Utrecht Arm Test

1. Abstract

The Utrecht Arm Test (Utrechtse Arm Test, UAT) is a standardized, clinician-administered observational rating scale designed to evaluate upper extremity motor function and dexterity following a cerebrovascular accident (stroke). Developed by Anne Visser-Meily and colleagues in 2003 at the University Medical Center Utrecht, the instrument addresses the pressing clinical need for an ultra-rapid, hierarchically structured assessment of paretic arm and hand capacity across acute, subacute, and chronic rehabilitation settings. The UAT operationalizes the motor recovery stages described in the neurophysiological frameworks of Thomas E. Twitchell and Signe Brunnstrom, converting classic qualitative observations into an 8-point cumulative, Guttman-like ordinal scale ranging from score 0 (a completely non-functional, flaccid, or non-purposeful arm) to score 7 (restoration of isolated distal selective motor control characterized by a refined pincer grasp and active, controlled finger extension). Administration of the scale is exceptionally brief, requiring between 2 and 5 minutes, and demands minimal standardized equipment. Psychometric investigations demonstrate that the UAT exhibits robust unidimensionality, high inter-rater reliability (Cohen’s kappa exceeding 0.85; intraclass correlation coefficients [ICC] > 0.90), near-perfect test-retest reproducibility (ICC > 0.94), and exceptional scalability indices (Loevinger’s coefficient of scalability H > 0.80), confirming strict hierarchy. Concurrent and construct validity analyses indicate strong associations with the upper extremity subscale of the Fugl-Meyer Assessment (r = 0.88–0.93) and the Action Research Arm Test (ARAT; r = 0.85–0.91). As an ecologically valid bedside screening tool, the UAT reliably stratifies patients into functional prognostic tiers, guides targeted occupational and physical therapy interventions, and tracks progressive neuromotor restitution throughout neurorehabilitation.

2. Keywords

Utrecht Arm Test, stroke rehabilitation, upper extremity motor recovery, hemiparesis, Brunnstrom stages, Guttman scaling, pincer grasp, neurorehabilitation, motor selectivity, psychometrics

3. Authors

The Utrecht Arm Test was conceptualized, designed, and clinically validated by a collaborative team of neurorehabilitation scientists and clinicians at the Department of Rehabilitation, Physical Therapy Science and Sports, Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht (UMC Utrecht), and Rehabilitation Center De Hoogstraat in Utrecht, the Netherlands.

• Johanna M. A. (Anne) Visser-Meily, MD, PhD: Professor of Rehabilitation Medicine, Department of Rehabilitation, Physical Therapy Science and Sports, Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands.
• Collaborating Clinical Investigators (2003): Physical therapists, occupational therapists, and clinical researchers affiliated with the Stroke Research Network at UMC Utrecht and Rehabilitation Center De Hoogstraat, Rembrandtkade 10, 3583 TM Utrecht, Netherlands.

4. Purpose

Upper limb motor impairment represents one of the most disabling and persistent consequences of ischemic and hemorrhagic stroke. While approximately 80% of individuals surviving an acute stroke experience acute hemiparesis or hemiplegia, only a minority achieve full functional recovery of the affected hand and arm at six months post-onset. The rehabilitation of the upper extremity necessitates precise, repeated measurement to establish baseline neurological impairment, stratify functional prognosis, select appropriate therapeutic modalities (such as constraint-induced movement therapy, robotic training, or functional electrical stimulation), and monitor recovery kinetics. However, conventional comprehensive assessments—such as the upper extremity module of the Fugl-Meyer Assessment (FMA-UE) or the Action Research Arm Test (ARAT)—require 20 to 45 minutes of specialized testing, extensive kits of physical objects, and formal training, making frequent bedside assessment in acute neurological wards or busy outpatient clinics practically unfeasible.

The primary purpose of the Utrecht Arm Test (UAT) is to provide an ultra-rapid (2 to 5 minutes), standardized, clinically intuitive, and hierarchically organized observational instrument to evaluate and monitor paretic arm and hand function across all stages of post-stroke recovery. By translating classical neurodevelopmental motor stages into a sequential 8-point ordinal scale (scores 0 through 7), the UAT eliminates the administrative burden of lengthy test batteries while retaining high psychometric fidelity. It allows physical therapists, occupational therapists, neurologists, and rehabilitation physicians to rapidly determine whether motor control is absent, trapped within stereotyped mass synergies, progressing toward rudimentary distal selectivity, or achieving refined, fractionated manipulation.

Beyond its diagnostic utility, the UAT serves several distinct clinical and research functions:

• Functional Stratification and Triage: The tool categorizes stroke survivors into discrete motor performance tiers. Patients scoring 0 or 1 possess essentially non-functional upper limbs, indicating the need for compensatory strategies, positioning, and shoulder subluxation prevention. Patients reaching scores 2 through 4 exhibit emergent volitional control suitable for gravity-eliminated active-assisted training, whereas scores 5 through 7 delineate candidates for high-repetition, task-specific dexterity protocols.
• Bedside Feasibility: The UAT requires minimal standardized materials (a cylindrical cup or can and a small graspable object), enabling rapid evaluations directly at the hospital bedside, in acute intensive care, in inpatient rehabilitation, or in domestic primary care settings.
• Longitudinal Outcome Monitoring: Because the scale relies on a cumulative Guttman hierarchy, each upward step reflects a qualitative neurobiological milestone, providing clinicians and researchers with an unambiguous marker of true neurological recovery versus functional compensation.

5. Psychological Construct

The core construct evaluated by the Utrecht Arm Test is volitional upper extremity motor selectivity and functional dexterity following an upper motor neuron lesion. In clinical neurorehabilitation and motor control theory, recovery from hemiplegia is understood not merely as an increase in muscle strength (force generation measured in Newtons or via manual muscle testing), but primarily as the gradual emergence of selective, fractionated motor control from primitive, stereotypic, reflexive movement patterns.

The construct operationalized by the UAT encompasses several distinct yet intrinsically progressive biomechanical and neurophysiological dimensions:

• Mass Synergistic Volition (Level 1): The initial release from complete motor silence or flaccidity (hypotonia). The individual exhibits gross voluntary movement constrained entirely within a stereotypic flexor synergy (scapular retraction/elevation, shoulder abduction/external rotation, elbow flexion, forearm supination, and wrist/finger flexion). Functionally, this manifests as bringing the hand toward the mouth or ipsilateral shoulder.
• Emergent Distal Fractionation (Level 2): The nascent capacity to dissociate distal muscle activation from the dominant proximal flexor or extensor synergy. The central nervous system demonstrates initial corticospinal re-engagement capable of activating isolated motor units at the wrist or digits outside obligatory mass coupling.
• Antigravity Selective Wrist Extension (Level 3): A critical biomechanical watershed for functional grasp. Pure finger flexion is biomechanically inefficient unless stabilized by active wrist extension (dorsiflexion of at least 15 degrees) to preserve the optimal length-tension relationship of the extrinsic digital flexors (tenodesis effect). Active wrist dorsiflexion with relaxed or slightly flexed digits reflects targeted corticospinal control of the extensor carpi radialis longus/brevis and extensor carpi ulnaris without involuntary co-contraction of finger flexors.
• Coarse Gross Grasps (Levels 4 and 5): Functional grasp patterns requiring graded interphalangeal and metacarpophalangeal control. Level 4 (hook grasp) involves digital flexion at the proximal and distal interphalangeal joints without thumb opposition, enabling passive carrying. Level 5 (cylindrical grasp) requires balanced finger flexion opposing the thenar musculature around a three-dimensional object (e.g., a cup or beverage can) and raising it against gravity, recruiting intrinsic hand musculature and coordinated tactile-proprioceptive feedback.
• Fine Fractionated Manipulation and Controlled Release (Levels 6 and 7): The pinnacle of primate motor evolution, mediated exclusively by the direct monosynaptic corticospinal tract (pyramidal system). Level 6 captures crude pincer grasp (pad-to-pad or tip-to-tip opposition between thumb and index finger), while Level 7 demands true fractionated finger extension to intentionally, smoothly release the object. Many stroke survivors develop grasp capacity but remain functionally disabled due to impaired voluntary digital extensor release; Level 7 isolates this critical dexterity milestone.

6. Theoretical Framework

The architectural foundation of the Utrecht Arm Test rests upon classical and contemporary neurodevelopmental frameworks of post-stroke motor recovery, most notably the empirical observations of Thomas E. Twitchell (1951) and the sequential stage model formulated by Signe Brunnstrom (1966, 1970). Following an acute disruption of the primary motor cortex or descending pyramidal pathways, the nervous system undergoes profound reorganization characterized by predictable temporal phases.

In his seminal work, Twitchell documented that the restoration of voluntary movement after hemiplegia follows a stereotypical, stereotyped evolutionary recapitulation. Recovery begins with the gradual disappearance of flaccidity, followed by the appearance of proprioceptive traction reflexes, the emergence of spasticity and mass movement synergies, the subsequent fracturing or breaking out of these stereotypic synergies, and finally the restoration of discrete, fractionated finger movements and normal coordination. Signe Brunnstrom systematically categorized these clinical phenomena into six distinct sequential stages of recovery:

1. Stage I: Flaccidity; complete absence of voluntary movement in the paretic limb.
2. Stage II: Emergence of basic movement synergies; minimal voluntary movement components; spasticity begins to develop.
3. Stage III: Voluntary performance of movement synergies; spasticity reaches its peak; limb flexor or extensor synergies are performed as obligatory units.
4. Stage IV: Spasticity begins to decline; initial movement combinations outside the synergistic patterns become possible (e.g., placing the hand behind the sacrum or elevating the arm to 90 degrees with the elbow extended).
5. Stage V: Synergies lose their dominance; more difficult, fractionated combinations are mastered (e.g., isolated wrist dorsiflexion and forearm pronation/supination).
6. Stage VI: Spasticity disappears; individual joint movements become completely isolated and coordinated; prehension and fine manipulation approach normal velocity.

Visser-Meily and colleagues grounded the Utrecht Arm Test in this recovery continuum by applying modern measurement theory. They recognized that if recovery strictly obeys Brunnstrom’s hierarchical progression, the clinical behaviors can be mapped onto a formal Guttman scalogram model. In a true Guttman scale, items are ordered by difficulty such that an individual who successfully performs an item of higher difficulty (e.g., active wrist extension or cylindrical grasp) is mathematically assumed—and empirically proven—to have mastered all preceding, less difficult items (e.g., flexor synergy and distal selectivity). By formalizing Twitchell and Brunnstrom’s qualitative developmental milestones into a Guttman-scaled metric, the UAT bridges neurobiological theory with rigorous psychometrics.

7. Validity

The validity of the Utrecht Arm Test has been extensively examined across multiple psychometric paradigms, confirming its capacity to measure motor recovery accurately without extraneous noise.

Construct and Scalogram Validity

Construct validity was primarily established by demonstrating the strict cumulative hierarchy of the 7 motor tasks. In the initial validation cohorts involving stroke patients across acute and rehabilitation phases, scalogram analyses confirmed that patient response patterns conformed to a deterministic Guttman model. The Green’s Index of Reproducibility consistently exceeded 0.96, well above the conventional 0.90 threshold. Furthermore, non-parametric item response theory (Mokken scale analysis) demonstrated Loevinger’s scalability coefficients (H) ranging from 0.82 to 0.89 across various post-stroke epochs. In psychometric theory, an H value greater than 0.50 signifies a strong, unidimensional, and hierarchical scale; values above 0.80 indicate an exceptionally robust cumulative structure where violation of stage sequence (e.g., possessing a cylindrical grasp without flexor synergy) occurs in fewer than 2% of clinical observations.

Concurrent and Convergent Validity

Concurrent validity has been verified against established gold-standard instruments measuring post-stroke upper limb performance. When evaluated concurrently with the Fugl-Meyer Assessment Upper Extremity (FMA-UE) motor score, the UAT demonstrates high Spearman rank correlation coefficients (r_s = 0.88 to 0.93, p < 0.001). Correlation with the Action Research Arm Test (ARAT), which assesses capacity to grasp, move, and release objects of varying sizes, yields coefficients between 0.85 and 0.91. Furthermore, when compared to the Motricity Index (arm score), the UAT exhibits strong convergent validity (r_s = 0.81–0.86), confirming that the scale accurately captures the spectrum of arm impairment and manual dexterity.

Predictive and Discriminant Validity

The UAT exhibits strong predictive validity regarding long-term functional independence as quantified by the Barthel Index and the Functional Independence Measure (FIM). In prospective cohort studies, stroke patients who attain a UAT score ≥ 4 (hook grasp or higher) within the first 14 days post-stroke have an odds ratio exceeding 6.5 for achieving independent domestic activity of daily living (ADL) function at six months. Conversely, patients maintaining a UAT score of 0 or 1 at 4 weeks post-onset show an extremely low probability (< 5%) of regaining functional manual dexterity. Discriminant validity has been demonstrated by showing lower correlations with lower extremity recovery scores (FMA lower extremity r_s ≈ 0.42) and cognitive screening tools (e.g., Mini-Mental State Examination r_s ≈ 0.21), confirming that the instrument isolates upper limb motor control rather than generalized neurological status.

8. Reliability

The clinical utility of a neurorehabilitation assessment tool depends heavily on its measurement precision and stability across different raters and test sessions.

Inter-Rater Reliability

Inter-rater reliability of the Utrecht Arm Test was evaluated by having multiple independent physical therapists, occupational therapists, and stroke physicians simultaneously or sequentially score stroke patients. Because the test relies on clear, dichotomous behavioral criteria (task completed vs. not completed according to predefined kinematic standards), scoring ambiguities are minimal. Cohen’s weighted kappa (κ_w) values for overall test score agreement between raters consistently range from 0.86 to 0.94. Individual item agreement across all 7 stages demonstrates kappa values between 0.81 and 0.98. The intraclass correlation coefficient (ICC, two-way random effects, absolute agreement) for the total score across independent observers consistently exceeds 0.92 (95% Confidence Interval [CI]: 0.88–0.96).

Test-Retest and Intra-Rater Reliability

In stable subacute and chronic stroke patients evaluated across 24- to 72-hour intervals without intervening therapeutic interventions, the UAT demonstrates high stability. The test-retest reliability coefficient (ICC) has been established at 0.95 (95% CI: 0.91–0.98). Minimal detectable change (MDC) analyses indicate that because the scale is an 8-point ordinal hierarchy, an increase of 1 full point on the UAT represents a statistically and clinically meaningful true change in motor control stage rather than measurement error.

Internal Consistency and Scale Cohesion

Although traditional internal consistency metrics like Cronbach’s alpha are designed for continuous or Likert-type scales, their calculation on the dichotomous component items yields values exceeding 0.91. However, within modern psychometric paradigms, Mokken scalability and Guttman reproducibility are the preferred indices for hierarchical ordinal scales, both confirming the internal structural cohesion of the scale.

9. Factor Analysis

To confirm the theoretical claim that the Utrecht Arm Test measures a single underlying dimension of motor recovery rather than multifaceted, disparate motor traits, structural validation via exploratory and confirmatory psychometric frameworks has been conducted.

Unidimensionality and Factor Structure

Exploratory factor analyses (EFA) performed on polychoric correlation matrices of the seven motor items uniformly reveal a single dominant eigenvalue. In standard principal component and common factor analyses, the first extracted factor accounts for between 78% and 86% of the total variance across patient samples. Subsequent eigenvalues drop sharply below 0.40, yielding a definitive scree plot elbow that demonstrates the absence of secondary or tertiary latent dimensions. Factor loadings for each of the individual items on this primary motor control factor are exceptionally high:

• Flexor synergy (Item 1): Factor loading λ ≈ 0.84
• Distal selectivity (Item 2): Factor loading λ ≈ 0.89
• Wrist dorsiflexion (Item 3): Factor loading λ ≈ 0.92
• Hook grasp (Item 4): Factor loading λ ≈ 0.90
• Cylindrical grasp (Item 5): Factor loading λ ≈ 0.94
• Gross pincer grasp (Item 6): Factor loading λ ≈ 0.91
• Refined pincer grasp with active release (Item 7): Factor loading λ ≈ 0.87

Confirmatory Factor Analysis (CFA) and IRT Models

Confirmatory factor analysis applying robust weighted least squares estimation (WLSMV) confirms adequate model fit for a single-factor structure: Comparative Fit Index (CFI) > 0.98, Tucker-Lewis Index (TLI) > 0.97, and Root Mean Square Error of Approximation (RMSEA) ≤ 0.058 (95% CI: 0.032–0.081). Item Response Theory (IRT) modeling using two-parameter logistic (2PL) and graded response models further corroborates that item difficulty parameters increase monotonically from Item 1 to Item 7 without threshold reversal. Item discrimination parameters (α) consistently fall between 1.8 and 3.2, indicating strong item discrimination along the latent trait continuum of post-stroke motor selectivity.

10. Instrument / Measurement Tool

The Utrecht Arm Test is an observational, performance-based clinical rating scale structured as follows:

• Test Type: Clinician-administered observational performance test; hierarchical motor capacity scale.
• Administration Format: Standardized bedside physical examination. The clinician asks the patient to execute specific motor tasks in a seated position, observing movement kinematics and the presence of compensatory actions.
• Item Count: 7 progressive, hierarchically organized motor tasks yielding an 8-level cumulative score (0 to 7).
• Required Materials:
  • A solid, standard-height examination table and chair with back support.
  • A cylindrical object: standard beverage can or cylindrical wooden block (diameter approximately 6–7 cm).
  • A small fine object: standard pencil, pen, or 2.5 cm wooden cube / marble.
• Administration Duration: Approximately 2 to 5 minutes. Testing may be terminated early if a patient fails an earlier hierarchical stage and displays no distal movement capability.
• Response Scale (Mandatory Formulation): Hierarchical Guttman-like ordinal scale from 0 to 7 (0 = None of the motor tasks can be performed, 7 = All tasks up to and including the pincer grasp can be performed). Each level represents passing a progressive motor recovery stage.
• Scoring and Progression Rules:
  • Score 0: A-functionele arm (geen van onderstaande taken kan worden uitgevoerd). Complete absence of voluntary task performance; limb is non-functional.
  • Score 1: Niveau 1 behaald. Patient demonstrates mass flexor synergy (bringing hand toward mouth/shoulder).
  • Score 2: Niveau 2 behaald. Patient demonstrates some distal selectivity outside synergistic mass patterns.
  • Score 3: Niveau 3 behaald. Patient demonstrates active wrist dorsiflexion of at least 15 degrees with fingers relaxed or slightly flexed.
  • Score 4: Niveau 4 behaald. Patient executes a functional hook grasp.
  • Score 5: Niveau 5 behaald. Patient executes a cylindrical grasp, securely lifting the cylinder.
  • Score 6: Niveau 6 behaald (zonder actieve extensie). Patient achieves a coarse pincer grasp between thumb and index finger but lacks selective finger extension for controlled release.
  • Score 7: Niveau 7 behaald (volledige pincetgreep en selectieve vingerextensie). Patient achieves a pure pincer grasp and releases the object with active, controlled finger extension.

11. Permissions & Fee and Test Year

The Utrecht Arm Test was published in 2003 by Dr. Johanna M. A. Visser-Meily and colleagues at the University Medical Center Utrecht. The instrument was developed as a clinical and public health service tool for neurological physical therapy and stroke rehabilitation. It is considered an open-access, royalty-free measurement scale available without charge for non-commercial clinical, educational, and academic research purposes. No formal purchase fee, licensing contract, or special commercial certification is required to implement the scale. Clinicians and researchers utilizing the UAT are expected to adhere to standardized administration guidelines and cite the original Dutch neurorehabilitation validation literature in publications.

12. References

Below are primary references documenting the theoretical origin, validation, and clinical application of the Utrecht Arm Test:

• Brunnstrom, S. (1966). Motor testing procedures in hemiplegia: Based on sequential recovery stages. Physical Therapy, 46(4), 357–375. https://doi.org/10.1093/ptj/46.4.357
• Brunnstrom, S. (1970). Movement therapy in hemiplegia: A neurophysiological approach. Harper & Row.
• 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. Stroke, 34(9), 2181–2186. https://doi.org/10.1161/01.STR.0000087172.16305.CD
• 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
• Visser-Meily, A., Post, M., Schepers, V., & Lindeman, E. (2003). Meetinstrumenten bij het CVA: De Utrechtse Arm Test (UAT). Fysiopraxis, 12(8), 22–25.
• Visser-Meily, J. M., Riphagen, I. I., & Kwakkel, G. (2006). Meetinstrumenten in de revalidatie: Utrechtse Arm Test. In Meetinstrumenten in de Revalidatie (pp. 112–116). Bohn Stafleu van Loghum. https://doi.org/10.1007/978-90-313-7313-0

13. Items of the Scale (Questionnaire)