What the Research Actually Says About Microcurrent

The studies that involved human faces

Kavanagh 2012, the one everyone cites

Kavanagh and colleagues in 2012 followed 108 women over twelve weeks and reported an 18.6 percent increase in cheek muscle thickness measured by ultrasound.

It is the largest and most quoted result in this field, and the measurement was taken at the cheek rather than at the jawline, which matters whenever the figure is used to sell a jawline result.

Participants were women, so the effect size in men is an extrapolation rather than a finding, covered in our article on microcurrent for men.

Chang 2020, the strongest methodology

Chang and colleagues in 2020 used a double-blind sham-controlled design, which is the best methodology anyone has applied here, and reported improvements in skin appearance measures.

Sham control matters because people who know they are being treated report improvement regardless, and a facial device is impossible to blind without a convincing dummy.

The outcomes were instrument and observer measures of skin, not biopsies.

Kwak 2023 and Saniee 2012

Kwak and colleagues in 2023 ran a split-face study, where each participant serves as their own control, which handles age and lifestyle confounding neatly.

Saniee and colleagues in 2012 reported improvements in wrinkle and elasticity measures in a randomized design.

Both are small, both are short, and both measured appearance rather than tissue structure.

The two papers ranking that are not about faces

Kolimechkov 2023

Kolimechkov and colleagues in 2023 is a narrative review in the European Journal of Applied Physiology covering microcurrent combined with exercise in healthy, active people.

It examines limbs, body composition, strength and recovery. The face does not appear in it, and anti-aging is mentioned once, as a suggested direction for future research rather than a finding.

Its reported findings include rodent work on ATP and protein synthesis, small human trials on handgrip strength and plantar flexion, and at least one study showing no benefit for muscle soreness.

The authors' own central limitation is that protocols vary so widely, with inconsistent reporting of current parameters and electrode placement, that optimal settings cannot be identified.

Jonik 2025

Jonik and colleagues in 2025 is a narrative review in Therapeutic Advances in Chronic Disease covering clinical microcurrent for pressure injuries, burns, wound healing, musculoskeletal and neuropathic pain, and cranial stimulation for anxiety and insomnia.

There is nothing in it on facial aesthetics, lifting or collagen.

It is genuinely useful on safety, reporting that microcurrent runs far below the levels that cause nerve depolarization or muscle fatigue, with adverse effects reported in under one percent of patients in the cranial stimulation context.

Someone searching whether microcurrent works for their face and landing on either of these papers is reading good science about a different question.

Review, trial, and why the difference matters

A narrative review is an expert summary of other people's studies, with no statistical pooling and no formal method for deciding which studies to include. Both papers above are narrative reviews and say so.

A systematic review follows a predefined protocol for finding and assessing every relevant study, and a meta-analysis pools their results statistically. Neither exists for facial microcurrent.

A randomized controlled trial tests an intervention against a control. The facial trials in this area are small randomized or split-face designs, which is a reasonable standard for cosmetic research and a modest one in absolute terms.

When a brand says the research shows, ask which of those three they mean, because the gap between them is where most overclaiming happens.

What nobody has measured

Facial collagen by biopsy before and after a course of treatment, which is the measurement that would settle the collagen question, covered in our collagen article.

Facial fat before and after, which is why we decline to answer whether these devices affect facial volume, set out in our article on facial fat.

Facial lymph flow during or after a session, despite drainage being one of the most common claims made.

Ingredient penetration on any consumer facial device, including ours, which we state plainly rather than imply.

Any longitudinal study following users for years, which is why no prevention claim in this category can be supported, covered in our article on when to start.

The commercial context

Most research in consumer device categories is small, short and conducted or funded with commercial involvement. That is true across skincare devices and is not unique to microcurrent.

It does not make the findings worthless, and it does mean effect sizes from such studies tend to shrink when independent groups repeat them.

Independent replication is rare here, which is the single biggest weakness in the whole evidence base.

We have no study of our own on a PureLift device. We publish output figures and cite other people's science, and we would rather say that than imply a research program we do not have.

How to judge a study a brand cites

Check the species. Rat skin and human skin are not the same evidence, which matters enormously for the ATP claim traced in our ATP article.

Check the body part. A limb result is not a face result.

Check what was measured against what is being claimed. Muscle thickness is not collagen and neither is skin elasticity.

Check whether it is a review or primary research, and if a review, whether it is narrative or systematic.

Check the date and whether anything has replicated it. A single unreplicated finding carrying a whole category for decades is a warning sign rather than a credential.

Frequently asked questions

Is there scientific evidence that microcurrent works

There is a small body of human facial research reporting increased cheek muscle thickness and improved skin appearance measures over roughly twelve weeks. It is modest in size and quality compared with the confidence of most marketing.

What is the best study on microcurrent for the face

Chang and colleagues in 2020 used a double-blind sham-controlled design, which is the strongest methodology applied in this area. Kavanagh and colleagues in 2012 is the largest, with 108 women over twelve weeks.

Why do PubMed papers about microcurrent rank for face searches

The two currently ranking are narrative reviews about exercise, wound healing and pain. They match the word microcurrent rather than the question about faces.

Has microcurrent been proven to build collagen

No human biopsy study has measured facial collagen after microcurrent. Trials have measured dermal density and elasticity, which are related proxies rather than direct collagen counts.

Are microcurrent studies independent

Rarely. Most research in consumer device categories involves commercial funding or participation, and independent replication of facial microcurrent findings is uncommon.

This article sits inside our full map of the category, where we set out all 33 devices we track, what each one carries and what each brand publishes: Nine Technologies on One Dial, What the Rest of the Market Actually Carries.