Restless legs syndrome (Willis-Ekbom disease)

Your patients already describe it to you. They say “cramps”, “restlessness”, “poor circulation”, “pins and needles that keep me awake”. Restless legs syndrome affects about 3 % of adults to a degree that warrants treatment, and its diagnosis is entirely clinical: four questions, plus a fifth that does all the work.

A synthesis based on the 2014 IRLSSG criteria, the JAMA review of February 2026, the 2025 guideline of the American Academy of Sleep Medicine, the systematic review of non-drug interventions by Harrison et al. and the REST general population study. Every figure carries its source where it is written.

Clinical summary

“Restless legs syndrome is poorly recognised by doctors and is consequently often misdiagnosed and poorly managed.”
Manconi et al., Nature Reviews Disease Primers, 20218

Why does this syndrome go unnoticed in patients we see every week?

Because it has no word in patients' vocabulary, because its prevalence changes threefold depending on the question asked, and because the complaint that brings people to the clinic is almost never the one that carries the diagnosis.

A patient never says “I have an urge to move my legs, brought on by rest and relieved by movement, with worsening in the evening”. They say they have cramps. Restlessness. Pins and needles. That it takes them in the evening in front of the television, that they have to get up, that their partner complains because they fidget. They often say they sleep badly, and they rarely connect the one to the other.

This condition has the further peculiarity of being invisible at the moment you examine it. The patient consults in the morning or the afternoon; the symptoms occur in the evening and at night. The clinical examination is normal, and it should be. There is no test, no manoeuvre, no imaging that makes the diagnosis. The 2026 JAMA review is explicit on this point: restless legs syndrome is diagnosed from the history, and polysomnography is not recommended for the diagnosis5.

A prevalence that depends on the question asked

This is the least well understood point of this condition, and it is not a methodological detail: it governs the way every published figure is read.

The REST study surveyed 16,202 adults and used 15,391 complete questionnaires. It measured three different things in the same subjects3 :

Remember the clinical reading of this graph: when an article announces “10 % of the population”, it is applying a broad definition; when it announces “2 to 3 %”, it is counting the patients who need treatment. Both are true. A physiotherapist with a caseload of 300 people statistically meets about twenty with symptoms, and about eight in whom the distress would warrant management.

Underdiagnosis, measured

The REST study did not stop at prevalence. It asked the 416 people meeting the criteria for clinically significant distress whether they had mentioned it to a doctor, and what they had been told.

It is those 316 patients that make this article of interest to a physiotherapy practice. They consult. They describe. They leave with no name for what they have, and often with a prescription for sessions for “heavy legs” or “circulatory problems”.

Who is affected, and in what settings

Certain settings concentrate the condition to the point where thinking of it becomes routine. The 2026 JAMA review gives the ranking5, and several of these settings are ones the physiotherapist follows over the long term.

Three of these settings deserve underlining for a physiotherapy practice. Multiple sclerosis comes top with 27.5 %. Peripheral neuropathy, diabetic in particular, at 21.5 %, and it is precisely the condition with which the syndrome is most often confused. Parkinson's disease at 20 %, where the picture is complicated by dopaminergic treatment itself.

What are the diagnostic criteria, and why did they go from four to five?

The four historical criteria describe the disease perfectly. Their fault is not that they miss patients: it is that they catch some who do not have it. The fifth criterion exists for that, and its exact wording matters.

The version in force

The reference criteria are those of the International Restless Legs Syndrome Study Group (IRLSSG), revised in 2014 and published in Sleep Medicine1. They are still those in force in 2026: the international classification of sleep disorders in its third revised edition (ICSD-3-TR, 2023) takes them up, and the 2026 JAMA review describes them identically5.

The 2014 revision introduced four changes, only one of which alters the diagnostic approach day to day1 :

• the addition of a fifth essential criterion, differential diagnosis, with the explicit aim of improving specificity by requiring that the symptoms not be confused with those of other conditions;
• the addition of a clinical significance qualifier, which identifies the patients who need treatment;
• the addition of course qualifiers: chronic-persistent or intermittent form;
• the merging of the paediatric and adult criteria into a single set.

Why four were not enough: the measurement

The demonstration came from a family case-control study conducted by the Johns Hopkins team. Of 1,255 relatives contacted, 1,232 could be classified by an expert in the syndrome by means of a validated telephone interview, richer than the four criteria alone: open questions on the quality of the sensations, the relieving strategies, the nature of the triggering situations2.

“A definitive diagnosis of restless legs syndrome therefore requires excluding these other conditions, which may be more common in the population than the syndrome itself.”
Hening et al., Sleep Medicine, 20092

How do you tell restless legs syndrome from the other painful legs in the evening?

This is the chapter that makes this article useful in practice. Five entities sound alike in the history and are separated on simple points: provided you know which, and ask the right questions rather than listening to the spontaneous complaint.

A leg that hurts in the evening is a commonplace reason for consulting. The patient says “cramps” whatever the mechanism, because that is the word available. Five main entities compete for this ground, and the World Health Organization formally separates them: restless legs syndrome carries the code 7A80 in ICD-11, and sleep related leg cramps a separate code, 7A8226.

The measured discriminators

The study by Hening et al. did not merely count false positives: it compared point by point the answers of true patients with those of subjects who ticked the four criteria without having the disease. That table is, to our knowledge, the only published quantification of these discriminators2.

Two practical lessons come out of this table, and they are counter-intuitive.

Relief by movement discriminates nothing. 88 % of true patients and 91 % of false ones say they are quickly relieved by moving: the difference is not significant. Yet it is the criterion everyone remembers as the disease's signature. It is necessary for the diagnosis, it is not discriminating.

What does discriminate is how the relief is obtained and how long the symptom lasts. In a mimic, a change of position is enough half the time and the symptom does not come back; the patient then stays settled. In a true patient, they have to get up, walk, and it starts again as soon as they sit down. That is what the two sharpest lines of the table express: more movement is needed to obtain relief, and the symptom lasts longer2.

Pain is rather an argument against. 47 % of mimics describe pain, against 23 % of true patients. A patient who insists on the word “pain” rather than on the words “urge to move” should prompt a search for something else, which does not rule out the painful forms, which do exist.

The differential table of painful legs in the evening

Venous insufficiency and peripheral artery disease: two opposite traps

These two entities deserve to be dealt with together because they mislead in opposite directions, and because the site does not yet devote a page of its own to them: their clinical side is covered in peripheral neuropathy for the diabetic foot, and in chronic exertional compartment syndrome for the triage of exertional calf pain.

Venous insufficiency is the most dangerous false friend, because it shares with the syndrome its evening timing and its relief by walking. Two elements unmask it. First the position: it is prolongedstanding or sitting, with the legs down, that makes it worse, whereas the syndrome is brought on by lying at rest. Then and above all, elevating the legs relieves venous insufficiency, and elevating means increasing rest, which ought to make restless legs syndrome worse. That dissociation is decisive and it is tested in the chair, in fifteen seconds.

Associated signs to look for: pitting evening oedema, varicose veins, haemosiderin staining, worsening in heat and at the end of the day. The review by Youn and Lee recalls that this condition is “frequently overlooked” through underestimation of its extent and unfamiliarity with its presentations15.

Peripheral artery disease is the simplest trap to rule out and the most serious to miss. Its description is the point-by-point opposite of that of the syndrome: the pain arises on exertion and settles at rest. Polonsky and McDermott give the operational definition: calf pain on exertion, which does not begin at rest, and which settles in under ten minutes of stopping14.

Akathisia: look at the prescription

Drug-induced akathisia shares with the syndrome an urge to move relieved by movement. Three elements separate them in practice.

• The temporal link with a drug. That is the main element. Antipsychotics, antiemetics such as metoclopramide, certain antidepressants. The AASM guideline in fact places review of the prescription as the very first step of management, explicitly naming antidopaminergic, serotonergic and antihistamine drugs7.
• The absence of a circadian rhythm. Akathisia has no marked evening preference: it occupies the whole day.
• The global character of the restlessness. The patient does not describe their legs: they describe an inner tension, an inability to stay put, that makes them pace or rock.

This distinction has a direct consequence. In a patient on an antipsychotic, a suggestive picture may be both at once : the antidopaminergic drug worsens a pre-existing syndrome while also being able to induce akathisia. It is not for the physiotherapist to decide, but it is for them to report the chronology if they have spotted it.

A four-step approach

Why does restless legs syndrome warrant a ferritin test?

Because it is the only common, accessible and correctable secondary cause, and because the thresholds that trigger supplementation are not those applied to the rest of the population.

Iron deficiency is central to the accepted pathophysiology. The reference review in Nature Reviews Disease Primers places brain iron deficiency among the main mechanisms under study, alongside dysfunction of the dopaminergic and nociceptive systems and alterations of the adenosinergic and glutamatergic pathways8. Not all of these hypotheses are established; the link with iron, for its part, has demonstrated therapeutic consequences.

The figure to remember: 23.9 % of patients with iron deficiency anaemia have restless legs syndrome5. One patient in four. On adult iron deficiency taken in itself (its causes, its work-up, its treatments), a recent review takes stock20.

Thresholds that are not those of the general population

This is the most important point of this chapter, and the one most often missed. An iron work-up may be read as “normal” by the laboratory reference values while still warranting supplementation in a patient who has this syndrome. The AASM 2025 good practice statement is explicit: “these iron supplementation recommendations are different from those for the general population”7.

One point about sampling conditions, often neglected and yet decisive: the AASM asks that the test be done in the morning, avoiding any supplement and any iron-containing food in the preceding 24 hours7. A patient already taking iron who is tested without that precaution will return an uninterpretable result.

What supplementation actually brings

The evidence is solid for the intravenous route and more modest for the oral route. The international IRLSSG group retained 31 publications out of 299 examined, including four class I studies in adults : one for intravenous iron sucrose and three for ferric carboxymaltose6. Ferric carboxymaltose at 1,000 mg is effective in moderate to severe forms for a ferritin below 300 µg/L, and can be used first line in adults. Oral iron, at 65 mg of elemental iron, is judged possibly effective for a ferritin of 75 µg/L or below: the shade of vocabulary reflects a lower level of evidence6.

The 2025 AASM guideline takes up that hierarchy: intravenous ferric carboxymaltose carries a strong recommendation, while oral ferrous sulfate, low molecular weight iron dextran and ferumoxytol receive conditional recommendations7.

The other secondary causes

Iron is not the only one. The 2026 JAMA review and the international group's report list the settings in which a syndrome appears or worsens519 : end-stage renal disease, pregnancy, peripheral neuropathy, inflammatory diseases. The spondyloarthritis case presented later illustrates a composite mechanism: the inflammatory anaemia of rheumatic diseases and the iron deficiency that accompanies it may contribute to the appearance of the syndrome17.

To this must be added the iatrogenic causes, which are the easiest to correct. The AASM places correction of aggravating factors second in its good practice statement: alcohol, caffeine, antihistamines, serotonergic and antidopaminergic drugs, and untreated sleep apnoea7. The JAMA review names in particular the centrally acting H1 antihistamines such as diphenhydramine5: present in over-the-counter sleeping aids, which patients take precisely because they sleep badly.

Why is sleep often the real complaint?

Because it is what drives people to consult, because it is what degrades quality of life, and because the way the patient connects their insomnia with their legs is the best diagnostic discriminator available.

Restless legs syndrome is classified by the World Health Organization among the sleep-related movement disorders, in the chapter on sleep-wake disorders26. That classification is not administrative: it says where the damage lies.

The mechanics are simple. The symptoms peak in the evening and at night, precisely when the patient settles down to sleep. The urge to move prevents falling asleep; if they wake, it prevents falling back asleep. The 2026 JAMA review puts it thus: the symptoms “may lead to difficulty falling asleep, staying asleep, or returning to sleep”5.

The figure that should stay in your memory

In the mimics study, two lines concern sleep and they do not say the same thing2.

• Sleep disturbance, all causes together: 87 % of true patients, 58 % of mimics. The gap is highly significant, but sleeping badly is common in both groups.
• Sleep disturbance that the patient attributes to their legs: 29 % against 5 %. A ratio of nearly six.

It is the clearest discriminator in the table. It is not enough to ask “do you sleep well?”, almost everyone answers no. The question that counts is: “What stops you sleeping?”, asked openly, without prompting the answer. The patient who spontaneously answers “my legs, I have to move them” has just supplied the most discriminating element of the interview.

An impact comparable to that of a chronic disease

The REST study measured the quality of life of the patients affected using the SF-36 questionnaire. The scores were significantly below population norms and lay at the level of those of patients with other chronic diseases3. The 2026 JAMA review adds measured psychiatric comorbidity: depression in 30.4 % of patients, and suicidal ideation or self-harm with an incidence of 0.35 cases per 1,000 person-years5.

That last figure deserves to be read correctly: it is a low incidence in absolute terms, but a measured one, reported in a reference review. It is a reminder that a condition that deprives people of sleep for years is not a benign nuisance.

What can the physiotherapist actually bring?

Data exist, they are favourable, and they are fragile. To present them otherwise would be dishonest, and the systematic review that gathers them says so itself.

The reference is a systematic review of randomised controlled trials of non-drug interventions, conducted by the physiotherapy department of Monash University and published in Disability and Rehabilitation9. Its method and its limits must be stated before its results: 442 articles examined, eleven trials retained. Eleven trials for the whole non-drug field.

“Few studies were identified and the quality of the evidence was not high.”
Harrison, Keating and Morgan, Disability and Rehabilitation, 20199

The authors add a warning one does not often read from the pen of physiotherapists, and which does them credit: “Non-drug interventions in restless legs syndrome may produce placebo effects, and rehabilitation professionals should control for that possibility in their future investigations”9.

What works, what does not

The review distinguishes two outcomes, and a modality can be effective on one without being so on the other9.

• Superior to control on syndrome severity: repetitive transcranial magnetic stimulation, exercise, compression devices, positional release (counterstrain), infrared therapy, standard acupuncture.
• Ineffective on severity: vibrating pads, cryotherapy, transcranial direct current stimulation.
• Improve some sleep outcomes without improving severity: vibrating pads, cryotherapy, yoga, compression devices, acupuncture.

Exercise, trial by trial

The first randomised controlled trial of an exercise programme dates from 2006. Its results are favourable and its sample size calls for caution: 41 patients randomised, 28 who actually started, and 23 who finished: eleven in the exercise group, twelve in the control group. The programme lasted twelve weeks, with three weekly sessions combining aerobic work and lower limb strengthening. The improvement was significant on the international group's severity scale (p = 0.001)10.

A second trial, run by the same Australian physiotherapy team as the systematic review, tested stretching and lower limb fatiguing exercises designed to provoke therapeutic tremors, against discussion groups. Sample size: eighteen participants, nine against nine, over six weeks11.

In haemodialysis patients, the data are more plentiful. A meta-analysis brought together fifteen randomised controlled trials and concludes to a favourable effect of physical training on the syndrome (p < 0.001), on depression and on fatigue. An important caveat: this is a dialysis population, in whom the syndrome is secondary: those results do not transfer mechanically to the primary syndrome12.

Pneumatic compression, the cleanest data in the field

Methodologically it is the most solid trial that rehabilitation has on this subject: prospective, randomised, double-blind, against a sham device. Thirty-five subjects wore a pneumatic compression device, active or sham, for at least one hour a day before the usual time symptoms appeared, for one month13.

In the active group, the severity score fell from 14.1 ± 3.9 to 8.4 ± 3.4 (p = 0.006) and the Johns Hopkins scale from 2.2 ± 0.5 to 1.2 ± 0.7 (p = 0.01), with greater improvement than the sham group across all the variables measured, including daytime sleepiness and fatigue. The most telling result: complete relief in a third of the subjects in the active group, and in none of the sham group13.

Modalities and level of evidence

In practice, what a programme looks like

The available data do not allow an optimal protocol to be defined, no comparative trial of exercise regimens exists, and the authors of the dialysis meta-analysis call for precisely those trials12. The only programme tested against control in the primary syndrome is Aukerman's, whose parameters can be taken up since they are the only ones to have been evaluated10 :

• Frequency: three sessions a week.
• Duration: twelve weeks, that is the point at which the effect was measured, not a magic threshold.
• Content: aerobic work combined with lower limb strengthening.
• Progression: progressive conditioning, as described in the trial.

Two clinical points of attention, which come from common sense and not from trials. First, intense exercise in the evening is reported by many patients as making things worse ; in the absence of data, it is reasonable to place sessions during the day and to listen to the patient on this point. Second, exercise also serves sleep and mood, two dimensions impaired here: the dialysis meta-analysis in fact finds an effect on depression and fatigue at least as clear as on the syndrome itself12.

Which medical treatments, and what changed in 2025?

A major reversal has taken place: the drugs that had been first line for twenty years no longer are. The physiotherapist does not prescribe, but they follow treated patients, and they are well placed to spot the complication that prompted that reversal.

The American Academy of Sleep Medicine guideline published in January 2025 is the reference text7. It rests on a systematic review and an assessment of the evidence using GRADE methodology. On the European side, the German sleep and neurological societies published their own guidelines in 2024, in abbreviated form18.

Augmentation: the complication the physio can spot

It is the reason for the reversal, and it is a phenomenon whose description makes it recognisable in the clinic.

Augmentation is an iatrogenic worsening of the syndrome induced by the dopaminergic treatment itself. Its annual incidence is 7 - 10 % on these drugs5. The Nature Reviews Disease Primers review already described it in 2021 as a serious worsening occurring with prolonged use, even though dopaminergic treatment is “initially very effective”8. The phenomenon is now the subject of work entirely devoted to it, including an observational study of a Spanish cohort22.

What should alert you in a patient treated for months or years:

• the symptoms start earlier in the day than before the treatment, that is the most characteristic sign;
• they spread to other territories, arms, trunk;
• they are more intense than before treatment was started;
• the interval before onset after settling at rest shortens ;
• the patient reports that they “need to increase the dose”, and that each increase relieves briefly before it all starts again.

Which special situations should you know about?

Three populations change the reading of the picture: the pregnant woman, the child, and the patient already followed for a neurological or renal disease. In each, the syndrome is commoner and its management differs.

Pregnancy

This is the situation in which the prevalence rises fastest and in which it is most often put down to something else. The 2026 JAMA review gives 22 % in the third trimester5, that is nearly one woman in four in late pregnancy. More recent work has measured the prevalence during pregnancy and postpartum21.

Three elements explain that concentration: iron requirements rise and iron deficiency is common, hormonal changes intervene, and the mechanical discomfort of late pregnancy blurs the interpretation. A pregnant woman who says she sleeps badly because of her legs is very often heard through the lens of cramps and circulation, when the question of the syndrome arises.

The child and the adolescent

The paediatric criteria were merged with the adult ones in 20141, which simplifies the approach but does not settle the main difficulty: a child describes their sensations poorly, and those around them spontaneously translate it into “growing pains”.

The question of whether some of the children labelled with “growing pains” in fact have restless legs syndrome has been raised in the literature since 200225, taken up in a review in 201524, and again in 202523. It is not settled, but it is established enough to change how you listen: faced with a child with painful legs in the evening, looking for the urge to move and for relief by movement costs two questions.

On the iron side, the paediatric threshold differs: supplementation is started for a ferritin below 50 ng/mL, by the oral or intravenous route7. The evidence of efficacy in children remains insufficient to conclude, for the oral as for the intravenous route6.

The detail of this paediatric differential is covered in our article on growth-related conditions of the lower limb in children and adolescents.

End-stage renal disease

Prevalence of 24 %5. It is the population in which the exercise data are the fullest: fifteen randomised controlled trials in a meta-analysis concluding to a favourable effect on the syndrome, on depression and on fatigue12. A physiotherapist working in a dialysis unit therefore has, for that particular population, a fuller evidence base than for the primary syndrome.

The AASM guideline contains specific lines: gabapentin, intravenous iron sucrose for a ferritin below 200 ng/mL with a transferrin saturation under 20 %, and vitamin C, all as conditional recommendations7. Note that the ferritin threshold there differs from that of the general population of affected patients.

Chronic neurological diseases

Here, multiple sclerosis comes top of all the settings recorded, with 27,5 %5. Then Parkinson's disease follows at 20 %, in a particular configuration: the patient is already receiving dopaminergic treatment, which makes any worsening delicate to interpret, is it the course of the disease, an associated restless legs syndrome, or augmentation? That question is for the neurologist, but the chronology reported by the physiotherapist who sees the patient every week often sheds more light on it than a six-monthly consultation.

Lastly, peripheral neuropathy, at 21.5 %, poses the opposite problem: these two entities appear in each other's differential and frequently coexist. A neuropathic patient can have both, and an abnormal neurological examination does not rule out an added syndrome. That is the fifth criterion in all its difficulty: it requires that the symptoms not be solely accounted for by the other condition1.

What do concrete clinical cases teach us?

Two published observations, chosen because each illustrates a different mechanism of underdiagnosis. These are genuine cases reported in the indexed literature, with their identifiers.

How do you apply all this in the clinic?

Three moments: spotting it during an assessment done for something else, the interview that confirms or rules out, and the handover that decides what comes next.

Spotting it, without looking for it

This syndrome is not screened for: it is noticed. The patients affected come for low back pain, knee osteoarthritis, post-operative rehabilitation. The sentence that should catch your ear is always from the same family: “in the evening, I have to move my legs”, “I get restlessness”, “I sleep badly, it works away in my legs”, “my wife says I fidget all night”.

Two questions are enough to decide whether to go further:

1. “When it takes you, do you have pain, or do you have an urge to move?”
2. “What stops you sleeping?”, without prompting the answer.

Confirming, in three minutes

Handing over, by naming it

This is the act with the most value, and it costs only a few lines. A letter or message to the family doctor containing:

• the five criteria, taken one by one, with what the patient answered;
• the impact on sleep, specifying that the patient attributes it to their legs;
• the mimics you have ruled out and on what grounds, that is what distinguishes a reasoned referral from a hunch;
• mention of an iron work-up, ferritin and transferrin saturation, recalling that the interpretation thresholds are specific to this condition7 ;
• where relevant, the drugs you have spotted that could make it worse.

Recall the order of magnitude that justifies this care: in the REST study, 81 % of the patients affected had mentioned their symptoms to a general practitioner and 6.2 % had received the diagnosis3. The difference does not lie in the doctor's competence: it lies in the way the information reaches them. “His legs hurt in the evening” and “he meets the five criteria for restless legs syndrome, with insomnia attributed to the legs, mimics ruled out” do not trigger the same consultation.

Frequently asked questions

Restless legs syndrome and Willis-Ekbom disease, are they the same thing?

Yes. They are two names for the same entity. “Willis-Ekbom disease” honours Thomas Willis, who gave a description of it in the 17th century, and Karl-Axel Ekbom, who characterised it in the 20th. The 2014 international criteria in fact use the double name in their title1. The World Health Organization retains “restless legs syndrome” as the official label of code 7A8026.

Is a blood test needed to make the diagnosis?

No for the diagnosis, yes for management. The diagnosis is entirely clinical and rests on the five criteria1 ; polysomnography is not recommended for making it5. Once the diagnosis is raised, however, the iron work-up is part of the standard assessment: the AASM asks that ferritin and transferrin saturation be measured regularly in every patient with a clinically significant syndrome7.

Can a patient have both a neuropathy and restless legs syndrome?

Yes, and it is common: 21.5 % of patients with peripheral neuropathy also have this syndrome5. The fifth criterion does not require the absence of any other condition, but that the symptoms not be accounted for solely by it1. In practice, a neuropathic patient who also describes an urge to move appearing at rest in the evening, relieved by walking and recurring on becoming still again, probably has both.

Are evening stretches useful?

There is no controlled trial specific to stretching in this indication. The systematic review of non-drug interventions does not identify it as a separately evaluated modality9. They are risk-free and many patients do them spontaneously because movement relieves, which is criterion 3, not a therapeutic effect. Offering them is reasonable; presenting them as a treatment is not.

Does magnesium work?

It is not established. Magnesium appears neither among the recommended treatments nor among those explicitly advised against in the 2025 AASM guideline7: there is simply not enough evidence to decide. What is established, on the other hand, is the role of iron, with precise thresholds. Directing a patient towards a ferritin test is well founded; advising them to take magnesium is not.

Why does my patient say their treatment works less well than before?

Two possible explanations, and they call for opposite responses. It may be the natural course of the disease. It may also be augmentation, an iatrogenic worsening induced by the dopaminergic treatment itself, whose annual incidence is 7 to 10 %5. The sign that points to it: the symptoms start earlier in the day than before the treatment, sometimes spread to the arms, and each dose increase relieves only briefly. That information must go back to the prescriber, without ever changing or stopping the treatment, since abrupt withdrawal risks severe rebound.

Can it be cured?

The primary syndrome is a chronic condition, classified by the 2014 criteria as chronic-persistent or intermittent1. The secondary forms, for their part, can regress when their cause is corrected: that is exactly what the iron deficiency anaemia case shows, where iron supplementation markedly improved the symptoms16. That is one more reason for the iron work-up not to be put off.

Can a physiotherapist make this diagnosis?

They can recognise it, document it and refer it; the medical diagnosis belongs to the doctor. That distinction is not a form of words: it is useful to the patient. A physiotherapist who writes “the five criteria are met, with insomnia attributed to the legs, and I have ruled out cramps, venous insufficiency and claudication” hands over usable clinical work. The REST figures show that it is precisely at that step that the diagnosis is lost3.