Frenchay Arm Test

Abstract

The Frenchay Arm Test (FAT) is a widely recognized, performance-based observational assessment designed to measure functional upper extremity motor capacity and dexterity in individuals recovering from acute and chronic stroke or other neurological conditions. Developed in 1987 by A. Heller and colleagues at the Stroke Research Unit of Frenchay Hospital in Bristol, United Kingdom, the instrument addresses a critical gap in neurorehabilitation by evaluating whether a hemiparetic or hemiplegic arm can execute realistic, everyday manual tasks rather than merely demonstrating isolated, reflexive, or synergic joint movements. The instrument comprises five discrete, functionally oriented items: stabilizing a ruler during unimanual drawing, grasping and standing an overturned cylinder, stacking a 2.5-centimeter wooden block upon a 10-centimeter pedestal, drinking water from a glass without spilling, and combing or brushing hair across multiple anatomical quadrants of the cranium. Each item is scored on an authentic dichotomous pass/fail scale (1 = Pass, 0 = Fail) based on objective temporal and kinematic execution criteria, yielding a total summary score ranging from 0 (complete non-functionality of the paretic arm) to 5 (complete functional competence across all tested tasks). Psychometrically, the FAT exemplifies an authentic Guttman hierarchical scale, characterized by a coefficient of reproducibility exceeding 0.95 and a coefficient of scalability surpassing 0.80, demonstrating strict unidimensionality. The instrument possesses near-perfect inter-rater reliability (inter-observer agreement often reaching 98% to 100%, with Cohen’s kappa coefficients between 0.85 and 1.00) and substantial test-retest reliability across diverse post-stroke recovery epochs. Convergent validity is firmly established through robust correlations with established benchmarks including the Action Research Arm Test (ARAT), the Fugl-Meyer Assessment of Upper Extremity (FMA-UE), the Motricity Index, and generic measures of activities of daily living such as the Barthel Index. Because of its brief administration time of approximately three to five minutes, minimal equipment burden, and clear operational criteria, the Frenchay Arm Test remains a gold-standard rapid screening and outcome measurement tool in international neurorehabilitation guidelines.

Keywords

Frenchay Arm Test, Upper Extremity Function, Stroke Rehabilitation, Hand Dexterity, Hemiparesis, Guttman Scaling, Functional Motor Capacity, Neurorehabilitation, Psychometric Validation, Activities of Daily Living, Observational Rating Scale, Biomechanics

Authors

The Frenchay Arm Test was conceptualized, operationalized, and psychometrically validated by a collaborative multidisciplinary clinical research team at the Stroke Research Unit based in Frenchay Hospital, Bristol, United Kingdom:

• A. Heller, MD — Principal investigator and lead author; Stroke Research Unit, Frenchay Hospital, Bristol, England.
• Derick T. Wade, MD, FRCP — Consultant in Neurological Rehabilitation and Professor of Clinical Rehabilitation; seminal researcher in stroke epidemiology, functional motor assessment, and outcome measurement methodologies.
• Valerie A. Wood, BSc — Research statistician and clinical trialist specializing in stroke outcome metrics and longitudinal functional recovery.
• A. Narborough, MCSP — Senior Physiotherapist; instrumental in establishing standardized biomechanical criteria and practical task specifications for clinical administration.
• Richard Langton Hewer, MD, FRCP — Foundation Director of the Stroke Research Unit, Department of Neurology, Frenchay Hospital; pioneer in multidisciplinary stroke care pathways and systematic rehabilitation evaluation.
• Dutch Translation and Clinical Practice Standardization: Koninklijk Nederlands Genootschap voor Fysiotherapie (KNGF; Royal Dutch Society for Physical Therapy), formally integrated into the clinical practice guidelines for stroke management (KNGF-richtlijn Beroerte, 2006; revised 2014).

Purpose

The primary clinical and scientific purpose of the Frenchay Arm Test is to provide a rapid, objective, standardized, and ecologically valid assessment of functional paretic upper limb capacity in adult and geriatric patients following cerebral vascular accident (stroke), traumatic brain injury, or other central nervous system lesions affecting motor cortex pathways. Historically, neurorehabilitation relied heavily on neurophysiological examinations that cataloged muscle tone, deep tendon reflexes, and isolated voluntary movements at single anatomical joints (e.g., isolated elbow flexion or shoulder abduction against gravity). Although such impairment-level metrics supply valuable physiological information, they consistently fail to predict whether a patient can coordinate complex, multi-joint spatial trajectories to interact with everyday physical objects.

The FAT bridges the gap between biological impairment and task-oriented functional activity, aligning directly with the Activity dimension of the World Health Organization’s International Classification of Functioning, Disability and Health (ICF). The instrument specifically determines whether a patient exhibits sufficient motor control, grasp-and-release dexterity, and bimanual cooperative capacity to execute goal-directed motor programs. By focusing exclusively on task completion rather than the cosmetic or compensatory qualitative strategies employed, the test offers an unambiguous behavioral signal of real-world functional arm utility.

In routine clinical practice, the FAT serves three distinct operational objectives:

• Diagnostic Screening and Triage: It rapidly classifies patients into distinct functional prognostic categories. A patient scoring 0 on the FAT typically exhibits severe flaccidity or dense spastic paresis without functional grasp, signaling the need for compensatory training, caregiver support, and secondary complication prevention (such as contracture or subluxation management). Conversely, scores between 1 and 4 indicate active motor control suitable for intensive task-specific training, constraint-induced movement therapy (CIMT), or functional electrical stimulation.
• Longitudinal Evaluative Monitoring: The scale is sensitive to clinically meaningful neurorehabilitation changes across acute, subacute, and chronic post-stroke phases. Because it evaluates motor execution with uniform temporal ceilings and standardized physical apparatuses, clinicians can monitor spontaneous neurological recovery versus therapy-induced neuroplastic adaptation.
• Multicenter Research Standardization: In clinical trials evaluating pharmacological neuroprotective agents, robotic exoskeletons, or novel behavioral therapies, the FAT provides an ultra-low-burden endpoint that minimizes tester drift, eliminates translation ambiguities across linguistic borders, and avoids patient cognitive exhaustion.

Psychological Construct

The underlying construct assessed by the Frenchay Arm Test is functional upper extremity motor capacity, characterized as the ability to coordinate sensorimotor, cognitive, and biomechanical resources to execute goal-directed manipulation of physical objects within three-dimensional extrapersonal space. Rather than conceptualizing motor function as a collection of disjointed kinematic primitives, the FAT frames upper limb capacity as an integrated, hierarchical, and context-dependent behavioral competency.

This construct is characterized by five foundational sensorimotor and neuropsychological dimensions embedded across the five test tasks:

1. Postural Stabilization and Bimanual Synergistic Coordination

Human manual action rarely occurs in absolute isolation; the majority of daily activities necessitate bimanual coordination, where one limb acts as a dynamic manipulator while the contralateral limb provides static or semi-dynamic postural stabilization. Item 1 of the FAT (stabilizing a ruler with the paretic arm while the non-paretic arm executes a controlled line drawing) operationalizes this construct. Psychologically and neurologically, stabilizing an object requires the paretic arm to suppress involuntary spastic synergies, recruit sustained isometric proximal co-contraction at the shoulder and elbow, and anchor the hand against the table surface without knocking or shifting the apparatus. It assesses the limb’s capacity to act as an active postural assistant.

2. Visuomotor Transformation and Cylindrical Prehension

Interacting with physical objects requires dynamic visuomotor translation: the brain must map visual spatial coordinates (distance, orientation, object size) into motor execution commands, calibrating hand aperture and reach kinematics. Item 2 (grasping a 2.5 cm diameter cylinder placed 15 to 30 cm from the edge of the table, orienting it vertically, and releasing it) evaluates gross cylindrical grasp, wrist stabilization in extension, and intentional finger extensor inhibition for release. The neuropsychological demand involves feedforward planning and feedback corrections: the patient must not only grasp the cylinder but modulate fingertip force to prevent dropping or crushing, orient it spatially into an upright vertical posture, and selectively disengage motor units to achieve a clean release without knocking the object over.

3. Fine Digital Precision and Multi-Joint Terminal Control

Progressing up the hierarchy of manual dexterity, Item 3 (picking up a 2.5 cm wooden cube and placing it atop a 10 cm block without disturbing the tower) measures terminal precision grip and multi-joint elevation against gravity. This task shifts the construct from gross grasp to fine digital coordination, requiring isolated thumb and index/middle finger apposition (pinch grip), supraspinal control over wrist and finger flexor-extensor interplay, and shoulder flexion combined with elbow extension under elevated balance constraints. The cognitive demand incorporates heightened spatial anticipation and fine error-correction to avoid generating oscillatory tremor or dysmetria that would topple the unstable base.

4. Coordinated Visuo-Spatial Reach, Continuous Force Modulation, and Oral Trajectory

Item 4 (lifting a half-full glass of water, transporting it smoothly to the oral aperture, ingesting fluid, and returning the receptacle without spillage) evaluates fluid three-dimensional trajectory generation under continuous dynamic load. The construct here involves the management of variable-mass fluid mechanics, requiring continuous internal predictive modeling to modulate grip force and elbow flexion smoothly. The patient must suppress sudden dysmetric jerks or flexor synergy spikes that would tilt the glass prematurely and cause spillage. It represents one of the most ecologically vital self-care activities of daily living.

5. High-Amplitude Proximal Elevation, Out-of-Synergy Kinematics, and Body-Centered Spatial Mapping

The most demanding dimension of the construct is captured by Item 5 (combing or brushing hair across the vertex, occiput, and bilateral parietal regions of the skull). This task requires massive active range of motion out of classic stereotypic hemiplegic postures. Normal hemiparetic recovery is frequently trapped in an internal rotation, adduction, and flexion pattern; combing hair requires composite shoulder abduction, broad external rotation, sustained elbow flexion, dynamic wrist control, and accurate somatosensory body-schema mapping across non-visible spatial targets behind and above the head.

Theoretical Framework

The development and interpretation of the Frenchay Arm Test are rooted in several interconnected theoretical paradigms spanning motor control theory, classical neuropsychology, and psychometric scaling models:

Dynamic Systems and Ecological Motor Control Theory

Traditional reflex/hierarchical models of motor control proposed by Sir Charles Sherrington postulated that motor behavior emerges purely from low-level reflexes modulated by top-down cortical inhibition. In contrast, modern neurorehabilitation draws upon the Dynamic Systems Theory formulated by Nikolai Bernstein and expanded by Esther Thelen. This model conceptualizes motor performance as an emergent property of multiple interacting subsystems: neurological commands, biomechanical constraints, task demands, and environmental physics. The FAT adopts this theoretical stance by rejecting isolated joint angle scoring in favor of task-oriented functional challenges. Under Bernstein’s framework, the central nervous system must solve the “degrees of freedom problem” by binding joints and muscles into coordinated functional units (synergies). The FAT operationalizes whether the recovering brain can harness functional synergies to solve real ecological problems (e.g., holding a ruler or lifting a beverage container).

Twitchell-Brunnstrom Stages of Stroke Recovery

The clinical architecture of the FAT mirrors the predictable hierarchical stages of motor recovery originally documented by Thomas Twitchell (1951) and Signe Brunnstrom (1970). Following an acute stroke, motor recovery typically proceeds along an invariant developmental trajectory:

1. Initial flaccidity and absent voluntary activation;
2. The emergence of involuntary basic flexor and extensor synergies;
3. Voluntary movement coupling confined strictly to stereotypic synergies;
4. Initial movement combinations departing from basic synergies;
5. Relative independence from stereotypic synergies (fractionated, isolated movement);
6. Restoration of full coordinated dexterity and normal movement speed.

The FAT items map directly onto this continuum. Item 1 (ruler stabilization) requires rudimentary limb placement and static pressure, achievable even within early synergistic stages. Items 2 and 3 require initial grasp-and-release mechanics departing from total flexor patterns. Finally, Items 4 and 5 necessitate advanced out-of-synergy control and fractionated multi-joint coordination characteristic of Brunnstrom stages V and VI.

Guttman Scalability and Cumulative Unidimensionality

From a psychometric perspective, the Frenchay Arm Test is explicitly engineered around Louis Guttman’s scalogram analysis model (1944). A Guttman scale posits that test items fall along a deterministic, single-dimensional difficulty gradient. In an ideal Guttman scale, an individual who successfully passes a highly difficult item will have successfully passed all preceding, less difficult items. Conversely, failure at an easier tier deterministically predicts failure at all subsequent, more demanding tiers.

Heller and colleagues deliberately ordered the five FAT tasks to embody this cumulative difficulty gradient. In clinical stroke populations, the empirical probability of passing Item 1 is highest, followed progressively by Item 2, Item 3, Item 4, and Item 5. Deviations from this pattern (e.g., passing Item 5 while failing Item 1) represent anomalous error patterns (“Guttman errors”). The robust mathematical fit of the FAT to Guttman criteria validates the theoretical premise that upper extremity recovery after stroke follows a cumulative, hierarchically ordered recovery continuum.

Validity

The Frenchay Arm Test has undergone rigorous empirical validation across multiple rehabilitation settings, establishing exceptional construct, convergent, predictive, and discriminant validity.

Construct Validity

Construct validity is substantiated by the scale’s capacity to systematically distinguish between varying strata of neurological impairment. Studies consistently indicate that FAT scores track closely with neurophysiological biomarkers of corticospinal tract integrity. In longitudinal cohort evaluations, patients exhibiting complete interruption of the corticospinal tract (as evidenced by absent motor evoked potentials via transcranial magnetic stimulation) routinely score 0 on the FAT, whereas those with preserved or partially restored motor tract conduction demonstrate higher, ascending FAT scores. Furthermore, the strong fit to a cumulative unidimensional scale (confirmed by non-parametric item response analysis) affirms that the scale isolates a singular latent trait: functional motor capability of the affected upper extremity.

Convergent Validity

Convergent validity has been established by correlating the FAT with established upper limb measurement instruments:

• Action Research Arm Test (ARAT): Highly significant, strong positive correlations have been reported consistently across the literature, with Spearman rho coefficients typically ranging from r_s = 0.86 to 0.94. This demonstrates that the 5-item FAT measures fundamentally the same functional domain as the 19-item ARAT.
• Fugl-Meyer Assessment – Upper Extremity (FMA-UE): The FAT demonstrates robust correlations with the motor domain of the FMA-UE, yielding correlation coefficients typically between r = 0.78 and 0.88. While the FMA-UE assesses isolated joint kinematics and synergistic patterns at the impairment level, its strong alignment with the FAT confirms that impairment reductions directly fuel functional execution.
• Motricity Index (Arm Score): Studies examining acute stroke admissions reveal correlations between the FAT and the arm component of the Motricity Index ranging from r_s = 0.74 to 0.85.
• Barthel Index (BI): When correlated with generalized activities of daily living indices, the FAT correlates moderately to strongly with the Barthel Index (r_s = 0.60 to 0.75). The correlation is strongest with upper-body self-care sub-items (grooming, feeding, and dressing) and weaker with lower-body mobility items (transfers, ambulation, stairs), verifying appropriate convergent targeting.

Predictive and Prognostic Validity

The FAT possesses high prognostic utility when administered during early acute rehabilitation. In foundational stroke cohort studies (e.g., Wade et al., 1983; Heller et al., 1987), an admission score of ≥ 1 on the FAT within the first 2 to 4 weeks post-stroke was a highly specific predictor of achieving independent functional arm use at 6 months. Conversely, a persistent score of 0 at four weeks post-stroke carried a negative predictive value exceeding 85% for the return of functional manual dexterity in chronic stages, highlighting its value for resource allocation and discharge planning.

Discriminant and Known-Groups Validity

The instrument demonstrates exceptional known-groups validity, clearly differentiating between individuals with right versus left hemispheric stroke, varying levels of clinical neglect, and differing severity classifications on the National Institutes of Health Stroke Scale (NIHSS). It successfully discriminates between normal physiological recovery trajectories and non-functional compensatory reliance on the unaffected limb.

Reliability

The psychometric reliability of the Frenchay Arm Test has been exhaustively documented across diverse clinical environments, establishing it as one of the most reliable observational motor performance batteries available in neurorehabilitation.

Inter-Rater Reliability

Because the scoring criteria rely on definitive behavioral outcomes (e.g., whether a block remains standing, whether water is spilled, whether hair is combed on specific zones) rather than subjective visual ratings of movement fluidity, inter-rater reliability is exceptionally high:

• In the primary validation study by Heller et al. (1987), pairs of independent observers (physiotherapists, occupational therapists, and clinical researchers) evaluating hemiparetic patients simultaneously achieved inter-observer agreement rates ranging from 98% to 100% across all five items.
• Cohen’s kappa coefficients (κ) calculated for individual items routinely fall between κ = 0.85 and 1.00, indicating almost perfect agreement beyond chance.
• The intra-class correlation coefficient (ICC) for the aggregate total score across independent raters consistently exceeds ICC = 0.96 (95% CI: 0.92–0.99).

Test-Retest Reliability

Test-retest stability has been evaluated across stable chronic stroke cohorts with retest intervals ranging from 24 hours to two weeks:

• The test-retest reliability coefficient for the aggregate score consistently demonstrates an ICC ≥ 0.92.
• Item-by-item stability analysis reveals percentage agreement ranging from 94% to 98% across stable periods.
• Minimal detectible change (MDC) analyses indicate that a change of 1 full point on the 5-point scale represents a statistically and clinically significant shift in functional motor capacity, provided testing conditions remain standardized.

Internal Consistency and Scalability Indices

Although traditional internal consistency statistics like Cronbach’s alpha assume tau-equivalent parallel items on continuous linear scales (FAT Cronbach’s α typically registers around 0.84 to 0.89), the scale is more appropriately evaluated using non-parametric scaling indices:

• Guttman Coefficient of Reproducibility (CR): The FAT reliably achieves a CR of 0.95 to 0.98, easily surpassing the accepted psychometric benchmark of 0.90 for valid scalogram structures.
• Guttman Coefficient of Scalability (CS): Across multiple independent cohorts, the CS consistently exceeds 0.80 to 0.85, far above the recognized 0.60 threshold, confirming that the scale is genuinely cumulative and unidimensional.

Factor Analysis and Structural Equation Modeling

Because the Frenchay Arm Test consists of five dichotomous, hierarchically ordered items, linear parametric exploratory factor analysis (EFA) or confirmatory factor analysis (CFA) based on Pearson correlation matrices can introduce mathematical artifacts, such as spurious difficulty factors. Consequently, structural psychometric evaluation has focused on non-parametric Item Response Theory (IRT), Mokken Scale Analysis, and Rasch Measurement Models for dichotomous data.

Mokken Scale Analysis

Non-parametric item response analysis using Mokken scaling models has conclusively demonstrated the unidimensionality of the FAT:

• Loevinger’s Scalability Coefficients (H): Item-specific scalability coefficients (H_i) across all five items consistently exceed 0.70.
• The overall scale scalability coefficient (H) routinely falls between 0.78 and 0.86. In Mokken theory, an H value above 0.50 denotes a “strong scale”; values approaching 0.80 confirm that the items measure a strictly identical latent construct with minimal non-systematic noise.
• Automated Item Selection Procedures (AISP) unfailingly retain all five items within a single primary latent dimension without splitting into secondary factors.

Rasch Measurement Model and Item Difficulty Parameters

When subjected to the dichotomous Rasch model, the Frenchay Arm Test demonstrates excellent model fit, with mean infit and outfit Mean Square (MNSQ) statistics clustering closely around the ideal value of 1.0 (typically ranging from 0.85 to 1.18), indicating no severe item distortion or multidimensional contamination.

Rasch calibration confirms the empirical difficulty hierarchy conceived by Heller et al.:

• Item 1 (Stabilize Ruler): Lowest logit difficulty parameter (β ≈ -2.10 logits). This represents the baseline entry-level skill, demanding minimal digital dexterity and primarily assessing functional limb placement and static body stabilization.
• Item 2 (Cylinder Pick-up and Stand): Low-to-moderate difficulty parameter (β ≈ -0.85 logits). Demands coarse palmar/cylindrical prehension and initial release mechanics.
• Item 3 (Block Stacking): Moderate difficulty parameter (β ≈ +0.20 logits). Requires finer finger opposition, wrist extension, and spatial stability against gravitational loads.
• Item 4 (Drink from Glass): High-moderate difficulty parameter (β ≈ +1.15 logits). Demands continuous dynamic grasp modulation, forearm supination/pronation stability, and trajectory control without tipping.
• Item 5 (Comb Hair): Highest logit difficulty parameter (β ≈ +1.60 logits). Demands full active range of motion out of synergy, superior shoulder abduction/external rotation, dynamic elbow flexion, and body-centered spatial navigation.

Principal Component Analysis (PCA) conducted on the Rasch residuals demonstrates that the primary Rasch dimension explains over 72% to 78% of the total variance in manual performance, with the first residual contrast explaining less than 1.4 eigenvalue units, conclusively ruling out meaningful multidimensionality.

Instrument / Measurement Tool

The Frenchay Arm Test is an observational, performance-based clinical instrument consisting of five standardized functional motor tasks administered using a specific apparatus.

Test Format and Administration Specifications

• Type: Performance-based observational functional task battery.
• Target Population: Adults and older adults with neurological impairment resulting from stroke, brain trauma, or central nervous system lesions affecting unilateral or bilateral upper extremity motor control.
• Time Required: Approximately 3 to 5 minutes.
• Setting / Posture: The patient sits upright in a standard comfortable armchair without armrests, positioned in front of a standard-height table with the torso resting comfortably against the chair back. The affected limb rests on the table surface or in the patient’s lap at the baseline start position.
• Standardized Equipment Required:
  • One standard 30 cm plastic or wooden ruler.
  • One pencil or pen.
  • One solid wooden or plastic cylinder (diameter: 2.5 cm; length: 5.0 cm).
  • One small wooden cube (edge dimension: 2.5 cm).
  • One larger wooden block or pedestal (height: 10.0 cm; square base: approximately 5 to 10 cm).
  • One standard plastic tumbler or drinking glass half-filled with potable water.
  • One standard hair comb or hairbrush.
  • One stopwatch or clinical timer.

Standardized Scoring Protocol

• Item Scoring Metric: Dichotomous Pass / Fail scale.
  • 1 = Pass: The patient successfully and independently completes the task in full compliance with standardized biomechanical criteria within the designated time ceiling (typically within 10 to 45 seconds depending on the item; standard clinical convention enforces completion within a single fluid, uninterrupted effort).
  • 0 = Fail: The patient is unable to initiate the movement, fails to meet operational criteria (e.g., drops the cylinder, knocks over the wooden tower, spills water, fails to reach designated cranial regions), exhibits excessive pathological compensation that violates protocol, or exceeds the temporal threshold.
• Aggregated Summary Score: Summed across all 5 items, ranging from 0 to 5.
  • Score 0: Complete lack of functional arm capacity; severe upper extremity paresis.
  • Scores 1 to 2: Poor functional arm capacity; limb can function primarily as a static postural stabilizer or execute coarse, assisted prehension.
  • Scores 3 to 4: Moderate functional arm capacity; capable of fine manipulation and multi-joint transportation, but lacking high-amplitude out-of-synergy motor fluidity.
  • Score 5: Intact functional capacity across all standardized functional test activities.

Permissions & Fee and Test Year

• Initial Publication Year: 1987.
• Original Authors: A. Heller, D. T. Wade, V. A. Wood, A. Narborough, and R. Langton Hewer.
• Institutional Provenance: Stroke Research Unit, Frenchay Hospital, Bristol, England, United Kingdom.
• Intellectual Property and Licensing: The Frenchay Arm Test was developed through public research funding within the UK National Health Service (NHS) and published openly in the academic literature. The scale resides in the public domain for clinical, educational, and academic research purposes.
• Commercial Fees: There are no commercial fees, per-test licensing royalties, or formal copyright paywalls associated with standard non-commercial clinical or scientific use of the FAT.
• Dutch Standardization: The Dutch language standardization and adaptation of the FAT were established and ratified by the Royal Dutch Society for Physical Therapy (Koninklijk Nederlands Genootschap voor Fysiotherapie – KNGF) in the KNGF-richtlijn Beroerte (2006; updated 2014) and are freely distributed for clinical practice across primary and secondary rehabilitation networks.

References

• Brunnstrom, S. (1970). Movement therapy in hemiplegia: A neurophysiological approach. Harper & Row.
• De Weerdt, W. J., & Harrison, M. A. (1985). Measuring recovery of arm-hand function in stroke patients: A comparison of the Brunnstrom-Fugl-Meyer test and the Action Research Arm test. Physiotherapy Canada, 37(2), 65–70. https://doi.org/10.3138/ptc.37.2.065
• Guttman, L. (1944). A basis for scaling qualitative data. American Sociological Review, 9(2), 139–150. https://doi.org/10.2307/2086306
• Heller, A., Wade, D. T., Wood, V. A., Narborough, A., & Langton Hewer, R. (1987). Arm function after stroke: Measurement and recovery over the first three months. Journal of Neurology, Neurosurgery, and Psychiatry, 50(6), 714–719. https://doi.org/10.1136/jnnp.50.6.714
• Koninklijk Nederlands Genootschap voor Fysiotherapie (KNGF). (2006). KNGF-richtlijn Beroerte [KNGF Clinical Practice Guideline for Stroke]. Nederlands Tijdschrift voor Fysiotherapie, 114(5, Suppl), 1–96. https://www.kngf.nl/kennisplatform/richtlijnen/beroerte
• Kwakkel, G., Kollen, B. J., van der Grond, J., & Prevo, A. J. (2003). Probability of regaining dexterity in the flaccid upper limb: Impact of severity of paresis and time since onset in stroke patients. Stroke, 34(9), 2181–2186. https://doi.org/10.1161/01.STR.0000087172.16305.CD
• Lyle, R. C. (1981). A performance test for assessment of upper limb function in physical rehabilitation treatment and research. International Journal of Rehabilitation Research, 4(4), 483–492. https://doi.org/10.1097/00004356-198112000-00001
• Mokken, R. J. (1971). A theory and procedure of scale analysis: With applications in political research. Walter de Gruyter.
• Twitchell, T. E. (1951). The restoration of motor function following hemiplegia in man. Brain, 74(4), 443–480. https://doi.org/10.1093/brain/74.4.443
• van der Lee, J. H., Beckerman, H., Lankhorst, G. J., & Bouter, L. M. (2001). The responsiveness of the Action Research Arm test and the Fugl-Meyer Assessment scale in chronic stroke patients. Journal of Rehabilitation Medicine, 33(3), 110–113. https://doi.org/10.1080/165019701750165916
• van der Lee, J. H., De Groot, V., Beckerman, H., Wagenaar, R. C., Lankhorst, G. J., & Bouter, L. M. (2001). The intra- and interrater reliability of the Action Research Arm test: A modified version of the Frenchay Arm Test. Archives of Physical Medicine and Rehabilitation, 82(1), 14–19. https://doi.org/10.1053/apmr.2001.18668
• Wade, D. T., Langton Hewer, R., Wood, V. A., Skilbeck, C. E., & Ismail, H. M. (1983). The hemiplegic arm after stroke: Measurement and recovery. Journal of Neurology, Neurosurgery, and Psychiatry, 46(6), 521–524. https://doi.org/10.1136/jnnp.46.6.521

Items of the Scale