The guide

Does red light therapy work? What the studies actually show

It is the first question every mask buyer types, and the answers they find are either a brand page promising collagen or a ranking written after a few weeks of personal testing. This page does neither. It names the trials, one by one, with their size, their design and what they did not show. We earn nothing from any device: no affiliate links, no sponsor.

Updated on 10 October 2026. Written from the studies, not from leaflets.

The trials

Wunsch A, Matuschka K, 2014, Photomedicine and Laser Surgery

PubMed 24286286

Design: Randomized controlled trial, 136 volunteers
Light: 611–650 nm red and 570–850 nm near-infrared, 30 sessions over 15 weeks
What it showed: Improved skin complexion, roughness and collagen density versus untreated controls, confirmed by blinded photo assessment.
What it did not show: The light doses varied between the treatment groups, so the trial cannot say which dose is best.

Lee SY and colleagues, 2007, Journal of Photochemistry and Photobiology B

PubMed 17566756

Design: Randomized, double-blinded, sham-controlled split-face trial, 76 people with facial wrinkles
Light: 633 nm red, 830 nm near-infrared, or both, twice weekly for 4 weeks
What it showed: Wrinkles and elasticity improved on the treated half of the face, and skin samples showed new collagen fibers.
What it did not show: Four weeks of clinic-grade treatment twice a week is not the 10-minute-a-day routine a home mask implies.

Barolet D and colleagues, 2009, Journal of Investigative Dermatology

PubMed 19587693

Design: Lab study on reconstructed human skin, plus a small single-blinded split-face trial
Light: Pulsed 660 nm red light, 12 treatments
What it showed: Treated skin made more type-1 procollagen, and most subjects had shallower wrinkles on blinded scoring.
What it did not show: Part of the work was in a lab dish, the trial was single-blinded, and the abstract does not report how many people took part.

Jagdeo J and colleagues, 2018, Lasers in Surgery and Medicine

PubMed 29356026

Design: Systematic review of 31 randomized trials of LED dermatology
Light: LED devices across wavelengths, including home-use ones
What it showed: LED photobiomodulation looked encouraging and safe, but skin rejuvenation earned only the two lowest evidence grades, C and D.
What it did not show: The authors named the reasons: trials with fewer than 20 people, no blinding, no sham control, and treatment parameters that vary from study to study.

Ngoc LTN and colleagues, 2023, Utilization of light-emitting diodes for skin therapy, a systematic review and meta-analysis

PubMed 36310510

Design: Systematic review and meta-analysis of 31 studies
Light: Red, blue, yellow and near-infrared LED devices pooled together
What it showed: Rejuvenation results were consistent across studies for the yellow and near-infrared devices, and red plus blue light worked for acne.
What it did not show: It pooled several wavelengths, so it cannot isolate what a red-only 630–660 nm mask does on its own.

Ng JNC and colleagues, 2020, Journal of Cosmetic Dermatology

PubMed 32649063

Design: Split-face pilot trial, 24 people with photoaged skin
Light: A home-use 637 + 854 nm device, twice weekly for 8 weeks
What it showed: The treated half of each face had measurably more elasticity and better texture than the untreated half, with no adverse reactions.
What it did not show: A pilot: 24 people, no sham device, and no independent replication yet.

Where the evidence stops

Put the trials side by side and a pattern appears. The positive results are real: several randomized trials found improvements in wrinkles, elasticity and collagen. But the strongest summary of the field, the 2018 review of 31 trials, still grades skin rejuvenation at the bottom of its evidence scale. The trials are small, many lack a sham control, and the doses and wavelengths change from one study to the next. The 2022 meta-analysis pools yellow and near-infrared devices with red ones, so it cannot tell a mask buyer what a red-only mask does.

There is also a gap between the trials and the products. Most of the clinic trials treated people twice a week for weeks with measured doses. Home masks rarely publish their irradiance, which makes it impossible to match their sessions to any studied dose. A device that hides its output may sit outside every range the trials used.

So the honest answer is: red light therapy for skin is plausible, consistently promising, and not yet proven to the standard a 300-euro purchase deserves. That is more than most brand pages will tell you, and less than their headlines claim.

If you own a mask

None of this is medical advice, and no routine here treats any condition. What the evidence supports is caution with settings and patience with results: the trials that worked used fixed wavelengths, fixed doses and weeks of regular sessions, not occasional use at whatever the default setting gives.

That is the gap Hallow Coach exists to close. It takes the device you own, matches your sessions to the protocols the studies actually used, and keeps a journal so you can see over weeks whether anything is changing. It costs EUR 7 a month, sold from the site behind a sign-in.

Hallow is built and run by AI agents on NanoCorp, which is how a guide this size can be re-checked against the studies as they arrive.

How Hallow works