Restless Legs at Night: Why Your Legs Won't Settle…

It is 11:40 pm and your legs will not be still. Not because you are anxious or wired, but because there is an ache, a crawl, a pull somewhere deep in your calves that only gets better when you move. You shift, you stretch, you finally get up and pace the hallway. Ten minutes later you are back in bed, and it starts again.

Most people who live with this assume it is a quirk, a bad habit, or just something that happens at night. It has a name, a formal set of diagnostic criteria, and in a large share of cases a treatable explanation that nobody has ever tested for.

TL;DR: Restless legs syndrome (RLS) is a neurological condition defined by five criteria: an urge to move the legs usually with uncomfortable sensations, symptoms that begin or worsen at rest, relief with movement, a clear evening-and-night pattern, and symptoms not explained by another condition. It is common — population studies across North America and Western Europe report prevalence anywhere from 4% to 29% of adults, averaging around 14.5%, and it is more common in women and with age. The strongest known mechanism is brain iron deficiency, which is why the 2025 American Academy of Sleep Medicine guideline says every patient with clinically significant RLS should have ferritin and transferrin saturation tested, ideally in the morning and without iron supplements for 24 hours beforehand. Iron treatment is recommended when ferritin is ≤75 ng/mL or transferrin saturation is below 20%. The guideline also says the first step in management is removing things that make it worse: alcohol, caffeine, antihistamines, serotonergic and antidopaminergic medications, and untreated sleep apnea. Dopamine agonists, once the go-to treatment, have been downgraded because long-term use causes augmentation. Your watch cannot diagnose RLS and does not detect periodic limb movements, but it can show you the signature: fragmented sleep, more time awake after sleep onset, a higher sleeping heart rate, and depressed HRV. Those patterns are worth tracking, but a blood test and a clinician are what actually solve this.

YouTube: Related video

What RLS actually is (and how to tell it apart from ordinary restlessness)

RLS is not a sleep disorder in the way insomnia is, and it is not a circulation problem. It is a neurological condition with a sensory component and a movement component. The NIH describes it as an irresistible urge to move the legs, accompanied by sensations that people describe as aching, throbbing, pulling, itching, crawling or creeping. It usually affects both legs, sometimes one, occasionally the arms.

The international consensus criteria, updated in 2014, define it with five requirements. All five should be present:

An urge to move the legs, usually accompanied by or caused by uncomfortable and unpleasant sensations.
The urge or the sensations start or get worse during rest — lying down or sitting still.
Movement gives relief. Walking or stretching reduces the symptoms, at least while you keep moving.
A clear evening or night pattern. Symptoms are worse in the evening and night than in the morning, or only appear then.
It is not explained by something else. Leg cramps, arthritis, swelling, positional discomfort, habitual foot tapping and other conditions should be ruled out first.

That fourth criterion is why this is so often dismissed. Symptoms show up in the late afternoon and evening, and the NIH notes there is often a distinct symptom-free window in the early morning. So people wake up fine, feel fine all day, and only start pacing the floor at bedtime — which makes it easy to write off as restlessness or anxiety. That fifth criterion is new in the 2014 update, and it matters, because a lot of things feel like RLS without being it.

The other thing worth knowing is that RLS and sleep problems are not separate issues you happen to have together. Difficulty falling asleep and staying asleep is part of the clinical picture, and anything that reduces sleep tends to make the symptoms worse.

How common is it?

More common than most people assume, and the estimates are messier than you would expect.

A systematic review of population-based studies across 16 North American and Western European countries — 34 studies, including over 230,000 participants combined — found reported prevalence ranging from 4% to 29% of adults, averaging 14.5%. Rates were higher in primary care populations than in randomly sampled general populations (19.5% versus 12.3%), and consistently higher in women and in older adults.

The enormous range is a real finding, not a rounding error. Studies used different severity thresholds and different diagnostic procedures, and the review is upfront that detection rates remain low in practice. The number most often quoted in clinical settings is somewhere around 5-10% of adults having at least mild symptoms, with a smaller group having symptoms severe enough to need treatment.

So if this sounds familiar, you are not a rare case. You are in a large group that mostly has not been formally diagnosed.

The iron connection, which is the part most people never hear about

This is the most actionable thing in this article.

Brain iron deficiency is the leading mechanistic explanation for RLS. The iron in question is not necessarily your blood haemoglobin — you can have completely normal haemoglobin and still have the low brain iron availability that drives symptoms. That is why ferritin (your stored iron) and transferrin saturation matter more here than a standard blood count.

The 2025 AASM clinical practice guideline makes this a good practice statement, not a suggestion:

Every patient with clinically significant RLS should have serum iron studies tested, including ferritin and transferrin saturation. Ideally in the morning, and avoiding all iron-containing supplements and foods for at least 24 hours before the draw, because those values swing otherwise.
Iron treatment should be considered when ferritin is ≤75 ng/mL or transferrin saturation is under 20%. That threshold is much higher than the usual "your iron is fine" cutoff in a routine blood panel, and it is the single most common reason people get told their iron is normal when it is not in the range that matters for RLS. Below 75, either oral or IV iron is reasonable; between 75 and 100, IV iron only.
In children, the threshold is ferritin below 50 ng/mL.

The IRLSSG iron treatment guidelines found that ferric carboxymaltose 1000 mg is effective for moderate to severe RLS in people with serum ferritin below 300 µg/l, and suggested it could even be considered first-line for adults. Oral iron (65 mg elemental iron) appears possibly effective at ferritin ≤75 µg/l — but oral iron absorption is often poor, which is one reason IV iron shows up so often in this condition.

We have written separately about how iron status quietly shapes your recovery data — iron deficiency without anemia explains the normal-haemoglobin trap, and why low ferritin shows up in your resting heart rate covers what the wearable side looks like.

Two things worth flagging. First, do not self-prescribe high-dose iron. Iron overload is a genuine risk, and the thresholds here are clinical decisions. Second, if you take iron, absorption is meaningfully better every other day than daily, and vitamin C helps while coffee, tea and calcium near the dose hurt.

Why it gets worse at night

RLS follows a circadian rhythm. Symptoms are mildest in the morning and climb through the afternoon into the evening, which is nearly the opposite of how most people expect a movement disorder to behave.

The practical consequence is a trap: the harder you try to fall asleep, the longer you lie still, and lying still is exactly the condition that makes symptoms worse. People who fight it end up with something that looks like insomnia — long sleep latency, frequent awakenings, and a growing dread of bedtime.

There is also a documented movement signature called periodic limb movements of sleep (PLMS): repetitive, stereotyped leg movements that occur during sleep, often every 20 to 40 seconds. PLMS are common in people with RLS, and the 2026 Continuum review notes an elevated periodic limb movement index is frequently seen in RLS patients. You do not feel them. Your bed partner often does.

What your watch can and cannot see

Let me be direct about the limits, because this is where wearables get oversold.

What it cannot do: no consumer watch detects periodic limb movements. Apple Watch does not measure leg movements in bed, and it does not diagnose or screen for RLS. A study validating actigraphic PLM detection used bilateral ankle actigraphy sampled at high frequency and compared it against same-night polysomnography in 166 patients, using a machine-learning classifier to separate mild from severe PLM index (<15 versus ≥15 movements per hour, with finer severity bands up to ≥50). That is a research-grade setup with a sensor on each ankle, and even so the authors present it as a proof of concept. Your wrist watch is not that.

What it can do, and this is not nothing: show you the downstream signature of a bad night, repeatedly.

Sleep fragmentation. More wake-ups after sleep onset, more time awake in bed, less consolidated sleep. If your sleep is fragmented in the same pattern most nights, that is worth investigating.
Sleeping heart rate. A restless night usually shows up as a higher average heart rate during sleep, because movement and arousals push it up.
HRV. Expect it to sit lower on the nights your legs were worst. This is a correlation, not a diagnosis, but it is measurable.
Sleep onset timing. If your recorded time-to-sleep keeps stretching out after the same bedtime, the pattern itself is information.

The honest framing: your watch is a symptom diary that fills itself in automatically. It tells you this is happening often enough to matter. It cannot tell you why.

A two-week log is more useful than a year of vague memory. Write down bedtime, a 0-10 rating of leg symptoms at bedtime, how long it took to fall asleep, and how many times you woke. Line that up against your watch's sleep and overnight heart rate data, and you have something a clinician can actually work with. That is the same logic behind Century: one daily score built from sleep, overnight heart rate and HRV, so you are reading a trend rather than guessing from a single bad night.

The things that reliably make it worse

The 2025 AASM guideline is unusually blunt here. The first step in managing RLS is not medication. It is removing the factors that exacerbate it:

Alcohol, especially in the evening.
Caffeine, particularly late in the day.
Antihistamines — yes, including the ones people take to sleep.
Serotonergic medications, such as many antidepressants.
Antidopaminergic medications used for nausea and some psychiatric conditions.
Untreated obstructive sleep apnea. If you snore heavily and wake unrefreshed, this needs to be sorted out, because the two conditions overlap and treating one often improves the other. Our guide to what snoring actually means covers the red flags.

Pregnancy is another big one: RLS is common in pregnancy, and treatment decisions there need their own risk-benefit conversation.

If you take any of the medication classes above and have RLS symptoms, do not stop them yourself. Bring the list to your doctor and ask whether an alternative exists. That conversation is often more effective than anything you can buy.

What actually helps

Start with the basics that have real evidence:

Get ferritin and transferrin saturation tested properly. Morning draw, no iron supplements for 24 hours beforehand. This is the highest-yield action on this page.
Move, in the evening, on purpose. A walk or some gentle mobility work before bed gives your legs the movement they are asking for on your terms, instead of at 1 am.
Warmth before bed. A warm bath or shower, or a heat pack on the calves, is a classic and low-risk intervention.
Consistent sleep and wake times. Irregular schedules make circadian symptoms worse. Sleep consistency matters more than most people think.
Leg massage and stretching in the evening. Not a cure, but consistently reported as helpful and essentially free.
Review your medications, as above.
Magnesium. Popular for this, thin on evidence. If you want the honest read on what the research supports, our magnesium deep dive goes through it. The AASM specifically recommends against valerian for RLS, so that one is off the table.

Then, if symptoms are clinically significant, there are real treatments:

Gabapentinoids. Gabapentin enacarbil, gabapentin and pregabalin all carry strong recommendations in the 2025 AASM guideline for adults with RLS, with moderate certainty of evidence. These are now the first-line pharmacological option, and they are especially useful when sleep is the main problem.
IV iron. Ferric carboxymaltose has a strong recommendation for adults with appropriate iron status. Low molecular weight iron dextran and ferumoxytol are conditional options.
Oral iron at ferritin below 75 ng/mL carries a conditional recommendation.
Bilateral peroneal nerve stimulation, a device-based treatment, has a conditional recommendation for adults, and the 2026 Continuum review notes it can be used as adjunct therapy in refractory cases.
Opioids, including extended-release oxycodone, have a conditional recommendation. Registry data has been reassuring, but this is a decision for a specialist.

And what is out: dopamine agonists. Levodopa, pramipexole, ropinirole and rotigotine are all now recommended against for standard use, because long-term use causes augmentation — the treatment starts making the condition worse, with symptoms arriving earlier in the day, spreading to other body parts, and intensifying. Cabergoline gets a strong recommendation against. This is one of the clearest reversals in sleep medicine in the last decade, and if you were put on a dopamine agonist years ago and your symptoms keep worsening, that history is worth a fresh conversation.

When to actually see someone

Book an appointment if:

Symptoms happen two or more nights a week, or interfere with falling or staying asleep.
You are tired during the day in a way that affects work or driving.
You have never had ferritin and transferrin saturation checked.
You snore, gasp, or wake unrefreshed, which raises the sleep apnea question.
You are pregnant, or you have kidney disease, both of which raise RLS risk and change treatment.
You are already on a dopamine agonist and your symptoms are getting worse or starting earlier in the day.

RLS is underdiagnosed and undertreated mostly because people assume it is not worth mentioning. It is. In many cases the workup is a blood test, some honest questions about your evening habits, and a conversation about which medication classes to avoid.

Bottom line

Restless legs at night is a real neurological condition with formal diagnostic criteria, a prevalence somewhere in the range of one in ten adults, and a mechanism that involves iron in the brain rather than restlessness or bad habits. The highest-value move is unglamorous: test ferritin and transferrin saturation properly, with the thresholds that apply to RLS rather than the ones from a routine panel.

Your watch cannot see the leg movements themselves, but it can show you the pattern of fragmented sleep, higher sleeping heart rate and lower HRV that goes with them — which is enough to stop you dismissing it. Track the nights, bring the data, and let a clinician handle the diagnosis and the iron.

Century AI helps you understand your body with a daily health score, recovery score, and sleep insights — using the watch you already wear.