How to Read a Facial Device Study Without Being Fooled
Medically reviewed by
4 independent reviewers
Andrew Conrad Barile, PT, DPT
Doctor of Physical Therapy (DPT) | Licensed Physical Therapist (PT) | CEO and Founder, Xtreem Pulse LLC
Read bio
Dr. Andrew Conrad Barile is a Doctor of Physical Therapy and the CEO and Founder of Xtreem Pulse LLC, the engineering company behind PureLift. He earned his Doctorate in Physical Therapy from Daemen College and is a licensed physical therapist.
He brings over two decades of clinical and entrepreneurial experience across physical therapy, craniosacral therapy and medical device innovation, built on a working knowledge of human anatomy and muscle physiology.
For PureLift LAB he reviews how current is delivered across all ten levels, from sub-sensory nanocurrent and skin-level microcurrent to motor-level EMS, and how the Infuse pass fits into the same ten-minute routine.
Bertica M. Rubio, M.D.
Board-Certified Physician, Dartmouth Medical School | Medical Director, Antiaging Regenerative Medicine Clinic
Read bio
Dr. Bertica M. Rubio is a board-certified physician and Medical Director of the Antiaging Regenerative Medicine Clinic in Redlands, California. She earned her Doctor of Medicine from Dartmouth Medical School and completed her pediatrics residency at UC Irvine Medical Center.
With decades of clinical experience, she specialises in age management medicine, regenerative medicine, wound healing and growth factor therapies, and her practice integrates evidence-based medical science with advanced aesthetic treatment.
For PureLift LAB she reviews articles for medical accuracy across the whole dial, from the gentle nanocurrent and microcurrent settings through to muscle-level EMS and the Infuse pass.
Daniel Grinberg, MD, FACS
Board-Certified Otolaryngologist | Head and Neck Surgeon | Fellow, American College of Surgeons | Mount Sinai
Read bio
Daniel Grinberg, MD, FACS is a board-certified otolaryngologist and head and neck surgeon at ENT and Allergy Associates in West Nyack, New York. He earned his medical degree from Columbia University College of Physicians and Surgeons.
He completed his otolaryngology residency at New York University Medical Center, serves as Assistant Clinical Professor at Mount Sinai School of Medicine, and is a Fellow of both the American College of Surgeons and the American Academy of Otolaryngology.
For PureLift LAB he brings a wider clinical lens, connecting at-home facial stimulation from microcurrent to EMS to the anatomy underneath, with the same rigour we apply to every device specification.
Prof. Dr. med. Ivo Buschmann
Chair of Angiology, Medizinische Hochschule Brandenburg | Clinic Director, University Clinic for Angiology
Read bio
Prof. Dr. med. Ivo Buschmann is Chair of Angiology at the Medizinische Hochschule Brandenburg Theodor Fontane and Clinic Director of the University Clinic for Angiology at the Brandenburg University Hospital.
He trained at the University of Hamburg, was a Max Planck Society Fellow at the Max Planck Institute for Heart and Lung Research, and held senior consultant posts at the Charite Universitatsmedizin Berlin before his appointment as Chair in 2016.
One of Europe's leading authorities on arteriogenesis, with over 150 peer-reviewed publications and US and EU patents, he reviews for PureLift LAB how the evidence on electrical stimulation is read and where its limits lie.
Share
Almost every facial device brand cites something that looks like science, and most of it is a consumer survey wearing a lab coat.
This article compares the two on what decides the result you actually see, such as:
- One, was anything measured
- Two, was there a control group
- Three, how many people and for how long
- Four, who paid for it
- Five, was this device tested, or the category
and many more!
Apply them to our claims as well as anyone else's; we have tried to write this so that you can.
Key Points:
"96 percent of users agreed" is a satisfaction survey, not a measurement of anything on a face.
No control group means no way to separate the device from time, expectation and better lighting.
Twenty participants is a pilot; a hundred over twelve weeks is a study.
A study of the category is not a study of the product, and brands blur the two constantly.
Look for who funded it, and treat an undisclosed funder as the answer to that question.
One, was anything measured
The most common claim in this category is a percentage of users who agreed with a statement. That records an opinion after a purchase, which is a survey of hope as much as of anything else.
A measurement is a number taken off a face with an instrument: muscle thickness on ultrasound, a contour distance, a wrinkle depth score read by someone who does not know which group the photograph came from.
If the claim contains the words agreed, felt, noticed or reported, you are reading a survey.
Two, was there a control group
Without a comparison, everything that happens over twelve weeks is credited to the device: the better skincare people adopt when they start a routine, the improved sleep, the flattering second photograph, and simple expectation.
A sham-controlled study gives one group an inactive device, which is the only way to separate the effect from the belief. Chang and colleagues did exactly that in a double-blind design (10.3390/ijerph17113783).
A split-face design is the other good answer, treating one side and using the other as the control, which removes the person as a variable entirely. Kwak and colleagues used one (10.52660/jksc.2023.29.6.1505).
Three, how many people and for how long
Twenty participants over four weeks is a pilot, and a pilot tells you a study is worth running rather than telling you a result.
Kavanagh and colleagues followed 108 women for 12 weeks (10.1111/jocd.12007). That is the shape to compare against: enough people that individual variation averages out, and long enough for a tissue change to occur.
Be suspicious of any facial claim measured at two weeks, because muscle and collagen do not move that fast whatever the intervention.
Four, who paid for it
Industry funding does not invalidate a study and it does change how carefully you read it. What matters is disclosure: a paper that names its funder and its authors' affiliations is behaving properly.
A brand citing a study it commissioned, never published, and will not show you is telling you something by the omission.
Check for corrections too. Papers occasionally publish disclosures after the fact, and a corrected paper is not a discredited one; an uncorrected undisclosed conflict is worse.
Five, was this device tested, or the category
This is the one that catches everyone, including brands acting in good faith.
The studies above tested facial electrical stimulation as a category, under defined protocols, with research equipment. None of them tested a consumer device you can buy, ours included.
So the honest phrasing is that the category has been shown to do something under study conditions. The dishonest phrasing puts a brand name in front of that sentence, and it is worth noticing which one you are reading.
Six, does the number mean what it sounds like
A percentage improvement needs a baseline. Improving a score from 2 to 3 on an eleven-point scale is a fifty percent improvement and is also almost nothing.
Statistical significance means the result is unlikely to be chance; it does not mean the result is large enough to see in a mirror. Those are separate questions and a headline usually answers only the first.
Watch for the switch from an average to a maximum, where a study's best responder becomes the number on the box.
Seven, was the intervention isolated
If a study tests a device that combines stimulation with light, iontophoresis and cooling all at once, the study cannot tell you which of the four produced the change.
That is not a flaw in the study, which may have been asking a different question; it becomes a problem only when a brand quotes the result as evidence for one of the four.
Eight, does anyone disagree
A field where every published paper agrees is usually a field with a small literature. Papaiordanidou and colleagues report a case where frequency modulation did not deliver the expected benefit (10.1371/journal.pone.0084740), and it sits alongside Downey and colleagues, who found that it did (PMID 21996798).
A brand that cites the supporting paper and never mentions the contrary one is curating rather than informing.
Applying it here
Everything we cite is category evidence rather than a PureLift trial, and we say so on every page that cites it.
The specifications are the part that is specific to our devices, and those are checkable: 7 mA at 500 ohms for Face and Pro with a 4 microsecond pulse, 7.7 mA for Pro Edition, 9 mA for Pro Plus and GLOW, cleared under K190269, K230506, K221443 and K243587 respectively.
Ask any brand for those four things. The answer, or the absence of one, tells you a good deal before you read a single study.
For the full breakdown, see how the nine PureLift technologies work.