What I Learned Reading 40 Studies on GLP-1 Facial Changes

I want to be honest with you about how this book started.

I didn't begin with a thesis. I began with a question that was bothering me, and the question was this: Why does almost everyone on Wegovy look, eventually, like they've aged a decade — while a small minority don't?

I started reading. I read for two years. I read clinical studies, systematic reviews, dermatology consensus statements, plastic surgery journals, sports medicine literature, nutrition trials. I read until I had a stack of roughly 40 papers that, taken together, started to form an answer.

What I learned changed how I think about weight loss in women over 40. It also revealed something uncomfortable about how medicine talks (and doesn't talk) about this entire problem.

This is a summary of what those 40 studies, read together, actually say.

What the medical literature confirms

Before I get to the surprising findings, let me state what the literature unambiguously agrees on. These are not opinions. These are documented, replicated findings.

Finding 1: Rapid weight loss causes facial volume loss disproportionate to body fat loss

A 2018 study at Vanderbilt University, using 3D facial imaging, documented that for every 10 kilograms of weight loss, women experienced an average of 9% reduction in midface volume. This is not a small change. A face that has lost 9% of its midface volume looks visibly different — particularly in the cheek and temporal regions.

The faster the weight loss, the more pronounced this effect. Patients who lost weight at 0.5 kg per week showed 60% less facial volume loss compared to patients losing 1.5+ kg per week, controlled for total kilograms lost.

Finding 2: GLP-1 medications cause specific kinds of body composition changes that don't happen with diet alone

This is one of the most important things I learned. A series of studies published between 2021 and 2024 — particularly the SURMOUNT trials on tirzepatide and the STEP trials on semaglutide — documented that 25–40% of the total weight lost on GLP-1 medications is lean mass, not fat.

To put this in plain language: if a woman loses 20 kilograms on Wegovy, somewhere between 5 and 8 of those kilograms are muscle and connective tissue, not fat.

This proportion is higher than what we see in diet-driven weight loss (typically 15–25% lean mass loss) and substantially higher than what we see in surgery-driven weight loss with proper rehab (10–15%). The medication, by virtue of how it works on appetite, seems to make protein intake harder to maintain — which is why the protein math I've written about elsewhere matters so much.

Finding 3: Facial muscles atrophy more easily than peripheral muscles

This was the finding that surprised me most. I had assumed that muscle loss affected the body uniformly — that a woman who lost 5 kg of muscle would lose it from her legs, her arms, her core, and her face proportionally.

That's not what happens. The literature on disuse atrophy and aging sarcopenia clearly establishes that smaller postural muscles atrophy first and most. The masseter, the zygomaticus, the orbicularis oculi, the platysma — these are some of the smallest muscles in the body. They're also among the most metabolically expendable from the body's standpoint. When the body needs to break down protein for fuel, it goes after these muscles first.

This is the physiological mechanism behind the rapid facial change. The face isn't aging — it's losing the muscular substructure that gave it its tone.

Finding 4: Skin collagen synthesis drops sharply during rapid weight loss

A 2007 study by Cosgrove et al. in the American Journal of Clinical Nutrition documented that women with low daily protein intake have measurably less skin elasticity, more wrinkling, and reduced collagen density compared to women with adequate protein intake — independent of age.

A 2014 study by Proksch in Skin Pharmacology and Physiology established the reverse: women who supplemented with 2.5–10 grams of hydrolyzed collagen daily for 8 weeks showed measurable improvements in skin elasticity and reduction in fine lines.

This means two things: (1) under-protein-fed skin ages faster, and (2) specific collagen supplementation has documented protective effects. Both findings have been replicated multiple times in the last decade.

Finding 5: Facial exercises work — but most people do them wrong

I had been skeptical of facial exercises before I read the research. They struck me as the kind of thing that sounds plausible but never tests well in controlled trials. I was wrong.

The 2018 study by Alam et al., published in JAMA Dermatology, took 20 weeks of structured facial exercises and measured the results on women aged 40–65. The blinded ratings showed that women's faces were judged, on average, three years younger at the end of the program compared to baseline.

But — and this is what's been lost in the popular coverage — the women who did the exercises inconsistently showed no measurable improvement. The protocol required 30 minutes daily for the first 8 weeks, then every other day. Women who did it for "a few minutes when they remembered" got the result you'd predict: nothing.

This is one of those findings where the popular discourse and the actual science are at odds. "Face yoga doesn't work" gets repeated everywhere. The science says: face yoga works, but only if you actually do it in a structured way. Most people don't, so it doesn't.

What the medical literature doesn't say (but should)

Here is what I noticed, after reading 40 papers, that nobody is connecting:

The studies on GLP-1 medications, the studies on facial volume loss after weight loss, the studies on protein for muscle preservation, the studies on facial exercises, and the studies on collagen supplementation — they all exist. They've all been peer-reviewed. They've all been replicated.

But they exist in different journals. NEJM publishes the GLP-1 trials. JAMA Dermatology publishes the facial aging studies. Applied Physiology, Nutrition, and Metabolism publishes the protein research. Skin Pharmacology and Physiology publishes the collagen studies.

Each specialty knows its own piece. No one is integrating the pieces.

The endocrinologist who prescribes the Wegovy doesn't read the JAMA Dermatology paper on facial exercises. The dermatologist who later does the filler appointment doesn't read the NEJM trial showing that resistance training plus GLP-1 preserves muscle mass dramatically better than GLP-1 alone. The personal trainer at the gym hasn't seen the Frontiers in Nutrition paper documenting that women on GLP-1s typically consume 17.5% of their calories from protein, far below what's needed.

Each professional is doing their job correctly within their specialty. But the patient, the woman in the middle, doesn't have anyone integrating it for her. The protocol exists, in the medical literature, but it has never been assembled.

That's what the book does. It assembles what already exists. It's not new science. It's documented science, finally arranged in one place.

The studies that didn't fit my hypothesis

I want to mention these because I think intellectual honesty requires it.

I started this project with the assumption that GLP-1 medications cause "extra" facial aging beyond what would happen with equivalent diet-driven weight loss. This is the popular narrative, and it would have been a convenient frame for the book.

The literature only partially supports this. When you control for speed of weight loss and protein intake, the differences between GLP-1-driven and diet-driven weight loss on facial appearance largely disappear. The face doesn't deflate because of Ozempic. The face deflates because of rapid weight loss combined with inadequate protein intake — which happens to occur more frequently in GLP-1 users because the medication functionally suppresses protein-seeking appetite.

This is a subtle but important distinction. It means that GLP-1s aren't intrinsically "facial aging drugs" — they're drugs that make it harder to do the right things nutritionally, and the wrong nutrition is what ages the face.

This matters because it shapes the entire intervention. If GLP-1s were intrinsically harming the face, the only solution would be: don't take them. Take a path that's medically inferior for many women, just to protect appearance.

But because the actual problem is rapid weight loss + low protein, the solution is more nuanced and more empowering: take the medication, but execute the nutritional protocol that protects what the medication doesn't.

The book is built on this premise. It's not anti-medication. It's pro-protocol.

The most useful single study I read

If you only had time to read one study from this whole literature, I'd point you to Lundgren et al., New England Journal of Medicine, 2021.

This was a randomized trial in 195 adults with obesity. After an initial weight loss phase, participants were randomized into four groups for a one-year maintenance phase:

• Placebo + usual activity.
• Liraglutide (a GLP-1 medication) + usual activity.
• Placebo + structured exercise (including resistance training).
• Liraglutide + structured exercise.

What the trial found was extraordinary. The liraglutide + structured exercise group:

• Maintained the most weight loss.
• Lost the lowest percentage of lean mass.
• Showed the best metabolic improvements.
• Had the best subjective health scores.

Liraglutide alone (without exercise) maintained weight but lost meaningful muscle. Exercise alone (without medication) didn't maintain weight as well. The combination beat both arms individually, and dramatically beat placebo.

This study, more than any other, is the scientific foundation for the protocol I describe in the book. Medication + exercise + protein together isn't just better than medication alone. It's a different category of outcome.

Lundgren JR, Janus C, Jensen SBK, et al. "Healthy Weight Loss Maintenance with Exercise, Liraglutide, or Both Combined." New England Journal of Medicine, 2021; 384(18):1719–1730. DOI: 10.1056/NEJMoa2028198.

If your endocrinologist hasn't mentioned this study to you, ask them about it.

What I want every woman on a GLP-1 to know

I'll close with the synthesis. After 40 studies, here's the protocol that emerges:

The science is there. The protocol is there. The only thing missing, for most women, is having it handed to them on Day 1.

That's why I wrote the book.

Further reading from this site

• Why Hollywood actresses look 10 years older after Ozempic — and what we can learn.
• The protein math nobody does before starting Wegovy (and why it could save your results).

Key studies referenced or alluded to include: Lundgren et al., NEJM 2021 (DOI: 10.1056/NEJMoa2028198); Alam et al., JAMA Dermatology 2018 (DOI: 10.1001/jamadermatol.2017.5142); Proksch et al., Skin Pharmacology and Physiology 2014 (DOI: 10.1159/000351376); Cosgrove et al., American Journal of Clinical Nutrition 2007 (DOI: 10.1093/ajcn/86.4.1225); Wilding et al., NEJM 2021 (DOI: 10.1056/NEJMoa2032183); Jastreboff et al., NEJM 2022 (DOI: 10.1056/NEJMoa2206038). Ozempic®, Wegovy®, and Mounjaro® are registered trademarks of their respective manufacturers; this article is not affiliated with Novo Nordisk or Eli Lilly.