Action Research Arm Test (ARAT)

Key Takeaways

19-item observational measure assessing upper limb function through grasp, grip, pinch, and gross motor movements in stroke, TBI, and neurological populations.

Standardized administration protocol requires specific positioning, equipment setup, and scoring on a 4-point ordinal scale (0-3 per item, max 57).

High interrater reliability when administered with training. MCID ranges 5-6 points for stroke populations depending on baseline severity.

Pabau’s digital forms and measurement tracking features help clinics document ARAT scores longitudinally and integrate results into structured patient records.

The Action Research Arm Test (ARAT) is a standardized 19-item measure used across stroke rehabilitation, traumatic brain injury recovery, and other neurological conditions to quantify upper limb motor control. This guide covers the observational protocol, item-by-item scoring, and score interpretation, plus how to record ARAT assessment results in patient progress notes.

The free Action Research Arm Test PDF above gives you a standardized scoring sheet to rate each item at the bedside and track recovery over time.

What is the Action Research Arm Test?

The Action Research Arm Test (ARAT) is a 19-item observational measure developed by Ronald Lyle in 1981 as a modified version of the Upper Extremity Function Test to assess functional performance of the upper extremity in patients with neurological impairment. It evaluates a clinician’s ability to measure specific changes in limb function among individuals who sustained cortical damage resulting in hemiplegia, commonly following stroke.

The test measures four domains of upper limb function:

• Grasp: lifting six graded objects — wooden blocks, a cricket ball, and a sharpening stone — from the table onto a raised shelf.
• Grip: pouring water between two glasses and displacing tubes and a washer.
• Pinch: lifting a ball bearing and a marble using the thumb opposed to the ring, middle, and index fingers.
• Gross movement: bringing the hand behind the head, on top of the head, and to the mouth.

Each item is scored on a 4-point ordinal scale (3 = performs the task normally, 2 = completes the task but takes abnormally long or with great difficulty, 1 = performs the task partially, 0 = can perform no part of the task), with a maximum total score of 57 points across 19 items.

Physical therapists, occupational therapists, and neurorehabilitation teams use the ARAT as both an evaluative tool during initial assessment and as a longitudinal outcome measure to track a patient’s recovery trajectory.

How to administer the test?

Standardized administration is critical for reliable and valid results. The ARAT protocol follows the Yozbatiran et al. (2008) standardized approach, which specifies exact positioning, object specifications, and instruction wording. Follow these step-by-step administration instructions:

1. Patient setup: Seat the patient in a firm-backed chair without armrests. The affected arm should rest on a table with the forearm in neutral position, with a shelf positioned about 37cm above the table for the grasp subtest. Ensure the patient’s vision is unobstructed and they can comfortably reach the test objects.
2. Grasp subtest (6 items): Present wooden cubes in four sizes (10cm, 7.5cm, 5cm, and 2.5cm), a cricket ball (about 7.5cm), and a sharpening stone (about 10 x 2.5 x 1cm). Starting with the largest cube, ask the patient to lift each object from the table onto the raised shelf and set it down.
3. Grip subtest (4 items): Ask the patient to pour water from one glass into another, then displace a large alloy tube (about 2.25cm diameter), a small alloy tube (about 1cm diameter), and a washer over a bolt. Score each item on completion and control, not on grip strength alone.
4. Pinch subtest (6 items): Using the thumb opposed in turn to the pads of the ring, middle, and index fingers, ask the patient to lift a ball bearing (about 6mm) and then a marble (about 1.5cm) — two objects x three fingers for six items in total.
5. Gross movement subtest (3 items): Score three tasks in order: placing the hand behind the head, placing the hand on top of the head, and bringing the hand to the mouth.

Administration is hierarchical: attempt the hardest item in each subtest first. If the patient scores a 3, credit the remaining, easier items in that subtest with full marks and move to the next subtest. If the two hardest items in a subtest both score 0, score the rest of that subtest 0 as well. This shortcut is what keeps full administration to roughly 5-15 minutes.

For clinical documentation best practices, record the exact positioning setup and any patient limitations (e.g. pain, tremor) that influence performance. Use digital assessment forms to capture standardized responses and score calculations in real time, reducing transcription errors.

Action research arm test scoring: An item-by-item breakdown

Scoring rates each of the 19 items on a 4-point ordinal scale: 3 (performs the task normally, within the time limit), 2 (completes the task but takes abnormally long or with great difficulty), 1 (performs the task partially), and 0 (can perform no part of the task).

Administration follows Lyle’s hierarchical decision rule described above, so a top score on the hardest item in a subtest carries through to the easier items below it.

Record individual item scores and calculate subtotal scores per domain (grasp, grip, pinch, gross movement). Total ARAT score ranges 0-57, with higher scores indicating better upper limb function. Use longitudinal measurement tracking to compare serial assessments across weeks or months of rehabilitation.

Interpreting test scores and clinical significance

Score interpretation depends on baseline severity and clinical population, but the ARAT itself is a continuous measure with no officially validated severity cutoffs. StrokEngine and Physiopedia both note that fixed mild, moderate, and severe bands aren’t part of the validated scale. Read a score against the patient’s own baseline function, then weigh any change against the MCID and MDC below.

The minimal clinically important difference (MCID) — the smallest change representing meaningful recovery — is approximately 5.7 points (about 10% of the 57-point range) for chronic stroke populations, per van der Lee et al. (2001), as cited by the Shirley Ryan AbilityLab RehabMeasures Database. Treat improvements at or above that threshold as signaling genuine functional recovery the patient and clinician can both detect.

Don’t confuse the MCID with the minimal detectable change (MDC), the smallest change that exceeds measurement error. A shift below the MDC may just be noise from the rater or the setup, while the MCID sits above it and marks change the patient actually notices. When reading an action research arm test score interpretation, weigh the change against both, not one in isolation.

Patient outcome tracking workflows should capture baseline ARAT scores and set recovery benchmarks at MCID intervals, helping teams identify which patients respond to intervention and who may need protocol adjustments.

Test reliability and validity across clinical populations

Action research arm test reliability and validity are both well documented. The ARAT demonstrates high interrater reliability when raters follow the Yozbatiran et al. (2008) standardized administration protocol. Reported intraclass correlation coefficients (ICC) for interrater reliability sit at or above 0.96 across every subtest and the total score, meeting gold-standard reliability thresholds — see the Shirley Ryan AbilityLab RehabMeasures Database for the full reliability data.

The test is validated across multiple populations: stroke (ischemic and hemorrhagic), traumatic brain injury (TBI), multiple sclerosis, and spinal cord injury. Construct validity is strong, with action research arm test scores correlating closely with the Fugl-Meyer Assessment and the Box and Block Test in stroke patients. The ARAT is sensitive to change over the first 3-6 months post-stroke, making it ideal for tracking acute-phase rehabilitation outcomes.

A shorter modified Action Research Arm Test has been developed for rapid screening, but the full 19-item version remains the reference standard when you need defensible reliability and validity over a course of rehabilitation.

Occupational therapy practice settings frequently use the ARAT alongside the Fugl-Meyer Assessment (Upper Extremity) to triangulate upper limb recovery. Integration into structured patient records ensures all assessment results, treatment notes, and progress metrics live in one searchable location for care continuity.

Best practices for ARAT administration and documentation

Standardize the environment: Use a quiet, well-lit room with minimal distractions. Fatigue reduces performance validity, so schedule assessment during peak energy hours. Ensure the patient has rested at least 30 minutes prior.

Train staff on scoring: Raters must understand the distinction between “movement with some property abnormal” (score 1) and “movement near normal” (score 2). Ambiguous borderline scores warrant discussion and consensus scoring when possible. Physical therapy documentation platforms should flag scoring caveats (e.g. “patient reported pain during grasp subtest, may have suppressed true capacity”).

Document contextual factors: Record pain levels, spasticity, sensory deficits, or fatigue that could influence performance. This context helps clinicians interpret score changes: a drop in ARAT score paired with a rise in spasticity signals a mechanism (increased tone) rather than true loss of motor control.

Use serial assessment: Schedule assessment sessions at consistent intervals (baseline, 2 weeks, 6 weeks, 12 weeks) to build a trajectory. Plot scores on a chart to visualize recovery rates and plateau points.

Integrating ARAT into clinical workflows

Modern rehabilitation teams use the ARAT as one outcome measure within a broader assessment battery. In occupational therapy and physical therapy settings, clinicians combine it with the Fugl-Meyer Assessment (motor control), the Box and Block Test (gross manual dexterity), the Nine-Hole Peg Test (finger dexterity), and the Modified Ashworth Scale (spasticity) to build a complete upper limb profile.

Related performance measures such as the Wolf Motor Function Test, the Arm Motor Ability Test, the Motricity Index, and a broader muscular strength test capture overlapping but distinct aspects of recovery, so choose the combination that fits your caseload.

AI-powered clinical documentation can automatically prompt clinicians to record ARAT results during post-session note generation, ensuring no assessment data is overlooked in the clinical workflow. Outcome measure interpretation frameworks guide clinicians in translating raw scores into actionable treatment modifications.

For practices managing multi-location caseloads or team-based care, centralized patient outcome tracking ensures every clinician sees the same assessment history, reducing duplicate testing and accelerating clinical decision-making.

Key references and resources

Official resources: The Shirley Ryan AbilityLab RehabMeasures Database hosts peer-reviewed evidence summaries on ARAT reliability, validity, and MCID values. APTA evidence-based practice resources include ARAT guidance for TBI populations.

Training and certification: Many universities and rehabilitation centers offer formal ARAT training workshops to achieve standardized administration and scoring. Investing in staff training dramatically improves interrater reliability and clinical confidence in score interpretation.

Outcome tracking: Document baseline ARAT scores, set MCID-based recovery milestones, and monitor longitudinal progress. Practices using structured patient record systems integrate ARAT data with treatment notes, automating progress reports and enabling outcome-driven quality improvement.

Putting the ARAT to work in your practice

The Action Research Arm Test only earns its place in a rehabilitation program when it’s administered the same way every time — the right objects, the right shelf height, the correct hierarchical shortcuts, and scores read against a real MCID rather than a rounded guess. Get that standardization right and the ARAT becomes a reliable signal of recovery you can defend to patients, families, and referring physicians.

Practice management software like Pabau helps that consistency survive contact with a busy caseload. Digital assessment forms keep the 19-item scoring sheet, subtotals, and total score in the same structured fields every time, so a score entered by one clinician reads the same way to the next.

Paired with longitudinal tracking, serial ARAT results sit alongside treatment notes in one patient record, making it easier to spot a plateau, flag a candidate for a protocol change, or pull a clean progress report for a referring physician.

Streamline outcome measurement with Pabau

Capture ARAT scores, track recovery trajectories, and generate outcome reports instantly within your clinical documentation workflow.

Book a demo

Continue your research

Need a structured framework for measuring arm function over time? Return-to-practice protocols show how practices establish and track rehabilitation milestones.

Want to integrate assessment data into patient progress notes? Improve patient engagement by sharing ARAT results and recovery graphs with patients during follow-up appointments.

Want to see how practices document a different standardized physical exam? Anterior drawer test documentation follows the same structured, repeatable format as the ARAT.

Need a template for a quick functional performance check? Hexagon agility test scoring sheets standardize a timed movement assessment from setup through scoring.

Frequently asked questions