Perimenopause weight gain: what the scale hides

Perimenopause weight gain is real, but the scale is a poor witness to it. In the largest longitudinal study of body composition across the transition, weight climbed steadily with age and did not accelerate when the transition began — while the rate of fat gain roughly doubled and lean mass started to fall. Two changes moving in opposite directions can leave the number under your feet looking almost unchanged. And nothing on your wrist measures fat, muscle, or metabolic rate, so much of this article is about what to stop asking your data for.

Why the scale can sit still while the body changes

The SWAN body composition study followed 1,246 women with repeated DXA scans — mean age 47.1 at baseline, mean age 52.2 at the final menstrual period. About two years before that final period, the rate of fat gain doubled and lean mass began to decline. Both trajectories continued until roughly two years after, then flattened out.

Body weight, measured in the same women, did something different: it rose linearly through premenopause with no acceleration at the transition, and its climb flattened afterwards. The lead author put it plainly — weight gain did not accelerate during the transition, although fat gain did.

That is a strange and useful result. The trade-off — fat arriving, lean tissue leaving — is the part specific to this window, and body weight is the single measurement most likely to cancel it out. If you have been weighing yourself daily and concluding that nothing is happening, or that everything is, you have been reading the one number designed to blur the two.

The “metabolism collapses at 40” story does not survive measurement

The most quoted explanation for midlife weight is that metabolism slows down. The best available measurement says otherwise. Pooling doubly-labelled-water data from more than 6,400 people across 29 countries, aged eight days to 95, researchers found that once daily energy expenditure is adjusted for fat-free mass, it holds flat through adulthood — roughly age 20 to 60 — and only declines after that. There is no cliff at 45.

Meanwhile, a four-year longitudinal study that measured the same women annually through the transition did record falling resting metabolic rate and falling physical-activity energy expenditure. Both findings are correct, and putting them side by side is the whole point: total daily burn can drop while the burn per unit of lean tissue does not change. Lose lean mass, move a little less, and you spend less energy in a day without your body having become less efficient at anything.

The difference is not academic. One version says your metabolism betrayed you and there is nothing to be done. The other names two specific, measurable things — how much lean tissue you carry and how much you move — and neither of them is a verdict.

Where it goes is a separate question from how much

In that same four-year study, all the women gained subcutaneous fat. Only the ones who became postmenopausal over the four years showed a significant increase in visceral fat, the deep abdominal kind that sits around the organs.

Redistribution can run well ahead of any change in total weight, which is why the cheapest instrument in this article outperforms the expensive one. A tape measure at the waist, taken the same way on the same morning each month, tracks the thing that is actually moving. Your own line over six months is the comparison that carries information — not a population chart, and not last Tuesday.

Why the calorie number on your watch cannot referee this

Two things are worth knowing about the energy figures your watch reports.

First, accuracy. In a Stanford comparison of seven wrist devices against indirect calorimetry in 60 adults, heart rate was measured well — the Apple Watch had the lowest error at about 2% — but not one of the seven produced energy expenditure estimates with less than 20% error. Same wrist, same sensor, radically different reliability depending on which number you read.

Second, and less often mentioned: those calorie figures are partly computed from the profile you entered. Age, height, weight and sex feed the estimate along with heart rate and motion. Raise the weight in your Health profile and the estimated burn rises, without anything about you having changed. A number that moves when you edit a settings field is not evidence about your metabolism. These are estimates, not lab measurements — useful for spotting a change in your own activity over weeks, useless as an energy balance ledger.

What the wrist does register: body fat and warm nights

There is one place where body composition and wearable data genuinely meet, and it is not the one people expect.

For years the assumption ran the other way — that carrying more fat meant more estrogen from peripheral conversion and therefore fewer hot flashes. SWAN found the opposite. Among 1,776 women assessed by bioelectrical impedance, each standard deviation higher percentage of body fat came with higher odds of reporting vasomotor symptoms (odds ratio 1.27, 95% CI 1.14–1.42). The association survived full adjustment and was not explained away by reproductive hormones, which fits a thermoregulatory reading: fat is insulation, and insulation makes heat harder to shed.

So the composition change and the warm, broken nights travel together, and the nights are the part your watch can put a date on — through overnight wrist temperature relative to your own recent baseline, and the pattern described in night sweats and wrist temperature. The traffic runs in both directions, too: disturbed sleep rises from roughly 16–42% before the transition to 39–47% during it, and broken sleep and body composition each influence the other. Your data shows the night. It does not show the cause.

How Perigee reads this

We built Perigee around a comparison your watch can actually support: you against your own recent weeks. There is no body-composition estimate in the app, and there will not be one, because the wrist cannot produce it — that is the same honesty we apply to every signal on the science page. What it can do is keep a dated record of what was measured — overnight temperature, resting heart rate, sleep timing — beside a line you wrote about how the week went, which is exactly the material that is hardest to reconstruct from memory in a ten-minute appointment. If you are setting this up for the first time, how to track perimenopause with an Apple Watch covers what to turn on.

One small thing to try this week

Measure your waist once, write the number down with the date, and put a reminder in for four weeks from now. That is the whole exercise. Two points a month apart, taken the same way, will tell you more about this particular window than a month of daily weigh-ins — and if what you find is weight change alongside deep fatigue or heavy bleeding, that is a conversation and a blood test, not a tracking project. Bringing a dated record rather than an impression is also what makes that appointment go faster.