The Sandman stands on a moonlit cobbled street holding an empty sack beside a well of silver bubbles, as a few windows go dark with sleep.
Back to the library

Energy & sleep

Sleep

Researchers have measured sleep before and after courses of hyperbaric oxygen and Alpha-Stim. Hyperbaric oxygen studies point to better sleep scores, mostly in people who were sleeping badly to begin with, though none used a sham comparison. The main Alpha-Stim insomnia trial, run at Walter Reed in the United States, found no clear overall benefit.

Evidence so far
Early
Reading timeAbout 12 minutes

Start here

Why this is being studied

Poor sleep is rarely just about sleep. It often comes with stress, pain, illness or a nervous system that will not settle at night, and it feeds back into all of them. Two of the things we offer at Breathe have been studied for sleep. Hyperbaric oxygen has a randomised trial in people with long-standing insomnia at high altitude, seven small trials in Parkinson's disease and a large group from Israel who completed a full course. Alpha-Stim, a small device that sends a gentle current through ear clips, has a sham-controlled insomnia trial in the US military. The results so far are encouraging for hyperbaric oxygen and mixed for Alpha-Stim, and neither has yet been tested in a large sham-controlled trial of everyday insomnia. If you stop breathing at night or wake up gasping, see our sleep apnoea page and speak to your GP.

Where the evidence stands

Flag text

The mechanism

What is going on, and why this might help

What is going on in the body

Good sleep depends on a well-timed body clock and a stress response that can switch off at night. When the nervous system stays on high alert, through stress, pain, illness or low oxygen, it becomes harder to fall asleep and stay asleep. Lack of sleep then raises stress hormones and inflammation, which makes the next night harder still.

Where hyperbaric oxygen comes in

Breathing oxygen under pressure dissolves far more of it into the blood plasma. Researchers propose that repeated sessions improve blood flow and energy production in under-performing brain tissue, reduce inflammatory signalling and help settle the stress response. At altitude, the simpler explanation is that a course corrects an oxygen shortfall that was disturbing sleep. Seven small trials in Parkinson's disease (pooled by Tan and colleagues in 2024) also reported longer, deeper sleep when hyperbaric oxygen was added to medication, though none used a sham.

Where Alpha-Stim comes in

Alpha-Stim sends a very small alternating current through ear clips. Brain recordings with similar devices show a shift towards alpha waves, the slower rhythm linked to calm wakefulness, which may make it easier to wind down. A 2023 review of sham-controlled trials found a benefit for sleep in people with anxiety, but it rested on only three small trials. When insomnia itself was tested directly at Walter Reed, the overall result was not clear.

How solid is this

For hyperbaric oxygen, the results are consistent but come from trials without a sham. For Alpha-Stim, the evidence for sleep on its own is mixed and small. Neither has yet been tested in a large sham-controlled trial of everyday insomnia, so we describe this as early.

Explainer illustration

The studies

Every study is laid out the same way: what the researchers wanted to find out, who took part, what they actually did, what they found, why it might work and where the limits are. No jargon, and a link to the original if you want to read it yourself.

Each study is laid out the same way: what the researchers asked, who took part, what they did, what they found, why it might work, and where the limits are. The clinician layer at the end of each one carries the numbers.

Paper 1

Hyperbaric oxygen treatment for chronic insomnia at high altitude: a prospective, randomized, open-label, parallel-group trial

Sun et al. · Travel Medicine and Infectious Disease · 2025 · randomised controlled trial, n=80 (hyperbaric oxygen)

Read the paper

What they asked

Sleeping badly is one of the most common problems for people who live or work at high altitude, where each breath carries less oxygen. This trial asked whether a short course of hyperbaric oxygen could improve sleep in adults whose insomnia had become long-standing at altitude.

Who took part

Eighty adults with chronic insomnia at high altitude were randomised one to one, half to hyperbaric oxygen and half to a control group that did not have it. They were young, with a median age of 25. The trial was registered in China (ChiCTR2100046917).

What they did

The oxygen group had a 10-day course of hyperbaric oxygen. The published abstract does not give the pressure or session length. Sleep was measured with two standard questionnaires: the Pittsburgh Sleep Quality Index (PSQI, scored 0 to 21, where higher means worse sleep) and the Insomnia Severity Index (ISI). Fatigue, anxiety and depression were also scored.

What they found

After 10 days, the average PSQI score was 4.6 in the oxygen group against 9.1 in the control group, an adjusted difference of about four points. A PSQI of 5 or below is usually counted as good sleep. Insomnia severity scores were 5.0 against 9.8. Fatigue, anxiety and depression scores also improved more in the oxygen group. No serious side effects were reported in either group.

Why it might work

At altitude the body is short of oxygen through the night, which fragments sleep and keeps the stress response switched on. A course of hyperbaric oxygen may temporarily correct that shortfall and calm the stress response. This reasoning is specific to altitude and may not carry over to insomnia at sea level.

Limits

The trial was open-label: everyone knew who was having treatment, and the control group had no sham, so expectation could have influenced questionnaire answers. Follow-up was immediate, so we do not know how long the effect lasted. The participants were young people at altitude, which is a very different group from most people with poor sleep in the UK.

Paper 2

The effect of hyperbaric oxygen therapy on sleep quality across diverse patient populations

Doenyas-Barak et al. · Frontiers in Neurology · 2026 · retrospective cohort, n=395 (hyperbaric oxygen)

Read the paper

What they asked

Many people say they sleep better during a course of hyperbaric oxygen, but most trials have not measured it. This team went back through the records of everyone who had completed a full course at their centre in Israel to see whether sleep scores changed, and who changed most.

Who took part

395 adults: 180 having hyperbaric oxygen for healthy ageing, 92 with long COVID and 123 with PTSD. Fourteen people with incomplete questionnaires were left out.

What they did

Everyone had 60 sessions at 2 ATA, 90 minutes each, five days a week. Sleep was scored with the PSQI questionnaire before and after the course. There was no comparison group.

What they found

Average sleep scores improved in all three groups: from 7.1 to 6.5 in healthy ageing, 8.4 to 7.7 in long COVID and 12.1 to 10.9 in PTSD (lower is better). Falling asleep faster, fewer disturbances and better self-rated sleep were the most consistent changes. The clearest pattern was that people who slept badly at the start improved, while people who already slept well changed very little. Use of sleep medication did not change.

Why it might work

The authors link the changes to effects of repeated sessions on brain blood flow, inflammation and the stress response, drawing on their earlier imaging work. This study did not test those mechanisms.

Limits

With no control group, the authors describe the results as associations. The changes are modest, and some of the improvement in poor sleepers may be regression to the mean (extreme scores tend to settle on retesting), expectation or the routine of daily visits. Sleep was self-reported only, with no overnight measurement. The data come from the authors' own treatment centre.

Paper 3

Efficacy of cranial electric stimulation for the treatment of insomnia: a randomized pilot study

Lande and Gragnani · Complementary Therapies in Medicine · 2013 · randomised double-blind sham-controlled pilot, n=57 (Alpha-Stim)

Read the paper

What they asked

Insomnia is very common among serving military personnel, and many want to avoid sleeping tablets. This pilot trial at Walter Reed National Military Medical Center in Maryland asked whether a week of Alpha-Stim could help service members sleep better than an identical device that delivered no current.

Who took part

Fifty-seven active-duty service members already receiving mental health care, all with significant insomnia, were randomised: 28 to an active device and 29 to a sham. Only 11 were women. Forty-four completed all five sessions.

What they did

Everyone used an Alpha-Stim with ear clips for 60 minutes a day for five days. Active devices were set at 100 microamps, below the level most people can feel, and the manufacturer coded the devices so that neither participants nor researchers knew which was which. Participants kept sleep diaries recording how long it took to fall asleep, how long they slept and how often they woke, after each session and again 3 and 10 days later.

What they found

Overall there was no clear difference between active and sham for falling asleep, waking in the night or total sleep. The active group's sleep time leaned in the right direction after three sessions, about 43 minutes more against 19 minutes less with sham, but this did not reach statistical significance. In a later look at men only, those using the active device reported roughly 50 to 60 minutes more sleep on some nights, but this had faded by the follow-up days. Two people stopped early, one with a headache and one reporting worse sleep.

Why it might work

The proposed idea is that the gentle current shifts brain activity towards a calmer, slower rhythm, making it easier to wind down. The authors suggest a week may simply be too short, or the dose too low, to change sleep reliably.

Limits

A small pilot with few women, sleep measured only by diary, and a short course. The men-only result was not planned in advance, so it needs testing in a new trial. The authors themselves say the findings do not establish Alpha-Stim as a treatment for chronic insomnia. The trial was run under a research agreement with the device manufacturer, which also supplied and coded the devices.