Written and maintained by CASRAI Editorial Board
Last updated
The Fugl-Meyer Assessment of Sensorimotor Recovery After Stroke (FMA)—developed by Axel R. Fugl-Meyer, Lisbeth Jääskö, Inger Leyman, Sigyn Olsson, and Solveig Steglind at the University of Gothenburg in 1975 (Scandinavian Journal of Rehabilitation Medicine, 7[1]:13–31)—is the gold-standard performance-based impairment index for evaluating post-stroke hemiplegic motor recovery. Grounded in Twitchell and Brunnstrom’s sequential stages of motor recovery (from flaccidity through spasticity and obligatory limb synergies to isolated out-of-synergy voluntary movement), the 66-point Upper Extremity subscale (FMA-UE) and 34-point Lower Extremity subscale (FMA-LE) are designated by the international Stroke Recovery and Rehabilitation Roundtable (SRRR) (Kwakkel et al., 2017) as core body-function outcome measures alongside acute stroke severity grading on the NIH Stroke Scale (NIHSS) and postural control testing on the Berg Balance Scale.
How Is the 226-Point Fugl-Meyer Assessment of Sensorimotor Recovery After Stroke (FMA-UE and FMA-LE) Scored?
Every item across the full 226-point Fugl-Meyer Assessment is scored by direct physical or occupational therapist observation on a strict 3-point ordinal scale:
0 = Cannot perform(voluntary movement, reflex, or sensation is absent)1 = Performs partially(movement is incomplete, slower, or deviates into abnormal flexor/extensor synergy)2 = Performs fully(faultless execution equivalent to the unaffected contralateral side)
The full 226-point instrument is divided into five independent clinical domains, though clinical trials most frequently administer the 100-point Motor Domain (FMA-UE + FMA-LE) as a standalone primary or secondary endpoint:
| FMA Domain / Subscale | Number of Items | Max Score | Subcomponent Breakdown |
|---|---|---|---|
| 1. Upper Extremity Motor (FMA-UE) | 33 items | 66 points |
• Shoulder, Elbow & Forearm (18 items, 0–36 pts): Reflex activity (4 pts), flexor synergy (12 pts), extensor synergy (6 pts), movement combining synergies (6 pts), movement out of synergy (6 pts), normal reflex activity (2 pts) • Wrist (5 items, 0–10 pts): Stability at 15° dorsiflexion, repeated flexion/extension, circumduction • Hand (7 items, 0–14 pts): Mass flexion, mass extension, hook grasp, thumb adduction, pincer grasp, cylinder grasp, spherical grasp • Coordination / Speed (3 items, 0–6 pts): Finger-to-nose test (5 repetitions, timed vs. unaffected side): tremor (2 pts), dysmetria (2 pts), speed (2 pts) |
| 2. Lower Extremity Motor (FMA-LE) | 17 items | 34 points |
• Hip, Knee & Ankle (14 items, 0–28 pts): Reflex activity (4 pts), flexor synergy in supine (6 pts), extensor synergy in supine (8 pts), movement combining synergies in sitting (4 pts), movement out of synergy in standing (4 pts), normal reflex activity (2 pts) • Coordination / Speed (3 items, 0–6 pts): Heel-to-shin test (5 repetitions): tremor (2 pts), dysmetria (2 pts), speed (2 pts) |
| 3. Balance | 7 items | 14 points | 3 sitting balance items (0–6 pts) and 4 standing balance items (0–8 pts) |
| 4. Sensation | 12 items | 24 points | Light touch (4 items, 0–8 pts) and joint position / proprioception (8 items, 0–16 pts) across upper and lower limbs |
| 5. Passive Joint Motion & Joint Pain | 22 + 22 items | 44 + 44 points | Passive range of motion (0–44 pts) and absence of joint pain during passive movement (0–44 pts) |
For upper-limb stratification in stroke trials (Woytowicz et al., 2017; Woodbury et al., 2013), baseline FMA-UE (0–66) severity is commonly categorized as Severe (0–28), Moderate (29–42), and Mild (43–66).
How Does the 19-Item, 57-Point Lyle Action Research Arm Test (ARAT) Measure Activity-Level Upper-Limb Capacity Compared with Body-Function FMA-UE?
Under the World Health Organization (WHO) International Classification of Functioning, Disability and Health (ICF) framework and the Stroke Recovery and Rehabilitation Roundtable (SRRR) consensus (Kwakkel et al., 2017), the FMA-UE and the Action Research Arm Test (ARAT) measure two fundamentally different levels of post-stroke recovery:
- FMA-UE (ICF Body Function / Motor Impairment Level): Evaluates whether the patient can execute isolated joint movements and escape pathological flexor/extensor synergies without carrying or manipulating real-world objects (apart from basic grasp test objects). Distinguishes true neurological restitution (restoring pre-stroke movement patterns) from behavioral compensation.
- Action Research Arm Test (ARAT; Lyle, 1981; Yozbatiran et al., 2008 standardized protocol — ICF Activity Capacity Level): Evaluates whether the patient can grasp, transport, and release standardized physical objects onto a 37 cm high shelf—allowing some trunk or proximal shoulder compensation at sub-maximal score levels. The 19-item ARAT (total score range 0–57) rates each item on a 4-point ordinal scale (0–3):
0 = Cannot perform any part of the test1 = Performs test partially2 = Completes test, but takes abnormally long (5.0–60.0 seconds) or has great difficulty/compensatory trunk motion3 = Performs test normally in <5.0 seconds with normal posture and hand control
The 19 ARAT items are organized into four hierarchical Guttman-scaled subscales (meaning if a patient scores
3on the first, hardest item of a subscale, they are credited3on all easier items in that subscale; if they score0on both the first and second/easiest item, they score0for the remaining items in that subscale):- Grasp Subscale (6 items, 0–18 points): Lifting wooden blocks (10 cm, 2.5 cm, 5 cm, 7.5 cm cubes), a 7.5 cm cricket ball, and a sharpening stone.
- Grip Subscale (4 items, 0–12 points): Pouring water from one glass to another, displacing 2.25 cm and 1 cm alloy tubes, and placing a washer over a bolt.
- Pinch Subscale (6 items, 0–18 points): Picking up a 6 mm ball bearing and a 1.5 cm marble using thumb-ring finger, thumb-middle finger, and thumb-index finger pinch grips.
- Gross Movement Subscale (3 items, 0–9 points): Placing the hand behind the head, on top of the head, and to the mouth.
What Are the Stroke Rehabilitation MCID Benchmarks for FMA-UE, ARAT, Box and Block Test (BBT), and 9-Hole Peg Test (9HPT)?
In neurorehabilitation clinical trials—where performance outcomes (PerfO) are captured alongside electronic clinical outcome assessment (eCOA) instruments and neurological baseline scales such as the Glasgow Coma Scale—investigators pair FMA-UE and ARAT with timed manual dexterity tests: the Box and Block Test (BBT) (Mathiowetz et al., 1985; counting how many 2.5 cm wooden cubes a patient transfers across a partition in 60 seconds) and the 9-Hole Peg Test (9HPT) (Kellor et al., 1971; Mathiowetz et al., 1985; timing in seconds how fast a patient places and removes 9 pegs one at a time, suited for mild hemiparesis with preserved fine pincer dexterity):
| Instrument | WHO ICF Domain | Score / Unit | Minimal Detectable Change (MDC₉₅) | Minimal Clinically Important Difference (MCID) |
|---|---|---|---|---|
| FMA-UE (Fugl-Meyer et al., 1975) | Body Function (Impairment) | 0–66 points (33 items × 0–2) |
3.2–5.2 points |
Chronic stroke: 4.25–7.25 points (Page et al., 2012)Subacute stroke: 9–10 points (Arya et al., 2011) |
| FMA-LE (Fugl-Meyer et al., 1975) | Body Function (Impairment) | 0–34 points (17 items × 0–2) |
3.57–4.0 points |
Chronic stroke: 6.0 points (Pandian et al., 2016) |
| ARAT (Lyle, 1981) | Activity Capacity (Upper Limb) | 0–57 points (19 items × 0–3) |
3.0–3.5 points |
Chronic stroke: 5.7 points (~10% of scale)Acute/subacute stroke: 12 points dominant / 17 points non-dominant (Lang et al., 2008) |
| Box and Block Test (BBT) (Mathiowetz et al., 1985) | Activity Capacity (Gross Manual Dexterity) | Blocks transferred in 60 seconds |
5.5 blocks/min (Chen et al., 2009) |
5–6 blocks per 60 seconds in post-stroke hemiparesis |
| 9-Hole Peg Test (9HPT) (Kellor et al., 1971) | Activity Capacity (Fine Finger Dexterity) | Completion time in seconds (or pegs/sec) |
32.8 seconds in stroke (~20% change in MS) |
Floor effect in moderate-severe hemiparesis; used when ARAT ≥ 40 or in multiple sclerosis (≥20% time change) |
Frequently Asked Questions
How is the 226-point Fugl-Meyer Assessment of Sensorimotor Recovery After Stroke (FMA: 66-point Upper Extremity FMA-UE and 34-point Lower Extremity FMA-LE) scored?
Every item on the FMA is scored 0 (cannot perform), 1 (performs partially), or 2 (performs fully). The 100-point motor scale combines the 33-item Upper Extremity scale (FMA-UE, 0–66 points across shoulder/elbow/forearm, wrist, hand, and finger-to-nose coordination) and the 17-item Lower Extremity scale (FMA-LE, 0–34 points), while Balance (14 pts), Sensation (24 pts), Passive Joint Motion (44 pts), and Joint Pain (44 pts) bring the total instrument to 226 points.
How does the 19-item, 57-point Lyle Action Research Arm Test (ARAT) measure activity-level upper-limb capacity compared with body-function FMA-UE?
Whereas the FMA-UE measures WHO ICF body-function motor impairment (ability to move joints out of abnormal flexion/extension synergies), the 19-item ARAT (0–57 points, scored 0–3 per item across Grasp, Grip, Pinch, and Gross Movement subscales) measures ICF activity-level functional capacity by testing whether the patient can pick up, transport, and release real-world objects of varying sizes and weights.
What are the stroke rehabilitation MCID benchmarks for FMA-UE, ARAT, Box and Block Test (BBT), and 9-Hole Peg Test (9HPT)?
In chronic stroke rehabilitation trials, the Minimal Clinically Important Difference (MCID) is 4.25–7.25 points for the FMA-UE (vs. 9–10 points in subacute stroke), 5.7 points for the ARAT (vs. 12–17 points in acute/subacute stroke), and 5–6 blocks per minute on the 60-second Box and Block Test, while the 9-Hole Peg Test has an MDC₉₅ of 32.8 seconds in mild post-stroke hemiparesis.







