Condition guide · Neurological rehabilitation
Parkinson's Disease Physiotherapy
What physiotherapy actually changes in Parkinson's, measured against doing nothing — and the honest limits of what a review of 39 small trials can and cannot tell you.
In plain words. Parkinson's disease affects the brain's control of movement, producing some combination of slowness, stiffness, tremor and impaired balance. It is progressive, and there is no cure. What physiotherapy offers is not a treatment for the underlying disease process, but a way of maintaining and improving the movement, balance and function that the disease affects.
The clearest way to know whether that actually works is to compare it with doing nothing at all. A Cochrane review did exactly that, pooling 39 trials.
Key points
- Physiotherapy improved walking speed compared with no intervention (0.04 m/s, 95% CI 0.02 to 0.06). [1]
- It improved balance — Berg Balance Scale 3.71 points better (95% CI 2.30 to 5.11) — and functional mobility, measured by Timed Up & Go (0.63 seconds faster). [1]
- Clinician-rated disability improved on the UPDRS motor score (5.01 points better, 95% CI 3.72 to 6.30). [1]
- No difference was found in falls or patient-rated quality of life. [1]
- For several outcomes — gait speed, balance, disability — the improvements were at or near what is considered a minimally important change, not merely statistically detectable. [1]
- No physiotherapy technique was shown better than another. A separate Cochrane comparison of 43 trials found insufficient evidence to support any one approach over the rest. [2]
What is actually happening
Parkinson's disease results from progressive loss of dopamine-producing cells in a part of the brain that helps initiate and regulate movement. The consequence is not weakness in the ordinary sense — the muscles work — but a disruption in how movement is planned, scaled and sequenced. That shows up as smaller, slower steps, reduced arm swing, difficulty initiating movement, and stiffness.
Balance is affected because the automatic postural adjustments that keep us upright rely on the same systems. That is why falls risk is a central concern in Parkinson's rehabilitation, even though the Cochrane review found no reduction in falls specifically. [1] Physiotherapy works around the disrupted automatic system by using external cues, conscious attention to movement, and repeated practice — strategies that can partially bypass what the disease has affected.
What it feels like
- Slower, smaller steps, and a feeling that walking takes more conscious effort than it used to.
- Stiffness and reduced arm swing on one or both sides.
- Tremor at rest, often starting in one hand.
- Difficulty starting to walk, or "freezing" — a brief inability to move the feet, often in doorways or turning.
- Balance that feels less secure, particularly turning or on uneven ground.
- Symptoms that can fluctuate through the day, particularly in relation to medication timing.
How it is diagnosed
Parkinson's disease is diagnosed clinically by a neurologist, based on the pattern of symptoms and examination findings, sometimes supported by imaging to exclude other causes. Physiotherapy assessment follows that diagnosis and focuses on a different question: which specific movements, balance reactions and functional tasks are affected, and by how much — using standardised measures such as the Berg Balance Scale, Timed Up & Go and gait speed, the same measures used in the trials cited here.
How physiotherapy and rehabilitation help
The Cochrane review's own framing is useful: benefit was found in most outcomes over the short term (under three months), and for several of them — speed, balance, disability — the size of the change approached what is considered clinically meaningful, not just statistically present. [1] That is a more precise claim than "physiotherapy helps", and it is the one this page makes.
What did not improve was falls frequency and patient-rated quality of life. [1] That does not mean physiotherapy is pointless for fall prevention — falls are notoriously hard to shift in any single trial because they are relatively rare events — but it means the evidence here should not be oversold on that specific point.
What a programme involves
| Component | What it is for | Evidence |
|---|---|---|
| Gait training, often with external cues visual lines, rhythmic sound, verbal counting |
Using cues to bypass the disrupted automatic step-generation system and improve walking speed and step size. | Supported Gait speed improved 0.04 m/s (95% CI 0.02 to 0.06) vs no intervention [1] |
| Balance training | Directly practising the reactions and postures that keep you upright. | Supported Berg Balance Scale 3.71 points better (95% CI 2.30 to 5.11) [1] |
| Functional mobility practice sit-to-stand, turning, transfers |
Practising the specific tasks that make up daily movement. | Supported Timed Up & Go 0.63 seconds faster (95% CI 0.21 to 1.05) [1] |
| Strength and general exercise | Maintaining the physical capacity that movement strategies are built on. | Contributed to the pooled result Bundled within "physiotherapy" in the review, not isolated [1] |
| A specific technique or school of approach e.g. one named method over another |
Various physiotherapy schools claim particular advantages. | No technique shown superior Insufficient evidence to support one approach over another, across 43 trials [2] |
| Falls prevention specifically | Reducing the frequency of falls. | Not shown in this review No significant difference in Falls Efficacy Scale [1] |
What a course of treatment looks like
Assessment against standardised measures — gait speed, Timed Up & Go, balance scales — both to guide the programme and to track change over time. Then a programme combining gait training, balance work and functional practice, individualised to the specific difficulties found.
Because Parkinson's is progressive and the evidence shows benefit particularly in the short term, ongoing or periodically repeated physiotherapy input, rather than a single course, fits the pattern of the disease better than a one-off block of treatment.
Assessment drives the programme. The Cochrane network meta-analysis found most types of exercise beneficial compared with no exercise, but could not establish that any one type is clearly best. [3]
What the evidence supports — and what it does not
Supported
- Physiotherapy over no intervention for gait speed, walk test distance, freezing of gait, functional mobility, balance and clinician-rated disability. [1]
- Benefit sizes approaching clinical importance for several key outcomes, not merely statistical significance. [1]
- A broad range of techniques — the review found no evidence that one physiotherapy approach outperforms another, which supports individualising the method to the person rather than seeking a single "correct" school. [2]
Not supported
- Claiming physiotherapy reduces falls — no significant difference was found on the Falls Efficacy Scale. [1]
- Claiming an improvement in patient-rated quality of life — the PDQ-39 summary index showed no significant difference. [1]
- Claiming one named technique or brand of therapy is superior — the comparative review found insufficient evidence to distinguish between approaches. [2]
- Presenting physiotherapy as slowing disease progression — these trials measured function and movement, not the underlying disease process.
What the most recent evidence adds
There is a profession-specific guideline. The American Physical Therapy Association published a clinical practice guideline on physical therapist management of Parkinson disease in 2022, developed by physiotherapists and a neurologist from systematic reviews of the evidence. [4] A Spanish version is published alongside it as a supplement, which matters for reach.
Most exercise works; no single type is established as best. The Cochrane network meta-analysis was updated in 2024 and now includes 154 randomised trials with 7837 participants, mostly people with mild to moderate disease and no major cognitive impairment. It found evidence of benefit for most types of physical exercise compared with a passive control group, on both motor sign severity and quality of life. [3] The honest caveat is in the study sizes: a mean of 51 participants per trial, ranging from 10 to 474. A ranking built from many small trials is a weak basis for telling you that dance beats treadmill or the reverse.
The practical consequence is liberating rather than disappointing. If most types work and none is proven superior, the right exercise is largely the one you will keep doing — which makes preference, access and enjoyment legitimate clinical considerations rather than soft ones.
Continuing matters more than the format. A meta-analysis of 10 studies and 663 people examined physiotherapy sustained over the longer term and found favourable effects on motor symptoms in the off-medication state (SMD -0.65, 95% CI -1.04 to -0.26) and on levodopa equivalent dose (-0.49, -0.89 to -0.09). Aerobic exercise specifically improved off-state motor symptoms (-0.42, -0.64 to -0.20). [5] Quality of evidence was low for aerobic exercise and very low for multidisciplinary rehabilitation, so hold the size of those numbers loosely — but the direction supports treating physiotherapy as ongoing management rather than a block of sessions that finishes.
Balance and falls are a large part of this. If dizziness or a spinning sensation is part of the picture rather than unsteadiness alone, that may be a separate and very treatable problem — see BPPV.
Falls are among the most serious consequences of Parkinson's. The general falls evidence, including what happens when a programme stops, is covered in falls prevention in older adults.
How certain is this?
Evidence grade: Low to moderate. The review pooled 39 trials and 1,827 participants, which is a substantial evidence base for a neurological rehabilitation question. But the trials were at a mixed risk of bias, with unreported allocation concealment and probable detection bias in many — a real concern when outcomes such as balance scores can be influenced by an unblinded assessor's expectations. [1]
The comparative review of different techniques is, if anything, more limited: with 43 trials averaging only 39 participants each, poor reporting of randomisation and concealment, and such varied interventions that no quantitative meta-analysis could even be performed, the honest conclusion is that we do not yet know which specific approach works best — not that they are all equal. [2]
What to expect
The evidence describes short-term benefit, typically measured within trials of under three months. [1] Parkinson's is a progressive, long-term condition, so the realistic expectation is periodic physiotherapy input over years, adjusted as the disease and its impact on movement change, rather than a single course that resolves things permanently.
Response varies between individuals, and symptoms can fluctuate through the day in relation to medication. A good rehabilitation programme is coordinated with the medical management of the condition, not separate from it.
Seek prompt medical review for
Contact the treating medical team promptly if there is:
- A fall resulting in injury, or a marked increase in fall frequency.
- Sudden worsening of movement, confusion, or a new symptom — this can indicate an infection or a medication issue and needs medical review, not only physiotherapy.
- Difficulty swallowing, choking on food or drink, or unexplained weight loss.
- Significant new low mood, hallucinations or confusion — these can occur in Parkinson's and are medically important, not something to manage alone.
- Chest infection symptoms — people with Parkinson's are at higher risk of aspiration.
This list is not exhaustive. Physiotherapy works alongside medical management of Parkinson's, not instead of it.
Common questions
Is physiotherapy actually worth it for Parkinson's?
Compared with no intervention, yes on the outcomes measured: gait speed, walk distance, freezing of gait, functional mobility, balance, and clinician-rated disability all improved, with several changes approaching what is considered clinically important. [1] It did not improve falls frequency or patient-rated quality of life in this review, which is worth knowing honestly rather than glossing over.
Which type of physiotherapy is best — is there a special method?
No named technique has been shown superior to the others. A review comparing 43 trials of different physiotherapy approaches found insufficient evidence to recommend one over another. [2] Be cautious of any programme marketed as uniquely effective on that basis — individualise the approach to the person instead.
Will physiotherapy prevent falls?
This review did not find a significant reduction in the Falls Efficacy Scale. [1] Balance and functional mobility did improve, which is plausibly relevant to falls risk, but the direct falls outcome was not significantly different, and that should not be overstated.
Does physiotherapy slow down the disease?
These trials measured movement, balance and function — not the underlying loss of dopamine-producing cells. There is no claim here, and no evidence cited here, that physiotherapy alters disease progression.
How often should physiotherapy happen?
The trials mostly measured benefit over the short term, under three months. [1] Given Parkinson's is progressive, ongoing or periodically repeated input — reviewed and adjusted as the condition changes — fits the evidence better than a single block of treatment.
Interpreting the Berg Balance Scale figures on this page
Berg Balance Scale · Outcome measure
A 14-item observed balance scale scored 0 to 56. Rater reliability is excellent; the change you can defend in one patient is much larger than most people assume.
Full detail, psychometrics and 7 verified sources →
Not validated below a score of 20, and not a falls screen on its own.
References
- Tomlinson CL, Patel S, Meek C, et al. Physiotherapy versus placebo or no intervention in Parkinson's disease. Cochrane Database of Systematic Reviews. 2013 Sep 10;2013(9):CD002817. doi:10.1002/14651858.CD002817.pub4 PMID 24018704 Systematic review
- Tomlinson CL, Herd CP, Clarke CE, et al. Physiotherapy for Parkinson's disease: a comparison of techniques. Cochrane Database of Systematic Reviews. 2014 Jun 17;2014(6):CD002815. doi:10.1002/14651858.CD002815.pub2 PMID 24936965 Systematic review
- Ernst M, Folkerts AK, Gollan R, et al. Physical exercise for people with Parkinson's disease: a systematic review and network meta-analysis. Cochrane Database of Systematic Reviews. 2024 Apr 8;4(4):CD013856. doi:10.1002/14651858.CD013856.pub3 PMID 38588457 Systematic review and network meta-analysis
- Osborne JA, Botkin R, Colon-Semenza C, et al. Physical Therapist Management of Parkinson Disease: A Clinical Practice Guideline From the American Physical Therapy Association. Physical Therapy. 2022 Apr 1;102(4):. doi:10.1093/ptj/pzab302 PMID 34963139 Clinical practice guideline
- Okada Y, Ohtsuka H, Kamata N, et al. Effectiveness of Long-Term Physiotherapy in Parkinson's Disease: A Systematic Review and Meta-Analysis. Journal of Parkinson's Disease. 2021;11(4):1619–1630. doi:10.3233/JPD-212782 PMID 34366377 Systematic review and meta-analysis
About this guide
If you need assessment
This page explains. It does not diagnose.
Parkinson's disease is diagnosed and medically managed by a neurologist. Physiotherapy assessment identifies which specific movements, balance reactions and functional tasks are affected and builds a programme around them. Physiotherapist India publishes information and takes no bookings. Ask a qualified physiotherapist to examine your own case before acting on anything written here.