Balance Exercises for Stroke Patients to Prevent Falls

Balance exercises meaningfully improve stability, walking ability, and fall prevention for people recovering from stroke, and the evidence behind them is extensive. A large scoping review covering 31 studies found that balance training improved balance ability and reduced the incidence of falls in stroke patients across all recovery stages.¹ What makes the topic more interesting than a simple “yes, they work” is that the specific type of exercise, the dose, the timing, and whether technology or cognitive challenges are layered in all change the size of the benefit and who benefits most.

Why Stroke Disrupts Balance in the First Place

Stroke damages brain regions that coordinate movement, sensation, and postural control. The result is usually one-sided weakness, altered muscle tone, and impaired sensory feedback from the affected limbs. But the problem goes deeper than weak muscles. Staying upright depends on continuous back-and-forth communication between the brain’s motor cortex and the body’s postural system. Research using nonlinear dynamics analysis has shown that stroke alters the temporal structure of postural sway, meaning the brain-body coupling that keeps you balanced becomes disorganized in ways that differ from person to person.² Two people can look equally unsteady on a clinical balance test but have very different underlying control problems. That variability helps explain why no single exercise program works best for everyone and why rehabilitation often needs to be tailored.

The good news is that the brain rewires itself. Neuroscientific research has demonstrated that neuroplastic changes in the cortex and the broader central nervous system are directly linked to motor skill retraining, and that task-oriented practice is the key driver of that neural reorganization.³ In practical terms, repeatedly practicing a challenging balance task signals the brain to build new pathways around the damaged area. This is the neurological reason balance exercises work at all, and why passive treatments or bed rest cannot substitute.

Core Stability Training

The trunk is the platform everything else builds on. If your core cannot hold your body steady, your legs and arms have a shaky foundation for every other movement. A systematic review found moderate evidence that trunk training exercises improve both trunk performance and dynamic sitting balance in stroke patients.⁴ A separate meta-analysis of core stability exercises added on top of standard physiotherapy confirmed moderate-quality evidence for improved trunk control, dynamic balance, and walking speed.⁵

What counts as core stability work in this context is not the gym-style “six-pack ab routine.” It involves exercises that challenge the deep trunk muscles to activate in anticipation of movement: seated reaching tasks, controlled weight shifts, bridging, and exercises that require resisting perturbation while sitting or standing. One randomized trial compared a developmental kinesiology-based core protocol against conventional core exercises in people with hemiparetic stroke. Both groups improved their Berg Balance Scale and Trunk Impairment Scale scores, but only the developmental kinesiology group maintained their gains in fall-related self-efficacy through a two-year follow-up.⁶ That long-term maintenance is worth noting because many balance interventions show short-term improvements that fade once training stops.

Strength Training and Its Limits for Balance

Weakness on the affected side is one of the most obvious problems after stroke, so strength training seems like an intuitive fix. And it does help, up to a point. One study of progressive resistance training in people at least a year post-stroke found that lower-limb strength improved roughly 68% on the affected side, and Berg Balance Scale scores improved about 12%.⁷ Eccentric-focused resistance training has also shown benefits for functional mobility specifically, with one trial finding greater improvements in the Timed Up and Go test for the eccentric group compared to conventional training.⁸

However, a systematic review and meta-analysis of resistance training in stroke rehabilitation found a notable caveat: while resistance training was beneficial for muscle force, motor function, and gait parameters, it was not significantly different from other therapies on balance and postural control specifically.⁹ Strength gains alone do not automatically translate into better balance. If your muscles are stronger but you have not practiced the specific postural reactions that prevent a fall, you are stronger but not necessarily steadier. This is why rehabilitation programs that combine strengthening with balance-specific tasks tend to outperform pure strength programs for stability outcomes.

Task-Specific Balance Training

The principle is straightforward: if you want to get better at balancing, practice balancing. Task-specific training structures exercises around the exact activities a person struggles with, such as standing up from a chair, reaching for objects while standing, stepping over obstacles, or turning around. A randomized trial in subacute stroke patients compared task-specific trunk and balance training against conventional physiotherapy and found that the task-specific group showed greater improvements in trunk control, postural alignment, and functional reach.¹⁰ Patient-centered task-oriented training has also been shown to improve balance, daily activity performance, and self-efficacy, which refers to how confident someone feels about not falling.¹¹

Confidence matters more than you might expect. Research has shown that depressive symptoms moderate the relationship between physical capacity, balance self-efficacy, and real-world participation. Among people with low depression, balance self-efficacy effectively bridged the gap between physical ability and social participation. But in those with high depression, that link broke down, meaning improvements in physical balance did not translate into actually doing more in daily life.¹² This finding underscores that addressing psychological health alongside physical training can determine whether balance gains actually change someone’s life.

Virtual Reality and Robot-Assisted Training

Virtual reality (VR) has become one of the more studied technology-assisted approaches to balance rehabilitation. A systematic review and meta-analysis found that stroke patients receiving VR-based exercise showed marked improvements across several standard balance measures compared to controls, including the Berg Balance Scale, the Timed Up and Go test, and the Functional Reach Test.¹³ A separate meta-analysis confirmed that VR training was more effective for gait and balance recovery than conventional therapy alone, with the largest benefits seen when VR was added on top of standard rehab and when the total practice time was matched between groups.¹⁴

The appeal of VR is partly motivational. Balance exercises can be repetitive and tedious, which hurts adherence. Gaming elements, visual feedback, and immersive environments can keep people engaged through more repetitions than they would tolerate in a traditional gym setting. One trial comparing immersive VR with traditional physiotherapy in chronic stroke patients found the VR group demonstrated superior effects on the Timed Up and Go test, walking speed, stride length, and balance confidence.¹⁵

Robot-assisted gait training is a separate but related technology. A meta-analysis of robotic devices for balance found that exoskeleton-based robot-assisted therapy produced significantly better balance recovery than conventional therapy alone.¹⁶ Wearable robot-assisted gait training in subacute stroke patients showed significant improvements in motor ability, balance symmetry, and walking parameters after eight weeks, with gains maintained at a 16-week follow-up.¹⁷ These tools are most useful for patients who are too impaired to do high-repetition balance practice on their own, since the device bears some of the body weight and guides leg movements.

Dual-Task Training and Perturbation-Based Approaches

Falls in the real world rarely happen during a quiet, focused balance exercise. They happen when you are walking and talking, carrying a cup of coffee, or scanning a busy sidewalk. Dual-task training addresses this by having patients practice balance or walking while simultaneously performing a cognitive task such as counting backward or naming words. A randomized trial found that a dual-task exercise program reduced the risk of falls by 25% and injurious falls by about 22% over a six-month follow-up, with improvements in dual-task walking performance largely maintained eight weeks after training ended.¹⁸ A more recent trial compared cognitive-motor multi-task training against cognitive dual-task training and found that only the multi-task group showed improvements in dual-task Timed Up and Go scores, gait assessment, and balance confidence at both post-treatment and follow-up.¹⁹

Perturbation-based balance training takes a different but complementary approach: instead of having you practice voluntary movements, the therapist or a motorized platform delivers unexpected pushes, pulls, or surface shifts so you practice catching yourself. This trains reactive balance, the quick stepping and weight-shifting responses that actually prevent a fall when you trip or get bumped. One trial in chronic stroke patients showed that those who received perturbation training took fewer extra steps to recover from a push compared to the control group.²⁰ Another proof-of-concept study found that participants took larger, more effective recovery steps after training, and these improvements were retained six weeks later.²¹ A more recent study found that perturbation training significantly improved balance and walking efficiency, especially in patients with severe balance impairments, outperforming traditional rehabilitation methods.²²

Tai Chi as a Balance Intervention

Tai chi stands out among mind-body practices because its slow, controlled weight shifts, single-leg stances, and turning movements directly challenge the balance system. A meta-analysis of tai chi combined with conventional rehabilitation found substantially greater improvements across multiple balance measures compared to conventional rehabilitation alone, including a mean difference of about 6 points on the Berg Balance Scale and roughly 2 seconds faster on the Timed Up and Go test.²³ A 6-point improvement on the Berg Balance Scale is clinically meaningful because it crosses the threshold that typically indicates a real change in fall risk. Tai chi also improved self-reported confidence in not falling, which, as discussed earlier, matters for whether balance gains carry over into daily activities.

One practical advantage of tai chi is that it requires no equipment and can be adapted for people who need a chair or a wall for support. Group classes also offer a social component that can help address the isolation many stroke survivors experience. The slow pace makes it accessible across a wide range of impairment levels, from people who can stand independently to those who practice seated versions.

How Much Exercise Is Enough

The dose question turns out to have a surprisingly specific answer. A large meta-analysis of randomized controlled trials recommended that stroke patients aim for a minimum of eight weeks of exercise, at least three times per week, for more than 60 minutes per session, with a target of at least 180 minutes per week. Programs meeting those parameters were consistently more effective at improving Berg Balance Scale scores than shorter or less frequent regimens.²⁴

A Bayesian network meta-analysis added nuance by examining total exercise dose in metabolic equivalent minutes per week. Balance and functional movement improvements peaked at roughly 1,800 to 2,200 MET-minutes per week, after which the benefits plateaued or declined with greater variability between individuals.²⁵ That same analysis found that different exercise types hit their peak benefit at different doses: aerobic-integrated training reached peak effects at moderate loads, making it good for early rehabilitation, while motor control training maintained steady effects across a broader range, supporting longer-term programs. VR training showed localized peaks at low-to-moderate doses but increased uncertainty at higher levels.²⁶ The practical takeaway is that more is not always better, and that the ideal weekly volume depends somewhat on the type of training being done.

Do These Exercises Actually Prevent Falls

This is the question that matters most. Improving a score on a balance test is encouraging, but the real endpoint is fewer falls and fewer injuries. The evidence here has evolved substantially. An older systematic review from 2010 concluded that no exercise intervention had been shown to reduce falls after stroke, with the only effective intervention being vitamin D supplementation for institutionalized female stroke survivors.²⁷ That finding was discouraging and often cited in the literature.

More recent evidence paints a different picture. A 2025 systematic review and meta-analysis found that physical exercise reduced the number of falls by about 39% among people with stroke. The programs that were most effective were those focused specifically on balance training and those integrating physical exercise with cognitive components, such as dual-task or mind-body exercise. Walking training alone, combined balance-and-walking training, and pure strength training were less effective at preventing falls.²⁸ The shift from the 2010 finding likely reflects both better-designed exercise programs in recent trials and the accumulation of more data, not a change in how falls work.

Home-Based Programs

Access to outpatient rehabilitation is a real barrier for many stroke survivors. Transportation, fatigue, cost, and geography all get in the way. The evidence on home-based balance programs is reassuring: a systematic review with meta-analysis found high-quality evidence that home-based exercises produce similar effects on walking speed and balance compared to equivalent doses of center-based exercises.²⁹ A randomized clinical trial of telerehabilitation versus in-clinic therapy found that the telerehabilitation group adhered to about 98% of their sessions, compared to about 93% for the in-clinic group, and functional improvements were statistically equivalent. The trial formally demonstrated that telerehabilitation was not inferior to in-clinic therapy.³⁰

One home-based rehabilitation trial did find mixed results when looking at specific tests: walking on a figure-eight course improved significantly in the home program group, but comfortable walking speed, fast walking speed, and quality-of-life scores did not differ significantly from controls.³¹ The figure-eight walking test involves turns and changing direction, which challenges balance more than walking in a straight line, so the home program may have been better at training real-world balance demands. Home programs appear to work best when they include structured progressions and some form of therapist check-in, whether by phone, video, or periodic in-person visits.

Adjuncts That May Boost the Effect

Non-invasive brain stimulation techniques, such as transcranial direct current stimulation or repetitive transcranial magnetic stimulation, are being studied as add-ons to physical exercise. A review of 25 studies found that these approaches are effective in supporting recovery of gait, balance, and lower-limb motor function after stroke.³² The idea is that stimulating the motor cortex while the patient practices balance tasks amplifies the neuroplastic response. These techniques are not stand-alone treatments and are not yet standard in most rehabilitation settings, but they represent a growing area where technology may eventually be layered onto conventional balance training.

Vestibular rehabilitation, which targets the inner-ear balance system and its connections to the brain, is another adjunct gaining attention. Dizziness is a common but often overlooked complaint after stroke, and evidence is growing that vestibular rehabilitation can improve balance and walking performance in stroke patients who experience it. An overview of the current evidence base describes consistent positive effects on ambulatory performance when vestibular-specific exercises are added to general rehabilitation.³³

How Balance Is Measured and Why It Matters to You

If you or a family member is going through stroke rehabilitation, you will almost certainly encounter the Berg Balance Scale. It is a 14-item test where a therapist watches you do things like stand on one foot, turn around, and pick an object up off the floor, scoring each task on a 0-to-4 scale. A score below about 45 out of 56 is generally considered to indicate elevated fall risk. Force plate measurements offer a more detailed picture by tracking how ground reaction forces shift under your feet while standing, revealing weight-bearing asymmetry and the coordination between your two limbs.³⁴ Force plate assessments can identify specific control problems, such as excessive reliance on the unaffected leg or poor coordination between sides, that standard clinical tests may miss.³⁵

Understanding these measurements can be useful because they give you a way to track progress objectively. A two-point change on the Berg Balance Scale might not sound like much, but in the context of a 56-point scale where every point represents a specific physical task you can now do, it often maps onto a meaningful gain in daily function. Asking your therapist what your current scores are, and what a realistic target looks like for your situation, turns abstract rehabilitation goals into something concrete you can work toward.