Strength training for perimenopause: what it changes, and what it doesn't

Most of what gets written about training through perimenopause is built on a version of the hormones that isn’t accurate.

The pitch is everywhere now. Lift heavy, protect your bones, fix your symptoms, get your body back. Some of that is well supported. Some of it is being sold a long way past what anyone has actually shown.

The women I coach through this don’t ask whether they should lift. They ask whether it will work. That deserves a straight answer, including the parts where the honest answer is no, or not reliably, or nobody has properly looked.

Almost everything starts from the wrong picture

The standard story is that estrogen declines through perimenopause, so you train around a hormone that shrinks a little every month.

That isn’t what happens. For the first years of the transition estrogen doesn’t fall in a line — it swings, and it often runs higher than it did in your thirties rather than lower. In one study tracking ovulatory cycles through the transition, more than a third of them showed a second estrogen surge dropping into a cycle that was already underway. Persistently low estrogen is a late-perimenopause and postmenopause pattern. It is not the early one.

Two things follow from that, and the second one saves you money.

Any program sold on “declining estrogen” is built on a premise that doesn’t hold for the part of the transition most women are actually in. And a single blood test can’t tell you where you are. The clinical staging system doesn’t use hormone levels at all — it stages you on your cycle pattern, because the levels move too much for one morning’s reading to mean anything.

The window you can’t see until it has closed

Bone is where the timing matters, and the timing is strange.

SWAN is the large US study that scanned the same women repeatedly across the transition. What it found is that bone loss doesn’t track your age. It tracks your final period. Loss began roughly a year before the last period and slowed about two years after it. Inside that three-year window women lost around 2% of spinal bone density a year — about 7% in total. Across the whole decade the spine lost around 10.6%. So the majority of a decade’s bone loss happens in about three years.

Now the part that decides what you do about it. Menopause is confirmed backwards: twelve months with no period. Which means the fast-loss window opens a year before an event you can only identify a year after it has happened. By the time you can name the window, you’re about two years inside it.

There is no version of this where you wait for a signal and then respond to it. If you’re in your mid-forties and your cycles have started changing, the working assumption has to be that the window is open or close to it.

The scale is the wrong instrument

The same study scanned body composition, and the result is the one I most often have to talk women through.

Starting about two years before the final period, the rate of fat gain roughly doubled — from around a quarter of a kilo a year to nearly half. Lean mass began falling at the same point. Across the transition the proportion of fat rose by 3.6% and lean fell by 1.9%.

What the study did not find was accelerated weight gain. The scale didn’t move any faster than it had been. The composition underneath it changed. As the lead author put it, these changes aren’t captured by total weight.

So if you are monitoring this transition with a bathroom scale, you will watch a number stay roughly where it was while the thing you actually care about moves underneath it — and you’ll conclude nothing is happening. Track what you can lift, track your waist, and get a DXA scan if one is available to you. Those three see what the scale can’t.

What lifting is actually shown to do

Here is the strongest piece of evidence, stated with its population attached, because the population is the whole argument.

LIFTMOR took 101 postmenopausal women, average age 65, all of them with low bone mass. Twice a week, thirty minutes, supervised. After a learning phase they worked at five sets of five at over 80% of their one-rep max in the deadlift, squat and overhead press, plus jumping. Eight months later spinal bone density was up 2.9% while the control group lost 1.2%. At the hip the training group roughly held their ground while the controls lost about 2%. Nobody got hurt.

Two honest qualifications. Those women were 65 and postmenopausal — you are not them. And a 2–3% gain is real but modest next to a 7% loss. Lifting changes the slope. It does not cancel the transition.

Then the harder admission. When researchers went looking for exercise trials run specifically in perimenopausal women, they found two. Both were low quality. Neither improved bone density or lean mass. Neither used loads within sight of LIFTMOR’s — one was essentially calisthenics.

My read is that this reflects how little anyone has studied women in the transition rather than evidence that loading fails them. Most exercise trials have simply excluded this group. But I’m not going to tell you that lifting is proven to prevent perimenopausal bone loss, because it isn’t, and you’ll hear that claim made confidently by people selling programs.

The body-composition side is on firmer ground. In one trial, fifteen weeks of supervised resistance training three times a week produced measurable reductions in visceral fat on MRI — the fat packed around the organs, which is the kind that carries the metabolic risk, and the kind the mirror is worst at reporting. Strength, muscle and physical function improve reliably across this whole literature. Those are not small things to buy.

What it probably won’t do

Hot flushes. This gets promised constantly and the evidence doesn’t support it.

The largest and cleanest trial randomised 248 women to three months of supervised exercise or no change. Flushes fell by 2.4 a day in the exercise group and 2.6 a day in the control group. No difference. The Menopause Society’s current position statement lists exercise as not recommended for vasomotor symptoms.

There is one result pointing the other way, and it happens to involve lifting specifically: 58 postmenopausal women did fifteen weeks of resistance training and reported about 2.7 fewer moderate-to-severe flushes a day than controls. That’s a single small unblinded trial from one group, and a larger one is still running.

So it might help you. Just don’t buy a barbell for it. Train for what it does to your bones, your muscle and your capacity, and treat anything that happens to your flushes as change you didn’t pay for.

Sleep and mood are more defensible. In that same trial that found nothing for flushes, sleep quality, insomnia symptoms and depressive symptoms all improved modestly — independently of whether the flushes changed. Modestly is the right word. It isn’t treatment for a depression that’s getting worse, and the risk of a major depressive episode genuinely does rise through this transition.

Two things worth dropping

Training to your cycle. The underlying evidence is weak even in women with regular cycles — a review of 78 studies found differences between cycle phases that were trivial at most, and a 2025 review concluded that periodising training by phase isn’t currently warranted. In perimenopause the premise collapses entirely. Your cycles are irregular by definition, and a surge can land in the middle of one. You cannot sync to a phase you can’t locate.

Reading one hormone panel as a verdict. Covered above, and worth repeating because it gets sold as a service. The levels swing. One reading tells you about one morning.

What I’d actually have you do

The trials that produced results all look boringly similar, which is useful.

On recovery: you will be told that perimenopause means you now need some fundamentally different recovery model. There’s no good evidence for that. The handful of studies are small, postmenopausal and contradict each other. What’s defensible is adjusting sessions to how you’re actually performing on the day — which is what good coaching does for everyone, and matters more here only because your sleep is more likely to be disrupted.

If you’re starting from nothing, learn the lifts first. If you want the longer version of what changes after fifty, I’ve written that here, and the bone-specific detail here.

The conversation that isn’t mine to have

I’m a strength coach. There is a part of this that sits with your doctor, and pretending otherwise would be doing you harm.

The Menopause Society’s position is that for women under 60, or within ten years of menopause, without contraindications, the balance favours hormone therapy for troublesome flushes and for the prevention of bone loss. In SWAN, women who started it during the transition lost about 0.4% less bone per year than those who didn’t.

Those are the facts as the clinical bodies state them. Whether any of it applies to you is a conversation with a doctor, and it is not one I’m qualified to have with you. What I will say is that lifting and that conversation are not alternatives to each other. The women who do best at this do both.

Go to a doctor rather than a coach first if symptoms started before 40, if bleeding is very heavy or prolonged, if your mood is deteriorating rather than fluctuating, or if you already have an osteoporosis diagnosis or have broken something in a minor fall. That last one isn’t a reason not to lift — every woman in LIFTMOR had low bone mass and none of them fractured — but it means the first conversation is medical and the second one is about supervision.

Train the window while it’s open.

The same programming my members run in Palmetto Bay — squats, hinges, presses and carries, loaded properly and progressed every week. Works at any gym with a barbell and a rack.

See the program → First month free. $54/month after, cancel anytime.

This article is general information about exercise and health. It isn’t medical advice and isn’t a substitute for your doctor’s assessment of your own situation.