Twenty-Three Years of Unnecessary Suffering: What the Women’s Health Initiative Actually Showed — and What the Medical Establishment Did With It
If you are a woman 40 to 50, your choice to pay attention to or ignore this information will - not may, WILL dramatically impact your quality of life.
by James Lyons-Weiler, PhD
SPECIAL COMPANION YOUTUBE VIDEO ON THE INTERESTING STUFF CHANNEL. SUBSCRIBE, LIKE, SMASH THE BUTTON AND SHARE OUR NEW COMPANION CHANNEL.
On November 10, 2025, the United States Food and Drug Administration quietly did something it almost never does: it reversed itself.
The FDA announced the removal of what it described as “misleading warnings” on hormone replacement therapy, stating in language that should be read carefully by every woman over forty in this country that “estrogen is a key hormone for women’s health where every single part of a woman’s body depends on estrogen to operate at its best — including the brain, bones, heart, and muscles.”
That sentence took twenty-three years to come out of a federal regulatory agency. Those twenty-three years have a body count — not of deaths from hormone therapy, but of preventable fractures, preventable cardiovascular events, preventable cognitive decline, and an incalculable accumulation of unnecessary suffering by women who were told, on the authority of science, to stop their treatment. Or to never start it.
The story of how this happened is the kind of story this publication exists to tell. It involves a landmark study whose design was unsuited to the question it was used to answer, relative risk inflation dressed up as settled science, a medical establishment that moved faster to alarm than to correct, and a generation of women who paid the cost.
What the Women’s Health Initiative Actually Showed
The Women’s Health Initiative, launched in 1991, was one of the largest randomized controlled trials ever conducted in the United States — more than 160,000 participants, federally funded, designed to address the leading causes of morbidity and mortality in postmenopausal women. The hormone therapy arm enrolled 16,608 women with intact uteruses who were given either a combination of conjugated equine estrogen plus medroxyprogesterone acetate — a synthetic progestogen — or a placebo.
In July 2002, that arm was terminated early, at a median follow-up of 5.2 years, after the data safety monitoring board identified what it classified as a small but statistically significant increase in breast cancer, cardiovascular events, and stroke in the hormone group relative to placebo.
The headlines were immediate, global, and unequivocal: hormone therapy causes breast cancer and heart disease. Prescriptions dropped precipitously. An estimated 50 to 75 percent of women taking hormone therapy at the time stopped within months of the announcement. Medical education shifted. Guidelines changed. A generation of women approaching the menopausal transition was counseled away from the most evidence-backed intervention for managing it.
Here is what the press releases did not say.
The average age of participants in the Women’s Health Initiative hormone therapy trial was sixty-three. Only 3.5 percent of participants were between fifty and fifty-four years old — the age group in which women actually face the decision about whether to start hormone therapy. More than half were over sixty at enrollment, and a significant proportion were more than a decade past the onset of menopause. These were not women navigating a perimenopause transition. They were elderly postmenopausal women, many with pre-existing cardiovascular risk, being started on hormone therapy for the first time years after their estrogen levels had already collapsed.
Readers who follow the primary literature on research methodology will recognize the classification error immediately. This study was not about the population it was used to make decisions for. The inferential leap — from findings in 63-year-old women with a decade of estrogen deprivation behind them, to recommendations for 49-year-old women in the early stages of the perimenopausal transition — is not a conservative extrapolation. It is a category error. The biology is different. The vascular environment is different. The bone and brain adaptation that follows estrogen withdrawal is different depending on how long it has been ongoing. Treating these populations as equivalent for policy purposes was not supported by the study design, and it was not defensible on mechanistic grounds.
The Buried Number
There is a number from the original WHI analysis that received almost no attention in the public coverage of 2002, and that deserves considerably more attention than it has gotten since.
The absolute risk increase for breast cancer in the combined hormone group — the number that actually describes what happened to real women in real terms — was approximately eight additional cases per ten thousand women per year, compared to placebo. That is a relative risk increase that translates, in absolute terms, to a risk that is smaller than the baseline absolute risk increase associated with drinking one alcoholic beverage per day, or with being sedentary, or with being obese.
This is not a dismissal of that risk. Breast cancer is devastating, and any honest reckoning with hormone therapy must include this finding in the conversation. But it is a required correction to the way the risk was communicated. Relative risk, stripped of its denominator, is a rhetorical instrument. When the baseline rate is small, a relative risk of 1.26 can be presented as a twenty-six percent increase in breast cancer — which is how it was widely framed — or as eight additional cases per ten thousand women per year in a specific, older population — which is what it actually meant. These are not equivalent framings. The first drives panic. The second permits informed decision-making.
The benefits that were simultaneously present in the data — significant reduction in hip fractures, significant reduction in colorectal cancer, meaningful relief from vasomotor symptoms that were, for many women in the trial, entirely separate from why they were enrolled — received substantially less coverage. The risk was the headline. The benefit was the footnote.
The Timing Hypothesis: What the WHI Could Not Answer and What It Was Used to Answer Anyway
The single most important development in menopausal medicine since 2002 is what researchers now call the timing hypothesis, or the critical window concept. The central finding is this: the benefit-risk profile of hormone therapy is substantially different depending on when, relative to the onset of menopause, it is initiated.
Women who start hormone therapy within approximately ten years of the onset of menopause, or before approximately age sixty, show cardiovascular outcomes that are neutral to favorable. The vascular endothelium, still responsive to estrogen after a short period of withdrawal, benefits from restoration of estrogen signaling. The brain, in which estrogen plays critical roles in energy metabolism and neurotransmitter function, benefits from sustained signaling during the window when adaptation can still occur. Bone, which loses density at its most rapid rate in the first five to seven years after menopause, can be substantially protected.
Women who start hormone therapy a decade or more after menopause onset — as the WHI participants largely did — show a different profile, because the vascular and neural adaptations to estrogen withdrawal have already occurred, and the introduction of exogenous estrogen into a system that has reorganized itself in its absence produces different biological effects.
This is not a post-hoc rationalization invented to protect a preferred conclusion. It is mechanistically coherent, biologically plausible, and supported by the subsequent re-analyses of the WHI data itself, which consistently show that the subgroup of women closest to menopause onset had the most favorable outcomes. It is the reason the FDA’s 2025 updated labeling endorses initiation within ten years of menopause or before age sixty as the appropriate framing for clinical decision-making.
The timing hypothesis also renders the WHI’s policy implications essentially inapplicable to the population for whom they were most consequentially applied. A study in which 3.5 percent of participants were in the fifty to fifty-four year age range, conducted with oral conjugated equine estrogen and a synthetic progestogen formulation that carries substantially higher thromboembolic and breast cancer risk than transdermal body-identical estradiol with micronized progesterone, cannot tell us what happens to forty-eight-year-old women who start transdermal estradiol at the first signs of the perimenopause transition. It simply cannot. The study was not designed to answer that question. Yet for more than two decades, the answer was treated as settled.
Who Benefits When Women Do Not Treat
This publication has consistently applied a straightforward analytical lens to questions about why evidence gets suppressed, delayed, or misapplied: follow the economic incentive structure and ask who benefits from the current state of affairs.
The abandonment of hormone therapy following the 2002 WHI headlines created a substantial unmet need. Millions of women stopped taking a treatment that, for most of them, was working. The symptoms that returned — the hot flashes, the disrupted sleep, the accelerating bone loss, the mood instability, the cognitive changes — did not go away. They were now being treated symptomatically, with individual pharmacological agents targeting individual symptoms: antidepressants for mood instability, sleep aids for disrupted sleep, bisphosphonates for bone loss, statins for cardiovascular risk. Each of these is a separate prescription. Each generates separate revenue. The systemic solution — hormone therapy — was replaced by a collection of targeted interventions that addressed downstream consequences of estrogen deficiency without addressing the estrogen deficiency itself.
I am not asserting that this outcome was engineered. What I am asserting is that the economic consequence of the 2002 panic deserves to be part of the analysis of how the panic was sustained for so long in the face of accumulating evidence that the original interpretation was incomplete. The pattern is familiar to anyone who has spent time studying how research findings get translated — or fail to get translated — into clinical practice. The finding that disrupts a profitable status quo faces a higher evidentiary bar than the finding that supports it. That bar is not written anywhere. It does not need to be.
The Evolutionary Frame: What Estrogen Was Doing in the First Place
There is a deeper question underneath the WHI story that menopausal medicine rarely asks, and that evolutionary biology can partially answer: why do women have a menopause at all?
Among mammals, humans are unusual — perhaps unique at the scale of human longevity — in experiencing a complete cessation of reproductive function while a substantial portion of life remains. The leading evolutionary hypothesis, known as the grandmother hypothesis, proposes that post-reproductive women in ancestral populations contributed to the survival and reproductive success of their daughters’ offspring, and that this contribution was significant enough to have been selected for over long timescales. The biology that makes this possible — an extended post-reproductive life supported by a gradual hormonal transition rather than a rapid collapse — is not an accident or a failure. It is, on the evolutionary hypothesis, a feature.
What estrogen was doing in the bodies of ancestral women aged forty to sixty was not simply supporting reproduction. It was supporting the full suite of somatic functions on which their continued contribution to kin fitness depended — cognitive sharpness, cardiovascular resilience, physical strength, bone integrity. The estrogen receptors distributed throughout the brain, heart, bone, and gut of a forty-five-year-old woman are not vestigial. They are signals of a system that was still meant to function.
The modern perimenopausal transition occurs in a context that differs radically from the ancestral environment in which that system was calibrated: higher refined carbohydrate load, greater insulin resistance, lower physical activity, chronic sleep disruption, environmental endocrine disruption, and profound social isolation relative to the multigenerational community structures in which female midlife was embedded for most of human history. Each of these mismatches compounds the biology of the transition. The women who fare best through this window are, consistently in the data, the ones whose lives most closely approximate the conditions the system was designed for: nutrient-dense whole food, substantial physical activity including load-bearing exercise, adequate sleep, and strong social integration.
The pharmaceutical response to the menopause transition — whether that is hormone therapy, or its symptomatic alternatives — cannot substitute for the environmental conditions that the underlying biology requires. It can only work with whatever foundation exists. This is a point that menopausal medicine, understandably focused on individual pharmacological decisions, tends not to make explicitly.
The Pattern This Is Part Of
The WHI story is not an isolated event in the history of medical evidence. It is an instance of a pattern that readers of this publication have seen documented across multiple domains: a large, expensive, federally funded study is conducted in a population that does not match the clinical target, using a formulation or intervention that does not match the clinical practice being evaluated, producing findings that are communicated in relative rather than absolute terms, and those findings are translated into guidelines and clinical practice with a speed and thoroughness that is never matched by the subsequent corrections.
The 2002 WHI results changed clinical practice within months. The subsequent analyses — demonstrating the timing hypothesis, demonstrating the role of formulation, demonstrating the differential outcomes by age and proximity to menopause — accumulated over a decade and a half and produced guideline changes that are, to this day, not uniformly reflected in the practice of the average primary care physician. The asymmetry is not coincidental. The alarm travels faster than the correction. And the correction is the harder story to tell, because it requires acknowledging that the alarm was, in significant ways, wrong.
The FDA’s November 2025 announcement is a formal acknowledgment of exactly this. It is welcome. It is two decades late. And its translation into actual clinical practice — into conversations between clinicians and perimenopausal women who deserve the full picture — will require sustained advocacy.
That is what this publication is for.
A Question the Field Has Not Answered
Here is what I find most striking, twenty-three years after the WHI results were published: we still do not have a large, prospectively designed, adequately powered randomized controlled trial of the intervention that is now recommended — low-dose transdermal estradiol with micronized progesterone, initiated in women aged forty-five to fifty-five within five years of the onset of the perimenopause transition — with hard cardiovascular, cognitive, and fracture endpoints followed for twenty years.
The WHI was designed in 1991. The formulations it tested were the formulations of 1991. The population it enrolled reflected the clinical practice of 1991. The study that should have been designed once the timing hypothesis emerged — a study actually addressing the question that matters for the forty-eight-year-old woman sitting across from her clinician right now — has not been done.
I will leave you with that observation, because what it implies about the priorities of the institutions that fund clinical research is a larger question than this article can close.
The companion video to this article, covering the practical options for women in the perimenopause window — hormonal and non-hormonal — is available on the Interesting Stuff YouTube channel. Link here!
If you are a clinician, a researcher, or a woman navigating this transition who found value in this analysis, the work of building this kind of science communication is supported by paid subscribers to Popular Rationalism. The depth of analysis here — the primary literature, the incentive structure questions, the evolutionary framing — is what your subscription makes possible.
Forthcoming: the insulin resistance-estrogen connection in perimenopause — and why metabolic health may be the most undervalued variable in the entire conversation.
James Lyons-Weiler, PhD, is the founder and president of IPAK-EDU and Editor-in-Chief of Science, Public Health Policy, and the Law. He publishes Popular Rationalism on Substack.
I don’t know… I had a terrible time during menopause. I about lost my mind. However, I didn’t want to take the horse hormones, or any other pharmaceutical product. That’s just me.
I in no way judge those that feel they need this type of intervention.
Made it through and feel like my normal self again. Took a good eight years…
I went to natural remedies following the information in the book The Pause. I did not want to take horse estrogen, which I understand takes a rather cruel way to get that. I did fine. For eons women have gone through the pause without medical intervention, often taking natural remedies. I see no reason to medicate a natural event. As the article says, nutrition has a lot to do with how menopause goes.