A patient came to my clinic last month convinced she had restless legs syndrome. Her husband had told her she “thrashed around all night.” Her sleep tracker had flagged her as having “very restless sleep” for months. She’d tried magnesium, iron, even a brief trial of pramipexole prescribed by another doctor. Nothing helped.

She didn’t have RLS. She had untreated obstructive sleep apnoea with frequent arousals causing whole-body movement. Once we put her on appropriate therapy, the restlessness disappeared.

This is one of the most common diagnostic muddles I see. The vocabulary patients use is similar — restless, fidgety, can’t stay still — but the underlying conditions and the appropriate workups are very different. Let’s pull them apart.

What restless legs syndrome actually is

Restless legs syndrome (RLS), now formally called Willis-Ekbom disease, is a sensorimotor condition with four core diagnostic criteria. An urge to move the legs, usually accompanied by uncomfortable sensations. The urge gets worse at rest. The urge gets better with movement. The urge is worse in the evening or at night.

That’s it. All four have to be present. If your patient says “I can’t keep still in bed” but doesn’t describe an urge to move driven by uncomfortable sensations, you don’t have RLS. The sensations are often described as creeping, crawling, pulling, or like soda water in the legs. It’s not pain, exactly, though severe cases can be painful. It’s a deep, distressing need to move.

The condition affects roughly 5-10% of adults to some degree. It’s heritable in a meaningful proportion of cases, more common in women, and gets worse with age. It’s also a clinical diagnosis — sleep studies don’t diagnose RLS, though they can show the related periodic limb movements during sleep.

What “restless sleep” usually means

When patients or partners describe restless sleep, they’re usually talking about one of these:

Frequent arousal-related body movements. The patient turns over, repositions, briefly partially wakes, returns to sleep. This pattern is normal in small doses and pathological in larger ones. The most common driver is sleep-disordered breathing — apnoeas and hypopnoeas trigger arousals, arousals trigger movement.

Periodic limb movements during sleep (PLMS). Stereotyped, rhythmic leg movements occurring during sleep, every 20 to 40 seconds. These can occur with or without RLS, and they can also occur as a consequence of OSA. Whether they’re clinically significant depends on whether they’re causing arousals and whether the patient is symptomatic during the day.

Anxiety and hyperarousal. A patient who’s psychologically wound up at bedtime will often have a fragmented, restless sleep with frequent position changes. The partner observes the movement, the patient feels unrefreshed.

Pain. Musculoskeletal pain, neuropathic pain, fibromyalgia — anything that hurts when you lie still — produces restless sleep without any underlying RLS or sleep-disordered breathing.

Medication effects. Some antidepressants, particularly SSRIs and venlafaxine, can produce or worsen periodic limb movements. Stimulants taken too late in the day produce the same. The patient or their GP rarely connects the dots.

How I work through it in clinic

The first move is always to slow down and ask about the sensations themselves. “When you’re trying to fall asleep, do your legs feel uncomfortable in a way that makes you have to move them?” If the answer is no, it’s not RLS, no matter what the wearable or the partner says.

If the answer is yes, I work through the four diagnostic criteria. Then I check the mimics — peripheral neuropathy, akathisia from medication, leg cramps, positional discomfort. I check iron status. Ferritin under 75 micrograms per litre is widely accepted as a threshold for trial of iron supplementation in RLS, and it’s a common reversible factor.

If the patient describes restless sleep without the sensorimotor features of RLS, the workup shifts entirely. Sleep history, snoring, witnessed apnoeas, daytime sleepiness, weight history, neck circumference. A home sleep test or in-lab study depending on the clinical picture. Often the answer is straightforward OSA and the “restlessness” resolves with treatment.

Where wearables muddy the water

I’m not anti-wearable. They’ve helped patients notice patterns that brought them to clinic earlier. But they’ve also created a new diagnostic noise problem. Devices flag “restless sleep” using accelerometer data interpreted by a black-box algorithm. The label is meaningful in some contexts and useless in others. It tells you something happened. It doesn’t tell you what or why.

Patients increasingly come in with months of wearable data and a self-diagnosis. Part of clinical work now is gently recalibrating expectations — the data is useful, but it’s a starting point, not an answer. The Sleep Health Foundation has put out good general guidance on this for the public.

What changes the treatment

The reason this matters is that the treatments diverge sharply.

RLS responds to iron repletion where appropriate, then to alpha-2-delta ligands like gabapentin or pregabalin as first-line pharmacotherapy. Dopamine agonists are still used but augmentation — where the symptoms worsen and spread with long-term treatment — has made guidelines more cautious. Lifestyle changes around caffeine, alcohol, and certain medications matter too.

Restless sleep due to OSA needs OSA treatment — CPAP, mandibular advancement splint, weight management, surgery in selected cases. Treating it as RLS does nothing helpful.

Restless sleep due to pain or anxiety needs the underlying condition addressed, not a sleep diagnosis bolted on top.

The takeaway for patients

If you think you might have RLS, the question to ask yourself is whether you have an urge to move your legs driven by uncomfortable sensations, worse in the evening, relieved by movement. If yes, get a proper assessment. If no, but you or your partner notice you move a lot at night, get a sleep assessment — the most common cause isn’t RLS, it’s sleep-disordered breathing.

Either way, treating yourself based on a wearable label or a Google search is the slow route. The right diagnosis is the foundation of the right treatment, and it’s worth the time to get it sorted properly.