Menopause tracked with abrupt shifts across body systems in 300 million lab tests from a million women

Menopause is usually described as a slow fade: a few years of irregular cycles, then a gradual settling into life after periods. In a study of 300 million laboratory tests from routine medical care, published on October 12 in Nature Medicine, researchers argue that for many body systems the change is closer to a step. Bone, liver, kidney, blood fat and hormone measures shifted abruptly around a woman’s final period, the researchers report, while the hormonal run-up began more than a decade earlier.

The team, led by Uri Alon at the Weizmann Institute of Science in Israel, calls this a “physiological cliff”, and that word has drawn as much attention as the data. Martha Hickey, a menopause specialist at the University of Melbourne who heads menopause services at The Women’s Hospital in Victoria and was not involved in the work, called the cliff framing “unnecessarily alarming” for women. “Physiological cliff is not a scientific or medical term,” Hickey said.

Both things can be true. The study is a large and inventive attempt to see what happens around the final period in ordinary medical records, and it is also a statistical reconstruction from one-time tests, with real limits on what it can show about any one woman.

Why menopause is hard to pin to a date

Menopause has a strict definition, and it can only be applied in hindsight. It is defined retrospectively, 12 months after the final menstrual period, according to the US Preventive Services Task Force, and the few years before it, when cycles become irregular, are known as perimenopause or the menopausal transition.

That makes the final period the natural anchor for studying what menopause does to the body, and also an elusive one. Most of what is known comes from studies that follow women year by year and record when their periods stop. Those studies are necessarily small, and the paper’s authors argue that they have tended to characterize menopause as a gradual transition. Huge stores of routine blood test results exist, but such large biomedical datasets remain underutilized, the authors write, because a woman’s age at her final period is almost never written down alongside them.

How the menopause study rebuilt the years around the final period

Alon’s group borrowed an idea from astronomy, where mathematical methods are used to sharpen images blurred by a telescope. Here the blur comes from the fact that women reach their final period at different ages, which in the US can vary from 45 to 55 years or older. Averaging lab results by age smears out any sudden change, because at any given age some women are well past menopause and others have years to go. The researchers’ framework rebuilds the path of each measure around the final period by removing the blurring caused by variation in the age at which women reach it. In effect, it works backward from the known spread of menopause ages to what a typical woman’s results would look like if everyone were lined up by the date of her last period.

The team ran the method on two large collections of routine lab results, one American and one Israeli, covering 1999 to 2024: more than 300 million tests from over a million women in all. Men’s results went through the same analysis for comparison. The work used retrospective data and self-reported menopause information, meaning records gathered for other purposes and analyzed afterward. The method does not need each woman’s own date: it works from the spread of menopause ages across the population.

Which lab tests jumped at menopause

In the American records, 52 of 90 laboratory tests shifted abruptly around the final period, by margins unlikely to be chance. The list ran from sex hormones and bone measures to liver and kidney tests, blood chemistry and blood fats. The pattern was much the same in the Israeli data. Men of similar ages showed no such jumps. In an interview with Newsweek, Alon said the team saw “gradual changes starting about a decade before menopause on average and accelerating in the years before the final menstrual period and then a jump at the final period.” On timing, Alon was careful: “We can’t say for sure how rapid it is but it looks like it’s within the year of the final menstrual period.”

Many of the changes point toward the direction of increased disease risk. Alon singled out bone density, liver damage tests, and blood fats including triglycerides that count among the risk factors for heart disease. Not every shift was unwelcome. “The silver lining is an improvement in iron and hemoglobin markers and a reduction of anemia and the reason is the stop of menstrual bleeding,” Alon added.

The hormones themselves moved much earlier. Anti-Mullerian hormone, which tracks the remaining supply of eggs, began falling rapidly about 15 years before the final period, and estradiol, the main form of estrogen, about 12 years before. Follicle-stimulating hormone, which rises as the ovaries respond less, increased more rapidly around 14 years before the final period. The authors’ own summary puts it as dysregulation that began more than a decade before the final period.

What the menopause data showed about hormone therapy

Women who were taking menopausal hormone therapy showed smaller changes in many measures, above all in bone, liver and kidney tests. The authors describe the shifts as markedly attenuated by menopausal hormone therapy, though the effect is something that will need to be confirmed in future studies.

Alon was blunter about the hormone therapy result. The finding “is observational and based on a small minority of women that take [menopausal hormone therapy], and therefore should be taken with a grain of salt,” Alon told Newsweek. According to Hickey, only about 9% of the women were using hormone therapy. Because they chose it, they may simply have been healthier to begin with, a pattern researchers call the healthy user effect.

Guidance based on randomized trials points the same cautious way. In 2022 the US Preventive Services Task Force recommended against using combined estrogen and progestin for the primary prevention of chronic conditions after menopause, and against estrogen alone for the same purpose, having found no net benefit. Hormone therapy remains a treatment for symptoms such as hot flashes, which this study did not examine.

JoAnn Manson, a professor of medicine at Harvard Medical School who was not involved, pointed Newsweek to the question that remains open. “Whether early initiation of estrogen therapy could protect against and counter the adverse physiological changes of the menopause transition is unknown because most randomized trials begin treatment after this critical window and later in menopause,” Manson said.

Why some experts reject the menopause cliff

Hickey’s objection goes beyond the word. In comments reported by Australian media, Hickey pointed out that the study looked back at existing records and did not follow the same women over time, which ranks it relatively low in the hierarchy of evidence. Pinning down each woman’s menopause timing from lab records is difficult, and errors could shift where the apparent cliff falls. And blood tests are not random: people usually have them because a doctor has a reason to order one, so the data may lean toward women who already had health concerns.

Rebecca Thurston, who directs a women’s health research center at the University of Pittsburgh, raised a related problem. Because each test is a one-time snapshot, Thurston told Newsweek, the researchers “cannot rigorously distinguish menopause (e.g., reproductive aging) from chronological aging”, and they had very large amounts of missing data on key variables.

Hickey has argued for years against treating menopause chiefly as a deficiency. A 2024 paper Hickey led in The Lancet proposed moving away from focusing on menopause as an endocrine deficiency, and toward helping women manage it as active partners in their care.

How the menopause results compare with long-running studies

Some of the abruptness is not new, and the best existing evidence supports it for certain measures. The Study of Women’s Health Across the Nation, known as SWAN, is a prospective study of the menopausal transition that enrolled 3,302 US women from five ethnic groups and examined them every year.

In one SWAN analysis of 862 women whose final period could be dated, bone density loss began 1 year before the final period and slowed, without stopping, two years after it. Of the 10.6% of spine bone density lost over a decade, about 7.4 percentage points, roughly 70% of the total, went in those three years. Another SWAN analysis, of 1,054 women, found that total cholesterol, LDL cholesterol (the kind that builds up in arteries) and a related protein rose sharply within a year of the final period. Its authors concluded that women experience a unique increase in lipids, or blood fats, at that time.

But the same cholesterol analysis found that the other heart risk factors it measured, including blood pressure, blood sugar and markers of inflammation, changed in a straight line, a pattern its authors called indicative of chronological aging. Whether the Nature Medicine study’s much longer list of abrupt changes holds up against direct, year-by-year measurements like these is one of the questions it leaves open.

Limits of reconstructing menopause from routine lab tests

Reconstructed timelines. No woman in the main analysis was followed through her own final period. The timelines are statistical reconstructions from one-time tests, and they rest on the method’s assumptions about how menopause ages are spread, which draw on self-reported menopause information that can be imprecise.

Lab values only. The study tracked test results. It did not count fractures, heart attacks or liver disease, so how far the jumps translate into illness is not shown.

Who was in the data. The problems Hickey raised run through everything above: routine tests come from people who had a reason to see a doctor, and the women on hormone therapy chose it.

Population averages. The study describes the typical pattern across a population. It does not say how abruptly any one woman’s tests change, or what the changes mean for her own risk of disease.

What the menopause cliff finding means for women

Alon has expressed the hope that the results “don’t cause stress or fear of menopause”. Andrew Shelling, a reproductive health researcher at the University of Auckland, said the results should be “viewed as hypothesis-generating rather than conclusive”. How the years around menopause affect a particular woman’s health is assessed by her doctor. Do not stop or change a prescribed medicine without talking to your doctor.

Routine tests from a million women show change clustered around the final period; whether it deserves the word cliff, and what it means for health, is not yet settled.