A weak or paralysed arm after stroke (hemiplegia or hemiparesis) can improve with intensive, task-specific practice, especially in the first three months, and several add-on treatments have reasonable evidence: constraint-induced movement therapy, mirror therapy, electrical stimulation and, for some people, robotic arm training and rTMS. None of them replaces hundreds of daily repetitions of reaching, grasping and real tasks, and none can promise full recovery. This guide explains what is realistic and how to make the most of the recovery window.
If the arm suddenly becomes weaker, or new facial droop, slurred speech, loss of balance or vision loss appears at any point, call 108 or 112. These are signs of a new stroke, not a rehabilitation problem.
How Common Is Arm Weakness After Stroke?
Very. In the South London Stroke Register, upper limb weakness was present in 77.4% of 1,259 first-ever strokes at the time of maximum deficit, the single most common impairment recorded (Lawrence et al., 2001).
Recovery is less predictable than for walking. In a cohort of 102 patients whose arm was initially flaccid (completely limp), about 38% regained some useful hand function by six months and 11.6% recovered fully (Kwakkel et al., 2003). In the Copenhagen Stroke Study, 79% of people with mild arm weakness regained full arm function, compared with 18% of those with severe weakness (Nakayama et al., 1994). The severity of the weakness in the first weeks is the strongest predictor, which is why early assessment by a rehabilitation team matters.
Why Is the Arm Harder to Recover Than the Leg?
Three reasons come up again and again in the research.
- Hand control depends on direct brain-to-spinal-cord pathways. Fine finger movements rely heavily on the corticospinal tract; when a stroke damages it, the hand suffers most, while walking is partly driven by more automatic spinal circuits.
- Learned non-use. Early attempts to use the weak arm fail and are frustrating, so the person learns to do everything with the good hand. The weak arm then gets even less practice and the brain devotes less of its map to it. This phenomenon was first described in animal research and later confirmed in people after stroke (Taub et al., 2006).
- Practice doses are often too small. When researchers observed 312 therapy sessions, functional arm practice occurred in only half of the sessions that addressed the arm, and the average was just 32 repetitions per session (Lang et al., 2009). Animal studies of brain recovery use hundreds of repetitions.
Intensity, repetition and real-task practice are therefore the foundation; technology is useful only if it adds to that dose.
What to Expect by Stage
| Stage | What is typical | Focus of therapy |
|---|---|---|
| First days to 2 weeks | Arm often flaccid; shoulder at risk of slipping out of joint (subluxation) and pain | Positioning, supported arm, gentle range of movement, early assessment |
| Weeks 2–4 | Flickers of shoulder and elbow movement may appear; tone may begin to rise | Many repetitions of assisted reach; mirror therapy; NMES if some contraction |
| Months 1–3 | Fastest gains; prognosis for hand function is usually clear by 4 weeks | Task-specific practice, CIMT if wrist and fingers can extend, spasticity control |
| Months 3–6 | Slower gains; some people still improve hand use | Functional tasks at home, bimanual tasks, work and hobby goals |
| Beyond 6 months | Improvement possible with structured, intensive programmes, usually in function rather than raw strength | Intensive blocks, adaptive strategies, prevent contracture and shoulder pain |
In the Copenhagen study, 95% of people reached their best arm function within 6 weeks if the weakness was mild and within 11 weeks if it was severe (Nakayama et al., 1994), but trials such as EXCITE, below, show targeted programmes can still produce change months later.
Which Treatments Have Good Evidence?
Repetitive, task-specific training
The 2026 American Heart Association/American Stroke Association (AHA/ASA) guideline gives its strongest recommendation to repetitive, progressively graded practice of functional tasks (Richards et al., 2026). NICE similarly recommends repetitive task training for arm weakness (NICE NG236, 2023). In plain terms: cups, keys and buttons, hundreds of times.
Constraint-induced movement therapy (CIMT)
CIMT restrains the stronger hand in a mitt for much of the day while the weak arm practises intensively. The landmark EXCITE trial randomised 222 people 3 to 9 months after stroke to two weeks of CIMT or usual care. The CIMT group's time to complete arm tasks fell by 52%, compared with 26% in the control group, and the gains persisted for at least a year (Wolf et al., 2006).
A Cochrane review of 42 trials (1,453 people) found CIMT produced a modest improvement in arm movement but did not convincingly reduce disability, and most trials were small (Corbetta et al., 2015). CIMT suits people who already have some movement: NICE suggests considering it for people with about 20 degrees of wrist extension and 10 degrees of finger extension, and warns about fatigue, low mood and falls during the programme (NICE NG236, 2023). Modified, gentler versions are used in practice.
Mirror therapy
A mirror between the arms lets the person see the strong arm's reflection in place of the weak arm. A Cochrane review of 62 trials (1,982 people) found moderate-quality evidence that mirror therapy improves arm movement and daily activities, and low-quality evidence that it reduces pain, with no adverse effects reported (Thieme et al., 2018). NICE suggests about 30 minutes, at least five times a week for four weeks, started within the first six months.
Neuromuscular electrical stimulation (NMES)
NMES uses surface electrodes to make weak muscles contract. The 2026 AHA/ASA guideline says NMES is reasonable to reduce shoulder subluxation and improve arm positioning early after stroke (Richards et al., 2026). NICE advises against routine use for the hand and arm but suggests a trial when the person has some muscle contraction yet cannot move against resistance, continuing only if function improves (NICE NG236, 2023). A meta-analysis of 29 trials also found NMES modestly reduced spasticity and increased joint range (Stein et al., 2015).
Robotic arm training
Robotic devices can deliver many assisted repetitions. A Cochrane review of 45 trials (1,619 people) found small improvements in daily activities, arm function and strength (Mehrholz et al., 2018). However, the largest trial to date, RATULS, randomised 770 people with moderate to severe arm weakness and found that robot-assisted training did not improve arm function compared with usual care (Rodgers et al., 2019). Guidelines disagree: NICE says do not offer it, while the 2026 AHA/ASA guideline says it is reasonable to consider for moderate to severe weakness in addition to standard treatment. The honest reading: an optional way to add repetitions, not a breakthrough.
rTMS
Repetitive transcranial magnetic stimulation (rTMS) uses magnetic pulses over the scalp. European evidence-based guidelines rate low-frequency rTMS over the unaffected hemisphere as having definite (Level A) efficacy for hand motor recovery in the post-acute stage, and high-frequency rTMS over the affected hemisphere as probably effective (Level B), while cautioning that statistical benefit does not always mean a clinically large one (Lefaucheur et al., 2020). It is given alongside therapy, never instead of it. Read more about rTMS at SENSE Health.
Shoulder Pain and Subluxation: Red Flags for Families
The weak shoulder is vulnerable. In the population-based Lund Stroke Register (416 first-ever strokes), 22% of the 327 patients followed up developed shoulder pain within four months, most of it moderate to severe, and almost a third had shoulder pain by 16 months; impaired arm movement was the main predictor (Lindgren et al., 2007). Subluxation, where the arm bone drops partly out of the shoulder socket because the muscles cannot hold it, is a common contributor (Paci et al., 2005).
The 2026 AHA/ASA guideline recommends teaching patients and carers correct positioning and range-of-movement care, says slings and positioning devices are reasonable, and specifically advises against overhead pulley exercises because they can cause harm (Richards et al., 2026).
Tell the therapist or doctor if you notice:
- A visible gap or step at the top of the shoulder, or the arm hanging lower than the other side
- Pain when the arm is moved, dressed or lifted, or pain that disturbs sleep
- Swelling, colour change or warmth of the hand (possible complex regional pain syndrome)
- Increasing stiffness, a hand that will not open, or fingers curling tightly
Never pull the person up by the weak arm, and do not let it dangle during transfers.
Home Practice: What Families Can Do Every Day
Therapy sessions are the start; home repetitions are where much of the recovery happens.
- Position the arm every time the person sits or lies down. Support it on a pillow or table, palm down, shoulder slightly forward.
- Do the prescribed range-of-movement routine daily, moving each joint gently through its full range to prevent contractures.
- Count repetitions. Aim for the number your therapist sets and increase it weekly: reaching for objects, sliding a cloth, squeezing and releasing a soft ball.
- Use the weak hand in real tasks, even as a helper: holding the plate while eating, steadying a bottle while the strong hand opens it.
- Mirror therapy at home, if your therapist has taught it: about 30 minutes, five days a week.
- Report pain or skin problems early rather than pushing through.
Managing Spasticity in the Arm
Spasticity, the involuntary tightening that pulls the elbow, wrist and fingers into flexion, can block recovery if ignored. Botulinum toxin injections given by a doctor have strong evidence for reducing stiffness and easing care of the hand, though not for improving active use (Andringa et al., 2019). Our article on spasticity after stroke compares all the options.
The Honest Limits of Technology
Robots and stimulation devices can add repetitions, motivation and sometimes a measurable edge. They cannot replace the therapists, the hours of practice or the family's daily involvement. If a centre promises that a machine will restore full hand function, be sceptical, and ask instead how many hours of hands-on arm therapy your family member will receive each day.
Upper Limb Rehabilitation at SENSE Health
At SENSE Health in Kalyan Nagar, arm rehabilitation is led by our consultant physiatrist, Dr. Vaibhav Shivram Lokhande, with physiotherapists and occupational therapists delivering task-specific arm and hand training and NMES; rTMS is offered to selected patients after medical assessment and is charged per session. Inpatients have 24/7 doctors and a 24/7 emergency response team. Before admission you receive a written package sheet showing what the chosen package includes and what is charged separately, explained and signed before you decide. Our first 90 days after stroke checklist and our stroke rehabilitation guide cover the wider picture.
Getting Started in Kalyan Nagar
SENSE Health is at AVR Heights, 100 Feet Road, HRBR Layout 1st Block, Kalyan Nagar, Bengaluru 560043, close to Banaswadi, Kammanahalli and Hennur, and easy to reach from Indiranagar, Hebbal and Horamavu. Bring the discharge summary, scans and medicine list. Expect a free initial consultation with an assessment of shoulder, arm and hand movement, tone, sensation and pain, and a plan for OPD therapy, inpatient rehabilitation or home-based practice with family training.
OPD hours: Monday to Saturday, 9 AM to 7 PM.
Call +91 96633 34659, WhatsApp us, email listen@sensehealth.co.in, or get directions.
Sources
- Lawrence ES, et al. Estimates of the prevalence of acute stroke impairments and disability in a multiethnic population. Stroke. 2001.
- Kwakkel G, et al. Probability of regaining dexterity in the flaccid upper limb: impact of severity of paresis and time since onset in acute stroke. Stroke. 2003.
- Nakayama H, et al. Recovery of upper extremity function in stroke patients: the Copenhagen Stroke Study. Archives of Physical Medicine and Rehabilitation. 1994.
- Taub E, et al. The learned nonuse phenomenon: implications for rehabilitation. Europa Medicophysica. 2006.
- Lang CE, et al. Observation of amounts of movement practice provided during stroke rehabilitation. Archives of Physical Medicine and Rehabilitation. 2009.
- Richards LG, et al. 2026 Guideline for Adult Stroke Rehabilitation and Recovery: a guideline from the American Heart Association and American Stroke Association. Stroke. 2026.
- National Institute for Health and Care Excellence. Stroke rehabilitation in adults (NG236). NICE. 2023.
- Wolf SL, et al. Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: the EXCITE randomized clinical trial. JAMA. 2006.
- Corbetta D, et al. Constraint-induced movement therapy for upper extremities in people with stroke. Cochrane Database of Systematic Reviews. 2015.
- Thieme H, et al. Mirror therapy for improving motor function after stroke. Cochrane Database of Systematic Reviews. 2018.
- Stein C, et al. Effects of electrical stimulation in spastic muscles after stroke: systematic review and meta-analysis of randomized controlled trials. Stroke. 2015.
- Mehrholz J, et al. Electromechanical and robot-assisted arm training for improving activities of daily living, arm function, and arm muscle strength after stroke. Cochrane Database of Systematic Reviews. 2018.
- Rodgers H, et al. Robot assisted training for the upper limb after stroke (RATULS): a multicentre randomised controlled trial. Lancet. 2019.
- Lefaucheur JP, et al. Evidence-based guidelines on the therapeutic use of repetitive transcranial magnetic stimulation (rTMS): an update (2014–2018). Clinical Neurophysiology. 2020.
- Lindgren I, et al. Shoulder pain after stroke: a prospective population-based study. Stroke. 2007.
- Paci M, Nannetti L, Rinaldi LA. Glenohumeral subluxation in hemiplegia: an overview. Journal of Rehabilitation Research and Development. 2005.
- Andringa A, et al. Effectiveness of botulinum toxin treatment for upper limb spasticity poststroke over different ICF domains: a systematic review and meta-analysis. Archives of Physical Medicine and Rehabilitation. 2019.
This article is for general education and does not replace advice from your own doctor. If symptoms are sudden or severe, call 108 or go to the nearest emergency department.
