The dress fits. That part was true, and it arrived roughly when they said it would.

What arrives alongside it is harder to name. You notice it in a changing room in March, under the overhead light nobody would choose: your upper arms have gone soft in a way that has nothing to do with fat, your ponytail has lost its heft in your hand, and your face has a hollowness that reads as tired rather than slim. You hold up a photograph from last summer to check. You are not imagining it.

Here is the part nobody puts in the leaflet. Most of what is on that list will come back. One thing will not, and the months in which you can still protect it are the months you are standing in right now.

What comes back, and what does not

What you lose Does it return? What decides it
Hair Yes, on its own Settles within months of the weight settling. Ferritin, if it does not
Muscle Yes, if you go and get it Resistance training and 1.2 to 1.6 g of protein per kg daily
Bone No How much weight comes off. Past 10%, hip loss roughly doubles
Skin No Age, sun history, how much came off. Nothing sold for it changes this

That table is the whole argument. The rest of this explains where each line comes from, and what it asks of you while there is still time to act on it.

You cannot avoid this by picking a different drug

In the body-composition substudy of SURMOUNT-1, people on tirzepatide lost 21.3% of their body weight, roughly three quarters of it fat and one quarter lean tissue. The number that matters is not that one. It is that the placebo group, losing far less weight by diet alone, lost it in the same proportion.

Lean tissue leaves alongside fat whenever a body gets smaller, whether the appetite was quieted by an injection or by discipline. That closes off the two escape routes women are usually offered. Switching molecules will not save your muscle, and neither will going slowly: the main review of weight-loss speeds, covering thirteen studies from 0.2 to 3.2 kg a week, found broadly similar body composition outcomes at every rate, and did not examine skin at all.

What is left is the only variable you actually control, which is what you do during.

The hair grows back

Something real is happening here, and it is smaller than the internet suggests. A 2026 BMJ study of more than 12,000 patients found hair loss at 6.91 cases per thousand person-years against 5.04 in a comparison group, and a meta-analysis covering over a million exposures put the pooled odds at 1.40. In absolute terms that is two to three extra cases per thousand people per year.

One detail in that meta-analysis is worth more than any of the risk ratios. Alopecia areata, which is autoimmune, showed no increase whatsoever. What rose was telogen effluvium, the diffuse shedding that follows any physical stress, and it rose the longer people stayed on treatment. That is the signature of rapid weight loss and a thinner diet, not of a drug attacking follicles.

Which means the lever is nutritional, and the one most often found wanting is iron. Ask for a ferritin test at the same appointment where you discuss your dose. Our guide to the two labs worth requesting explains why that number tells you more than most.

From the Viva shop: CANPREV Iron Bis-Glycinate 25 (120 caps), $16.09. Chelated iron at a rebuilding dose, gentle enough that you will still be taking it in week three, which is where most iron supplements quietly lose people. Test first, then start.

The muscle comes back, but only if you go and get it

The raw figures look alarming. In the STEP 1 scan substudy, patients on semaglutide lost 15% of their body weight and 9.7% of their lean mass, a 60:40 fat-to-lean split.

Two things stop those numbers meaning what they appear to mean. A body scan does not measure muscle; it measures lean mass, which includes water, glycogen, connective tissue and the organs themselves, and a body carrying forty fewer pounds needs less fluid and less scaffolding to run. More importantly, strength did not follow the scan downwards. In SEMALEAN, which followed 106 people for a year, lean mass fell for seven months and then held, grip strength rose by 4.5 kg, and the proportion of patients meeting the definition of sarcopenic obesity fell from 49% to 33%.

The intervention that produces this is not sold in a bottle. A 2024 review in Diabetes Care found that supervised resistance training over ten weeks or more yields roughly 3 kg of lean mass and a 25% gain in strength. In the S-LITE trial, women who combined a GLP-1 with supervised training gained lean mass over the same period in which the drug alone lost it. Two sessions a week with weight in your hands is the single most effective thing available to you, and it is also the only one that changes the shape of your arms. Muscle under skin is the closest thing to a contour there is.

Pair it with 1.2 to 1.6 grams of protein per kilogram of body weight daily, spread across meals rather than banked at dinner, because appetite on these drugs is usually lowest in the evening.

From the Viva shop: ERGOGENICS Plant Protein + Greens, Vanilla (720g), $70.59. Twenty grams in water solves the real problem, which is getting protein in on the mornings when appetite has simply left the building. NOW Creatine Monohydrate Pure Powder (227g), $16.09 is five grams a day and the most studied training supplement there is, now as well evidenced for women in midlife as it ever was for athletes. More on the dose by decade here.

The bone does not come back

This is the line in the table that should change what you do this month, and it is the one that gets the least coverage.

The largest study yet, published in the Journal of Clinical Endocrinology and Metabolism in 2026, matched 255 GLP-1 users against 255 controls. It found significantly faster bone loss at the spine. At the hip, the excess appeared only in patients without diabetes, and both groups lost bone at a similar rate overall. The only randomised trial to examine this found nothing at any site, and no study has yet shown more fractures. Read honestly, that means the effect is real at the spine and unsettled everywhere else.

What the same study found without ambiguity is the pattern the whole article turns on. Bone loss at the hip tracked how much weight came off, and women who lost more than 10% of their body weight lost roughly double the bone density of those who lost less.

Professor Wendy Kohrt has spent thirty-five years researching women's bone. She calls menopause combined with rapid weight loss a perfect storm, and she notes that in her earlier work, the women who regained the weight did not regain the bone. That is expert judgement rather than trial data, and it should be weighed as such. It is also the reason the window is a window: fat comes back easily and famously, and the bone that left with it does not come back on the same terms.

If you are over fifty and starting a GLP-1, ask for a baseline bone scan. It costs one appointment, and it converts a vague worry into a number you can measure against later.

From the Viva shop: CANPREV Magnesium + Vegan D3 & K2 (90 veg caps), $32.95. The unglamorous trio that directs calcium where it is meant to go. Bone responds over months rather than weeks, which is the argument for starting it the week you start the injection rather than the year afterwards.

Neither does the skin, and nothing you can buy changes that

This is where the internet behaves worst, so here is the straight version.

Whether skin retracts depends mostly on things already decided. Elastin, unlike collagen, is barely replaced in adult skin. Decades of sun degrade the elastic network. Older skin remodels more slowly, and the more weight that comes off, the more surface there is left over.

The supplement currently marketed hardest for this is Pycnogenol, an extract of French maritime pine bark, and it deserves a straight answer because the gap between what it is sold for and what has been tested is wide. Everyone cites the same study, which followed 20 postmenopausal women for 12 weeks and reported 25% better skin elasticity. It had no placebo group, no blinding, and was funded by the company that makes the extract. More to the point, it has never been tested in anyone who has lost weight, for any skin outcome at all. A gene reading in twenty unblinded women is not a finding about a stomach that has deflated by a fifth of its owner's body weight.

Collagen peptides stand on firmer ground, though not the ground you want. A meta-analysis of 26 randomised trials in 1,721 people found real improvements in elasticity and hydration against placebo. Every one of those trials was conducted in ordinary ageing skin, and none in skin left slack by weight loss.

From the Viva shop: GENUINE HEALTH Clean Collagen Bovine (280g), $29.59, ten grams of hydrolysed collagen that dissolves into coffee without announcing itself. We stock it for the elasticity evidence in ageing skin, which is what it has actually been tested on, and we would rather say that plainly than sell you a promise about your stomach. DERMA E Anti-Aging Regenerative Day Cream (56g), $39.89 is the topical half of a sensible routine.

Is any of this different for women?

Start with something odd about this field. Almost everything known about these drugs is known from women: STEP 1 was 73% female, the scan substudy behind most of the muscle headlines was 73% female, and the 2026 bone study was 92% female. A women's magazine writing about GLP-1s is, unusually, writing about the population that was actually studied.

Three differences hold up. Women lose more weight, 9.6% against 7.2% for men in the trial with the most balanced enrolment, partly because at the same body weight women reach around a third higher drug exposure. They also feel it more, reporting nausea and vomiting at roughly two and a half times the male rate, with twice as many abandoning treatment because of it. For anyone still deciding whether to start, that is the most practical sentence on this page, and it is almost never said aloud. And the hair signal sits mostly with women, reported by 7.1% of women against 0.5% of men in tirzepatide's own trials.

One widely repeated difference does not survive the data. Muscle loss is not proportionally worse in women. Three quarters of the weight lost was fat in both sexes, and the statistical test for a sex difference in lean mass came back at 0.93, which is as close to nothing as this kind of number gets. What differs is the reserve rather than the rate. Women begin with less muscle and less bone, so an identical proportional loss lands nearer the thresholds that matter clinically. That is a reasonable inference and not a measured finding, and nobody has tested it yet.

One honest gap

A great deal has been written about GLP-1s in perimenopause, most of it confident. Almost none of it is tested. No trial has enrolled perimenopausal women as a defined group or measured whether falling oestrogen and rapid weight loss compound one another. Both independently take muscle and bone, so clinicians expect them to stack, and that expectation is reasonable. It is not yet a finding.

Where this leaves you

None of it is an argument for stopping, and the dress still fits.

It is an argument for treating these months as active ones rather than a period to be waited out. Two sessions with weights, protein at every meal, a ferritin number, and past fifty a bone scan. Your hair will sort itself out. Your skin will do what it was always going to do, and anyone selling you otherwise has not read the studies. The bone is the one keeping score.

Frequently asked questions

I am already six months in. Have I missed the window?

No, though the arithmetic changes. Bone loss is cumulative and what has gone has gone, which is an argument for measuring now rather than assuming. Muscle is different: the training response does not expire, and the S-LITE participants who gained lean mass were training alongside the drug rather than after it. Starting in month seven is worth considerably more than starting after you stop.

When does the hair shedding start, and when does it stop?

Telogen effluvium characteristically lags its trigger by two to four months, so shedding that begins in June usually reflects what your body was doing in March. It typically resolves over roughly six months once the trigger settles. Thinning that is still worsening past six months warrants a ferritin and thyroid check rather than more patience.

Do I need a gym for the resistance training to count?

No, but you do need progressive load, which is the part home routines usually drop. The trials that produced these results used supervised programmes two to three times a week. Adjustable dumbbells and a bench replicate the principle at home provided the weight goes up over time; bodyweight work alone tends to plateau once you can complete the sets comfortably.

How much calcium and vitamin D should I be getting?

Canadian guidance is 1,000 mg of calcium and 600 IU of vitamin D daily for women aged 19 to 50, rising to 1,200 mg and 800 IU from 51. Reaching the calcium figure through food is harder than usual on a GLP-1 because total intake falls, which is the practical case for counting it rather than assuming it.

Would a lower dose protect my muscle and bone?

The evidence says the cost tracks how much weight comes off rather than which molecule or dose delivered it, so a lower dose that produces less loss would be expected to cost less. That is a conversation to have with the clinician prescribing it, not a change to make on your own, and it is a genuine trade against the benefit you started for.

Which scan should I ask for?

A DXA scan of the hip and spine, which is the same measurement used in the studies described here. Ask for the numbers rather than only the verdict, because the value of a baseline is that a future scan can be compared against it.

Shop the story

Chosen by the Viva editors. Viva sells what it recommends and takes no outside commission on editorial picks. Prices in Canadian dollars and subject to change.

Nothing here replaces the conversation with the clinician prescribing your medication, and nothing here is a reason to change a dose or stop a treatment on your own.

Read next

Sources

This article is for information and does not replace individual medical advice.