Estrogen, Progesterone, and Testosterone: What Really Happens to Our Hormones During Perimenopause and Menopause?

When we talk about perimenopause and menopause, estrogen gets most of the attention. We often hear that menopause is essentially a story of “estrogen declining.”

Eventually, estrogen does decline. But that explanation misses much of what is actually happening during the years leading up to menopause. I'd like to walk through common hormonal changes women experience in perimenopause and menopause...

Progesterone often becomes less reliable first. Estradiol becomes more erratic before it becomes consistently low. Testosterone follows yet another trajectory—and while its average decline is more gradual, some women experience a substantial loss of testosterone as well.

Understanding these differences helps explain why symptoms can begin years before the final menstrual period, why one hormone may be helpful before another, and why I don't think of hormone therapy as simply “replacing estrogen.”

It also helps explain why progesterone may be the first hormone worth considering for many women in perimenopause, estradiol becomes increasingly important as the transition progresses, and testosterone may become relevant for a subset of women—particularly when androgen-related symptoms are present.

First, let's put these hormones into the same units

One reason it's difficult to intuitively compare our sex hormones is that laboratory reports use different units for each one.

• Estradiol is generally reported in pg/mL.

• Progesterone is usually ng/mL.

• Total testosterone is generally ng/dL.

If we convert them all to pg/mL, the relative concentrations are surprisingly different.

Consider reasonable illustrative values for a healthy younger woman during the mid-luteal phase:

Hormone	Typical example	Converted to pg/mL
Estradiol	100 pg/mL	100 pg/mL
Progesterone	10 ng/mL	10,000 pg/mL
Total testosterone	30 ng/dL	300 pg/mL

By mass concentration, that works out to approximately:

Estradiol : Progesterone : Testosterone = 1 : 100 : 3

That does not mean progesterone is 100 times more powerful or important than estradiol.

These hormones bind different receptors, have different affinities and signaling effects, circulate differently, and act very differently in different tissues. A serum mass ratio is therefore a useful teaching tool—not a physiologic target we should try to recreate with hormone therapy.

But putting them into the same units illustrates something important:

After ovulation, progesterone becomes an extraordinarily abundant circulating ovarian hormone.

And that progesterone surge is one of the first things that becomes less reliable as we move through perimenopause.

What a normal menstrual cycle actually looks like

During the first half of a normal menstrual cycle—the follicular phase—progesterone is relatively low.

An illustrative early/mid-follicular picture might be:

Estradiol: ~50 pg/mLProgesterone: ~0.5 ng/mL = 500 pg/mLTotal testosterone: ~30 ng/dL = 300 pg/mL

Estradiol then rises as a follicle matures and peaks around ovulation. After ovulation, something dramatic happens. The follicle that released the egg becomes the corpus luteum, which begins producing large amounts of progesterone. Progesterone may rise from less than 1 ng/mL to 5, 10, 15 or even 20+ ng/mL. At 10 ng/mL, that's 10,000 pg/mL of progesterone.

Estradiol also has a second, smaller rise during the luteal phase, while testosterone is relatively stable compared with the enormous cyclical changes in estradiol and progesterone.

So a healthy menstrual cycle isn't simply an “estrogen cycle.” It is a coordinated cycle in which estradiol rises substantially before ovulation and progesterone dramatically changes the hormonal environment after ovulation.

Perimenopause doesn't start with estrogen simply running out

This is one of the biggest misconceptions about perimenopause. Early perimenopause is not necessarily a low-estrogen state.

As ovarian reserve declines, inhibin B and AMH decrease and FSH begins to rise. Follicular recruitment becomes increasingly erratic. During ovulatory cycles, estradiol can remain normal—and in some studies, estradiol exposure actually increases during portions of the menopausal transition. At the same time, mean luteal progesterone decreases. PubMed

So early perimenopause can sometimes look less like:

↓ estrogen + ↓ progesterone

and other times more like:

normal/high/erratic estrogen + increasingly unreliable progesterone.

Why progesterone may be the first hormone that becomes helpful

Progesterone has a unique vulnerability during reproductive aging: we only make very large amounts of it when we ovulate.

As ovarian aging progresses, ovulation becomes less consistent. Even when ovulation occurs, luteal progesterone production may become less robust.

Research examining menstrual cycles across reproductive aging has demonstrated decreasing luteal progesterone as women progress through the menopausal transition. SWAN data also show that the proportion of cycles with robust luteal activity declines substantially in the years approaching the final menstrual period. PubMed

Eventually, anovulatory cycles become increasingly common. And without ovulation, there is no corpus luteum producing that enormous luteal progesterone surge.

Meanwhile, estradiol may still be plentiful.

That creates an interesting window in early perimenopause where a woman may not yet have significant estrogen deficiency, but she is no longer experiencing the same predictable monthly exposure to progesterone that she had in her younger reproductive years.

This is one reason micronized progesterone alone can be a useful starting point for some women in early perimenopause.

Progesterone is about more than periods

Progesterone isn't simply the hormone that prepares the uterus for pregnancy.

Oral micronized progesterone is metabolized into neuroactive steroids, including allopregnanolone, which positively modulates GABA-A receptors in the brain. This is one likely contributor to the sedating and sleep-promoting effects many women experience when oral micronized progesterone is taken at bedtime.

There is clinical evidence supporting this effect.

In a randomized placebo-controlled trial of 189 perimenopausal women, 300 mg of oral micronized progesterone at bedtime did not significantly outperform placebo on the study's primary overall vasomotor-symptom endpoint (hot flashes, for example). However, women receiving progesterone reported significantly greater improvements in night sweats and sleep quality, and there was no increase in depression or serious adverse events. PubMed Central (PMC)

That doesn't mean every woman in early perimenopause should take progesterone, nor does it establish that the commonly used 100–200 mg doses will produce identical effects to the 300 mg dose studied.

But it helps explain why I often think about progesterone relatively early when a still-cycling woman begins experiencing symptoms such as new sleep disruption, nighttime waking, cycle changes or increasingly difficult premenstrual symptoms, particularly when there is not yet a clear picture of estrogen deficiency.

I also think it's worth mentioning that a smaller subset of women feels worse with progesterone, particularly some women with significant premenstrual mood sensitivity. Neurosteroid sensitivity and the brain's response to changes in the progesterone metabolite, allopregnanolone, appear to be important in PMDD. But we don't have a reliable percentage of perimenopausal women who are genuinely “progesterone intolerant,” and I don't think this relatively specific phenotype should overshadow the fact that many women tolerate micronized progesterone well. It's worth mentioning that vaginal insertion of progesterone typically results in less allopregnanolone production, and may be much better tolerated than oral progesterone in this group of women.

As perimenopause progresses, unpredictability becomes the defining feature

Late perimenopause is not simply “less hormones.” It's often more hormonal volatility.

The STRAW+10 reproductive-aging criteria describe late menopausal-transition cycles as having extreme fluctuations in hormone levels and frequent anovulation. Estradiol may be quite low during one interval and surprisingly high during another. FSH can similarly move between reproductive and menopausal ranges. PubMed Central (PMC)

One cycle may still be strongly ovulatory and produce substantial progesterone.

The next may be anovulatory and produce very little.

This is why a single estradiol, FSH or progesterone measurement can be misleading during perimenopause. It is a snapshot of a system whose defining characteristic is increasing variability.

It's also why symptoms don't always correlate neatly with a single hormone level.

For some women, the fluctuation itself may be part of the problem. In these cases, at home on-demand hormone testing devices can help elucidate month-long trends as we trouble shoot individualized hormone support plans.

This is where estradiol increasingly enters the picture

As the transition progresses toward the final menstrual period, estradiol eventually begins spending more time at low levels. And this is when the familiar symptoms of estrogen deficiency often become increasingly prominent:

hot flashes, night sweats, disrupted sleep, vaginal dryness, pain with intercourse, urinary symptoms and, over the longer term, accelerated bone loss.

After the final menstrual period, mean estradiol continues declining for approximately two years before becoming relatively stable at its lower postmenopausal level. PubMed Central (PMC)

For women with bothersome symptoms and no contraindications, systemic estrogen therapy becomes an extremely effective option. Menopausal hormone therapy remains the most effective treatment for vasomotor symptoms and genitourinary syndrome of menopause and prevents bone loss and fracture. For most healthy symptomatic women younger than 60 or within 10 years of menopause onset, major menopause guidance considers the benefit-risk profile favorable when therapy is appropriately individualized. PubMed

So my conceptual progression is often:

• Early transition: consider whether declining/unreliable progesterone is part of the picture.

• Progressing transition: consider whether estrogen volatility or emerging estrogen deficiency is now contributing.

• Late perimenopause/menopause: estradiol often becomes increasingly important.

• And if systemic estrogen is prescribed to a woman who still has a uterus, adequate progesterone/progestogen is necessary for endometrial protection.

Menopause changes the hormone relationships dramatically

Once ovulation has stopped, the corpus luteum disappears from the equation.

Progesterone doesn't literally fall to zero, but the enormous cyclical ovarian production is gone.

Estradiol also becomes consistently low as ovarian follicular activity ends.

Testosterone behaves differently.

Here's an intentionally simplified illustration:

	Younger cycling woman, mid-luteal	Perimenopause	Untreated menopause
Estradiol	~100 pg/mL	Highly variable; sometimes quite high	often ~5–30 pg/mL
Progesterone	~10,000 pg/mL	Progressively less reliable; very low in anovulatory cycles	roughly ~100–300 pg/mL
Total testosterone	~300 pg/mL	Variable, often gradual decline	commonly ~100–300 pg/mL

These are illustrative values rather than reference ranges or treatment targets. Assay, age, cycle timing and individual physiology matter enormously.

But the scale is useful.

A representative younger luteal woman might have:

E2 100 : P 10,000 : T 300 pg/mL

while a representative untreated postmenopausal woman might have:

E2 15 : P 200 : T 200 pg/mL.

Progesterone has fallen enormously.

Estradiol has fallen enormously.

Testosterone typically declines notably.

Testosterone does not follow one universal menopausal trajectory

You'll sometimes hear that testosterone doesn't decline at menopause.

That's too simplistic.

At a population level, testosterone generally does not demonstrate the same dramatic, synchronized collapse around the final menstrual period that estradiol and progesterone do. Testosterone also declines across the reproductive years, meaning that some of the decline we associate with menopause is actually related to aging over a longer period. The international consensus statement on testosterone therapy specifically notes that testosterone concentrations decline during the reproductive years and may remain measurable well beyond menopause. OUP Academic

But population averages aren't individual women.

Some women arrive in menopause with testosterone concentrations substantially below their earlier levels. Ovarian androgen production also remains relevant after menopause, which becomes especially apparent when we look at women who have had their ovaries removed.

In a study of 2,251 postmenopausal women not using hormone therapy, women who had undergone bilateral oophorectomy had approximately 25% lower circulating testosterone than women who had experienced natural menopause—15.5 versus 20.8 ng/dL. Free testosterone showed a similar pattern. PubMed Central (PMC)

A more recent meta-analysis likewise found significantly lower testosterone after bilateral oophorectomy, reinforcing that the postmenopausal ovary continues to contribute meaningfully to circulating androgens. PubMed Central (PMC)

So I think the most accurate way to describe testosterone is:

Testosterone usually declines more gradually than estradiol and progesterone, but individual trajectories vary substantially, and some women experience a significant androgen decline.

Testosterone is also biologically important beyond the laboratory number

Androgen receptors aren't confined to sexual tissue. They are found throughout the body, and testosterone can also be converted locally into DHT or estradiol.

Women may notice changes in sexual desire or arousal during this transition, and some also report changes in energy, motivation, exercise response, muscle strength or recovery.

This is an area where it's especially important to separate biologic plausibility and clinical experience from what randomized trials have actually established.

The strongest evidence for testosterone therapy in women is for postmenopausal hypoactive sexual desire disorder (HSDD) after other contributors have been evaluated. Physiologic testosterone therapy improves several domains of sexual function in appropriately selected postmenopausal women. OUP Academic

Evidence is currently insufficient to recommend testosterone specifically for improving cognition, mood or general well-being, and long-term safety data are limited. OUP Academic

That distinction is important. “We have strong evidence for testosterone in HSDD” does not mean “testosterone only matters for libido.”

It means that HSDD is where clinical trials have currently established a treatment benefit.

When testosterone is used, the goal is physiologic female replacement—not creating supraphysiologic testosterone levels. International consensus recommendations specifically discourage preparations that produce supraphysiologic concentrations. OUP Academic

So should we be trying to restore an estrogen-to-progesterone ratio?

Not generally.

This is where I think the concept of hormone ratios can become misleading.

Remember our illustrative younger luteal woman:

Estradiol = 100 pg/mLProgesterone = 10,000 pg/mLTestosterone = 300 pg/mL

By mass, that's roughly:

1 : 100 : 3

But there is no evidence that a woman using HRT should have exactly 100 times as much progesterone as estradiol.

Estradiol and progesterone don't compete on a scale. They bind different receptors and exert different effects at dramatically different concentrations.

Likewise, testosterone being three times the mass concentration of estradiol doesn't mean the body needs three times as much testosterone therapy.

These ratios describe physiology. They are not HRT dosing formulas.

What I find useful about them is that they make reproductive aging easier to visualize.

The hormonal transition in one picture

Reproductive years

Estradiol cycles predictably. Ovulation produces an enormous progesterone surge every month. Testosterone is present at meaningful concentrations and changes comparatively little throughout each menstrual cycle.

Early perimenopause

Estradiol may still be robust—and may occasionally be higher than it was previously. Ovulation and luteal progesterone production become less reliable. Testosterone may gradually decline but generally isn't changing as dramatically from month to month.

Late perimenopause

Estradiol becomes increasingly volatile. Ovulatory cycles alternate with anovulatory cycles, so progesterone exposure becomes highly inconsistent. Symptoms of estrogen withdrawal or deficiency become increasingly common.

Menopause

Estradiol becomes consistently low. Corpus-luteum progesterone production disappears. Testosterone remains present but may be substantially lower than it was earlier in adulthood for some women.

And this is why hormone therapy doesn't have to start with all three hormones

I don't think there is one universal “menopause hormone cocktail.”

For a still-cycling 43-year-old whose primary change is suddenly poor sleep and increasingly irregular or difficult cycles, progesterone may be a reasonable first hormone to consider, particularly if she isn't showing clear symptoms of estrogen deficiency.

For another woman, hot flashes, night sweats, vaginal symptoms or other manifestations of estrogen deficiency may dominate the picture. Estradiol may be the most important intervention, with progesterone added for endometrial protection when the uterus is present.

For another woman—particularly later in the transition or after menopause—loss of sexual desire and other androgen-related symptoms may prompt an evaluation of testosterone alongside the many other factors that influence those symptoms. When true HSDD is present, testosterone has meaningful clinical-trial support. OUP Academic

And some women ultimately benefit from a combination of hormones.

The point isn't that progesterone should always come first, estrogen second and testosterone third.

It's that these hormones do not decline at the same time, at the same rate, or in the same way.

Hormone therapy should support physiology—not chase a ratio

Perhaps the most important thing to understand about perimenopause is that the hormonal system doesn't simply shut down all at once.

First, predictable ovulation begins to disappear.

Then estrogen becomes increasingly erratic.

Eventually ovarian estrogen production falls dramatically.

And testosterone follows its own, much more individualized trajectory.

That is why two women of exactly the same age can have completely different symptoms—and why their hormone therapy may appropriately look completely different.

I find hormone levels and ratios useful because they help us understand what physiology has changed.

But the goal of hormone therapy isn't to make a woman's laboratory values look exactly like they did at age 25.

The goal is to understand where she is in the transition, what symptoms she is experiencing, what risks and benefits apply to her, what new evidence is emerging related to HRT and longevity, and which hormone—or combination of hormones—actually makes sense for her.