The Benefits of Red Light Therapy: What the Research Actually Shows

Written by the Lumiglo research team. Published July 2026. Every figure below is sourced, and where the evidence is thin we say so.

You have seen the masks. You probably decided they cost too much for something you could not verify. So here are the benefits of red light therapy, each one with the trial behind it, the sample size, and the measured result.

Including the ones with no trial at all, because that half of the subject is where the money gets wasted.

What is in this guide

• The short answer
• What red light therapy is, and what it does to skin
• Red light therapy benefits, ranked by evidence
• What red light therapy has not been shown to do
• How good is this literature, honestly
• Face or body: why the same device does not do the same job
• At-home masks compared with in-office treatments
• How to read a spec sheet before you buy
• How often and how long, according to the trials
• A realistic timeline: week 4, week 12, month 6
• Safety, side effects, and who should be careful
• Where it fits in a skincare routine
• The bottom line
• Frequently asked questions

What red light therapy is, and what it does to skin

Red light therapy, also called photobiomodulation or low-level light therapy, uses specific wavelengths of visible red and near-infrared light to influence how skin cells behave. It uses no ultraviolet light, it does not heat the tissue, and it is non-invasive.

Red and near-infrared do different jobs

Two bands carry almost all the published evidence, and they are not interchangeable.

• Red, roughly 630 to 660 nm. Penetrates about 2 to 4 mm, which is the epidermis and the upper dermis. That is where fibroblasts sit, and fibroblasts are what produce collagen and elastin. This is the band behind the wrinkle and firmness research.
• Near-infrared, roughly 830 to 850 nm. Penetrates further, into deeper tissue and the vascular layer. Invisible to the eye, which is why people often assume the device has stopped working when the infrared mode is running.

The practical consequence is one that almost nobody spells out: red is the wavelength for facial skin, near-infrared is the wavelength for deeper tissue. A device that offers only near-infrared is not the right tool for fine lines, and a device that offers only red is not the tool for anything below the surface.

What happens inside the cell

The proposed mechanism is well described. Light at these wavelengths is absorbed by cytochrome c oxidase, an enzyme in the mitochondrial respiratory chain. That absorption is associated with increased ATP production, changes in reactive oxygen species signalling, and downstream fibroblast activity, which is where collagen and elastin come from.

Worth being precise here: the mechanism is plausible and supported, but a mechanism is not a result. Plenty of things change cellular signalling without changing what you see in the mirror. That is why the rest of this guide is organised around measured outcomes rather than around biology.

Red light therapy benefits, ranked by strength of evidence

Here is the table we wanted and could not find anywhere. Each row gives the benefit, the study behind it, how many people were in it, the protocol, the measured result, and how much weight we think it carries.

Fine lines and firmness: the strongest skin evidence

The 2014 trial is the one everyone cites, and rightly. It was randomized, it had a control group, and it measured collagen density with ultrasound rather than asking people how they felt. The treated group improved on wrinkle appearance and skin roughness. The control group did not.

The strongest evidence for a home-use mask rather than a clinic panel came in 2025, published in Medicine. Sixty participants, multicentre, randomized, double-blind, and controlled against a sham device, at 630 and 850 nm. Improvement in crow's feet was reported in 86.2 percent of the active group against 16.7 percent under the sham. A sham-controlled double-blind design is the closest this field gets to a proper test, and it is the study we would point to first.

Worth noting for the argument that runs through this whole guide: that trial delivered a maximum of around 10 mW/cm² per wavelength. It is not the powerful devices that carry the evidence.

A 2023 trial in Skin Research and Technology also ran on a home mask: twenty participants aged 45 to 70, fifteen women and five men, 630 nm, 15.6 J/cm², twelve-minute sessions twice a week for three months. Crow's feet depth fell 38.3 percent, dermal density rose 47.7 percent, pore diameter fell 32.8 percent, and the improvements were still measurable two to four weeks after participants stopped.

The caveat we would rather write ourselves than have you find: twenty participants is small, and that study was run by authors employed by the cosmetics group and the device manufacturer behind the mask being tested. That does not make it worthless, it makes it directional. Treat the direction as meaningful and the exact percentages as optimistic. The same caution applies to the 2014 trial, which was also manufacturer-funded and was not blinded.

The uncomfortable fact about consensus

In 2025 the Journal of the American Academy of Dermatology published an evidence-based consensus on photobiomodulation. Twenty-one experts, a systematic review, two Delphi rounds, revised until they agreed. They listed the conditions where photobiomodulation is an effective option: peripheral neuropathy, androgenic alopecia, several kinds of wound and ulcer, pain from diabetic foot ulcers, acute radiation dermatitis.

Skin rejuvenation is not on that list. Not rejected, not included. The individual trials exist and several are decent, but the field's own expert panel did not consider the body of evidence sufficient to name facial rejuvenation as a validated indication.

We sell a face mask, so that sentence costs us something to write. We would rather you read it here than find it later.

The irradiance the trials actually used

Here is the number that reframes the whole category. The red arm of the 2014 trial ran at 5.9 to 13.3 mW/cm². The 2025 sham-controlled trial ran at up to about 10 mW/cm² per wavelength. Published protocols cluster between 20 and 60.

So the masks advertising 100 or 128 mW/cm² are not several times more effective. They are several times further from what was tested. Photobiomodulation has a biphasic dose response: below a threshold nothing happens, above an optimum the effect flattens and then falls away. Reviews of the field note that negative studies more often come from overdosing than from underdosing.

Hair growth is the best-documented use, and it is not the face

If you rank the whole field by evidence quality, hair comes first, not skin. Red light for androgenetic alopecia has been through multiple randomized controlled trials, including head-to-head comparisons with minoxidil. That is a stronger literature than anything in facial rejuvenation. It also requires a different device: a cap or a comb, not a face mask.

What red light therapy has not been shown to do

This section exists because the pages ranking above us do not have one, and because it is the honest half of the subject.

Stanford Medicine published a review in which clinicians went through the popular claims and separated them. On their reading, the data does not currently support red light therapy for athletic performance, sleep, chronic pain, erectile dysfunction or dementia. Professor Zakia Rahman, a clinical dermatologist quoted in that piece, put it in a sentence worth keeping: there is real evidence that red light can change biology, but that is not the same as saying it is some kind of panacea.

That matters because two of the pages currently ranking on the first page for this exact question do claim cognitive benefits, and one claims weight loss. Two major hospital systems are publishing opposite conclusions on the same search results page. When you see a benefit listed with no study attached, that is usually why.

To be equally clear about our own category: an LED face mask is a general wellness device. It is not a treatment, it does not replace a dermatologist, and it does not do what an injectable does. Anyone telling you a mask is better than Botox is comparing two things that do not do the same job.

How good is this literature, honestly

A methodological review published in PLOS ONE in 2025 went through twenty-seven dermatological LED studies and counted. The median study had twenty-six participants. Thirty-seven percent were sponsored by the device manufacturer, with a further eleven percent run by commercial practices. A third had no non-light control group. A third had unblinded assessors. And not one of the twenty-seven validated the dose it claimed to deliver.

Doses across the studies varied by three orders of magnitude. That is the state of the field, and it applies to the studies in the table above as much as to any other.

There is one exception worth naming. A meta-analysis in JAMA Dermatology in 2025 pooled six studies and 216 participants on at-home LED devices, and found roughly a 45 percent reduction in inflammatory acne lesions. It is the only meta-level evidence on home devices specifically. It is about acne, not wrinkles, and its authors flag both heterogeneity and likely publication bias given how lightly this market is regulated.

Face or body: why the same device does not do the same job

Almost every guide on this subject says "skin" and then gives only facial examples. The distinction is physical and it changes what you should buy.

Red at 630 to 660 nm is absorbed in the first few millimetres. That is exactly where facial ageing happens: fine lines, tone, texture, pore appearance. Near-infrared at 830 to 850 nm travels further, which is why it appears in the research on deeper tissue rather than in the research on wrinkles.

So a facial mask is a red-light instrument first, with near-infrared as a useful addition. A full-body panel is a different instrument for a different depth. Buying a panel because it is more powerful and pointing it at your face is not an upgrade, it is a different treatment at a distance the studies did not use.

At-home masks compared with in-office treatments

The consensus across the medical sources is consistent: a clinic device is more powerful, better controlled, and usually delivers faster visible change. That is true and we are not going to argue with it.

What none of those sources mention is the part that decides the outcome in practice. A clinic course requires appointments, and the published protocols run over four to twelve weeks with two sessions a week. Adherence is the variable that separates people who see something from people who do not, and adherence is much easier at home in your own bathroom than in a diary.

The cost comparison is also less lopsided than it looks. In-office LED sessions are commonly quoted between 25 and 200 dollars each. At two sessions a week for three months, that is twenty-four sessions. Home devices run roughly 100 to 1,000 dollars once. Whichever way you run the arithmetic, the home device pays for itself inside one published protocol.

One number is worth carrying with you. A 2026 review in Cureus counted 148 products registered under the FDA product code for these devices, and only about ten masks holding a clearance backed by clinical study data. The category is far less regulated than its marketing implies, in both directions: a cleared device is not proven effective, and an uncleared device is not proven useless. It simply means nobody filed.

How to read a spec sheet before you buy

Everyone tells you home devices are weaker. Nobody tells you how to check. Five questions, in order of how much they matter.

1. Which wavelengths, in nanometres? You want 630 to 660 nm for facial skin, ideally with 830 to 850 nm alongside. A brand that will not print the numbers is telling you something.
2. What irradiance, in mW/cm², and measured at what distance? The second half of that question is the one that separates honest specs from headline numbers. A reading taken at the surface of the diode and a reading taken at the skin are not the same measurement. Published protocols cluster around 20 to 60 mW/cm².
3. What dose per session, in J/cm²? Irradiance multiplied by session length. Effective protocols in the literature land roughly between 4 and 18 J/cm². This is the number that actually predicts a result, and almost no brand publishes it.
4. How long is a session, and can you change it? A fixed thirty-minute cycle pushes you past the useful dose for no additional benefit.
5. What does the warranty cover, and for how long? Straps and controllers are what fail on these devices.

If you want to see those five questions applied to nine masks on sale in the United States, including ours and including the ones that refuse to answer, we did exactly that in our comparison of the best LED face masks of 2026.

How often and how long, according to the trials

Not according to us. According to the protocols that produced published results.

Two things jump out. Sessions are short, ten to twelve minutes rather than thirty. And the frequency that produced published results is twice a week. Three to five is a reasonable practical ceiling for a home mask, but nothing in the literature suggests daily use adds anything.

A realistic timeline: week 4, week 12, month 6

This is where most people quit, because they expected the wrong thing on the wrong date.

• Weeks 1 to 3. Nothing visible. What you may notice is that skin feels calmer and less reactive. That is not collagen, it is worth nothing as a measurement, and it is also the phase where most people stop.
• Week 4. The earliest point at which self-reported change appears in the literature. In the 90-patient series reported by the American Academy of Dermatology, eight sessions over four weeks produced improvement that participants could see, in over ninety percent of cases.
• Weeks 8 to 12. The window where instrument-measured change shows up. Both the 2014 and 2023 trials ran to roughly this length before measuring.
• Months 4 to 6. Where dermatology bodies place meaningful change for most people. This is a maintenance product, not a course you finish.

What happens if you stop? The 2023 trial measured participants two to four weeks after the last session and the improvements were still there. Beyond that, nobody has published good data. Treat it like exercise rather than like a procedure.

Safety, side effects, and who should be careful

Red light therapy has a good safety record. A 2019 study of maximum tolerated doses found mild adverse effects, principally temporary redness and some post-treatment hyperpigmentation, and no serious adverse events. That is a reassuring result, and it is not the same as saying nobody should take precautions.

Eyes

Red and near-infrared at these intensities are not considered hazardous to the eye, but the light is bright and it is uncomfortable. Close your eyes, lie down. If you are light-sensitive, use LED therapy goggles, and check whether your device includes them, because most do not.

Deeper skin tones

This is the precaution most guides skip. The American Academy of Dermatology recommends that people with darker skin tones speak to a dermatologist before starting, because of the risk of hyperpigmentation. There are also user reports of melasma worsening under LED masks. If you have melasma or a history of pigmentation changes, get advice before you buy, not after.

Medication and medical conditions

Several common medications increase photosensitivity, including isotretinoin and some antibiotics. Lupus and other photosensitive conditions are a reason to check first. If you are pregnant or breastfeeding, there is no specific evidence either way, which is exactly why the cautious answer is to ask your doctor rather than to assume.

What "FDA-cleared" means

It means a 510(k) submission established substantial equivalence to a device already on the market, under Class II. It is a safety and equivalence review. It is not a finding that the product works, and these files often contain no clinical trial. No LED face mask is FDA approved. Any brand using that phrase is misrepresenting its status.

Where it fits in a skincare routine

Not one of the ten pages we reviewed answers this, and it is the question you will actually have on Tuesday evening.

• Clean, dry skin, nothing else on it. Makeup, sunscreen and heavy oils sit between the light and your skin. Cleanse, dry, then the mask.
• Serums and moisturiser go after, not before. Skin is warm and receptive at the end of a session, and nothing is blocking the light during it.
• Retinol goes on after the session too, and on a different evening if your skin is reactive. Both are asking something of the same tissue.
• Give acids their own night. Glycolic and other exfoliating acids leave skin more sensitive. Alternate rather than stack.
• Evening rather than morning, for one unglamorous reason: it is the only time of day most people will actually lie still for ten minutes.

One more thing, and it is the part buyers mention that no spec sheet captures. Several describe the ten minutes as the piece they look forward to: lying down, eyes shut, phone out of reach. If you have spent years on creams that asked nothing of you and gave nothing back, a routine that makes you stop for ten minutes is not a small thing.

That last point is worth more than it sounds. Every trial in this article depends on people completing thirty sessions. The device that works is the one that fits the evening you already have.

Frequently asked questions

Is 10 minutes of red light therapy enough?

Yes, for a facial device in the usual power range. At around 27 mW/cm², ten minutes delivers roughly 16 J/cm², which sits inside the 4 to 18 J/cm² window used in published protocols. The home-mask trial published in 2023 used 15.6 J/cm² per session. Longer is not better: the dose response is biphasic, and past the optimum the effect declines.

Is there a downside to red light therapy?

Three, honestly. It is slow, and most people quit before week four. It requires consistency, two to five sessions a week for months. And the evidence is much thinner for some claimed benefits than the marketing suggests, particularly sleep, athletic performance and cognition, which Stanford clinicians describe as not supported by current data.

Why don't doctors recommend red light therapy?

Many do, for specific uses. What dermatologists resist is the way it is sold. The evidence for hair growth and for the appearance of fine lines is real; the evidence for the long list of other benefits attached to it commercially is not. Clinicians also point out that home devices are less powerful than clinic equipment and that buyers usually cannot verify the output they are getting.

How often should you use red light therapy?

The protocols that produced published results used two sessions a week over four to twelve weeks. Three to five sessions a week is a reasonable practical range for a home mask. Daily use is not harmful but adds little, and consistency over months matters far more than frequency in any given week.

Where should you not put red light therapy?

Avoid directly over the eyes without closing them, over active skin cancers or undiagnosed lesions, and over areas being treated with photosensitising medication. If you have melasma or a history of pigmentation changes, speak to a dermatologist before treating the face.

How long does it take for red light to tighten skin?

Instrument-measured changes in dermal density and elasticity appear in trials running eight to twelve weeks. Self-reported improvement can appear from week four. Dermatology bodies generally place meaningful change at four to six months. Anything faster than that is a claim, not a finding.

Is red light therapy better than Botox?

They do different things. An injectable relaxes the muscle that creates an expression line. Red light acts on the skin itself, on density, texture and tone. Comparing them is comparing two different mechanisms, and any brand claiming its mask replaces an injectable is overselling.

Does at-home red light therapy actually work?

The best evidence is a 2025 multicentre trial published in Medicine: sixty participants, randomized, double-blind, controlled against a sham device, with improvement in crow's feet reported in 86.2 percent of the active group against 16.7 percent under sham. A smaller 2023 trial on a home mask found measurable changes in dermal density and pore diameter after three months, but it had twenty participants and was run by authors employed by the manufacturer. Direction credible, exact percentages to be read with caution.

The bottom line

Three things hold up: the appearance of fine lines and skin firmness, dermal density, and hair growth, which has the strongest literature of all and needs a different device. Three do not: sleep, athletic performance and cognition, whatever a hospital blog may have told you.

The dose that produced those results is smaller than the marketing implies. Ten to twelve minutes, roughly 4 to 18 J/cm², twice a week, over four to twelve weeks. The trials ran at single-digit and low double-digit irradiance, not at three figures.

Which leaves you with a simpler question than the category pretends. Not which mask is most powerful, but which one publishes its numbers and fits the evening you already have.

Keep reading

• Best LED face masks of 2026, the specs brands publish and the ones they hide.
• Dull skin: seven things that actually restore radiance, where LED fits in an evening routine.
• How red light works at the cellular level, the mitochondrial mechanism without the marketing.
• Red light and hair, the same wavelengths, a different application.

Sources: Wunsch A, Matuschka K. Photomed Laser Surg. 2014;32(2):93-100. Zein R, Selting W, Hamblin MR. J Biomed Opt. 2018;23(12):120901. Huang YY, Chen ACH, Carroll JD, Hamblin MR. Dose-Response. 2009;7(4):358-383. Maghfour J, Mineroff J, Ozog DM, Jagdeo J, Lim HW. J Am Acad Dermatol. 2025;93(2):429-443. Grimes DR. PLOS ONE. 2025. Ershadi S, Barbieri JS. JAMA Dermatol. 2025. Couturaud V et al. Skin Res Technol. 2023. Jagdeo J et al. 2019. Brosseau L et al. Cochrane Database Syst Rev. 2005. American Academy of Dermatology, red light therapy guidance. Stanford Medicine, February 2025. Manufacturer specifications and public FDA 510(k) filings, consulted July 2026. This article is informational and is not medical advice. Lumiglo devices are general wellness devices and are not intended to diagnose, treat, cure or prevent any disease.