Parkinson's: what physiotherapy achieves – and what it does not – Knowledge

1. Why this article?

People who receive a diagnosis of Parkinson's usually hear about medication first. That is as it should be: there are effective drugs for Parkinson's, and for many people they are the most important help for years.

But there is a second treatment that belongs just as firmly in the picture and is talked about far less often: movement – guided, regular and targeted. It is not a pleasant extra alongside the tablet. It works on exactly the things the medication is weakest at – on balance, on falls, on freezing while walking, on safety in everyday life.

That is the reason for this article. It answers four questions:

• What is Parkinson's in the first place – and why is not every tremor Parkinson's?
• What does physiotherapy achieve, measured by what the studies show?
• Which kind of movement helps with which problem?
• How much of it is needed, and for how long?

The article is written for people affected and their families, not for professionals. Technical terms do appear, because you will meet them in your medical reports anyway – but each one is explained the first time it comes up. The numbers in square brackets refer to the reference list at the very end.

One note in advance: this article does not replace an examination. Parkinson's runs a different course in every person, and what your neurologist determines for your situation takes precedence.

2. Parkinson's is not always Parkinson's

Let us start with a distinction that is almost always lost in conversation – and that matters more in practice than it sounds.

2.1 «Parkinsonism» is a pattern of complaints, not a disease

Doctors use two words that sound similar and mean different things:

• Parkinsonism (also: parkinsonian syndrome) is a description. It says: this person moves in one particular altered way. It says nothing about the cause.
• Parkinson's disease (technically: idiopathic Parkinson's syndrome, «idiopathic» meaning «with no outwardly identifiable cause») is a specific illness – the most common, but not the only, cause of parkinsonism.

Think of the word «fever». Fever is a state, not a diagnosis. It can come from influenza, from a bladder infection or from something else entirely. What gets treated is not «the fever» but whatever lies behind it.

Parkinsonism is present when movements are slowed and at least one of the following signs is added [6]:

• Slowing (technically: bradykinesia). Movements become slower and smaller. Handwriting shrinks towards the end of the line, steps get shorter, the face moves less, the voice grows quieter. This «getting smaller» is the actual core sign.
• Tremor at rest (resting tremor). The hand shakes when it is doing nothing – resting on the thigh, while watching television. As soon as it reaches for a glass, the shaking usually stops. The much more common essential tremor behaves in exactly the opposite way: it gets worse on reaching and holding.
• Stiffness (rigidity). The muscle offers a constant, doughy resistance, even when someone else moves the limb. It feels different from a tense shoulder: it does not ease off when you relax.
• Unsteady balance (postural instability). The body no longer catches itself reliably when it is pushed off balance. In Parkinson's disease this sign usually appears only later – and if somebody falls early, that argues against classic Parkinson's disease and in favour of something else [6].

These four signs look very similar across very different causes. Same symptoms, many possible causes – which is why doctors look in four drawers.

2.2 Four drawers

First drawer: parkinsonism as a pattern of complaints. This is the umbrella term above. Somebody is slow, shakes, is stiff, has trouble with balance. That is all this term says – but it is the starting point of every assessment.

Second drawer: parkinsonism that responds to levodopa – usually Parkinson's disease. Levodopa is the most effective Parkinson's drug. The body converts it into dopamine, the messenger substance that is lacking in Parkinson's disease [6]. When the complaints improve substantially and lastingly on levodopa – smoother walking, less tremor, more capacity in daily life – that is a strong pointer towards Parkinson's disease. It is the most common cause of parkinsonism; a few other conditions respond partially, but less and less durably.

Third drawer: atypical parkinsonism. Here a different disease of the nervous system lies behind the picture. The best known carry abbreviations you might see in a medical report [6]:

• MSA – multiple system atrophy. Alongside the movement disorder, circulation and bladder are affected early: dizziness on standing up, sudden urinary urgency. Signs of a cerebellar disorder often come with it.
• PSP – progressive supranuclear palsy. Its hallmarks are early falls, usually backwards, and later difficulty moving the eyes downwards on purpose – which makes going down stairs dangerous.
• CBS/CBD – corticobasal syndrome. One side of the body is far more affected; on top of that, a hand «no longer knows how to do it», even though it would be strong enough.
• DLB – dementia with Lewy bodies. Here problems with memory and perception appear early, often with fluctuating alertness and with hallucinations.

What these diseases have in common: additional neurological signs appear, they progress more rapidly, and they respond poorly or only temporarily to levodopa [7].

Fourth drawer: secondary parkinsonism. Here there is an identifiable external cause – and this is the drawer where looking closely pays off most directly [6]:

• Medication that blocks the action of dopamine in the brain. This includes many drugs against psychosis and against nausea. This form is the one most often missed. It typically begins days to weeks after the drug is started – roughly nine out of ten cases within three months – and it often affects both sides of the body equally, whereas Parkinson's disease usually starts on one side.
• Circulatory damage in the brain, for instance after many small strokes.
• Toxins, brain injuries, infections and metabolic disorders.
• Normal pressure hydrocephalus – a disorder in which brain fluid builds up without the pressure rising much. Typical is the combination of an unsteady, broad-based gait, bladder weakness and mental slowing.

The principle in this drawer is: treat the cause. If a drug is producing the complaints, stopping or switching it can make them disappear. This is the only form of parkinsonism in which recovery is possible – one more reason to examine that possibility seriously instead of concluding prematurely: «that is just Parkinson's».

2.3 The drawers overlap

Reality is not as tidy as a list makes it look. Early on in particular, all four groups look confusingly alike, and misdiagnoses are common.

How movable these boundaries are is shown by a British study of 222 people with atypical parkinsonism [7]. When the specialists applied newer, more broadly framed criteria, the number of people diagnosed with PSP rose from 58 to 101 – almost a doubling. And just under half of those 101 did not have the classic form with early falls but one of the quieter variants. Put differently: a substantial share of these people would have carried a different diagnosis for years under the older criteria.

That is not a reason to distrust neurology, but a reason for patience. A careful assessment takes time, sometimes years, and the diagnosis is reviewed as things develop. The most important yardstick remains: how does it evolve, and how well does levodopa help?

2.4 Why this matters for physiotherapy

You may be thinking: surely the name of the disease makes no difference to the exercises. Unfortunately it does. The classification changes three things:

• The pace of planning. Atypical parkinsonism progresses more rapidly [7]. Aids, home adaptations and training for family members must then not wait until they become «necessary» – they come earlier.
• The timing of training. People who respond well to levodopa have good and bad hours through the day. It is then worth putting practice into the good hours (section 15). People who hardly respond do not have that rhythm – but their capacity fluctuates less.
• What to expect from the result. Almost all the numbers below come from studies of people with Parkinson's disease. For the atypical forms the evidence is thin (section 17). Treatment is still worthwhile there – but the goals shift from «getting better» to «staying safe».

And one more thing: if you are in physiotherapy and your therapist notices things that do not fit the diagnosis – early falls, balance that is much worse than the mobility, rapid progression – that belongs back with the doctor. Physiotherapy does not make diagnoses, but it sees people moving more often and for longer than any consultation does.

3. What the disease does to everyday life

From the technical terms back to the living room. What actually restricts people with Parkinson's is rarely what gets mentioned first in public.

Tremor is the best-known sign, but often not the worst. It is conspicuous, it is uncomfortable in company – but it interferes with life less often than the slowing does.

The slowing is the real opponent. It shows up everywhere a movement would need to have a certain size:

• Steps get shorter, walking more shuffling. The arm stops swinging on one side.
• Turning no longer happens on the spot but with many small steps – and that is the moment in which falls happen most often.
• The upper body hurries ahead of the feet, the steps become faster and shorter, until you would have to run to catch yourself.
• Getting up from the armchair, rolling over in bed, doing up shirt buttons take longer and longer.
• The voice grows quieter without your noticing. Relatives keep asking you to repeat things – while you feel you are speaking at a normal volume.

On top of that come the complaints that have nothing to do with movement, and which often weigh heavier than the movement disorder itself [6]: constipation, which can precede the tremor by years. Loss of the sense of smell – in up to nine out of ten of those affected. Dizziness on standing up because the blood pressure drops, which affects about half and contributes substantially to falls. Sleep disorders in which dreams are acted out. Exhaustion. Low mood and anxiety, affecting around a third. And, as time goes on, restrictions of memory and thinking in a considerable proportion.

This list is not here to frighten anyone. It is here to make two things clear. First: if you have one of these problems, you are not alone with it and it is not «just old age». Second – and this is what this article is about: a large part of this list can be influenced by movement. Not all of it, but more than people think.

4. Why movement in Parkinson's is something other than fitness

To understand why physiotherapy works here, and how it goes about it, an image of the brain helps.

Deep in the brain lies a group of nuclei that can be pictured as an automation unit. It makes sure that learned movements run off by themselves and have the right size. You do not need to think about how long your step should be while walking – the unit knows, and it sets the measure.

In Parkinson's disease, precisely those nerve cells die that supply this unit with the messenger substance dopamine [6]. The unit then keeps working, but it throttles down: it sets movements that are too small. The step length becomes too short, the handwriting too small, the voice too quiet. And – this is the heart of the matter – the gearbox no longer engages reliably when two things are running at once.

Three things follow from this, and they explain the whole rest of this article:

• The «too small» is a wrong setting, not a lack of strength. The muscles could manage bigger steps. They are simply not called up. That is why practising – deliberate, consciously large practising – can work, even though the disease itself does not go away.
• What the automation no longer delivers, attention can take over. There is a second route in the brain through which movements are controlled consciously. In Parkinson's it stays intact for a long time. This is exactly what the cues and strategies in section 11 rest on: you replace a broken automatic switch with deliberate action.
• The conscious route has little room. It only works as long as it is not occupied with something else. That is why gait deteriorates as soon as someone talks, calculates or carries a tray at the same time – and why «one thing at a time» in tricky moments is genuine advice and not condescension.

Physiotherapy in Parkinson's is therefore not simply sport for the ill. It is an attempt to work at three points at once: to make movements big again, to train the conscious route as a substitute, and to keep the body strong and fit enough to do that extra work at all.

5. What physiotherapy demonstrably achieves

5.1 The large overview

The most comprehensive compilation comes from the Netherlands. A group around Danique Radder and Bastiaan Bloem gathered every study in which a physiotherapy treatment for Parkinson's was tested against «no treatment» or a sham treatment: 191 studies with around 8000 participants [1].

Before the results, a word on what these studies actually measured. Four things:

• Movement signs – measured on a scale on which professionals rate tremor, stiffness, slowing and gait (abbreviated UPDRS, part III).
• Balance – for instance with the «timed up and go» test: get up from a chair, walk three metres, turn around, come back, sit down, timed in seconds.
• Walking – walking speed, step length, the distance covered in six minutes, and a questionnaire about freezing.
• Quality of life – a questionnaire on how much the disease restricts everyday life.

The result in one sentence: physiotherapy improved all four areas – but not every kind of treatment improved every area. Here are the main findings, each compared against groups that received nothing or a sham treatment [1]:

• Conventional physiotherapy (45 studies, a good 2600 participants) – that is, what is usually done in a practice, mostly a mixture of several components: improved movement signs, walking, and, uniquely among the categories, clearly also quality of life. Fear of falling and freezing decreased as well.
• Balance and gait training (28 studies, a good 1000 participants): improved movement signs, balance and walking. On the most demanding balance test the effect was large.
• Treadmill training (32 studies): improved above all walking – walking speed and walking distance – and did so more clearly than any other approach. It had no effect on balance or movement signs.
• Strategy training (14 studies), that is, practising with cues and consciously structured movement sequences: improved balance and walking speed.
• Dance (11 studies): improved movement signs, balance and walking – with the largest effect on movement signs of all the approaches.
• Tai Chi and Qigong (11 studies): improved movement signs, balance, walking speed and step length.
• Resistance training (17 studies): improved the distance walked in six minutes. Further effects could not be established in this analysis.
• Endurance training (5 studies): improved movement signs, balance and walking – with large effects, but on a narrow basis.
• Hydrotherapy (8 studies): improved balance and fear of falling.
• Movement games on screen (9 studies): improved balance and quality of life.
• Dual-task training (3 studies), that is, deliberately practising two things at once: improved none of the areas examined in this analysis.

Two qualifications belong with this straight away, and the authors name them themselves. First, the quality of many individual studies is mediocre: the groups are small, and only some of the studies describe cleanly how participants were allocated. Second – and this is the more important point – the approaches were not tested directly against one another, but each on its own against doing nothing. So you cannot read off this list that dancing is «better» than treadmill training.

5.2 The second large analysis – and its uncomfortable answer

It was precisely this gap that a group in Cologne set out to close on behalf of the Cochrane Collaboration. Cochrane is an international network that summarises medical studies according to particularly strict rules; its reviews are regarded as the most careful there are.

The group analysed 154 studies with just under 7900 participants and used a statistical method that allows treatments to be compared even when they have never been tested head to head [2].

The result has two halves.

First half: almost every kind of movement worked – measured on movement signs and on quality of life. Best supported were dance and gait, balance and functional training; for endurance training, hydrotherapy, resistance training and mindfulness-based forms such as Tai Chi, smaller benefits appeared with less certainty. For quality of life, hydrotherapy came out best.

Second half, and this is the actual message: between the kinds of movement there was hardly any difference. The authors say so explicitly – the choice of the kind of movement is probably secondary; what matters is that training happens at all [2].

That is good news, because it means: there is no secret right sport that you might be missing. The best kind of movement is the one you actually do twice a week – for years.

One qualification, for honesty's sake: this statement holds for the two things that were measured, that is, for the movement signs overall and for quality of life. It does not hold for individual problems. When it comes to freezing, to balance or to walking speed, particular approaches are indeed at an advantage – which is why the following sections each have a heading of their own. The Cochrane group states precisely this: their conclusion is compatible with individual symptoms being best treated by Parkinson-specific programmes.

5.3 How big is «big»?

A brief pause here, because numbers from studies are easily misunderstood.

On the movement scale that most of these studies work with, an improvement of about 2.5 points counts as the smallest change a person notices at all [2]. The larger effects in the Cochrane analysis were between 6 and 10 points – so clearly above that threshold. That is not a cure, but neither is it statistics without everyday meaning.

At the same time all these numbers are averages from groups. They describe what happens on average, not what will come out for you personally. And they come from people who volunteered for a study – who were therefore unusually motivated and closely supervised. In practice an effect is therefore often smaller than in the study report. The authors of the large meta-analysis warn themselves against transferring their numbers unchecked to everyone affected [1].

6. Walking: getting the step length back

Walking is the area in which physiotherapy achieves something most reliably – and the one people notice for themselves fastest.

The core of the problem is not speed but step length. Someone who tries to walk faster with Parkinson's mostly takes more steps, not longer ones – and slides even deeper into the shuffling pattern. That is exactly why the instruction in therapy is almost never «walk faster» but «take big steps».

What helps with that:

• Treadmill training. The treadmill solves the problem elegantly: it sets the pace, and the body adapts its step length by itself. In the large overview, treadmill training was the approach with the clearest effect on walking speed and walking distance [1]. For balance, by contrast, it did not help – anyone who needs both needs both.
• Walking to a beat. A metronome, a piece of music with a clear beat, or a companion counting along gives walking from outside the rhythm the automation unit no longer supplies (section 11).
• Nordic walking. The poles lengthen the step almost by themselves and restore the arm swing. In the overview this looked very good – though on the basis of only three small studies, which makes the statement considerably less certain than for the large categories [1].
• Dance. Music sets the beat, the figures force big steps, weight shifts and turns – and many people enjoy it, which counts for more in keeping going than any refinement of the programme. On freezing, however, dance showed no effect [1].
• Practising deliberately large movements. There are programmes built entirely on this: every movement is performed exaggeratedly large, until the exaggerated size feels normal again. That is a sensible idea and is widely used – the evidence for it, however, is still uncertain [2].

For home use the most important point fits into one sentence: a deliberately large step is the simplest and most effective self-help there is for walking – and it costs nothing but remembering to do it.

7. Freezing: when the feet stick to the floor

Hardly any sign is so hard to explain and so frightening as freezing – in technical language freezing of gait.

What happens. The feet suddenly stay on the floor although the head wants to set off. Sometimes they simply stick, sometimes the legs tremble on the spot, sometimes the steps become tiny and ever faster without the body moving forward. Most blocks last a few seconds, some longer.

It nearly always comes in the same situations: when setting off, when turning, in doorways and narrow places, just before the destination – such as in front of the chair or the toilet – and under time pressure. And it comes more often when you are doing something else at the same time or are agitated.

How common. In early stages of the disease about one in four people report such blocks; in advanced stages the proportion rises to as much as 90 per cent [3]. Freezing is one of the most important causes of falls.

What the studies show. An Italian group gathered every study in which physiotherapy was tested against freezing – 19 studies with 913 participants [3]. Their conclusion: physiotherapy reduces freezing, but the effect is small to moderate, and the quality of the data is mixed. What was found:

• Compared with no treatment at all, what worked above all were programmes for the home – that is, exercises the participants carried out themselves in their own surroundings. Exactly where the freezing happens.
• Compared with another treatment, what worked were watching movements – people watch videos of smooth walking and turning and then practise it – and treadmill training.
• The effect of movement observation still held about six weeks later at follow-up. Of all the approaches tested it was the one with the best rating, albeit only with moderate certainty.
• Training with cues alone showed no established effect on the freezing questionnaire in this analysis. That is surprising, because cues undoubtedly help at the moment of a block – more on that shortly in section 11. So it does not mean «cues are no use». It means: that practising with cues durably lowers the number of blocks has not been demonstrated.

What you can do at the moment of a block. The most important thing first: do not pull harder and do not push. Pushing or tugging makes it worse – and brings a risk of falling into play. What helps is switching to the conscious route:

• Stop and take the pressure off. One calm breath. Hurry lengthens the block.
• Shift your weight deliberately onto one leg before you release the other. Very often that alone is the solution: the foot is not glued to the floor, it is loaded.
• March on the spot – two or three times – and then set off.
• Step over something. A target on the floor – a joint in the tiles, a pattern in the carpet, a companion's foot or walking stick – gives the movement a measure the automation no longer supplies.
• Count or use march music. «One – two – one – two», out loud or inwardly.
• Bigger, not faster. The first step should be deliberately large, not quick.

Which of these tricks works for you is very personal. Trying them out belongs explicitly in physiotherapy: not discussed at a table but practised in the doorway, in the narrow hallway and in front of the chair – until you have the one trick that reliably works for you.

8. Balance and falls

Falls are the biggest worry for many of those affected and their families, and the worry is justified: most people with Parkinson's fall at least once in the course of the disease [4].

This is also where the clearest gap in drug treatment shows. Tremor, stiffness and slowing respond well to levodopa. Balance and trunk stability, by contrast, hardly do – and deep brain stimulation, the surgical option for difficult courses, does not improve that side either [6]. So whatever helps against falls has to come from somewhere else.

The most precise answer here comes from another Cochrane review: 32 studies with 3370 participants [4]. For exercise programmes in people with mild to moderate Parkinson's disease the result reads:

• The number of falls fell by about 26 per cent (12 studies, a good 1400 participants). The certainty of this statement is rated as moderate – which is a good value for this field.
• The number of people who fell at all fell slightly – by about 10 per cent.
• Whether this also produces fewer fractures cannot be said from the available data. The studies were too small for that.
• Quality of life improved slightly immediately after the programmes.

Three points of context worth knowing:

First, this holds for mild to moderate stages. For people with advanced disease or with marked memory problems the data are thin – and the few that exist are less encouraging. That does not mean training is pointless there. It means the goals shift: away from preventing every fall, towards safe routines, aids, a safe home and trained family members.

Second, education alone is not enough. In the analysis, a purely informational programme without training was not sufficient [4]. Knowing how to walk safely does not replace the ability to catch yourself.

Third, blood pressure belongs in the picture. About half of those affected have a blood pressure that drops on standing up [6]. If your vision goes black when you stand, you do not have a balance problem but a circulation problem – and that needs medical assessment, not training.

What makes balance training effective: it has to go to the limit. Training in which you never wobble trains nothing – balance only improves when it is challenged. In practice that means: narrow stance, standing on one leg, walking with changes of direction, turns, stepping over obstacles, reacting to a light nudge, stopping under control on command – all of it supervised and with something to hold on to. That is exactly why balance training is the part that works least well alone at home.

9. Strength and power

In Parkinson's, strength is often overlooked, because the disease is not called «weakness» after all. It still belongs in the programme, for two reasons.

The first reason is banal and important. People with Parkinson's move less, often years before the diagnosis. Less movement means less muscle – and that loss is an additional problem, which has nothing to do with the disease itself and can be reversed with training.

The second reason is power. That means how quickly you can produce force – not how much. When catching yourself after a stumble, exactly that counts: half a second too late and you fall despite having enough strength.

A small but instructive study from Miami looked into this [11]. Twenty-six people with mild to moderate Parkinson's disease trained for three months, twice a week, with moderate weights performed briskly, supplemented by balance and agility drills. By the end, not only strength and power had improved but also the slowing at arms and legs as rated by professionals – and quality of life, particularly in mobility and activities of daily living.

That is remarkable, because here it was not only a side effect that improved but a core sign of the disease. The study was small, and a single result of this size must be read cautiously. But it fits what is in section 4: if movements are «set too small and too slow», practising large and fast movements can turn the setting back somewhat.

Incidentally, in the large overview resistance training did not stand out clearly on movement signs, but it did on walking distance [1]; in the Cochrane analysis it showed smaller benefits [2]. So strength is not the one miracle cure in Parkinson's – it is the foundation on which the other components can work in the first place.

How strength training is built up, how heavy it has to be and why what counts is the effort and not the equipment, is set out at length in its own article, strength training. The same principles apply in Parkinson's, with three additions:

• Do not train to complete exhaustion. That applies to older people in general and here in particular: with complete exhaustion the execution becomes sloppy, and sloppiness is expensive when you have a balance disorder.
• Perform briskly wherever it is safe to do so. The weight may be moderate; what matters is that you get it moving fast. Lowering stays slow and controlled.
• Machines are often the better choice than free weights – not because they are more effective, but because they do not additionally challenge your balance. If you have to concentrate on balancing while you exercise, you are no longer training strength.

10. Endurance – and the big open question

Now comes the section with the most hope and the most blur in it. So let us be especially careful here.

The question is: can movement not only ease the symptoms but slow the progression of the disease itself?

Three pieces of research are worth knowing.

First, the American SPARX study [8]. 128 people with freshly diagnosed Parkinson's disease who were not yet taking medication were divided into three groups: treadmill at high effort (four times a week, at 80 to 85 per cent of maximum heart rate), treadmill at moderate effort (60 to 65 per cent) and a waiting list. After six months the movement scale had barely changed in the high-effort group, while in the waiting group it had worsened by a good 3 points. Moderate effort was not enough.

The qualification here matters, and the authors state it themselves: this study was a feasibility study. It was meant to test whether such training is safe and practicable at all – and whether a large trial was worth doing. It does not prove that hard endurance training slows the disease. That large trial is now under way; its results are still awaited.

Second, the Dutch Park-in-Shape study [9]. 130 people with mild Parkinson's disease who had been physically inactive trained for six months at home: three times a week for 30 to 45 minutes on an exercise bike linked to a screen game, with supervision from a distance. The comparison group did stretching exercises.

After six months the difference on the movement scale was 4.2 points in favour of endurance training – and measured in the phase in which the medication was working least. That is more than the 3.5 points that had been defined in advance as meaningful. This study was designed as double blind, which rarely succeeds in exercise research and makes it particularly credible.

And it shows something practically important: it worked at home. Not in a clinic, not with specialist equipment – on an exercise bike in the living room, with support by telephone and app.

Third, a Chinese long-term observation of Tai Chi [10]. 143 people with Parkinson's practised Tai Chi regularly, 187 others did no sport; both groups were followed for about three and a half years. In the Tai Chi group the scores on the movement scale worsened more slowly, complaints outside movement were fewer, and – the most interesting observation – the medication dose had to be increased more slowly.

The qualification here is a fundamental one: this was not a study with allocation by lot but an observation. People decided for themselves whether to do Tai Chi. Those who choose it and stick with it may well differ in other ways too – healthier, at an earlier point in their course, better connected socially. Part of the difference may therefore come from those differences and not from the Tai Chi. The observation is encouraging; a proof it is not.

What follows from this for you? Something cautious and nonetheless clear:

• That endurance training improves the symptoms is well established.
• That it slows the course of the disease is a well-founded hope with serious pointers behind it – but not yet a proof.
• Effort seems to play a part in this: in the SPARX study moderate loading was not enough. «Going for a walk» is healthy, but it is not the same as endurance training.

And how do you know the effort is right? Without a heart rate monitor a simple rule will do: at moderate effort you can still speak in full sentences but no longer sing. At high effort you can only get single words out. If you have a heart condition, an untreated high blood pressure or a blood pressure that drops on standing, have the loading cleared medically beforehand.

11. Cues and strategies: operating the automation by hand

This section describes what distinguishes physiotherapy in Parkinson's from sport in general.

11.1 Cues

A cue is a signal from outside or inside that triggers a movement and gives it a measure. It replaces what the automation unit no longer supplies [12]. There are three kinds:

• Hearing. A metronome, music with a clear beat, counting out loud, a companion's command. The most common is a beat slightly above your own habitual step frequency.
• Seeing. Stripes on the floor to step over. A walking stick turned upside down so that its handle lies as a hurdle in front of your foot. Joints in a tiled floor. A laser dot that some walking sticks project in front of the foot.
• Feeling. A vibration signal at the wrist, tapping on the thigh, shifting weight from one leg to the other in rhythm.

What is established and what is not. A clean distinction pays off here, because on the internet the two get mixed up:

• In the moment, cues work. The review on this topic summarises 24 investigations in 354 people with freezing: training with cues reduced blocks immediately afterwards and improved the gait pattern [12]. And in everyday life those affected report this anyway – the trick works.
• In the long run the evidence is weak. The same review notes that the effect often does not persist after training and transfers poorly to new situations [12]. And the meta-analysis on freezing found no established effect of cue training alone on the questionnaire [3].

The practical conclusion is therefore not «don't bother» but: cues are a tool for the moment, not a course you complete. They are practised where they are needed – in your own hallway, your own kitchen, at your own front door – and they stay in use permanently. A metronome on your phone is not something you «get over»; it is more like a pair of glasses.

11.2 Movement strategies

The second technique goes one step further. A movement strategy breaks a difficult, otherwise automatic action down into conscious single steps that you work through like a recipe.

The best example is getting up from a chair. Automatically it happens in one go. In Parkinson's it often fails – usually because the upper body does not come far enough forward, so the weight never arrives over the feet. As a recipe it looks like this:

• Feet well back under the chair, one foot slightly ahead.
• Slide to the front edge of the seat.
• Upper body far forward – «nose over toes».
• Count to three, then up in one movement, without taking a run-up.

The same principle can be applied to almost any difficult sequence: rolling over in bed, getting into the car, turning while walking (professionals recommend a wide arc rather than a turn on the spot), picking something up off the floor. In the large overview this strategy training improved balance and walking speed [1]. It is a core element of the European physiotherapy guideline for Parkinson's [14].

11.3 Two things at once

Because the conscious route has little room (section 4), it collapses as soon as something else is added. That is why some people with Parkinson's stop dead in the middle of a conversation or start to stumble when carrying a tray.

The practical advice is therefore: in tricky situations, one thing at a time. Do not talk while walking. Stand still before answering. Look at the route before carrying something. That is not condescension but the only way of keeping the limited room free for what matters.

Whether this dual-task ability can be trained on purpose is an open question. In the large overview, dual-task training alone improved none of the areas measured [1] – though on the narrow basis of only three studies. Anyone caring for someone at home is better served by the simple rule above.

12. Getting up, turning, sitting down

Professionals call them transfers: the moves from one position to another. They are constantly needed in daily life and become difficult early in Parkinson's – and they are the area in which good physiotherapy changes something noticeably fastest, because almost everything here can be solved with strategies.

• Getting up from a chair. See the recipe above. In addition: the chair may be high and have armrests – a low, soft armchair is the most difficult seat in the whole flat.
• Sitting down. Walk backwards to the chair until your calves touch it, only then support yourself with both hands and lower yourself slowly. Never while falling.
• Turning over in bed. Pull the knees up, head to the side, then take arms and shoulders to the same side. A satin sheet or a satin strip across the bed reduces friction considerably – a small trick with a large effect.
• Getting out of bed. First roll onto your side, then legs over the edge, then push yourself up with the lower arm. Do not pull straight upwards.
• Into the car. Sit down backwards onto the seat, then swing both legs round together. A plastic bag on the seat makes turning much easier.
• Turning while standing. In a wide arc, with deliberately large steps – do not pivot on the heel. Turning on the spot is one of the most common causes of falls.

These handholds are not scientifically spectacular, but they often decide whether somebody stays independent at home. They belong in every course of physiotherapy and – just as importantly – in the instruction of family members.

13. Posture, mobility, pain

Over time the body in Parkinson's draws itself together: the upper body tips forward, the head pushes ahead, the shoulders round, rotation of the trunk is lost.

Honesty is called for here. Stretching alone achieves little in Parkinson's. In the Cochrane analysis, flexibility training was the only category that did not stand out positively in the numbers [2] – and in several studies stretching served deliberately as the comparison treatment, precisely because no great effect was expected from it [9].

That does not mean mobility is unimportant. It means it does not come back through passive stretching but through active, large movements: consciously wide-reaching arms, trunk rotations sitting and standing, straightening up against gravity, exercises lying on your back. So through the same principle as everywhere in this article – big instead of small.

On pain: many people with Parkinson's have pain, often in the shoulder and the back. A stiff shoulder on the more affected side is sometimes even the very first sign of the disease, years before the diagnosis. Such complaints are treatable – but they need an examination of their own and not the explanation «that's the Parkinson's». Whether a shoulder is stiff because the muscle is permanently tensed, because the joint capsule has become adherent or because a tendon is irritated makes a large difference to the treatment.

14. Speaking, swallowing, handwriting – where other professions take over

Physiotherapy does not cover everything. Three areas belong explicitly to other professions, and it is worth knowing when to call them in.

• Speaking. The voice becomes quieter and more monotone, the words run together [6]. The tricky part: those affected hear themselves as normally loud. Speech and language therapy has effective programmes for this, working on exactly the same principle as gait training – speak deliberately exaggeratedly loudly, until the exaggerated level feels normal again.
• Swallowing. Choking increases, and saliva runs out of the mouth – not because more saliva is produced but because automatic swallowing becomes less frequent [6]. This too is a case for speech and language therapy. Frequent choking, especially on liquids, should not be left to run: it carries a risk of pneumonia.
• Hands and daily life. Buttons, handwriting, cutlery, safety in the home, aids – that is the territory of occupational therapy.

In the Netherlands this insight has been turned into an entire model of care: a network of professionals who have specialised in Parkinson's and coordinate with one another. The European physiotherapy guideline [14] and a large part of the research quoted here come from that setting. The same applies in spirit in Switzerland: it is worth asking for therapists who regularly treat people with Parkinson's – experience with the condition makes a difference in the choice of strategies.

15. Medication and training: using the good hours

A practical point that saves or ruins whole training plans.

Levodopa does not work evenly around the clock. Early in the disease the effect is usually steady. Later it becomes fluctuating: there are times when mobility is good – professionals say «on» – and times when the effect wears off and everything is hard again: «off». Some people feel this switch to within a quarter of an hour.

Three simple rules follow:

• Demanding practice belongs in the good phase. Typically it begins about an hour after taking the tablet. Training in the «off» phase mostly means practising frustration – and increases the risk of falling. Studies that set out to measure capacity plan their assessments in the good phase for exactly this reason [11].
• Arrange therapy appointments around your medication schedule, not the other way round. That is a legitimate request, and any practice familiar with Parkinson's knows it.
• Keep a simple record for a few days if you are unsure when your good times are: the time you take the tablet, and an hourly note on how your mobility is. That is at the same time the most useful thing you can bring to a neurology appointment.

This does not apply to every exercise, by the way. Balance, strength and endurance are sensibly trained in the «on» phase. Strategies for difficult situations, though – getting up, turning, releasing a block – have to be practised at some point in the «off» phase too. Because that is exactly when you need them.

16. How much, how often, for how long?

Now the question that is asked most often – and on which, honestly, research has least to offer.

An American group looked into exactly this: 46 studies with around 3900 participants, analysed for which type, which timing, which frequency and which duration work best [5]. The result is sobering and illuminating at the same time:

• What is customary: in 85 per cent of the studies the programme lasted 2 to 12 weeks; in just under 60 per cent training took place two or three times a week; in 85 per cent a session lasted 30 to 60 minutes.
• What can be derived from that: little. More frequent training brought no demonstrably greater benefit in the analysis. And the various approaches did not differ from one another.
• What was almost never studied: whether the effect lasts. Only about half the studies even looked – and not a single one followed participants for longer than 18 months.

In a disease that lasts decades, this last gap is the biggest. It also explains why professionals so often answer evasively when asked «how often?»: there is simply no robust answer. Anyone who gives you an exact number is giving you a reasoned estimate, not evidence.

What can nevertheless be said rests on what was actually done in the effective studies:

• Two to three times a week of targeted training, 30 to 60 minutes each time – that is the pattern with which the great majority of effective programmes worked [5].
• Plus endurance: in the two endurance studies, training was three to four times a week for 30 to 45 minutes [8][9].
• Programmes shorter than twelve weeks are probably too short to achieve anything meaningful on the movement signs; an analysis quoted in the large overview points that way [1].
• And then it continues. That is the most important point of all.

Why «and then it continues» is the whole heart of it. With a torn cruciate ligament, physiotherapy is an episode: you do it, you get better, you stop. In Parkinson's it is not. The disease progresses; a level of training you do not maintain is lost – and the disease keeps running in the meantime.

So the right question is not «how many treatment sessions do I need?» but «how do I get this into my life permanently?». And the answer to that is rarely a practice prescription alone. It is usually a mixture: physiotherapy in blocks when something new appears or something needs readjusting – and in between a home programme, a group, a choir, a dance class, a Nordic walking round, an exercise bike in the living room. The Cochrane analysis says just this: which kind of movement you choose is secondary [2]. Whether you are still doing it in five years is not.

17. Atypical and secondary parkinsonism: what is different here

Now the section people like to skip and which is the most important one for those it concerns.

Practically all the numbers in this article come from studies of people with Parkinson's disease. The large meta-analysis explicitly excluded studies with atypical or secondary parkinsonism [1]. That is scientifically clean – but it means that people with MSA, PSP or CBS are left with a far thinner basis.

How thin is shown by a review on PSP: of the treatment studies found, most were designed without a comparison group or with very few participants, and the authors conclude that robust evidence for improving gait and balance is lacking [13]. There are pointers towards benefit – established fact there is little.

Does that mean physiotherapy is not worthwhile there? On the contrary. Precisely because medication helps little in these diseases, non-drug treatment often remains the only one available. Only the goals shift – and that should be stated openly from the start, rather than promising an improvement that will not come:

• Safety before performance. In PSP, falls come early and hard, often backwards, and those affected regularly underestimate the danger. Here it is about safe routines, a safe home, aids introduced early – and not about improving walking speed.
• Early rather than late. Because deterioration runs faster [7], care has to stay ahead of the course. A rollator that comes three months too early is unpleasant. One that comes three months too late costs a fractured hip.
• Think in the particulars. In PSP, the restricted downward gaze – stairs and kerbs become dangerous, and obstacles on the floor are simply not seen. In MSA, the dropping blood pressure – exercises lying down followed by standing up need care and time.
• Family members are part of the treatment, not spectators. Back-sparing assistance, safe repositioning, two-person transfers – these are practised while things are still calm.

And in the fourth drawer, secondary parkinsonism, the rule from section 2.2 applies first of all: check the cause first. If a drug is causing the complaints, they may recede once it is stopped [6]. It would be bitter to train for years against something that one tablet fewer would also have solved.

18. What a programme looks like in practice

18.1 What comes first

A good course of treatment does not start with exercises but with a stocktaking. The European guideline [14] names five core questions for this, and everything else follows from them:

• Have you fallen in the past twelve months – and how often?
• Are there blocks while walking, and if so, in exactly which situations?
• Where in daily life is something most likely to go wrong: walking, turning, getting up, transfers?
• How much do you actually move in everyday life?
• What would you like to be able to do again?

A few simple measurements go with this – walking speed over a short distance, timed up and go with a stopwatch, balance tests, strength compared side to side. These numbers are not an end in themselves: they are the only way of telling in six months whether something has changed or whether you merely wish it had.

18.2 The components

A complete programme usually consists of five parts. Not every one is necessary for everybody – the selection follows from the stocktaking:

You can see that this quickly sounds like too much. It can be combined: a dance class covers endurance, balance, gait and large movements at once. Nordic walking covers endurance and step length. Tai Chi covers balance and large, slow movements. That is why group offerings in Parkinson's are not the budget version but often the smarter solution – the social side comes free, and groups are easier to keep up than exercise sheets.

18.3 The home programme

The home programme is not an appendix; for freezing in particular it is the most effective part: in the meta-analysis it was the programmes at home that showed an established effect against no treatment at all [3]. And the Park-in-Shape study showed that demanding endurance training works at home too, if somebody keeps in touch from a distance [9].

For it to work, three things are needed:

• Short and firmly anchored. Ten minutes tied to an existing habit – after breakfast, after the morning tablet – get done. Forty minutes «sometime during the day» do not.
• In the right place. The block happens in your hallway, not in the exercise room. So that is where it gets practised.
• Written down. A sheet with a few exercises, printed large, in a visible place. And a tick for each day done – that is not surveillance, that is what makes the difference.

18.4 How to measure for yourself whether it is working

In a progressive disease, «success» is not always «better». Sometimes «unchanged» is the best achievable result – and it is a real one. But you will not notice that if you never measured.

Take a few values you can repeat at home:

• Getting up from a chair five times without using your hands, with a stopwatch. The most practical strength test for the legs.
• Timed up and go: get up from a chair, walk three metres, turn, come back, sit down – in seconds.
• Walking speed over a measured distance at your usual pace.
• Standing on one leg with eyes open, until you have to put the foot down – only with something to hold on to within reach.
• A falls diary. A tick in the calendar for every fall and every near-fall. It sounds like little, but it is the single most informative measurement there is in Parkinson's.

Two notes on this. Always measure at the same time of day and at the same interval from taking your tablets – otherwise you are measuring the drug level and not your training state. And measure twice at the start, on two different days. The difference between those two values is your personal day-to-day variation; a later change is only worth something when it is clearly larger. After that, every three months is enough.

19. For family members

A short section of its own, because family members almost never appear in studies and almost always in daily life.

• Do not pull, do not push. During a block, pulling on the arm is the worst thing you can do. Instead, stand in front of the person, set a beat, or offer your foot as an obstacle to step over.
• Do not talk while walking when things are tricky. Save your question until you are both standing still. That is not rudeness, it is fall prevention.
• Allow time. What looks slow is usually the maximum possible speed. Take over only what really no longer works – a task taken over prematurely rarely comes back.
• Say what you see. People affected often do not notice themselves that their steps have got shorter and their voice quieter. A matter-of-fact remark is not criticism.
• Have someone show you how to help. Transfers, getting up after a fall, safeguarding while walking – that belongs in a physiotherapy session you attend. Improvising ends in injuring yourself.
• Think of yourself. Exhaustion in family members is the most common reason why care at home comes to an end. Using respite services early is not a failure.

20. Eight common misunderstandings

«Tremor means Parkinson's.»
No. The most common tremor of all is essential tremor, and it behaves in exactly the opposite way: it appears on holding and reaching and disappears at rest. With Parkinson's tremor it is the other way round [6]. And conversely: a considerable proportion of people with Parkinson's never have tremor at all.

«It's Parkinson's anyway, the cause makes no difference.»
It does. A parkinsonism caused by a drug can recede once the drug is stopped [6]. And an atypical parkinsonism runs a different course and needs different planning [7]. The classification determines the treatment (section 2).

«I'll do physiotherapy when it gets worse.»
This is the most common and most expensive wrong decision. Two of the most informative studies included people with mild and in part still untreated disease – and that is where the effects were clearest [8][9]. And habits you have built up while things were going well carry over into the harder years. The other way round hardly works.

«I have to find the right sport.»
The Cochrane analysis of 154 studies found hardly any differences between the kinds of movement and states that the kind is probably secondary [2]. For individual problems – freezing, balance, walking speed – there are indeed more suitable approaches. For the overall effect, what counts above all is that you stick with it.

«Going for a walk is enough.»
Walking is good and much better than nothing. But it trains neither balance at its limit nor strength nor endurance in the range in which the interesting study results were produced: in the SPARX study even moderate loading was not enough [8].

«I have no place and no equipment.»
The programmes that took place at home were the only ones with an established effect on freezing [3], and an exercise bike in the living room produced a clinically meaningful difference over six months [9]. What you need is a plan and somebody who checks in – not a gym.

«If I fall, only a rollator will help now.»
In mild to moderate disease, exercise programmes reduced the number of falls by about a quarter [4]. And an aid and training are not opposites – one secures daily life, the other secures the ability.

«If I do exercises, I'll need less medication.»
That is not how it should be turned around. Movement does not replace Parkinson's medication, and nobody should stop anything on their own account because of it. The observation that the dose had to be increased more slowly in regular exercisers comes from an observational study without allocation by lot [10] – interesting, but not a proof and certainly not an instruction to leave things out.

21. When to get in touch

Get medical advice promptly for:

• A new fall with injury – or several falls within a few weeks.
• Dizziness or blacking out on standing up. This is usually blood pressure and is well treatable.
• Frequent choking, especially on liquids, or repeated pneumonia.
• Newly appearing hallucinations, confusion or marked daytime sleepiness – often a side effect that can be adjusted.
• A rapid deterioration within weeks. That does not fit the usual course and often has an additional, treatable cause – an infection, for instance.
• If the complaints appeared after starting a new medication (section 2.2).

Raise in physiotherapy:

• You get stuck while walking – even if it is «only sometimes».
• You avoid routes, places or activities for fear of falling.
• Getting up from a chair or turning over in bed no longer works reliably.
• Your family say your steps have got shorter or your voice quieter.
• You have been doing the same thing for months without it having become more demanding.
• You have pain that nobody has looked at because «that's the Parkinson's».
• You do not know how things are meant to continue after the last prescribed treatment.

What to expect in our practice: a stocktaking with measured values – walking speed, getting up, balance, strength compared side to side, falls history –, from that a programme with clear goals, practice of strategies where you need them, a short home programme for in between, instruction for family members if you wish, and repeat measurements at fixed intervals. And, because that belongs to a long illness: thinking through how things go on without us – which group, which offering, which rhythm.

22. In summary

Parkinsonism is a pattern of complaints with several possible causes. Parkinson's disease is the most common, but there are also the faster-progressing atypical forms and the secondary forms with an identifiable cause – in which recovery is possible if the cause is removed [6][7].

Physiotherapy works. In the summary of 191 studies it improved movement signs, balance, walking and quality of life [1]. In mild to moderate disease, exercise programmes reduced the number of falls by about a quarter [4]. And for freezing there are effective approaches, above all exercises in your own home [3].

Which kind of movement you choose is secondary – that is the result of 154 studies [2]. For individual problems there are more suitable approaches, and they are in the sections above. For the overall effect: the best kind of movement is the one you will still be doing in five years.

What makes the difference can be put in five points:

• train purposefully two to three times a week, 30 to 60 minutes [5];
• include balance training that really goes to the limit;
• endurance at a noticeable effort, not just going for a walk [8][9];
• all movements deliberately large – steps, arms, voice, handwriting;
• strategies and cues for the difficult spots of your own daily life, practised at exactly those spots [12][14].

And what you do not need: the one right sport, a gym, expensive equipment – or the certainty that movement slows the course. That certainty does not exist yet. What does exist is strong evidence pointing that way, and firm proof that movement makes life with the disease better.

In a disease that lasts years and decades, the decisive factor is not the refinement of the programme. It is whether you are still training in five years. Everything in this article – the group rather than the exercise sheet, the home programme rather than the clinic, the dance class rather than the perfect selection of exercises – serves that one question.

References

All Digital Object Identifiers (DOIs) were individually verified against the Crossref register. The links in the reference list lead via the DOI service to the publishers' pages. Some of these are located outside Switzerland and the EU. When you click on them, your IP address is transmitted to the respective provider – this does not happen on our own site.

[1] Radder DLM, Silva de Lima AL, Domingos J, Keus SHJ, van Nimwegen M, Bloem BR, de Vries NM. Physiotherapy in Parkinson's Disease: A Meta-Analysis of Present Treatment Modalities. Neurorehabilitation and Neural Repair. 2020;34(10):871–880. https://doi.org/10.1177/1545968320952799

[2] 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;4(4):CD013856. https://doi.org/10.1002/14651858.CD013856.pub3

[3] Cosentino C, Baccini M, Putzolu M, Ristori D, Avanzino L, Pelosin E. Effectiveness of Physiotherapy on Freezing of Gait in Parkinson's Disease: A Systematic Review and Meta-Analyses. Movement Disorders. 2020;35(4):523–536. https://doi.org/10.1002/mds.27936

[4] Allen NE, Canning CG, Almeida LRS, et al. Interventions for preventing falls in Parkinson's disease. Cochrane Database of Systematic Reviews. 2022;6(6):CD011574. https://doi.org/10.1002/14651858.CD011574.pub2

[5] El Hayek M, Lobo Jofili Lopes JLM, LeLaurin JH, et al. Type, Timing, Frequency, and Durability of Outcome of Physical Therapy for Parkinson Disease: A Systematic Review and Meta-Analysis. JAMA Network Open. 2023;6(7):e2324860. https://doi.org/10.1001/jamanetworkopen.2023.24860

[6] Hayes MT. Parkinson's Disease and Parkinsonism. The American Journal of Medicine. 2019;132(7):802–807. https://doi.org/10.1016/j.amjmed.2019.03.001

[7] Jabbari E, Holland N, Chelban V, et al. Diagnosis Across the Spectrum of Progressive Supranuclear Palsy and Corticobasal Syndrome. JAMA Neurology. 2020;77(3):377–387. https://doi.org/10.1001/jamaneurol.2019.4347

[8] Schenkman M, Moore CG, Kohrt WM, et al. Effect of High-Intensity Treadmill Exercise on Motor Symptoms in Patients With De Novo Parkinson Disease: A Phase 2 Randomized Clinical Trial. JAMA Neurology. 2018;75(2):219–226. https://doi.org/10.1001/jamaneurol.2017.3517

[9] van der Kolk NM, de Vries NM, Kessels RPC, Joosten H, Zwinderman AH, Post B, Bloem BR. Effectiveness of home-based and remotely supervised aerobic exercise in Parkinson's disease: a double-blind, randomised controlled trial. The Lancet Neurology. 2019;18(11):998–1008. https://doi.org/10.1016/S1474-4422(19)30285-6

[10] Li G, Huang P, Cui S, He Y, Tan Y, Chen S. Effect of long-term Tai Chi training on Parkinson's disease: a 3.5-year follow-up cohort study. Journal of Neurology, Neurosurgery & Psychiatry. 2024;95(3):222–228. https://doi.org/10.1136/jnnp-2022-330967

[11] Ni M, Signorile JF, Balachandran A, Potiaumpai M. Power training induced change in bradykinesia and muscle power in Parkinson's disease. Parkinsonism & Related Disorders. 2016;23:37–44. https://doi.org/10.1016/j.parkreldis.2015.11.028

[12] Ginis P, Nackaerts E, Nieuwboer A, Heremans E. Cueing for people with Parkinson's disease with freezing of gait: A narrative review of the state-of-the-art and novel perspectives. Annals of Physical and Rehabilitation Medicine. 2018;61(6):407–413. https://doi.org/10.1016/j.rehab.2017.08.002

[13] Dale ML, Silva-Batista C, de Almeida FO, Horak FB. Balance and gait in progressive supranuclear palsy: a narrative review of objective metrics and exercise interventions. Frontiers in Neurology. 2023;14:1212185. https://doi.org/10.3389/fneur.2023.1212185

[14] Keus S, Munneke M, Graziano M, et al. European Physiotherapy Guideline for Parkinson's Disease. KNGF/ParkinsonNet; 2014. parkinsonnet.nl