Ingredient Beauty BenefitsRoute · Device-based
LED cap: the primary endpoint was missed, and significance arrived only after pooling three arms
A 26-week, multicentre, double-blind, sham-controlled trial — a rare quality of design for this layer. But 160 were randomised and 91 dropped out of the analysis, and the primary endpoint did not reach significance in any individual arm. Only a pooled analysis of the three active arms beat sham (p = .033, 28.5 hairs per cm²). Two device companies appear in the author affiliations.
Last checked Sep 21, 2026
Start here
- What it is
A cap or comb fitted with red or blue LEDs, used on the scalp.
- What it is used for
For hair loss. Usually a few minutes a day at home.
- How it is said to work
The light is said to raise cellular metabolism in the follicle. That account and what has been measured in people are separate things.
There are no numbers and no citations in this box. What has been established, and how far, is below.
At a glance
- Session length
- 10 minReported in a trial
- Pain
- Barely any
- Downtime
- Same day
- Sessions
- OnceNot a procedure counted in sessions but a device used daily. The trial above ran 26 weeks.
- How long it lasts
- 0 monthsOnly while it is being used. We have not verified data on what happens after stopping.
- Relative cost
- Lower
- How well established
- Not established
“Reported in a trial” means a number a study on this page actually reported. “Typical range” is not a measured value but roughly how it is usually done, and it varies by clinic and by person.
Cost is not an amount but a band: where this procedure falls when the ones on this site are lined up per session, or per area treated. Real prices differ several-fold by clinic and by how much is treated, so they are not given here. Surgery is measured on a different scale and cannot be compared with this band.
This is not how large the effect is but how well confirmed it is. It shows the highest grade among the findings on this page.
A cap of red and blue LEDs worn on the scalp, sold for a few minutes of use a day at home.
The trial on this page is well designed. 26 weeks, multicentre, randomised, double-blind, and sham-device controlled — compared against an identical cap that does not light up. As the PRP page says, trials in this layer that set up a proper inert control are rare.
But the result does not read the way that sounds.
The paper's conclusion is 'the LED caps were well tolerated and increased hair density'. Carry that sentence alone and three things disappear: the primary endpoint was missed, 91 of 160 left the analysis, and significance arrived only after three arms were pooled.
That is why this page exists.
This site does not deal in prices, clinics or reviews.
What is established
The primary endpoint did not reach significance in any individual arm
A 2025 trial in Dermatol Surg.
Adults aged 18 to 65 with androgenetic alopecia took part in a 26-week, multicentre, randomised, double-blind study, using the device for 10 minutes daily. Three active devices were compared against sham.
- 625 nm and 660 nm red light
- 425 nm blue light
- both blue and red
The primary endpoint was the mean change in non-vellus hair count from baseline to week 16.
Result: it did not reach statistical significance in any of the individual study arms.
One more thing belongs here. 160 subjects were randomised, and 91 were excluded from the per-protocol analysis.
The per-protocol population required completing 16 weeks, no major protocol violations, and at least 80% treatment compliance.
That is more than half of everyone randomised. And one of the exclusion criteria is whether they used it as instructed. Keeping only the people who managed ten minutes a day for 26 weeks means keeping people who may differ from the rest from the start.
We have no room beside the badge for this number, so we write it here. The participant count on the badge above is the 160 randomised.
Hence 'not established' — despite a good design. If anything, it is because the design was good that the missed primary endpoint stands out so clearly.
- Randomised controlled trial in people · 160 participants · 16 weeks · Compared with vehicle (same formula minus the active) · Funding manufacturer-funded · Endpoint clinical events (cancers, lesion counts) PMID 39679573
Only the pooled analysis of the three active arms beat sham
The second analysis of the same trial.
The investigators combined the three active arms into one and compared that against sham.
Result: the pooled group was significantly better than sham (p = .033). The difference was 28.5 hairs per cm².
How to read that.
First, this is not the primary endpoint's result. The primary endpoint compared each arm against sham, and it was missed. Pooling came afterwards.
Second, the three pooled arms were different devices — red only, blue only, and both. That different wavelengths act differently is the premise these devices are sold on, and pooling sets that premise down for a moment. One of the three may have done the work and carried the other two; or all three may have done a little. This analysis cannot separate them.
Third, p = .033. Below 0.05, but not by much.
Fourth, two device companies appear in the author affiliations — emitBio Inc. and KNOW Bio, LLC. We mark that on the badge as 'industry funding'. Industry money does not make a result wrong. But in a trial where the primary endpoint was missed and significance arrived in a later analysis, that fact reads heavier.
So this entry is 'not established' too.
The number 28.5 hairs per cm² is not itself small. The women's trials on minoxidil reported a difference of 10.9 hairs per cm² at 12 weeks. So we are not saying this device does nothing. We are saying no trial has yet produced that number as its primary endpoint.
- Randomised controlled trial in people · 160 participants · 16 weeks · Compared with vehicle (same formula minus the active) · Funding manufacturer-funded · Endpoint clinical events (cancers, lesion counts) PMID 39679573
Known risks
- The trial above reported that the devices were well tolerated.
- Do not shine it into the eyes — the standard instruction for these devices.
- It is a home device. Keeping it up daily drives the result, and in the trial above more than half the participants fell short of that bar and left the analysis.
- The things called hair loss are different conditions. If it is not androgenetic, this is not the right approach. Diagnosis comes first.
- There are options with thicker evidence. Minoxidil is the third item on this site to carry 'strong evidence'.
What is not established
- We could not include a trial that met its primary endpoint.
- Why 91 of 160 left the analysis is not broken down in the abstract.
- Which wavelength does the work is unknown. The analysis that reached significance pooled all three configurations.
- Results at a later primary timepoint than 16 weeks are unverified. The trial itself ran 26 weeks.
- Combined use with other approaches is not covered on this page.
- We have not yet opened the primary Korean regulatory classification, so it stays 'being verified'.
Whether to have it done is a decision to make with a doctor.
What this article relies on
- PubMed — 미국 국립의학도서관 문헌 데이터베이스National Library of Medicine · Checked on 2026-09-16