In short: controlled studies suggest red and near-infrared light can modestly improve wrinkles over 8–12 weeks. Most studies are small, many are funded by device makers, and the strongest ones used research or clinic devices rather than the masks sold in shops. Reviews disagree on how good the evidence for acne is.
A 2025 analysis of 27 LED skin studies found that none had independently measured how much light its device gave out, and doses varied more than a thousandfold between studies. That is why we show every light-output figure as the manufacturer's claim.
None of the masks currently in our directory has an independent, sham-controlled study of that exact device.
Studies
Original research. Check who funded it and whether there was a comparison (sham) group.
Wunsch A, Matuschka K · Photomedicine and Laser Surgery 2014;32(2):93–100
Design: Randomised to treatment groups; non-randomised, untreated control group (mostly company employees); no sham; treatment unblinded; wrinkle photos scored by blinded experts · 136 enrolled; 128 analysed (57 red, 48 broad-spectrum, 23 controls)
Device: Full-body and partial-body cabin units with gas-discharge/fluorescent lamps. NOT LED, NOT a face mask
Findings: Improved collagen density (ultrasound), skin roughness and expert-rated wrinkles vs the control group; high self-rated satisfaction; no serious adverse events
Funding: Fully funded by JK-Holding GmbH (manufacturer of the devices); PI paid by sponsor
- Not an LED mask: different light source, distance and dose
- Control group not randomised and received no sham
- Industry-funded
Bragato EF, … Fernandes KPS (10 authors) · Lasers in Medical Science 2025;40(1):170
Design: Randomised, sham-controlled, double-blind; 3 arms (3×/week, 2×/week, sham 2×/week) for 4 weeks · 95 women aged 45–60
Device: Red LED face mask (research device, not a named consumer product)
Findings: No difference between groups on the Wrinkle Assessment Scale. Image-measured glabellar and right periorbital wrinkle length reduced vs sham (p<0.001). Satisfaction (FACE-Q) higher than sham. 3×/week was no better than 2×/week.
Funding: Brazilian public research funders (CAPES, CNPq); authors declare no competing interests
- Short (4 weeks); red light only (no NIR)
- Mixed results: main wrinkle scale showed no difference, image measurements did
- Abstract read via PubMed; full text not yet read
Ablon G · Journal of Clinical and Aesthetic Dermatology 2025;18(10):25–32
Design: Single-arm, open-label; no sham or control · 30 enrolled, 29 completed; ages 14–45; mild-to-moderate acne
Device: Omnilux Clear (GlobalMed Technologies), home silicone mask, blue 415 + red 633 nm
Findings: Inflammatory and non-inflammatory lesion counts fell (p<0.0001); 86% improved ≥1 acne grade; mild transient side effects
Funding: Funded by GlobalMed Technologies (device maker)
- No control group, so improvement can't be separated from natural course or other factors
- Manufacturer-funded; small sample
- Internal inconsistency on side-effect counts; dose not reported
Park SH, Park SO, Jung J-A · Medicine (Baltimore) 2025;104(7):e41596
Design: Randomised, double-blind, sham-device controlled, 2 Korean centres. Sham: no 850 nm, 630 nm at one-tenth intensity. · 60 randomised; 59 analysed (29 active, 30 sham); ~93% female; aged 30–65; Fitzpatrick II–V
Device: "Easy Claire" (Y&J Bio, Korea), home-use; described as helmet-type in methods. Not sold in the UK as far as known.
Findings: Independent-rater crow's feet grade improved ≥1 point in 86% (active) vs 17% (sham) at 12 weeks (p<0.001). Differences appeared from week 8; participants' own ratings did not differ significantly between groups. No serious adverse events.
Funding: Funded by Y&J Bio (device maker); authors declare no conflicts
- Manufacturer-funded
- Small; mostly women; Korean population
- Several internal inconsistencies in reported numbers (withdrawals, demographics, analysis sets)
- Self-assessment showed no significant difference vs sham
- Not a UK-sold device
Lee SY, Park KH, Choi JW, et al. · J Photochem Photobiol B 2007;88:51–67
Design: Randomised, placebo (sham light) controlled, double-blind, split-face; 4 arms: 830 nm, 633 nm, 830+633 nm, sham · 76 adults with facial wrinkles
Device: Clinic LED panels (not a mask)
Findings: Wrinkle reduction up to 36% and elasticity up to 19% vs baseline in active arms; increased collagen and elastic fibres on biopsy; no adverse effects.
Funding: Not stated in abstract
- Clinic device, not a home mask
- Abstract only read; results reported vs baseline in abstract
Ranpariya M, Shakelly P, Tchack M, Rafiq B, Rao BK · Journal of Cosmetic Dermatology 2025;24(9):e70416 (letter)
Design: Cross-sectional analysis of top 100 TikTok posts for "LED Light Therapy Masks" (15 Oct 2024), rated with DISCERN
Findings: 63 of 100 posts by influencers, ~80% of which were promotional; only 8 by physicians. Information quality low across all groups (mean DISCERN 20–26 out of 80). Posts rarely mentioned contraindications, precautions or evidence.
Funding: None; no conflicts declared
- 100 posts on a single day
- Minor calculation inconsistencies
Reviews
Papers that pool or summarise many studies.
Posadzki P, Car J · JAMA Dermatology 2018;154(5):597–598 (Clinical Evidence Synopsis)
Design: Evidence synopsis summarising a systematic review of light therapies for acne (full text paywalled; abstract read)
Findings: Evidence for light therapies in acne is weak and inconclusive. Only red-light photodynamic therapy with a prescription drug (MAL-PDT) showed a small reduction in inflamed lesions, judged not clinically significant. Adverse effects poorly reported.
- About clinic light therapies incl. photodynamic therapy with a drug, not home LED masks
- Only the abstract was available to us
Ngoc LTN, Moon J-Y, Lee Y-C · Photodermatology, Photoimmunology & Photomedicine 2023;39(4):303–317
Design: Systematic review and meta-analysis of 31 studies (from 554 screened)
Findings: Pooled effect for red/blue LED in acne was large (SMD −2.42). Effects also reported for skin rejuvenation, herpes simplex and psoriasis; high heterogeneity for wound healing and atopic dermatitis.
Funding: Korean public research grants
- Abstract only read; risk-of-bias results not visible
- Does not separate home masks from clinic devices
- Pooled acne result conflicts in tone with the 2018 JAMA synopsis (gen_0002) — present both
Grimes DR · PLOS ONE 2025;20(9):e0332995
Design: Methodological analysis of 27 LED-only skin studies (2005–2024) taken from two earlier systematic reviews plus one later study. Not itself a systematic review; the author calls the studies a convenience sample. Converts each study's dose into the hours of sunlight that would give the same light at those wavelengths · 27 studies (acne 9, wrinkles 5, wound healing 3, psoriasis 3, redness/erythema 7); 14–105 patients each, median 26
Findings: Doses varied more than a thousandfold between studies, even for the same condition (0.1–126 J/cm², median 40.5), usually with little explanation of why. None of the 27 studies independently measured their device's light output; all took the maker's stated power on trust, without saying whether it meant electrical power or light output. 10 (37%) were sponsored by the device maker and 3 more (11%) were run by clinics selling the treatment. Only 9 (33%) had both a non-light comparison group and blinded assessment. Some treatments delivered less light at the relevant wavelengths than under an hour of ordinary daylight
Funding: Wellcome Trust; author declares no conflicts of interest
- Single author, one peer reviewer; not a systematic review, so the 27 studies are not the whole literature
- Looks at how studies were run and reported, not whether LED works; it does not show that LED has no effect
- Sunlight comparison assumes LED light spreads evenly around its stated wavelength, which the author says is a reasonable but simplifying assumption
- Mostly older studies with clinic or research devices; includes several Omnilux devices (Lee 2007, gen_0012, among them)
Maghfour J, Mineroff J, Ozog DM, Jagdeo J, Lim HW, et al. (21 panellists) · Journal of the American Academy of Dermatology 2025;93(2):429–443
Design: Expert consensus guideline. Systematic search of Embase and MEDLINE (June 2022), then a 21-member international panel (chosen on publication history) agreed recommendations through two Delphi survey rounds and two meetings
Findings: Panel concluded PBM is safe for adults and that red light PBM does not cause DNA damage. It named PBM an effective option for peripheral neuropathy, androgenic hair loss, several kinds of wound and pressure ulcers, diabetic foot ulcer pain and acute radiation dermatitis. Cosmetic skin rejuvenation (wrinkles) is not among the uses the abstract names as effective, although the guideline covers "medical and aesthetic applications"
Funding: Not stated in the abstract
- Abstract only read; full text is paywalled, so the panel's conflicts of interest, its rating of the wrinkle and acne evidence and any device or dose advice are not known to us
- Literature search ran to June 2022, so it predates the 2025 sham-controlled mask trials (gen_0006, gen_0010)
- Covers PBM broadly (lasers and LEDs, clinic devices), not home masks
- A consensus is expert agreement on the evidence, not new evidence
Hernández-Bule ML, Naharro-Rodríguez J, Bacci S, Fernández-Guarino M · International Journal of Molecular Sciences 2024;25(8):4483
Design: Narrative review (not systematic) of PBM, LED and low-level laser studies in dermatology from the previous six years, covering mechanisms, acne, wrinkles, wounds, psoriasis and radiation dermatitis
Findings: Authors conclude well-designed placebo-controlled trials are "still scarce" and that no firm clinical recommendations can be made, partly because studies use very different settings. Overall impression is "moderate results… with repeated sessions"; PBM is safe and can be an add-on to other treatments. The wrinkle trials it tabulates (8 studies, 2005–2023, 13–136 people) mostly used red 630–660 nm and/or near-infrared 830 nm, several with clinic Omnilux-type devices at doses of 66–126 J/cm². Safety section - red and near-infrared do not appear to cause DNA damage; skin of colour may be more sensitive to visible light (one trial found up to 50% lower maximum tolerated red-light dose) and can develop lasting hyperpigmentation; blue-violet light (400–440 nm) carries risks to eyes and skin at high doses. Advises choosing devices certified by relevant authorities.
Funding: No external funding; authors declare no conflicts of interest
- Narrative review, so studies were not selected or graded systematically
- Mixes LED, low-level laser and photodynamic therapy; most wrinkle trials used clinic devices, not home masks
- Its introduction calls light therapy suitable for pregnant women and children, which differs from mask makers' own warnings; follow the device instructions
Further reading
Plain-English explainers from health publishers. Useful context, not evidence.
Maureen Salamon; reviewed by Rachel Reynolds MD
Findings: Not a miracle fix; needs regular use over months; home devices vary in strength and dose is hard to know; optimal dose not established; caution for light-sensitive conditions/medications and for darker skin (pigmentation).
- Cites no specific studies
- US-focused (FDA-cleared advice)
Camille Noe Pagán, Leah Rosenbaum; medically reviewed by Stephanie S. Gardner MD
Findings: Home devices usually weaker than clinic devices; use eye protection; avoid with photosensitising medication; much research preliminary.
- US-focused
- Sources listed but mostly not linked inline
BBC News (Jennifer Meierhans)
Findings: Reports the ASA's 5 Nov 2025 rulings against Beautaholics, Invention Works BV (Silk'n), Luyors Retail and Project E Beauty LLC for acne and rosacea claims. LED masks must be registered with the MHRA to make medical claims about skin conditions; the MHRA told the BBC "there are currently no LED face masks currently registered". Registration can be checked on the MHRA's Public Access Registration Database (PARD).
- News report; the MHRA statement dates from November 2025 and registrations may have changed since
BBC News (Annabel Rackham)
Findings: Published 8 Mar 2025. Consultant dermatologist Dr Jonathan Kentley says PBM is mostly considered safe, but many studies were small, unstandardised and often manufacturer-funded, so firm conclusions are hard to draw. Dermatologist Dr Justine Kluk says she knows of no trials showing home masks deliver a clinic dose, and expects benefits to be "probably very modest"; for severe acne a mask is "unlikely to do enough on its own". Main differences from clinic devices given as LED strength, number of LEDs and distance from the skin.
- News feature; quotes individual experts, not new research
- Also quotes CurrentBody's chief executive, who sells LED masks
Research summaries are written by Luma Index from the papers (or their abstracts where the full text isn't available). They're not medical advice. If you have a medical condition, are pregnant, or take medicines that increase light sensitivity, talk to your GP or pharmacist first.