What outcomes appear best supported?
Fine lines and wrinkles
Several controlled human studies report improvement in selected facial or periocular wrinkle measures. The signal is credible, but not uniform: clinician scales, imaging, replicas and participant ratings sometimes disagree, and effect estimates belong to the studied protocol.
Texture and elasticity
Some studies found changes in instrumental texture or elasticity measures; others did not. This is better described as protocol-dependent evidence than a settled effect of red light generally.
Collagen-related outcomes
A small part of the human literature includes biopsy, histology or biochemical markers consistent with remodelling. That does not justify treating “boosts collagen” as a device-independent promise or as proof of a visible benefit for every user.
Pigmentation, redness and disease claims
These outcomes involve different conditions, mechanisms and often other wavelengths or photodynamic therapy. This synthesis does not support generalising its photoageing findings to acne, rosacea, pigmentation disorders or treatment of disease.
A 2018 systematic review found six randomised skin-rejuvenation LED trials and assigned that use a grade C recommendation, noting small samples, absent blinding or sham controls, and varied treatment parameters in parts of the literature.[1] A later meta-analysis pooled a favourable skin-rejuvenation result from six studies, but the authors’ own analysis raised publication-bias concerns and noted varied conditions and short follow-up.[2]
That combination supports a conditional conclusion: a genuine effect is plausible and has been observed in controlled human work, but its size and reproducibility for a specific consumer product are not settled.
What the human studies actually tested
This table prioritises studies that directly treated facial skin with red, near-infrared or closely relevant comparison protocols. It is not a vote count. A larger uncontrolled study can provide less causal confidence than a smaller credible sham-controlled trial.
| Study and design | Protocol | Main result | Transfer limit |
|---|---|---|---|
| Omnilux Revive randomised half-face study (2005) [source] Single-blinded randomised half-face study; one side treated and the other left untreated. | 633 nm 96 J/cm² reported · 20 minutes · Three times weekly · Three weeks (nine treatments) | Subjective and blinded photographic assessments suggested visible change, while objective hydration and elasticity measures did not show significant improvement. | This professional lamp is not the current Omnilux Contour Face mask. |
| Early Omnilux-system facial skin-rejuvenation study (2005) [source] Prospective before/after study; 38 people were selected and 31 completed, without a sham or untreated control group. | 633 nm and 830 nm sequential treatment 126 J/cm² at 633 nm and 66 J/cm² at 830 nm reported · Nine treatments across five weeks | A study associated with an Omnilux LED system evaluated repeated 633 nm and 830 nm treatment for facial rhytids and reported changes in clinical appearance measures. | The evidence is relevant to a platform history and protocol, not an independent current-product ranking. |
| Facial LED skin-rejuvenation trial (2007) [source] Prospective randomised, placebo-controlled, double-blinded split-face study with three active treatment groups and a sham group. | 633 nm, 830 nm, or sequential 830 nm plus 633 nm Twice weekly · Four weeks | The study applied specific 633 nm, 830 nm or combined LED settings to one side of the face twice weekly for four weeks and reported improvements from baseline in active groups across several measured outcomes. | The abstract foregrounds changes from baseline; it does not establish that every reported endpoint differed from sham. |
| Red versus broad-spectrum white LED wrinkle trial (2017) [source] Prospective randomised double-blinded active-comparator study. | 660 nm red versus 411–777 nm white Paper abstract reports 5.17 J per treatment; area-normalised interpretation is not assumed here · Daily · 12 weeks | Both active-light groups improved on one replica measure, but the study found no between-group difference and blinded dermatologists did not detect significant change. | Different outcome methods gave different conclusions. |
| Home-use LED split-face photoageing study (2020) [source] Prospective split-face pilot study; the left side was treated and the other side served as control. | 637 nm and 854 nm Twice weekly · Eight weeks | This small home-use split-face study reported treatment-side improvements in elasticity and texture after a defined 637/854 nm protocol. | The result does not establish equivalence with a current commercial mask. |
| Red versus amber periocular-wrinkle split-face trial (2023) [source] Randomised split-face active-comparator trial. | 660 nm red and 590 nm amber 7.5 mW/cm² reported · 3.8 J/cm² reported · Ten sessions across four weeks · Four weeks | Wrinkle volume improved from baseline on both red- and amber-treated sides, while hydration and viscoelasticity did not. | This was a wavelength comparison, not a mask comparison. |
| Facial PBM frequency sham-controlled mask trial (2025) [source] Randomised, double-blind, sham-controlled trial comparing two active treatment frequencies. | 660 ± 10 nm 6.4 mW/cm² reported · 8.05 J/cm² reported · 21 minutes · Two or three times weekly · Four weeks | The trial produced mixed findings: its clinical wrinkle scale did not differ between groups, while selected image-analysis measures and satisfaction favoured active treatment. | The paper does not establish one preferred treatment frequency for consumers. |
| Home-use LED mask crow’s-feet trial (2025) [source] Multicentre, randomised, double-blind, visually matched sham-controlled home-use trial; 59 participants were analysed. | 630 nm and 850 nm Maximum 10 mW/cm² for each active wavelength, with ±20% tolerance reported · 9 minutes · Five times weekly · 12 weeks | This multicentre double-blind trial compared one 630 nm and 850 nm home-mask protocol with a visually matched lower-output sham for 16 weeks and reported between-group differences in crow’s-feet assessments from week eight. | The PubMed abstract contains inconsistent irradiance-unit rendering; PLT does not reuse it as a shopping specification. |
The studies range from short professional half-face exposures to a 12-week wearable home-mask protocol. They differ in wavelength, radiant exposure, frequency, geometry, controls and outcome instruments. A 2025 methodology review of visible-LED dermatology research found wide exposure variation, incomplete reporting and no independent verification of stated fluence in the included studies.[3] Its scope was visible light, so it should not be stretched into a complete critique of near-infrared research.
Examples of why the result depends on the outcome measure
- The 2005 Omnilux Revive half-face study found favourable participant and blinded-photograph assessments, but no significant objective hydration or elasticity change.[4]
- A 2017 active-comparator trial found improvement from baseline in skin-replica wrinkles under both red and white LED exposure, no significant difference between them, and no significant change in blinded dermatologist assessments.[7]
- A 2025 sham-controlled 660 nm mask trial found no group difference on its Wrinkle Assessment Scale, while selected image-analysis measurements and satisfaction favoured active treatment.[10]
Why wavelength alone is not a transferable treatment
Positive studies have used red light around 630–660 nm, near-infrared around 830–854 nm, combinations, and—in one direct comparison—590 nm amber light. That variation does not identify a universal “best wavelength”. Nor do the studies establish that red alone, NIR alone or a combination is always superior: direct, adequately controlled comparisons are too sparse and protocol differences are too substantial.
A treatment is more than its wavelength label. Transferability depends on irradiance, radiant exposure, spectral bandwidth, treatment duration and frequency, working distance, contact or non-contact geometry, coverage and uniformity, participant characteristics, and the exact outcome definition. See PLT’s dose, irradiance, distance and time guide for the underlying quantities.
Nor does a study’s incident skin exposure establish energy delivered to a deeper biological target. Tissue absorption and scattering vary, and a penetration-depth estimate does not itself demonstrate a clinical effect. The facial studies reviewed here do not support a universal “NIR reaches deeper, therefore it works better” conclusion.
Five levels of device-evidence transferability
The studied hardware/version is the product being assessed. This is the closest fit, but study design, protocol and outcome quality still determine what the evidence supports.
The study used an earlier or professionally related system. It can inform technological history, not silently become proof for a current retail model.
The setup resembles the consumer product in more than wavelength—such as wearable facial geometry and schedule—but material optical and hardware differences remain.
The study and product share nominal wavelengths. This supports a spectral connection only, not device efficacy.
Mechanistic or non-facial evidence may explain plausibility, but is not direct evidence that a facial mask changes a consumer outcome.
The older Omnilux papers are therefore predecessor/platform evidence for the current Omnilux Contour Face, not exact-device trials.[5] Even the stronger 2007 split-face work tested professional Omnilux systems under a defined protocol, not today’s consumer mask simply because 633 nm and 830 nm also appear in its specifications.[6]
Form factor does not provide a shortcut. Flexible and rigid masks may differ in fit, gaps, comfort and spatial distribution, but the clinical literature does not establish either construction as inherently more effective. Those are mainly usability and geometry questions unless an exact device is tested. See panel versus mask for form-factor choice, then use the Australian LED face-mask comparison for current researched product trade-offs.
What PLT could verify for CurrentBody and TheraFace
CurrentBody Skin Series 2
PLT located and reviewed the full 167-page SGS report linked from CurrentBody’s research hub.[13] It identifies Series 2 and follows 35 healthy Chinese adults for 56 days: 10 minutes daily through day 28, then five times weekly through day 56.
The report presents multiple favourable within-participant changes from baseline. It has no control group, is not a peer-reviewed publication, and does not independently verify irradiance in the report. It therefore provides exact-device evidence with limited causal confidence—not proof of superiority or a clean estimate of treatment effect.
TheraFace Mask Glo
ClinicalTrials.gov verifies an open-label, single-group exact-device study sponsored by Therabody, with Eurofins CRL as collaborator and 115 enrolled participants.[15] Therabody reports that 104 completed 12 weeks and presents favourable outcomes.[14]
No registry results, full report or peer-reviewed paper was identified. The regimen combined red, red-plus-infrared and blue modes with vibration, so it cannot isolate the effect of red/NIR light. The study design is now verifiable; the outcome claims remain manufacturer-presented.
A reality check on common claims
- “Clinically proven”
- Meaningful only after asking: which exact device, study design, comparator, outcome and protocol? An uncontrolled exact-device report is clinical research, but it is not equivalent to a blinded sham-controlled trial.
- “Boosts collagen”
- Constrained human tissue findings make collagen-related activity plausible under specific protocols. They do not establish a universal visible benefit or quantify what every consumer mask will do.
- “Reduces wrinkles”
- Reasonably supported for some specific protocols and measures. The size, durability and relevance of the effect remain device-, protocol- and outcome-dependent.
- “Reverses ageing”
- Not supported by this literature. Studies measure selected appearance or skin-property outcomes over limited periods; they do not reverse biological ageing.
- “Optimal wavelength” or “clinically proven dose”
- Insufficiently established as a universal consumer rule. Positive studies use heterogeneous parameters, and equal nominal wavelength or calculated radiant exposure does not guarantee equivalent delivery or response.
- “More irradiance is better”
- Unsupported as a general quality rule. Irradiance is one protocol quantity and must be interpreted with time, area, geometry, measurement method and safety instructions.
- “Medical grade”
- A marketing phrase unless tied to a specific, verifiable regulatory or technical meaning. It is not an evidence hierarchy.
Safety, adverse events and what buyers should do with this evidence
Short facial studies generally reported few serious problems. Reports included transient erythema, isolated minor events and, in one older study, withdrawal after mild facial herpes simplex activation. Small samples, eligibility exclusions and short follow-up mean rare or long-term harms remain less certain. A broader oncologic-safety review found no clinical malignancy signal in the literature it assessed, while emphasising limited data and mixed preclinical findings.[16]
Eye risk depends on factors including intensity and exposure time, and photosensitising medicines can change risk; follow the exact device instructions rather than assuming all masks have the same eye-use rules.[17] Seek professional advice before use if you have an eye or skin condition, a history of photosensitivity, take potentially photosensitising medication, are using the device around a suspected lesion, or want to treat a medical condition.
What this means when choosing a mask
- Prefer evidence for the exact device; label predecessor/platform evidence honestly.
- Check whether a “clinical study” had a sham or untreated comparator.
- Match the studied outcome to the outcome you care about.
- Compare the study’s wavelength, output, session, frequency and geometry with the retail device.
- Look for objective measures, blinded assessment, follow-up and adverse-event reporting—not just participant satisfaction.
- Treat comfort, fit and adherence as practical value factors, not efficacy evidence.
- Use PLT’s Australian TGA, ARTG and RCM guide to keep regulatory status separate from clinical evidence.
Browse the evidence library →
How PLT evaluates research and product claims →
Read the health and site disclaimer →
Sources and references
- Jared Jagdeo et al.. Light-emitting diodes in dermatology: A systematic review of randomized controlled trials. Lasers in Surgery and Medicine. 2018-01-22. DOI: 10.1002/lsm.22791. PMID: 29356026. Accessed 2026-09-08.
- Le Thi Nhu Ngoc, Ju-Young Moon, Young-Chul Lee. Utilization of light-emitting diodes for skin therapy: Systematic review and meta-analysis. Photodermatology, Photoimmunology & Photomedicine. 2022-11-09. DOI: 10.1111/phpp.12841. PMID: 36310510. Accessed 2026-09-09.
- David R. Grimes. Methodological issues in visible LED therapy dermatological research and reporting. PLOS One. 2025-09-25. DOI: 10.1371/journal.pone.0332995. Accessed 2026-09-09.
- Jaideep Bhat, Jan Birch, Colin Whitehurst, Sean W. Lanigan. A single-blinded randomised controlled study to determine the efficacy of Omnilux Revive facial treatment in skin rejuvenation. Lasers in Medical Science. 2005-05-21. DOI: 10.1007/s10103-005-0330-5. PMID: 15909229. Accessed 2026-09-09.
- Brian A. Russell et al.. A study to determine the efficacy of combination LED light therapy (633 nm and 830 nm) in facial skin rejuvenation. Journal of Drugs in Dermatology. 2005-12-01. DOI: 10.1080/14764170500370059. PMID: 16414908. Accessed 2026-09-09.
- Seung Yoon Lee et al.. A prospective, randomized, placebo-controlled, double-blinded, and split-face clinical study on LED phototherapy for skin rejuvenation. Journal of Photochemistry and Photobiology B: Biology. 2007-07-27. DOI: 10.1016/j.jphotobiol.2007.04.008. PMID: 17566756. Accessed 2026-09-08.
- Chan Hee Nam et al.. The efficacy and safety of 660 nm and 411 to 777 nm light-emitting devices for treating wrinkles. Dermatologic Surgery. 2017-03. DOI: 10.1097/DSS.0000000000000981. PMID: 28195844. Accessed 2026-09-09.
- Janice Natasha C. Ng, Rungsima Wanitphakdeedecha, Chadakan Yan. Efficacy of home-use light-emitting diode device at 637 and 854-nm for facial rejuvenation: A split-face pilot study. Journal of Cosmetic Dermatology. 2020-07-10. DOI: 10.1111/jocd.13613. PMID: 32649063. Accessed 2026-09-09.
- Lidiane Rocha Mota et al.. Photobiomodulation reduces periocular wrinkle volume by 30%: A randomized controlled trial. Photobiomodulation, Photomedicine, and Laser Surgery. 2023-02. DOI: 10.1089/photob.2022.0114. PMID: 36780572. Accessed 2026-09-09.
- Erick Frank Bragato et al.. Role of photobiomodulation application frequency in facial rejuvenation: randomized, sham-controlled, double-blind, clinical trial. Lasers in Medical Science. 2025-03-31. DOI: 10.1007/s10103-025-04383-1. PMID: 40167796. Accessed 2026-09-09.
- Sang Hyun Park, Seong Oh Park, Jae-A Jung. Clinical study to evaluate the efficacy and safety of home-used LED and IRED mask for crow's feet: A multi-center, randomized, double-blind, sham-controlled study. Medicine. 2025-02-14. DOI: 10.1097/MD.0000000000041596. PMID: 39960921. Accessed 2026-09-08.
- CurrentBody Skin Series 2 clinical study summary. CurrentBody Skin. Accessed 2026-09-09.
- Series 2 SGS clinical assessment report QDCPCH24000497-02_EN-V2. SGS-CSTC Standards Technical Services (Shanghai). 2024-12-06. Accessed 2026-09-09.
- TheraFace Mask Glo clinical study summary. Therabody ANZ. 2025-10-14. Accessed 2026-09-09.
- Clinical study to evaluate the efficacy of a light therapy facial mask in improving skin health (NCT07054710). ClinicalTrials.gov, U.S. National Library of Medicine. Accessed 2026-09-09.
- Gregory E. Glass. Photobiomodulation: The clinical applications and oncologic safety of low-intensity light therapy. Aesthetic Surgery Journal. 2023-01-27. DOI: 10.1093/asj/sjad018. PMID: 36722207. Accessed 2026-09-09.
- Lasers, IPL sources and LED phototherapy in the cosmetic and beauty therapy industry. Australian Radiation Protection and Nuclear Safety Agency. Accessed 2026-09-08.