Does sleep meditation actually work? What the trials found

Sleep meditation is the default recommendation of the entire wellness industry, and the pooled numbers are blunter than the marketing: across 18 randomized trials of 1,654 people, meditation beat placebo-grade comparisons (effect size 0.33) and showed no clear advantage over proven sleep treatments (0.03, an estimate too uncertain to call in either direction). No bedtime technique we’ve ranked has a wider gap between what the research shows and what gets claimed on its behalf. We publish guided sleep meditations, so treat us accordingly. What follows is the whole picture, the unflattering parts included.

Start with the good trials

Two well-run studies are the ones everyone cites, and they deserve their reputation.

In a randomized trial published in JAMA Internal Medicine, 49 older adults with sleep disturbance got either a mindfulness program or sleep hygiene education: six weeks, two hours a week, plus homework. The mindfulness group’s Pittsburgh Sleep Quality Index scores fell from 10.2 to 7.4; the sleep hygiene group went from 10.2 to 9.1. The between-group difference was 1.8 points, effect size 0.89 (Black 2015).

In a trial of 54 adults with chronic insomnia, eight weeks of mindfulness-based stress reduction or mindfulness-based therapy for insomnia cut total wake time by about 44 minutes a night, against roughly one minute in the self-monitoring group (Ong 2014). Forty-four minutes is a real night’s difference.

If the story ended there, meditation would be the clear winner of this whole category. It doesn’t end there.

What happens when you pool everything

A 2019 meta-analysis gathered 18 randomized trials covering 1,654 participants and did something most reviews don’t: it split the comparison groups in two. Nonspecific active controls: things designed to match time and attention, controlling for placebo. And specific active controls: actual evidence-based sleep treatments.

Against nonspecific controls, mindfulness meditation helped, with an effect size of 0.33 and moderate-quality evidence. Against specific active controls, the effect was 0.03 — essentially nothing — with the confidence interval running from −0.43 to 0.49 (Rusch 2019).

A much larger review reached a bleaker conclusion. Screening nearly 19,000 citations down to 47 trials with 3,515 participants, it found moderate evidence that meditation helps anxiety, depression, and pain, and “low evidence of no effect or insufficient evidence of any effect” on sleep. It also found no evidence that meditation programs beat any active treatment, including drugs, exercise, and other behavioral therapies (Goyal 2014).

So: better than doing nothing, better than being told about sleep hygiene, and roughly a wash against treatments that actually work.

The mismatch nobody mentions

Here is the most important part, and it applies directly to what we make.

Go back and look at what those trials actually gave people. Black 2015: six weeks, two hours a week, in a room, with a teacher, plus assigned homework. Ong 2014: an eight-week structured course. These are training programs. Participants were learning a skill over months of accumulated practice.

That is not what “sleep meditation” means to most people now. It means pressing play on a twenty-minute track in bed. Those are different interventions that happen to share a word, and the evidence for the first does not transfer to the second.

The digital research doesn’t close the gap either. A 2025 review of 13 studies and 5,251 participants found digital sleep interventions improve sleep quality overall. But it pooled app-based mindfulness together with digital CBT-I, and CBT-I is the heavyweight in that mix (Lu 2025). You can’t extract “does the meditation track work” from a result like that.

To be blunt about our own category: we could not find a trial isolating a guided sleep meditation played at bedtime and showing it improves sleep. It might. It hasn’t been shown.

Meditation isn’t risk-free either

One more thing the wellness framing leaves out. A systematic review of 83 studies covering 6,703 participants found the prevalence of adverse events during or after meditation practice was 8.3% (most commonly anxiety, depression, and cognitive disturbances), and that these occurred in people with no previous mental health history (Farias 2020). That figure blends very different designs, though: 3.7% across the experimental studies, 33.2% across the observational ones.

For most people, a calm ten minutes before bed is exactly as harmless as it sounds. But “meditation can’t hurt” is not something the literature supports, and if turning attention inward at night makes your thoughts louder rather than quieter, that’s a documented response and not a personal failing.

So what is working in a good bedtime track?

Probably not the meditation part.

The best-supported bedtime mechanism in this whole area isn’t mindful awareness — it’s imagery distraction. When 41 people with insomnia were told to occupy their minds with vivid imagery at bedtime, they fell asleep faster and reported less frequent, less distressing pre-sleep thinking. A general instruction to distract themselves didn’t achieve it (Harvey 2002).

That’s an important distinction. Classical mindfulness asks you to notice your thoughts without engaging. Imagery asks you to go somewhere else entirely. For a mind that’s spinning at 11pm, the evidence favors somewhere else, which is also why sleep stories have a better mechanistic case than their thin direct evidence would suggest.

Most popular “sleep meditations” are, in practice, guided imagery with a calm voice: a beach, a forest path, a slow scan down through the body. If they work, that’s likely the reason, and it means the ones written as a place to go should beat the ones written as an awareness exercise.

What about sleep meditation podcasts?

They’re the free version of the same thing, and the practical advice is the same as for any bedtime audio: look for episodes long enough that they don’t end while you’re still awake, a host who doesn’t get louder for the ad read, and a feed that doesn’t autoplay something jarring next. A meditation that ends in silence at minute twelve, leaving you alert in the dark, has undone its own work.

Using it well

• Treat it as a wind-down habit, not a cure. The trial evidence supports meditation training as a general practice more than it supports a track at bedtime.
• Prefer somewhere-to-go over notice-your-thoughts. Body scans and guided imagery suit bedtime better than open-monitoring practice.
• Don’t lie there indefinitely. If you’re still awake and frustrated after twenty minutes or so, get up and do something dull in dim light.
• For chronic insomnia, ask about CBT-I. It’s the first-line treatment, and it’s the thing meditation didn’t beat in the pooled analyses.
• If it makes things louder, stop. That’s a documented response, not a sign you’re doing it wrong.

What we make

Slumber is guided sleep meditations and stories (the imagery-heavy kind, with breathing and muscle-relaxation exercises built in) in a library of 2,200+ tracks with new ones daily. Get Sleepy is the free podcast version of the same idea, easing in with soft music and a short meditation before the story starts; Premium removes the ads.

We’d rather you knew what the evidence does and doesn’t say before you pay for any of it. If a free podcast episode does the job, that’s a fine outcome.

---

Sources: Black et al. 2015 · Ong et al. 2014 · Rusch et al. 2019 · Goyal et al. 2014 · Lu et al. 2025 · Farias et al. 2020 · Harvey & Payne 2002. More on where meditation lands in the full ranking. This is general information, not medical advice — persistent insomnia is worth taking to a doctor.