-Acute Care++++++++Barthel Index (Modified Barthel Index, PET-MBI)++++++de Morton Mobility IndexFunctional Independence Measure (FIM)+++++++++Functional Status Score for the ICUKatz Index+++++Klein–Bell ADL Scale+++++++Lucerne ICF-Based Multidisciplinary Observation Scale++++++++++Melville-Nelson Self-Care Assessment+++++Hierarchical Assessment of Balance and Mobility++Performance Assessment of Self-Care Skills++++++++++Physical and Cognitive Performance Test for Assisted Living Facilities++++++++Rivermead Mobility IndexRonchi Brief Evaluation Battery++++Section GG, CARE item set+++++SOFMER Activity Score+Sunnaas Index of ADL++++++++Timed Movement BatteryTraffic Light System-BasicADL+++++ Content mapping of items across included activity outcome measures: self-care, domestic life, and other activities Methodological Quality of Measurement Property Studies Table 4 summarizes the methodological quality of included studies using the COSMIN Risk of Bias checklist. Across 83 articles, most ratings concerned internal consistency and reliability and were predominantly very good/adequate, particularly for BI, Modified Barthel Index (MBI), FIM, and DEMMI. In contrast, evaluations of structural validity, cross-cultural validity/measurement invariance, measurement error, and responsiveness were less frequent—often based on single studies—and were more commonly rated doubtful/inadequate because of small samples and suboptimal or insufficiently reported methods (e.g., limited sample sizes for factor analysis, inadequate reporting of hypotheses/statistics). Criterion validity was not assessed and is therefore not shown. Overall, BI/MBI, FIM, and DEMMI had the broadest psychometric evidence base, whereas many other instruments were supported by sparse and methodologically heterogeneous evaluations. Table 4. Overview of COSMIN-reviewed measurement propertiesIDMeasureContent validityStructural validityInternal consistencyCross-cultural validity/measurement invarianceReliabilityMeasurement errorCriterion validityConstruct validityResponsivenessTanaka, 201019)ABMS IIAdequateChang, 201823)AM-PACAdequateAdequateVery goodAdequateHou, 202024)AM-PACAdequateVery goodVery goodVery goodHoyer, 201825)AM-PACDoubtfulVery goodVery goodAdequateBowen, 201526)BIDoubtfulInadequateInadequateDoubtfulDoubtfulCano, 20065)BIAdequateCincura, 200927)BIAdequateAdequateCollin, 19883)BIAdequateAdequatede Haan, 199328)BIVery goodVery goodVery goodAdequatede Morton, 200829)BIVery goodEdwards, 199530)BIVery goodFormiga, 200531)BIDoubtfulFormiga, 200632)BIAdequateGonzález, 201833)BIVery goodAdequateAdequateVery goodGreen, 200134)BIAdequateAdequateHartigan, 201135)BIAdequateAdequateHsueh, 200136)BIDoubtfulVery goodAdequateDoubtfulHulsbæk, 201937)BIAdequateVery goodLaake, 199538)BIAdequateDoubtfulLekamwasam, 201139)BIAdequateAdequateVery goodDoubtfulInadequateAdequateLiu, 201540)BIVery goodMinosso, 201041)BIAdequateAdequateAdequateNicholl, 200442)BIAdequateAdequateO’Connor, 20044)BIVery goodOveisgharan, 200643)BIAdequateAdequateVery goodAdequatePrasad, 201844)BIVery goodDoubtfulAdequateRichards, 200045)BIAdequateDoubtfulSchlote, 200446)BIAdequateVery goodTaghizadeh, 202047)BIVery goodDoubtfulVery goodDoubtfulAdequateValero-Chillerón, 202348)BIVery goodWolfe, 199149)BIVery goodDoubtfulDos Reis, 202250)BIAdequateVery goodVery goodAdequateJanša, 200451)BI (Extended BI)DoubtfulAdequateAdequateAdequateAminalroaya, 202052)MBIVery goodAdequateDoubtfulAdequateFerfeli, 202453)MBIAdequateVery goodVery goodDoubtfulVery goodAdequateKüçükdeveci, 200054)MBIAdequateDoubtfulAdequateDoubtfulAdequateAdequateDoubtfulLeung, 200755)MBIAdequateVery goodVery goodDoubtfulVery goodShah, 198956)MBIAdequateAdequateShinar, 198757)MBIAdequateYang, 202158)MBIVery goodOhura, 201159)BI (PET-MBI)DoubtfulVery goodVery goodVery goodOhura, 201760)BI (PET-MBI)Very goodAdequatede Morton, 201061)DEMMIVery goodVery goodVery goodde Morton, 201162)DEMMIVery goodVery goodde Morton, 201320)DEMMIVery goodVery goodVery goodJezek, 202463)DEMMIVery goodVery goodSommers, 201664)DEMMIAdequateAdequateAdequateDoubtfulTrøstrup, 201965)DEMMIVery goodVery goodBrosseau, 199466)FIMVery goodVery goodAdequateHaas, 200267)FIMVery goodHamilton, 199468)FIMVery goodHeinemann, 199769)FIMAdequateLawton, 200670)FIMDoubtfulInadequateLundgren-Nilsson, 200571)FIMDoubtfulInadequateMaritz, 201972)FIMVery goodVery goodNaghdi, 201573)FIMDoubtfulAdequateVery goodInadequateAdequateDoubtfulAdequatePollak, 199674)FIMAdequateVery goodAdequateYamada, 200675)FIM (motor, short form)Very goodDoubtfulAlves, 201976)FIM (motor), FSS-ICUVery goodVery goodVery goodVery goodHsueh, 200277)FIM, BIVery goodVery goodVery goodHong, 201878)FIM, MBIAdequateWales, 201879)FIM, Sunnaas Index of ADLAdequateMacKnight, 199580)Hierarchical Assessment of Balance and MobilityVery goodAdequateAdequateÅsberg and Sonn, 198981)Katz IndexAdequateAdequateAzad, 201782)Katz IndexAdequateAdequateAdequateVery goodAdequateVery goodBrorsson, 198483)Katz IndexVery goodAdequateAdequateGerrard, 201384)Katz IndexVery goodVery goodAdequateKatz, 197085)Katz IndexAdequateAdequateKlein, 198286)Klein–Bell ADL ScaleAdequateVery goodAdequateVan de Winckel, 201987)Lucerne ICF-based Multidisciplinary Observation ScaleVery goodAdequateNelson, 200288)Melville–Nelson Self-Care AssessmentVery goodVery goodVery goodFerreira, 202189)Performance Assessment of Self-Care SkillsAdequateAdequateVery goodVery goodKeleman, 202590)Performance Assessment of Self-Care SkillsAdequateAdequateKoyama, 200691)Rivermead Mobility IndexVery goodAdequateMaeshima, 200592)Rivermead Mobility IndexDoubtfulDoubtfulBaldelli, 200493)Ronchi Brief Evaluation BatteryDoubtfulDoubtfulChang, 201522)Section GG/CARE Item SetAdequateAdequateDoubtfulVery goodAdequateHarmon, 202321)Section GG/CARE Item SetAdequateInadequateAdequateCharvolin, 202194)SOFMER Activity ScoreVery goodKorpelainen, 1997Sunnaas Index of ADLVery goodVery goodTanaka, 200895)Sunnaas Index of ADLVery goodAdequateCreel, 200196)Timed Movement BatteryDoubtfulAdequateAsplin, 202297)TLS-BasicADLAdequateAdequate Overview of COSMIN-reviewed measurement properties Psychometric Findings Table 5 summarizes the psychometric findings for the included key outcome measures by setting/phase. Full study-level results for all instruments are provided in the Supplementary Table. Overall, internal consistency and inter-rater or test–retest reliability were the most consistently reported properties and were typically in the acceptable-to-good range for the total scores of many instruments. In contrast, evidence for structural validity, cross-cultural validity/measurement invariance, measurement error, and responsiveness was more limited and heterogeneous. According to the COSMIN-modified GRADE ratings, the overall quality of evidence was generally moderate for the internal consistency and reliability of the BI/MBI, FIM, and DEMMI, but low or very low for most other measurement properties and for the remaining, less frequently evaluated instruments. Table 5. Summary of psychometric evidence for key activity outcome measures by setting/phaseMeasureSetting/phaseNumber of studiesContent validityStructural validityInternal consistencyCross-cultural validity/measurement invarianceInter-rater reliabilityTest–retest reliabilityMeasurement errorConstruct validityResponsivenessBIAcute/critical care7+0/−0/?0+1/−1/?0+1/−0/?0+0/−0/?1+2/−2/?0+0/−0/?0+0/−0/?1+1/−0/?0+0/−0/?0Post-acute/inpatient rehabilitation8+0/−0/?0+0/−0/?0+3/−0/?0+0/−0/?1+5/−0/?0+1/−0/?0+0/−0/?1+4/−0/?0+4/−0/?0Community/long-term care9+2/−0/?0+2/−1/?1+6/−0/?0+0/−0/?1+4/−0/?0+2/−0/?1+0/−0/?4+6/−0/?0+1/−0/?0MBIAcute/critical care3+0/−0/?0+2/−0/?0+1/−0/?0+0/−0/?0+0/−0/?0+1/−0/?0+0/−0/?0+2/−0/?0+1/−0/?0Post-acute/inpatient rehabilitation4+0/−0/?1+0/−1/?0+2/−0/?0+0/−0/?1+3/−0/?0+1/−0/?0+0/−0/?1+2/−0/?1+1/−0/?1Community/long-term care3+1/−0/?0+1/−0/?0+1/−0/?0+0/−0/?0+2/−0/?0+1/−0/?0+0/−0/?0+1/−0/?0+0/−0/?0FIMAcute/critical care3+0/−0/?0+1/−0/?0+1/−0/?0+0/−0/?1+1/−0/?0+1/−0/?0+0/−0/?1+2/−0/?0+0/−0/?0Post-acute/inpatient rehabilitation7+0/−0/?0+2/−0/?0+1/−0/?0+0/−2/?0+2/−0/?0+0/−0/?0+0/−0/?0+1/−0/?1+2/−0/?1Community/long-term care5+1/−0/?0+2/−0/?0+3/−0/?0+0/−0/?2+1/−0/?2+2/−0/?0+1/−0/?1+5/−0/?0+0/−0/?1DEMMIAcute/critical care5+0/−0/?0+1/−0/?1+0/−0/?0+0/−0/?0+1/−0/?0+1/−0/?0+0/−0/?1+5/−0/?0+4/−0/?0Post-acute/inpatient rehabilitation1+0/−0/?0+1/−0/?0+0/−0/?0+0/−0/?0+0/−0/?0+0/−0/?0+0/−0/?0+1/−0/?0+1/−0/?0Community/long-term care1+0/−0/?0+0/−0/?1+0/−0/?0+0/−0/?0+0/−0/?0+0/−0/?0+0/−0/?0+1/−0/?0+0/−0/?0Values indicate the number of studies rated as sufficient (+), insufficient (−), or indeterminate (?) for each measurement property based on Supplementary Table. Totals may be less than the number of studies because not all studies evaluated every measurement property. Criterion validity is not presented because it was not assessed in the included studies. Reliability includes κ or intraclass correlation coefficient as reported. Summary of psychometric evidence for key activity outcome measures by setting/phase Values indicate the number of studies rated as sufficient (+), insufficient (−), or indeterminate (?) for each measurement property based on Supplementary Table. Totals may be less than the number of studies because not all studies evaluated every measurement property. Criterion validity is not presented because it was not assessed in the included studies. Reliability includes κ or intraclass correlation coefficient as reported. When the evidence was viewed across care settings/phases, psychometric evaluations were concentrated in post-acute and inpatient rehabilitation. In these settings, BI, MBI, FIM, and DEMMI were most frequently evaluated, and responsiveness was more commonly examined over the course of inpatient rehabilitation. In acute/critical care settings, studies mainly focused on reliability and, in some cases, measurement error. In community and long-term care settings, psychometric work focused primarily on cross-sectional construct validity, whereas longitudinal evaluations of responsiveness were scarce. For BI, internal consistency and inter-rater or test–retest reliability were generally acceptable across stroke, geriatric, neurological, and other adult rehabilitation samples. Construct validity was supported by correlations with other activity or disability measures and by known-group or predictive analyses. Responsiveness was reported in several rehabilitation cohorts; however, floor effects in very dependent patients and ceiling effects in higher-functioning patients were recurrent concerns, particularly when monitoring change across a wide range of recovery levels. Evidence regarding measurement error and interpretability was comparatively limited. For MBI, internal consistency and reliability were also generally acceptable across stroke, mixed neurological, and long-term care samples. Construct validity was supported by associations with legacy activity scales and expected group differences. Compared with the original BI, MBI may provide improved sensitivity to change in some rehabilitation populations because of its expanded scoring levels; however, the evidence base remains smaller and setting-specific, and information on measurement error and responsiveness is still limited. FIM showed acceptable internal consistency and reliability for total or motor scores across several rehabilitation populations. Construct validity was generally supported by associations with disease-specific disability measures, other activity scales, length of stay, discharge destination, and level of care. Responsiveness of the FIM motor score was reported in several neurorehabilitation cohorts. However, Rasch-based studies identified cross-cultural differential item functioning and category threshold problems, indicating that raw FIM scores may not always be directly comparable across countries or settings without appropriate adjustment. DEMMI demonstrated favorable psychometric performance as a mobility-focused measure. Structural validity was supported by Rasch or Mokken analyses, reliability was generally acceptable, and construct validity and responsiveness were supported in acute, geriatric, hip-fracture, and critical care populations. However, because DEMMI focuses on mobility and does not assess self-care, it should be interpreted as complementary to broader activity measures rather than as a comprehensive activity index. DISCUSSION In this COSMIN-based systematic review, we identified 83 studies reporting on 20 disease-nonspecific, clinician-administered, or performance-based activity outcome measures used in adult rehabilitation settings. Across these instruments, BI/MBI, FIM, and DEMMI accounted for the largest number and breadth of psychometric evaluations. Content mapping showed that most measures focused on basic movements, mobility, and self-care, with BI and FIM covering a common core of transfers, walking, and personal care items, whereas the Section GG item set provided relatively broad coverage of basic movement, mobility, and self-care within a single framework, although psychometric evidence was limited. The methodological quality appraised with the COSMIN Risk of Bias checklist was generally very good or adequate for studies of internal consistency and inter-rater or test–retest reliability, whereas structural validity, cross-cultural validity, measurement error, and responsiveness were examined in fewer studies and were often rated as doubtful or inadequate. Consistent with these patterns, internal consistency and reliability estimates for BI/MBI, FIM, and DEMMI were within commonly accepted ranges and achieved at most moderate-quality evidence ratings, whereas evidence for construct validity, responsiveness, and other measurement properties was more limited. Overall, these findings indicate that only a limited number of generic activity measures (BI/MBI, FIM, and DEMMI) have moderate-quality evidence for internal consistency and inter-rater or test–retest reliability. A key contribution of this review is that it couples COSMIN-based appraisal with item-level content mapping and a setting/phase-specific synthesis. This combination clarifies both what each instrument measures and where along the care continuum its evidence base is strongest, thereby supporting pragmatic selection in adult rehabilitation. Several previous systematic reviews have examined activity-related outcome measures, particularly ADL and IADL questionnaires in older adults or specific clinical populations; however, most have done so within relatively narrow clinical or demographic frames. Pashmdarfard and Azad9) conducted a COSMIN-based review of ADL and IADL tools used in older adults and identified a range of generic and condition-specific questionnaires, including the BI, Katz Index, FIM, and Lawton IADL scale, with generally acceptable levels of reliability and at least some evidence for construct validity, usually through correlations with other functional scales or known-group comparisons. They also highlighted important gaps, noting that data on structural validity, responsiveness, measurement error, and interpretability were often sparse or methodologically weak. Taken together, this review showed that several widely used ADL/IADL instruments in older adults have at least acceptable reliability, but the overall evidence base for other key measurement properties remains incomplete.9) Conceptually, their work is similar to that of the present review in that both collate and appraise disease-nonspecific activity measures using the COSMIN methodology; however, they focused exclusively on self-reported or proxy-reported questionnaires in older adults. In contrast, the present review extends this line of work by including disease-nonspecific, clinician-administered, or performance-based activity measures used across a broad range of adult rehabilitation populations. Other reviews have addressed overlapping but more specific segments of activity constructs. For community-dwelling older adults, Hopman-Rock et al.8) reviewed ADL instruments and recommended the Functional Autonomy Measurement System (SMAF), five-item Katz list, Functional Independence and Difficulty Scale (FIDS), and BI as the best-supported measures. Kelbling et al.10) evaluated extended and instrumental ADL scales and concluded that, although the Lawton IADL scale and a performance-based IADL test had relatively stronger evidence, psychometric data for most instruments were still limited and heterogeneous.10) Across dementia care and post-stroke or brain injury rehabilitation, previous reviews found that only a few ADL/IADL or “real-life” mobility questionnaires—such as the Rivermead Mobility Index—showed reasonable internal consistency and structural validity, whereas responsiveness, measurement error, and evidence for most other tools were limited.7,11) Across these reviews, a consistent pattern emerges: a small number of widely used instruments have been examined repeatedly in specific populations and show at least acceptable reliability, whereas evidence for structural validity, responsiveness, measurement error, and minimally important change is limited and methodologically heterogeneous for many measures. Thus, although numerous ADL and IADL instruments have been developed and partially evaluated, most reviews remain anchored in specific age groups, diagnoses, or care settings, and few have synthesized activity measures across both mobility and self-care domains in a way that cuts across diagnoses, age groups, and rehabilitation settings. Against this background, our review offers a more cross-cutting synthesis by examining generic activity measures that jointly cover mobility and self-care across multiple diagnoses, age groups, and rehabilitation settings. From a clinical and research perspective, our findings highlight the need for generic activity measures that can be applied consistently across diagnoses, care settings, and stages of recovery, at a time when many rehabilitation patients live with multiple coexisting conditions, and continuity of functional assessment from the acute to the community phase is increasingly emphasised.1) Among the instruments identified in this review, the BI, MBI, FIM, and DEMMI are the most plausible candidates for such cross-cutting use; however, the BI, MBI, and FIM do not include basic movement items that are crucial for very early or severely disabled patients, whereas the DEMMI focuses on mobility and does not address self-care. More recent frameworks, such as the Section GG item set, provide broader combined coverage of basic movements, mobility, and self-care; however, their development within the US post-acute care payment and quality-reporting system entails constraints on use in other countries, and psychometric evidence from non-US settings remains limited.21,22) Our content mapping suggests that floor and ceiling effects may be reduced by broader coverage across the activity continuum, including core basic movement items for patients with lower functional levels and more demanding mobility and self-care items for those with higher functional levels. Taken together, these considerations suggest that future work should refine existing measures or develop new activity indices that offer wide coverage of core activity domains, robust measurement properties, feasibility in busy rehabilitation services, and systematic cultural adaptation, so that mobility and self-care can be tracked in a comparable way across diagnostic groups and along the rehabilitation continuum. Study Limitations This review had several limitations. First, we focused on adult rehabilitation and excluded pediatric populations; therefore, the findings cannot be generalized to children or adolescents. Second, we deliberately restricted inclusion to disease-nonspecific, clinician-administered, or performance-based instruments in which at least half of the items addressed basic movements, mobility, or self-care. As a result, we did not evaluate PROMs, disease-specific activity scales, or broader multidomain tools that may be relevant in certain rehabilitation contexts. Third, because of this content focus, our synthesis does not fully address instruments that primarily target instrumental activities of daily living, including domestic life and social participation, and the conclusions should not be extrapolated to broader constructs. CONCLUSION In this COSMIN-based systematic review, we synthesized evidence from 83 studies on 20 disease-nonspecific, clinician-administered, or performance-based activity measures used in adult rehabilitation. A small group of widely used instruments, particularly the BI/MBI, FIM, and DEMMI, showed at least acceptable internal consistency and inter-rater or test–retest reliability, whereas evidence for structural validity, responsiveness, measurement error, and other properties was generally limited or of lower certainty. Content mapping indicated that these measures differ in the extent to which they cover basic movements, mobility, and self-care, and that only a subset provides relatively broad activity coverage. These findings provide a practical basis for selecting measures by care phase and highlight priority targets for future disease-nonspecific activity measures.