Primary Image

Rehab Measures Instrument

Action Research Arm Test

Last Updated

Purpose

The Action Research Arm Test (ARAT) is a 19-item measure aimed at assessing functional performance of the upper extremity through observational means.

Link to Instrument

Acronym ARAT

Area of Assessment

Activities of Daily Living
Coordination
Dexterity
Upper Extremity Function

Assessment Type

Observer

Administration Mode

Paper & Pencil

Cost

Not Free

Actual Cost

$875.00

Cost Description

The cost of the kit is $875 including shipping within the U.S. and can be purchased at https://www.aratkits.com. The kit may also be purchased through http://saliarehab.com/arat.html. Correspondence with the Salia Rehab clinic site through email and phone call received a new pricing of $725 plus $110 shipping for the ARAT kit.

CDE Status

Not a CDE--last searched 9/20/2023.

Diagnosis/Conditions

  • Brain Injury Recovery
  • Multiple Sclerosis
  • Stroke Recovery

Key Descriptions

  • The ARAT is a 19 item?measure divided into 4 sub-tests (grasp, grip, pinch, and gross arm movement). Performance on each item is rated on a 4-point ordinal scale ranging from:
    3) Performs test normally
    2) Completes test, but takes abnormally long or has great difficulty
    1) Performs test partially
    0) Can perform no part of test
  • Lyle’s decision rules:
    1) Patients who achieve a maximum score on the first (most difficult) item are credited with having scored 3 on all subsequent items on that scale.
    2) If the patient scores less than 3 on the first item, then the second item is assessed.
    3) This is the easiest item, and if patients score 0 then they are unlikely to achieve a score above 0 for the remainder of the items and are credited with a zero for the other items.
    4) The maximum score on the ARTS is 57 points (possible range 0 to 57). A higher score indicates greater functioning in the upper extremity.
  • Items can also be summed (van der Lee et al., 2002).
  • See Yozbatiran et al. (2008) for standardized scoring protocol.

Number of Items

19

Equipment Required

  • Various sized wood blocks
  • Cricket ball
  • Stone
  • Jug and glass or two glasses
  • A small and large tube
  • Washer and bolt
  • Ball bearing
  • A marble
  • A chair without arm rests
  • A table
  • A plank
  • A tin lid
  • Kit can be bought at http://www.aratest.eu/ or www.saliarehab.com

Time to Administer

5-15 minutes

Dependent on the number of items performed

Required Training

No Training

Age Ranges

Adolescent

13 - 17

years

Adult

18 - 64

years

Elderly Adult

65 +

years

Instrument Reviewers

Initially reviewed by the Rehabilitation Measures Team in 2011; Updated by Cara Weisbach, PT, DPT and Wendy Romney, PT, DPT, NCS and the SCI EDGE task force of the Neurology Section of the APTA with references from the chronic stroke population in 2012; Updated by Irene Ward, PT, DPT, NCS and the TBI EDGE task force of the Neurology Section of the APTA in 2012. Updated by Maggie Bland PT,DPT,NCS and Nancy Byl PT,MPH,PhD, FAPTA and the STROKEDGE II Task Force of the Neurology Section of the APTA in 2016. Updated in 2023 by UIC Occupational Therapy Students Kimberly Martinez, Amanda Montoney, Leslie Everhart, and Alicia Ethen under the direction of Susan Magasi, PhD, Associate Professor, Departments of Occupational Therapy and Disability Studies, UIC.

Body Part

Upper Extremity

ICF Domain

Activity

Measurement Domain

Motor

Professional Association Recommendation

Recommendations for use of the instrument from the Neurology Section of the American Physical Therapy Association’s Multiple Sclerosis Taskforce (MSEDGE), Parkinson’s Taskforce (PD EDGE), Spinal Cord Injury Taskforce (PD EDGE), Stroke Taskforce (StrokEDGE), Traumatic Brain Injury Taskforce (TBI EDGE), and Vestibular Taskforce (Vestibular EDGE) are listed below. These recommendations were developed by a panel of research and clinical experts using a modified Delphi process.

For detailed information about how recommendations were made, please visit: 

Abbreviations:

 

HR

Highly Recommend

R

Recommend

LS / UR

Reasonable to use, but limited study in target group  / Unable to Recommend

NR

Not Recommended

Recommendations for use based on acuity level of the patient:

 

Acute

(CVA < 2 months post)

(SCI < 1 month post)

(Vestibular < 6 months post)

Subacute

(CVA 2 to 6 months)

(SCI 3 to 6 months)

Chronic

(> 6 months)

SCI EDGE

LS

LS

LS

StrokEDGE

R

R

R

Recommendations based on level of care in which the assessment is taken:

 

Acute Care

Inpatient Rehabilitation

Skilled Nursing Facility

Outpatient

Rehabilitation

Home Health

StrokEDGE

R

R

R

R

R

TBI EDGE

LS

LS

R

R

R

Recommendations based on SCI AIS Classification:

 

AIS A/B

AIS C/D

SCI EDGE

LS

LS

Recommendations for use based on ambulatory status after brain injury:

 

Completely Independent

Mildly dependant

Moderately Dependant

Severely Dependant

TBI EDGE

N/A

N/A

N/A

N/A

Recommendations for entry-level physical therapy education and use in research:

 

Students should learn to administer this tool? (Y/N)

Students should be exposed to tool? (Y/N)

Appropriate for use in intervention research studies? (Y/N)

Is additional research warranted for this tool (Y/N)

SCI EDGE

No

No

No

Not reported

StrokEDGE

No

Yes

Yes

Not reported

TBI EDGE

Yes

Yes

Yes

Not reported

Considerations

  • An instructional video is available at http://www.aratest.eu/
  • The ARAT and WMFT are highly correlated, so they may not provide significant levels of incremental validity.
  • A Rasch analysis in Chen et al. 2012) suggested revising the original 4-point scale into a 3-point scale. Tasks of “place hand behind head” and “place hand on top of head” showed poor item fit and item bias relevant to participant’s ages.
  • Sivan et al. (2011) examined outcome measures used in robot-assisted exercise trials in stroke. The ARAT had high/excellent test-retest reliability and inter-rater reliability, moderate construct validity and responsiveness, poor floor and ceiling effects, and a moderate burden overall. Also, for patients fewer than or more than six months post-stroke with moderate impairment (Fugl-Meyer > 25), the ARAT was a recommended measure of activity and participation.
  • Nordin et al. (2014) examined percent intra- and inter-rater reliability across 35 participants a median of 22 months post-stroke. Items 10, 11, 14, and 19 had some systematic disagreement within raters and items 1, 4, 17, and 19 between raters. Item 19 (hand to mouth) had the most disagreements. In general the greatest difficulties were deciding between a score of 2 or 3.
  • Croarkin et al. (2004) ranked different tests of upper extremity function based on psychometric properties. They ranked the ARAT as Level II: Established by evidence for inter-rater reliability and concurrent and convergent validity.
  • Velstra et al. (2011) examined how measures of upper extremity function related to the International Classification of Functioning, Disability, and Health. The ARAT addressed the mobility at several joints category under the Body Functions and Body Structures domain 17 times. Under the Activities and Participation domain 1-time fine hand use was covered, 6 times for grasping, 1 for turning or twisting the hands or arms, and 1 for drinking. One item was found not linked.
  • Burton et al. (2022) examined the ARAT-VR and the hand-tracking technology. The limitations in the technology resulted in a reduction in the number of items on the ARAT-VR, and an incapability of providing tactile feedback.
  • Burton et al. (2022) also note how the ARAT-VR is limited in its data among different populations. While a valid and reliable tool for Stroke patients, considerations must be made when looking at larger groups of individuals that occupational therapists serve.
  • Grattan et al. (2019) examined how the ARAT aggregate score does not indicate which items were easy, difficult, or optimally challenging for the participant. They suggested the score has limited clinical interpretability because of the lack of specific behavioral details regarding the participant’s abilities.
  • For individuals with upper limb spasticity, Pike et al. (2018) found limited evidence for ARAT’s psychometric properties of construct validity and responsiveness.
  • The ARAT is a multi-step assessment. Administering the ARAT could potentially lead to confusion or discouragement on the part of the client. The purpose and method of the ARAT should be clearly explained before its administration.

Mixed Populations

back to Populations

Test/Retest Reliability

Chronic and Acute Stroke, Multiple Sclerosis & Traumatic Brain Injury: (Platz et al., 2005; n = 23)

Test-Retest Reliability (after 7 days)

 

 

 

Action Research Arm Test:

 

 

 

 

Rating

ICC

rho

Grasp

Excellent

0.949

0.965

Grip

Excellent

0.947

0.955

Pinch

Adequate

0.894

0.897

Gross movement

Excellent

0.976

0.976

Total score

Excellent

0.965

0.968

Fugl-Meyer Test, arm section

 

 

 

Motor function

Rating

ICC

rho

A Shoulder/elbow/forearm

Excellent

0.954

0 944

B Wrist

Excellent

0.973

0.961

C Hand

Excellent

0.958

0.941

D Co-ordination/speed

Excellent

0.936

0.947

Total motor score

Excellent

0.965

0.951

Sensation

Adequate

0.806

0.672

Passive joint motion/joint pain

Excellent

0.946

0.883

Box and Block Test:

 

 

 

 

Rating

ICC

rho

Total

Excellent

0.963

0.973

 

Interrater/Intrarater Reliability

Chronic and Acute Stroke, Multiple Sclerosis & Traumatic Brain Injury: (Platz et al., 2005; n = 44)

Interrater Reliability (between 2 raters)

 

 

 

Action Research Arm Test:

 

 

 

 

Rating

ICC

rho

Grasp

Excellent

0.997

0.999

Grip

Excellent

0.964

0.958

Pinch

Excellent

0.999

0.999

Gross movement

Excellent

0.984

0.984

Total score

Excellent

0.998

0.996

Fugl-Meyer Test, arm section

 

 

 

Motor function

Rating

ICC

rho

A Shoulder/elbow/forearm

Excellent

0.989

0 984

B Wrist

Excellent

0.987

0.983

C Hand

Excellent

0.987

0.984

D Co-ordination/speed

Excellent

0.971

0.971

Total motor score

Excellent

0.997

0.995

Sensation

Excellent

0.979

0.969

Passive joint motion/joint pain

Excellent

0.983

0.980

Box and Block Test:

Excellent

0.993

0.993

 

Allotransplantation Recovery: (Ninkovic?, et al., 2015; n = 4; Mean Age = 41.25, age range = 24-54 years)

  • Excellent overall inter-rater (Cohen kappa coefficient = 0.88) and intra-rater (Cohen kappa coefficient = 0.82) reliability.
  • Excellent inter-rater and intra-rater reliability of the ARAT subtests: Cohen kappa coefficient = 0.87 (inter-rater) and 0.81 (intra-rater) for subtest A and 1.00 for subtests B, C, and D for both inter- and intra-rater reliability.

Criterion Validity (Predictive/Concurrent)

Concurrent validity:

Multiple Sclerosis: (Carpinella et al, 2014; [subjects with MS n = 21; mean age = 47.4 (9.0) years; male = 12]; [healthy volunteers n = 12; mean age = 44.3 (9.5) years; male = 7]; instrumented ARAT)

  • Excellent concurrent validity between ARAT score and z-scores related to the mean item duration (r?=?-0.823, p?<?0.001) and jerk index (r?=?-0.898, p?<?0.001).
    • The two instrumental parameters were also negatively correlated with the 9-Hole Peg Test score expressed as number of pegs per minute (Z_Duration: r = -0.776, p < .001; Z_Jerk: r = -0.765, p < .001)
    • The two parameters were able to discriminate between different levels of upper limb impairment, as demonstrated by the significant progressive increase of Z-Duration and Z-Jerk with increasing arm dysfunction. Taken together, these results suggested that instrumented ARAT is a valid tool for quantifying upper limb dysfunction in persons with MS.” (p. 12).

Construct Validity

Chronic and Acute Stroke, Multiple Sclerosis & Traumatic Brain Injury: (Platz et al., 2005; n = 56; mean age = 54, range = 13-92 years)

Construct Validity: correlational analysis (Spearman’s rho)

 

 

 

 

Fugl-Meyer motor

Action Research Arm Test

Box and Block Test

Action Research Arm Test

0.925

1

0.951

Fugl-Meyer motor

1

0.925

0.921

Fugl-Meyer sensation

0.239

0.298

0.285

Fugl-Meyer joint motion/pain

0.470

0.421

0.433

Box and Block Test

0.921

0.951

1

Motricity Index

0.861

0.811

0.798

Ashworth Scale

-0.422

-0.296

0.383

Modified Barthel Index

0.086

0.049

0.044

Correlational analysis were based on the (first) assessment of 56 patients

 

 

 

 

The above table indicate the ARAT's is strongly related to the:

  • Fugl-Meyer motor
  • Box and Block Test
  • Motricity Index

Negatively related to the Ashworth Scale, moderately related to the Fugl-Meyer sensation and joint motion/pain scales and Not related to the Modified Barthel Index

Content Validity

The ARAT is a modified version of the Upper Extremity Function Test (UEFT)

Face Validity

Multiple Sclerosis: (Carpinella et al, 2014; instrumented ARAT) 

  • The authors reported ARAT was selected for the study because, “It is one of the most widely used standardized measures for upper limb, it is relatively quick and, at the same time, it evaluates both arm and hand during the execution of functional tasks very similar to the activities of daily living.” (p. 10).

Floor/Ceiling Effects

Multiple Sclerosis: (Carpinella et al, 2014; instrumented ARAT) 

  • The authors reported a ceiling effect as a drawback for the ARAT. They explained the effect results in the impossibility for “detection of possible improvements induced by rehabilitation treatments in mild subjects scored at the top of the scale.” (p. 10).

Responsiveness

Traumatic Brain Injury: (Pike et al., 2018*; systematic review)

  • Studies that included participants with no identified limb spasticity were reviewed for their responsiveness and it was found that “the original ARAT [was] responsive to change over time in acute through to chronic stroke and in chronic TBI” (p. 459).

*Note: While systematic review is recent as of 2018, studies featured in this review were conducted prior to this time.

Stroke

back to Populations

Standard Error of Measurement (SEM)

(Simpson, 2013) A literature review identifying responsiveness data in patients post stroke (van der Lee, 2004 and Hsueh, 2002).

  • 1.3
  • Note: effect size for perceived effect ( e.g. MAL) were 1.66.2 times larger than the functional changes (measured ARAT or Wolf)

Stroke: (Grattan et al., 2019; n = 122; Mean Age = 57.2 (13.7); Mean Time Post CVA= 3 months)

  • SEM (calculated) for entire sample (n = 122): 2.35 

Minimal Detectable Change (MDC)

Stroke: (Simpson, 2013)

  • MDC90 = 3.0
  • MDC95 = 3.5

Stroke: (Daghsen et al., 2022; n = 47 [n = 22 for the construction cohort; n = 25 for the validation cohort]; Mean Age = 63 for construction cohort, 68 for validation cohort; Mean Time Post CVA = 3 weeks)

  • MDC on ARAT for entire group (n = 47): 7 points
  • MDC on Mini-ARAT for entire group (n = 47): 4 points

Stroke: (Burton et al., 2022; n = 30; Mean Age = 59.8 ± 10.8; male = 22; Mean Time Post CVA = 2.9 [0.4–14.1] months; shortened ARAT-VR (virtual reality version incorporating 13 of the original 19 ARAT items)

  • MDC for paretic hand of patients with stroke (n = 30): 4.0
    • Result was slightly superior to the traditional ARAT (MDC = 3.0) 

Minimally Clinically Important Difference (MCID)

Chronic Stroke: (van der Lee et al., 2001; n = 20; mean age = 62 (IQR = 52.5–71.8) years; median time since stroke = 3.6 years; mean ARAT score = 29.2 points)

  • MCID = 10% of the measures total range (i.e. 5.7 points)

 

Chronic Stroke: (van der Lee et al., 2001; n = 22, mean age = 58.5 years; mean time since stroke = 3.6 years; Median baseline ARAT score = 38.0 points)

  • MCID = 5.7

 

Acute Stroke: (Lang et al., 2008; mean age = 64 (14); time between stroke and first assessment = 9.5 (4.5) days)

MCID=12-17 points

MCID Raw Score:

 

 

 

 

 

 

 

MCID if Dominant Side Affected

 

 

MCID if Nondominant Side Affected

 

 

 

Raw Value

Percentage of Scale

Effect Size

Raw Value

Percentage of Scale

Effect Size

ARAT

12

21

0.78

17

30

1.10

 

Stroke: (Daghsen et al., 2022; n = 47)

  • MCID on ARAT for entire group (n = 47): 17 points
    • MCID was 14 points when the dominant side was affected and 19 points when the non-dominate side was affected).
  • MCID on Mini-ARAT for entire group (n = 47): 9 points
    • MCID was 8 points when the dominant side was affected and 9 points when the non-dominate side was affected).

Normative Data

Chronic Stroke: (van der Lee et al., 2001)

  • Mean (SD) intake ARAT score 29.2 (12.5)
  • Mean (SD) intake Fugl-Meyer Assessment score 49.2 (9.9)

Item norms (based on healthy elderly adults): 

Subtest:

Item

Time Limit (s)

Grasp

Block 2.5cm

3.6

 

Block 5cm

3.5

 

Block 7.5cm

3.9

 

Ball 7.5cm

3.8

 

Stone

3.6

 

Block 10cm

4.2

Subtest:

Item

Time Limit (s)

Grip

Tube 2.25cm

4.2

 

Tube 1cm

4.3

 

Place washer over bolt

4

 

Pour water from glass to glass

7.9

Subtest:

Item

Time Limit (s)

Pinch

Large marble first finger and thumb

3.8

 

Large marble second finger and thumb

3.8

 

Large marble third finger and thumb

4.1

 

Small marble first finger and thumb

4

 

Small marble second finger and thumb

4.1

 

Small marble third finger and thumb

4.4

Subtest:

Item

Time Limit (s)

Gross Movement

Move hand to mouth

2.4

 

Place hand on top of head

2.7

 

Place hand behind head

2.7

Time limits (mean + 2 SD of the performance times of 20 healthy elderly subjects)
If performance is slower than the time limit or if the patient loses contact with the back of the chair during performance, the score is 2 instead of 3.

 

Acute Stroke: (Beebe and Lang, 2009; mean age = 56.9 (10.2), times since stroke onset = 18.6 (5.6) days)

Normative Data:

 

 

 

 

1 month

3 months

6 months

ARAT

26.4 (23.9)

39.5 (19.7)

41.3 (20.8)

Grip Strength (kg)

9.2 (9.6)

14.0 (10.3)

15.4 (11.4)

9HPT (sec)

88.8 (40.2)

67.8 (41.7)

60.8 (39.7)

SIS: Hand function

19.9 (28.0)

48.4 (32.7)

43.9 (34.2)

9HPT = 9-Hole Peg Test
SIS = Stroke Impact Scale-Hand

 

 

 

Test/Retest Reliability

Stroke: (Burton et al., 2022; shortened ARAT-VR)

  • Excellent test-retest reliability between trials with the paretic hand: (ICC = 0.98)

Interrater/Intrarater Reliability

Acute Stroke: (Nijland et al., 2010; n = 40; mean age = 60 (13.6) years; median ARAT score = 38; times since stroke onset < 6 months; Dutch sample)

  • Excellent interrater reliability (ICC = 0.92)

Chronic Stroke: (Van der Lee et al., 2001)

  • Excellent Interrater Reliability (ICC = 0.995)
  • Excellent Intrarater Reliability (ICC = 0.989)

Chronic Stroke: (Yozbatrin et al., 2008)

Interrater Reliability

Action Research Arm Test

 

 

 

 

Rating

ICC

rho

Grasp

Excellent

0.9992

1.0

Grip

Excellent

0.996

0.99

Pinch

Excellent

0.997

0.98

Gross Movement

Excellent

0.978

0.93

Total Score

Excellent

0.9986

0.96

 

Intrarater Reliability

Action Research Arm Test

 

 

 

 

Rating

ICC

rho

Grasp

Excellent

0.98

0.93

Grip

Excellent

0.97

0.93

Pinch

Excellent

0.99

0.98

Gross Movement

Excellent

0.93

0.91

Total Score

Excellent

0.99

0.99

Stroke: (Page, 2015) Patients an average of 4.6 years since stroke with moderate upper extremity paresis

  • Excellent Intrarater reliability (ICC = 0.99, 95% CI 0.98-0.99)

Stroke: (Page, 2012) Patients greater than 12 months post-stroke with minimal upper extremity paresis enrolled in trial. Measurements were made before starting the trial, approximately 1 week apart.

  • Adequate Intrarater reliability (ICC = 0.71, 95% CI 0.53-0.89)

Stroke: (Daghsen et al., 2022)

  • Excellent intra-rater reliability for ARAT and Mini-ARAT: (ICC = 0.99, 95% CI: 0.98-0.99)

Chronic Stroke: (Spence et al., 2020; n = 20; Mean Age =  32, age range = 25-53; Physiotherapists rating the same patient post-stroke)

  • Adequate inter-rater reliability: (Kendall’s W = 0.71, p < 0.05) 

Internal Consistency

Acute Stroke: (Nijland et al, 2010)

  • Excellent Internal Consistency (Cronbach's alpha = 0.985)

Stroke: (Ferna?ndez-Solana, et al., 2022; n = 83; Mean Age = 61.81 (11.54); Mean Time Post CVA = 6 months [acute phase]; Spanish version)

  • Excellent: Cronbach’s alpha = 0.96
    • In addition, “Cronbach’s alpha with each of the suppressed elements ranged from 0.958 and 0.960, and the total correlation of corrected elements was greater than 0.42 in all cases” (p. 8).

Stroke: (Daghsen et al., 2022)

  • Excellent: Cronbach’s alpha for ARAT = 0.93*
  • Excellent: Cronbach’s alpha for Mini-ARAT = 0.87

Stroke: (Grattan et al., 2019)

  • Exellent: Person reliability (analogous to Cronbach’s alpha) = 0.98

Stroke: (Zhao et al., 2019; n = 44; Mean Age = 57.50 years, range = 22 – 80; male = 36; Median Time Post CVA = 3.0 months, range = 0.5 – 80.27; Chinese version of the ARAT (C-ARAT))

  • Excellent: Cronbach’s alpha = 0.98* (p < 0.001)

*Scores higher than 0.9 may indicate redundancy in the scale questions. 

Criterion Validity (Predictive/Concurrent)

Chronic Stroke: (van der Lee et al., 2001)

  • Evidence of concurrent validity confirmed by comparison with the upper limb subtest of the Fugl- Meyer Assessment and the Motor Assessment Scale.

 

Chronic Stroke: (Yozbatiran et al., 2008)

  • Excellent correlation between ARAT and arm motor score of the Fugl-Meyer (r = 0.94, p<0.01)

(Chen, 2012) Patients seen an average of 17.19 (±15.29) months post-stroke

  • Adequate predictive validity with the composite physical domain and hand domain of the Stroke Impact Scale (ρ = 0.45 and 0.58, p<0.001, respectively)
  • Excellent predictive validity with the performance time and functional ability scale on the Wolf Motor Function Test (ρ = -0.66 and 0.76, p<0.001, respectively), Motor Activity Log Amount of Use (30) and Quality of Movement (30) (ρ = 0.62 and 0.66, p<0.001, respectively)

(Page, 2015) Patients an average of 4.6 years since stroke with moderate upper extremity paresis

  • Excellent concurrent validity with the Wrist Stability and Hand Mobility Subscales of the Fugl-Meyer Assessment (0.67-0.74, p < .001)

(O’Dell, 2014) Community-dwelling volunteers seen an average (mean (SD)) of 4.1 (4.5) years post-stroke for upper extremity robotics training

  • Excellent concurrent validity ( 0.79, p = 0.001) with the 9-item version of the Arm Motor Ability Test.

(Wei, 2011) Twenty-seven stroke participants with moderate motor impairment in their affected upper extremity, an average of 4.92 ±0.45 years post-stroke.

  • Excellent concurrent validity (0.81 – 0.96, p < 0.01) with the Fugl-Meyer Assessment and Motor Status Scale.

(Lin, 2010) Fifty-nine stroke participants an average of 16.14 ± 13.95 months post-stroke engaging in upper extremity training or placebo.

  • Adequate concurrent validity (0.31 – 0.54, p < 0.05) with the Fugl-Meyer Assessment, Motor Activity Log-Amount of Use and Quality of Life, and Stroke Impact Scale Hand Function Domain.

(Chuang, 2012) Sixty-seven participants an average of 21.12 ± 13.63 months post-stroke had functional state of upper extremity skeletal muscle assessed with a Myoton-3 myometer to measure tone, elasticity and stiffness.

Pretreatment

  • Poor concurrent validity of the ARAT with muscle tone, elasticity and stiffness of the flexor carpi radialis (0.27, [p < 0.05], 0.02 [P>0.06], and 0.30 [p<0.05])
  • Poor concurrent validity with the ARAT and muscle tone, elasticity and stiffness of the extensor digitorum (-0.01 to -0.03) and the flexor carpi radialis (-0.07 to 0.10).

Posttreatment

  • Poor to Adequate concurrent validity of the ARAT with muscle tone, elasticity and stiffness of the flexor carpi radialis (0.29 [P<0.05], 0.03 [P>0.05) and 0.36,[ < 0.01])
  • Poor concurrent validity with the ARAT and the muscle tone, elasticity and stiffness of the extensor digitorum(-0.-8 to 0.19) and the flexor carpi ulnaris (-0.18 to 0.11)

(Edwards, 2012) Fifty-one post-stroke subjects as part of the VECTORS (constraint induced movement therapy) study assessed at Day 0 (9.5 ± 4.5), Day 14 (24.9 ± 10.6), Day 90 (110.8 ± 20.7).

  • Adequate to Excellent concurrent validity with the Wolf Motor Function Test, Functional Ability Score (WMFT FA)

 

Day 0

Day 14

Day 90

WMFT FA Function Score

0.745

0.827

0.863

WMFT FA Time Score

-0.641

-0.825

-0.772

WMFT FA Grip Score

0.702

0.631

0.553

 

(Lee, J., 2015) Fifteen subjects an average of 3.1 ± 2.3 years post-stroke who wore accelerometers while completing the ARAT

  • Poor concurrent validity (0.24) on the affected side, and Excellent concurrent validity on the non-affected side (0.91).

(Lee, G., 2015) Forty-three participants were recruited a mean 15.21 ± 3.32 months post-stroke for assessment.

  • An Adequate negative correlation was found (-0.41;p<0.05) with hypertonia (defined as a Modified Ashworth Scale Score of > 1+)
  • Adequate predictive validity (Cut-off ≤ 15.50/57 points, AUC .76 (95% CI 0.61-0.90, p <0.01). If a person had an ARAT score of ≤ 15.50/57 points, they had a 1.359 increased risk of having a Modified Ashworth Scale score of ≥ 1+ (p = 0.051, 95% CI 1.01-1.82)

(Li et al, 2015) A study with 95 individuals with first time stroke greater than one month post, < 2 on the Modified Ashworth and good mental status (MMST >24) reported significant and adequate concurrent and predictive correlations between arm kinematics (movement time) and the ARAT during reaching (with and without trunk constraint) .

Regression

Kinematic Variable

Adjusted R2

P value

 

Concurrent Validity Before Intervention

 

 

 

Trunk constraint

MT Pre

0.29

<0.001

 

Trunk Unconstraint

MT Pre

0.40

<0.001

 

Concurrent validity After Intervention

 

 

 

Trunk constraint

MT Pre

0.38

<0.001

 

Trunk Unconstraint

MT Pre

0.34

<0.001

 

Predictive Validity ARAT Post

 

 

 

Trunk constraint

MT Pre

0.33

<0.001

 

Trunk Unconstraint

MT Pre

0.36

<0.001

 

 

(Page, 2012) Patients greater than 12 months post-stroke with minimal upper extremity paresis enrolled in trial.

  • Excellent concurrent validy with the Wrist/Hand Subscales of the Fugl-Meyer Assessment.

(Hsieh, 2009) A cohort of 57 stroke participants (an average of 12.98±7.62 months post-stroke) with assessments completed prior to and after treatment in an effectiveness trial examining distributed constraint-induced therapy and bilateral arm training.

  • Poor predictive validity (Spearman ρ (95% CI)) between the ARAT and the Functional Independence Measure-Total Score (0.22 (-0.04, 0.46, not significant)) and the Functional Independence Measure-Motor Score (0.26 (0.00, 0.49, not significant)).

 

Predictive validity:

Stroke: (Ferna?ndez-Solana, et al., 2022; Spanish version)

  • Adequate predictive validity of the ARAT with the motor function dimension of the Fugl-Meyer Assessment of the Upper Extremity (FMA-EU) scale (r = 0.369 – 0.596)
  • Adequate to Excellent predictive validity of the ARAT with the total score of the FMA-EU scale (r = 0.343 – 0.609)
  • Excellent predictive validity of the Mini-ARAT at 3 weeks in predicting the Fugl-Meyer Assessment at 12 weeks in two different cohorts (Spearman’s coefficient (CI) = 0.68, 95% CI: 0.31-0.81, p < 0.01)—see table below.
    • “The predictive value of the Mini-ARAT in the independent cohort remained good, strengthening the validity of the Mini-ARAT since its predictive value could generalize to independent cohorts with different baseline characteristics” (p. 1265).

Spearman’s correlation coefficients between the ARAT, Mini-ARAT and the Fugl Meyer Assessment

Cohort

ARAT (95% CI)

Mini-ARAT (95% CI)

Construction

 

 

   FM* at baseline

0.85 (0.66-0.93)

0.79 (0.55-0.90)

   FM at 3 months

0.80 (0.56-0.92)

0.81 (0.54-0.92)

Validation

 

 

   FM at baseline

0.61 (0.23-0.99)

0.60 (0.22-0.98)

   FM at 3 months

0.59 (0.20-0.97)

0.58 (0.19-0.96)

*FM: Fugl Meyer scores; p < 0.001 for all correlations.

 

Concurrent validity:

Stroke: (Burton et al., 2022; shortened ARAT-VR)

  • Excellent concurrent validity between ARAT-VR and ARAT-19 scores (r = 0.84, p < 0.001)
  • Excellent concurrent validity between ARAT-VR and ARAT-13 scores (r = 0.83, p < 0.001)

Stroke: (Zhao et al., 2019; Chinese version of the ARAT (C-ARAT))

  • Excellent concurrent validity between the C-ARAT total score/all subscales and the Upper Extremity Session of the Fugl-Meyer Assessment (UE-FMA), the Wolf Motor Function Test – Functional Ability score (WMFT-FA), and the Wolf Motor Function Test – Time score (WMFT-Time)--see table below.
    • The total C-ARAT and all subscales exhibited strong negative correlations with the WMFT-Time.

Spearman’s correlation coefficients between C-ARAT Scores and those of UE-FMA, WFMT-FA, and WFMT-Time*

Variable

UE-FMA

WMFT-FA

WFMT-TIME

C-ARAT Total Score

0.824**

0.852**

-0.940**

C-ARAT Grasp Score

0.857**

0.873**

-0.894**

C-ARAT Grip Score

0.844**

0.917**

-0.927**

C-ARAT Pinch Score

0.773**

0.780**

-0.903**

C-ARAT Gross Score

0.665**

0.720**

-0.782**

*C-ARAT=Chinese version of the Action Research Arm Test; UE-FMA=Upper Extremity subscale of the Fugl-Meyer Assessment; WFMT=Wolf Motor Function Test; WFMT-FA=Wolf Motor Function Test-Functional Ability

**p < 0.001

 

Construct Validity

(Awert et al (2015) 50 patients post first stroke ( < 5 years), participating in outpatient rehabilitation, capable of completing a self assessment questionnaire Michigan Hand Outcomes Questionnaire (MHOQ) and strength testing (ARAT) .

  • Excellent correlations between the MHOQ and the ARAT for all patients (0.64; p<0.001) and for patients with arm impairments (0.60; P<0.000).

 

(Houwink, 2011) Twenty-one participants admitted to rehabilitation, with a stroke diagnosis occurring less than six weeks prior to admission.

  • Excellent cross-sectional (0.91, p < 0.001) and longitudinal, 3 months in between assessments, (0.71, p < 0.001) correlations with the Stroke Upper Limb Capacity Scale.

(Rabadi and Rabadi 2006) A study of 104 patients in an acute stroke rehabilitation (16±9 days on average post stroke) measured the performance on the ARAT within 72 hours of admission and 24 hours before discharge as well as the FMA National Institutes of Health Stroke Scale, FIM instrumental total score and FIM activities of daily living.

  • The Spearman rank correlation of ARAT was excellent with the FMA (rho=0.77 on admission and 0.87 on discharge, P<0.001), adequate with the FIM-ADL on admission and discharge (0.32; P<0.001) and adequate to poor for the FIM (0.33 admission and 0.21 discharge; P<0.001)

(Hsieh, 2009)

  • Excellent construct validity at pre-treatment and post-treatment was found between the ARAT and the Fugl-Meyer Assessment (0.73-0.74, p<0.01) and the Wolf Motor Function Test-Functional Ability Scale (0.68-0.77, p<0.01).
  • Adequate to Excellent construct validity was found between the ARAT and the Wolf Motor Function Test-Performance Time (0.58-0.63, p<0.01).
  • Poor construct validity was found between the ARAT and the Functional Independence Measure-Motor Score (0.27-0.39, p<0.01).

 

Convergent validity:

Stroke: (Ferna?ndez-Solana, et al., 2022; Spanish version)

Pearson correlation coefficients between ARAT and FIM-FAM, Lawton and Brody, and ECVI-38a

 

ARAT scale

 

FIM-FAM

 

Lawton and Brody

 

ECVI-38

Thick grip (left)

0.278*

           0.244*

-0.307**

Thick grip (right)

 

0.507**

 

           0.293**

 

-0.469**

Grip (left)

0.266*

           0.238*

-0.284**

Grip (right)

0.478**

           0.259*

-0.455**

Clamp (left)

0.853**

           0.168

-0.516**

Clamp (right)

0.172

           0.277*

-0.230*

Thick Mov (left)

0.485**

           0.255*

-0.469**

Thick Mov (right)

 

0.505**

 

           0.251*

 

-0.477**

Total ARAT (left)

 

0.237*

 

           0.214

 

-0.283**

Total ARAT (right)

 

0.506**

 

           0.283**

 

-0.469**

aARAT: Action Research Arm Test; FIM-FAM: Functional independence measure-functional assessment measure; ECVI-38: Stroke Quality of Life Scale

*Correlation is significant at the 0.05 level (bilateral)

**Correlation is significant at the 0.01 level (bilateral)

Content Validity

The ARAT is a modified version of the Upper Extremity Function Test (UEFT)

Stroke: (Burton et al., 2022)

  • The Burton et al. (2022) study explored the content validity of the 13 item ARAT-VR measure. The “median difficulty of the 13 ARAT-VR items (0[0 to ?1] to 0[0–1]) evaluated by healthcare professionals was rated as equivalent to the classical ARAT for all tasks except those involving the marbles (p. 6).

Floor/Ceiling Effects

Acute Stroke: (Lin et al., 2009; n = 53; mean age = 64; Taiwanese sample)

% of Individuals who Experienced Floor and Ceiling Effects:

 

 

Days Post Stroke

Floor

Ceiling

14

41.5

9.4

30

17.0

20.8

90

11.3

20.8

180

11.3

22.6

 

Acute Stroke: (Nijland et al., 2010)

  • Floor effects for scores < 3
  • Ceiling effects for scores > 54

(Edwards, 2012)

% of Individuals who Experienced Floor and Ceiling Effects:

Days Post Stroke

Floor

Ceiling

Day 0 (see time frame above)

5.9

3.9

Day 14

2

22

Day 90

2.1

33

 

(Hsueh, 2002) Forty-eight participants undergoing rehabilitation. At admission it was a median of 24 days (range 7-53) post-stroke.

% of Individuals who Experienced Floor and Ceiling Effects:

 

Admission to rehabilitation Floor

Discharge from rehabilitation Ceiling

ARAT Total

52.1%

7%

ARAT Grasp

70.8%

27.1%

ARAT Pinch

72.9%

16.7%

ARAT Gross Movement

52.1%

29.2%

 

Stroke: (Daghsen et al., 2022)

  • Daghsen and colleagues (2022) constructed the shortened version of the ARAT in such a way that it included “no more floor or ceiling effects than the original Action Research Arm Test total score” (p. 1263).

Stroke: (Zhao et al., 2019; Chinese version of the ARAT (C-ARAT))

  • Excellent: No floor effects were observed for the C-ARAT.
  • Adequate ceiling effect of 6.8% found for the C-ARAT.
    • “A significant ceiling effect of WMFT-time was observed when compared to C-ARAT, which indicated that C-ARAT may be a more optimal assessment tool to evaluate the people with mild stroke when compared to WMFT-time.” (p. 7).

Chronic Stroke: (Spence et al., 2020)

  • In post-stroke individuals, “the ARAT has its maximum ability when used in moderate degree of stroke severity, due to known ceiling effects. Therefore, if chronic stroke participants show mild symptoms it might be worth considering using an alternative due to ARAT’s inability to detect the smaller milder symptoms.” (p. 484). 

Responsiveness

Chronic Stroke: (van der Lee et al., 2002; n = 31 (RIQ = 52–66) years; mean baseline ARAT score = 30.27; > 1 year post stroke)

  • 1.2 points (using Lyle’s decision rule) 
  • 1.7 points (summing items)

 

Acute Stroke: (Lin et al., 2009; n = 53; mean age = 64; Taiwanese sample)

Responsiveness:

 

 

Days Post Stoke

Effect Rating

Interpretation

14–30

Small

Poor

14–90

Moderate

Adequate

14–180

Moderate

Adequate

 

Acute Stroke: (Beebe and Lang, 2009)

Responsiveness:

 

 

Measure

1–3 months

1–6 months

ARAT

0.55

0.63

9HPT

0.52

0.66

SIS-Hand

1.02

0.86

 

Acute Stroke (Lang et al., 2006; mean age = 64 (14), Admission NIHSS = 5.3 (1.8); time between stroke and first assessment = 9.5 (4.5) days)Acute Stroke: (Beebe and Lang, 2009)

 

Responsiveness of the ARAT

 

 

Method

Day 0 to Day 14

Day 0 to Day 90

Single population all demonstrated large effect sizes

 

 

 

ARAT total score

1.018

1.390

ARAT gross subscore

0.729

0.984

ARAT grasp subscore

1.042

1.224

ARAT grip subscore

1.017

1.324

ARAT pinch subscore

0.854

1.494

 

(Edwards, 2012)

Responsiveness:

Measure (see time frame above)

Day 0-14

Day 0-90

ARAT Total Score

1.017

1.390

ARAT Gross Motor Score

0.729

0.984

ARAT Grasp Score

1.042

1.224

ARAT Grip Score

1.017

1.324

ARAT Pinch Score

0.854

1.494

(Murphy et al, 2013) Kinematic movement analysis and clinically meaningful improvement in the upper extremity were evaluated with kinematic movement analysis of a drinking task and the ARAT in 51 subjects 9 days and 3 months post stroke.

Kinematics

Clinically meaningful improvement in ARAT (95%CI)

Movement Time

-5.16 (-2.4; -8.4)

 

Movement Units

-6.96 (-2.4; -11.4)

 

Trunk Displacement

-2.58 (-1.2; -4.2)

 

 

 

 

 

(Tong and Hu, 2011) Twenty seven patients chronic post stroke (avg.4.92 years) with low level arm function participated in a robotic training paradigm and responsiveness was measured with the FMA, MSS and ARAT using the standardized response mean (SRM) and the Guyatt’s Responsiveness Index (GRI).

  • There were no significant gains in the scores on the ARAT after treatment (25.00 [11.25] to 25.86 [10.82])
  • The responsiveness was low with SRM 0.22 and GRI 0.81
  • The responsiveness was lower for the ARAT than the FMA and MSS. In addition, the responsiveness was lower than 0.85 reported in a previous Hseih study (2009) The ARAT may not be as responsive in patients with greater upper limb impairments (e.g. baseline ARAT score for the Hseih study was 42.72 + 12.11 compared to the Wei study with a baseline score of 23.48 + 11.62).

(Rabadi and Rabadi 2006)

  • The SRM was 0.68 (admission score of 23 ±24 and discharge score 36±23
  • This SRM was lower than reported by vanderLee and Roord (2002) which included subjects with a higher level of upper limb function (ARAT score 30.27 at baseline with subjects > 1 year post stroke)

(Hsieh, 2009)

  • The standardized response mean (95% CI) of the ARAT was found to be 0.95 (0.75, 1.20, Wilcoxon Z = 4.64, p<0.01).
  • The ARAT had a smaller standardized response mean (difference in SRM (95%CI) when compared against the Fugl-Meyer Assessment (0.47 (0.09, 0.89, p<0.05)) and the Wolf-Motor Function Test-Functional Ability Scale (0.35 (-0.01, 0.78, p = not significant)). However, the ARAT had a greater standardized response mean when compared to the Wolf-Motor Function Test-Performance Time (0.57 (0.28, 0.86, p<0.05)).

(Lin, 2010)

  • The standardized response mean (95% CI) was found to be 0.79 (0.63-1.10, Wilcoxon Z = 5.76, p<0.001).

 

Stroke: (Pike et al., 2018*; n = 985 [+20 participants could not be differentiated as either stroke or TBI]; Mean Age = [>18 years]; Mean Time Post CVA = 46% of studies included participants ?6 months post injury; systematic review)

  • Of nine studies included in the best evidence synthesis stage of the study (three of which included participants with upper limb spasticity), the synthesis found there was a “positive moderate level of evidence for responsiveness” (p. 459).
  • Studies that included participants with no identified limb spasticity were reviewed for their responsiveness and it was found that “the original ARAT [was] responsive to change over time in acute through to chronic stroke and in chronic TBI” (p. 459).

*Note: While systematic review is recent as of 2018, studies featured in this review were conducted prior to this time.

Bibliography

Awert ,H.J, Keiser S, Kromme, CH, Vliet, TP, Vlieland, V, Meesters, J.J. (2015) “Validity of the Michigan Hand Outcomes Questionnaire in patients with stroke” Arch of Phys Med and Rehab.

Beebe, J. A. and Lang, C. E. (2009). "Relationships and Responsiveness of Six Upper Extremity Function Tests During the First Six Months of Recovery After Stroke." Journal of Neurologic Physical Therapy 33(2): 96-103

Burton, Q., Lejeune, T., Dehem, S., Lebrun, N., Ajana, K., Edwards, M. G., & Everard, G. (2022). Performing a shortened version of the Action Research Arm Test in immersive virtual reality to assess post-stroke upper limb activity. Journal of neuroengineering and rehabilitation, 19(1), 133.

Carpinella I, Cattaneo D, Ferrarin M (2014). Quantitative assessment of  upper limb motor function in multiple sclerosis using an instrumented  Action Research Arm Test. J Neuroeng Rehabil; 11(1): 67. 

Chen, H., Lin K., Wu, C., Chen, C. (2012). “Rasch validation and predictive validity of the action research arm test in patients receiving stroke rehabilitation.” Arch Phys Med Rehabil 93: 1039-1045.

Chuang, L., Wu, C., Lin, K. (2012). “Reliability, validity, and responsiveness of myotonometric measurement of muscle tone, elasticity, and stiffness in patients with stroke.” Arch Phys Med Rehabil 93: 532-540.

Croarkin, E. Danoff, J., Barnes, C. (2004). “Evidence-based rating of upper-extremity motor function tests used for people following stroke.” Phys Ther 84(1):62-74.

Daghsen, L., Fleury, L., Bouvier, J., Zavanone, C., Dupont, S., Hummel, F. C., & Rosso, C. (2022). Evaluation of a shortened version of the Action Research Arm Test (ARAT) for upper extremity function after stroke: The Mini-ARAT. Clinical rehabilitation, 36(9), 1257–1266.

Edwards, D.F., Lang, C.E., Wagner, J.M., Birkenmeier R.L., Dromerick, A.W. (2012). “An evaluation of the Wolf Motor Function Test in motor trials early after stroke.” Arch Phys Med Rehabil 93:660-668.

Ferna?ndez-Solana, J., Pardo-Herna?ndez, R., Gonza?lez-Bernal, J. J., Sa?nchez-Gonza?lez, E., Gonza?lez-Santos, J., Soto-Ca?mara, R., & Santamari?a-Pelaez, M. (2022). Psychometric Properties of the Action Research Arm Test (ARAT) Scale in Post-Stroke Patients-Spanish Population. International journal of environmental research and public health, 19(22), 14918.

Grattan, E. S., Velozo, C. A., Skidmore, E. R., Page, S. J., & Woodbury, M. L. (2019). Interpreting Action Research Arm Test Assessment Scores to Plan Treatment. OTJR : occupation, participation and health, 39(1), 64–73.

Houwink, A., Roorda, L.D., Smits, W., Molenaar, I.W., Geurts, A.C. (2011). “Measuring upper limb capacity in patients after stroke: reliability and validity of the stroke upper limb capacity scale.” Arch Phys Med Rehabil 92: 1418-1422.

Hsieh, Y., Wu, C., Lin, K., Chang, Y., Chen, C., Liu, J. (2009). “Responsiveness and validity of three outcome measures of motor function after stroke rehabilitation.” Strok 40:1386-1391.

Hsueh, I., Hsieh, C. (2002). “Responsiveness of two upper extremity function instruments for stroke inpatients receiving rehabilitation.” Clin Rehabil 16: 617-624.

Lang, C., Edwards, D., et al. (2008). "Estimating minimal clinically important differences of upper extremity measures early after stroke." Archives of physical medicine and rehabilitation 89(9): 1693.

Lang, C. E., Wagner, J. M., et al. (2006). "Measurement of upper-extremity function early after stroke: properties of the action research arm test." Arch Phys Med Rehabil 87(12): 1605-1610.

Lee, G.,An, A., Lee, Y., Lee, D., Park, D. (2015) “Predictive factors of hypertonia in the upper extremity of chronic stroke survivors.” J Phys Ther Sci 27: 2545-2449.

Lee, J., Park, J.H., Kim, Y. (2015). “Sensitivity of the accelerometer as a measurement tool for upper extremity movement by stroke patients: a comparison with the action research arm test.” J Phys Ther Sci 27: 1053-1054.

Li, K-y, Lin, K-c, Chen C-k, Liing R-j, WU, C-y, Chang, W-y (2015) “Concurrent and predictive validity of arm kinematics with and without a trunk restraint during a reaching task in individuals with stroke.” Arch of Phys Med and Rehab:96:1666-75

Lin, J.-H., Hsu, M.-J., et al. (2009). "Psychometric comparisons of 4 measures for assessing upper-extremity function in people with stroke." Phys Ther 89: 840-850.

Lin K-c, Chuang L-l, Wu C-y, Hsieh Y-w. (2010) Responsiveness and validity of three dexterous function measures in stroke rehabilitation. J of Rehabilitation Research and Development:;47 (4): 563-572

Lyle, R. C. (1981). "A performance test for assessment of upper limb function in physical rehabilitation treatment and research." Int J Rehabil Res 4(7333761): 483-492.

McDonnell, M. (2008). "Action research arm test." Aust J Physiother 54(3): 220.

Murphy, MA, Willen Cand Sunnerhagen KS.(2013) “Responsiveness of upper extremity kinematic measures and clinical improvement during the first three months after stroke.” 27 (9): 844-853

Nijland, R., van Wegen, E., et al. (2010). "A comparison of two validated tests for upper limb function after stroke: The Wolf Motor Function Test and the Action Research Arm Test." J Rehabil Med 42(7): 694-696.

Ninkovic?, M., Weissenbacher, A., Pratschke, J., & Schneeberger, S. (2015). Assessing the outcome of hand and forearm allotransplantation using the Action Research Arm Test. American journal of physical medicine & rehabilitation, 94(3), 211–221.

Nordin, A., Murphy, M.A., Danielsson, A. (2014) “Intra-rater and inter-rater reliability at the item level of the action research arm test for patients with stroke.” J Rehabil Med 46: 738-745.

O’Dell, M.W., Kim, G., Rivera, L., et al. (2013) “A psychometric evaluation of the arm motor ability test.” J Rehabil Med 45 (6):510-527.

Page, S.J., Levine, P., Hade, E. (2012). “Psychometric properties and administration of the Wrist/Hand Subscales of the Fugl-Meyer Assessment in minimally impaired upper extremity hemiparesis in stroke.” 93: 2373-2376.

Page, S.J., Hade, E., Persch, A. (2015). “Psychometrics of the wrist stability and hand mobility subscales of the Fugl-Meyer Assessment in moderately impaired stroke.” Phys Ther 95(1): 103-108.

Pike, S., Lannin, N. A., Wales, K., & Cusick, A. (2018). A systematic review of the psychometric properties of the Action Research Arm Test in neurorehabilitation. Australian Occupational Therapy Journal, 65(5), 449–471.

Platz, T., Pinkowski, C., et al. (2005). "Reliability and validity of arm function assessment with standardized guidelines for the Fugl-Meyer Test, Action Research Arm Test and Box and Block Test: a multicentre study." Clin Rehabil 19: 404-411.

Rabadi, MH, Rabadi FM. (2006) “Comparison of the action research arm test and the Fugl-Meyer assessment as measures of upper-extremity motor weakness after stroke.” Arch Phys Med Rehabil 87(7):962-6

Simpson, L.A., Eng, J.J. (2013) “Functional recovery following stroke: capturing changes in upper extremity function.” Neurorehabil Neural Repair 17(3): 240-250. Find it on PubMed

Sivan, M., O’Connor, R.J., Makower, S., Levesley, M., Bhakta, B. (2011) “Systematic review of outcome measures used in the evaluation of robot-assisted upper limb exercise in stroke.” J Rehabil Med 43:181-189.

Spence, N., Rodrigues, N. C. L., Nomikos, P. A., Yaseen, K. M., & Alshehri, M. A. (2020). Inter-rater reliability of physiotherapists using the Action Research Arm Test in chronic stroke. Journal of Musculoskeletal & Neuronal Interactions, 20(4), 480–487.   

van der Lee, J. H., Beckerman, H., et al. (2001). "The responsiveness of the Action Research Arm test and the Fugl-Meyer Assessment scale in chronic stroke patients." J Rehabil Med 33(3): 110-113.

Van der Lee, J. H., De Groot, V., et al. (2001). "The intra- and interrater reliability of the action research arm test: a practical test of upper extremity function in patients with stroke." Arch Phys Med Rehabil 82(1): 14-19.

van der Lee, J. H., Roorda, L. D., et al. (2002). "Improving the Action Research Arm test: a unidimensional hierarchical scale." Clin Rehabil 16(6): 646-653.

Velstra, I., Ballert, C.S., Cieza, A. (2011). “A systematic literature review of outcome measures for upper extremity function using the international classification of functioning, disability, and health as reference.” American Academy of Physical Medicine and Rehabilitation 3: 846-860.

Wei, X., Tong, K., Hu, X. (2011). “The responsiveness and correlation between Fugl-Meyer Assessment, Motor Status Scale, and the Action Research Arm Test in chronic stroke with upper-extremity rehabilitation robotic training.” International J of Rehabilitation Research 34 (4): 349-356.

Yozbatiran, N., Der-Yeghiaian, L., et al. (2008). "A standardized approach to performing the action research arm test." Neurorehabil Neural Repair 22(1): 78-90.

Zhao, J. L., Chen, P. M., Li, W. F., Bian, R. H., Ding, M. H., Li, H., Lin, Q., Xu, Z. Q., Mao, Y. R., & Huang, D. F. (2019). Translation and Initial Validation of the Chinese Version of the Action Research Arm Test in People with Stroke. BioMed research international, 2019, 5416560.