What the 2002 WHI Study Actually Found, and What Twenty Years of Research Has Said Since
"Eight additional cases per 10,000 women per year, in a population averaging 63 years of age, taking one formulation, did not" make the headlines. Dr. Brooks unpacks what a 26% relative risk increase actually meant in absolute terms — and why the sentence that outlived the study is still shaping care today.
Last updated: September, 2026
In July 2002, one clinical trial was halted early, and within months the way American medicine talked to women about hormones changed almost completely. I still meet women who were handed that decision secondhand, twenty years later, in a sentence that took four seconds: hormone therapy causes breast cancer, so we do not do that here. This post is for the woman who was told no without a conversation, and who wants to know whether the advice she received reflects what the evidence actually says. I write it as a physician who is board certified in internal medicine and who holds the Menopause Society Certified Practitioner (MSCP) credential, which means menopause care is something I was formally examined in rather than something I picked up along the way.
The sentence that outlived the study
The consequence here is not abstract. Menopausal hormone therapy use among postmenopausal women in the United States fell from 26.9 percent in 1999 to 4.7 percent in 2020, a decline of roughly 82 percent, according to an analysis of national survey data published in JAMA Health Forum in 2024. That is not a story about a treatment falling out of favor because something better replaced it. It is a story about a generation of women who were never offered an individualized conversation at all.
Some version of the following happens in my office nearly every week. A woman in her early fifties describes sleep that has not been intact in two years, night sweats, joint aches, and a fog she cannot think her way out of. She has already raised it with someone. She was told that hormones are dangerous, or that she should wait it out, or that this is simply what fifty looks like. Nobody asked about her cardiovascular history, her family history, when her periods actually stopped, or what her symptoms were costing her. The decision was made before the visit started.
What the Women's Health Initiative was actually built to answer
The Women's Health Initiative was not designed to answer the question most women are asking. It was a prevention trial. Its central question was whether hormone therapy, given to postmenopausal women broadly, would prevent chronic disease over the long term, particularly coronary heart disease.
That distinction matters because it shaped everything about who was enrolled:
The estrogen plus progestin trial enrolled 16,608 women, with a mean age of 63.3 years.
Only 33.2 percent of participants were between 50 and 59 years old. Roughly two-thirds were 60 or older, and 21.5 percent were between 70 and 79.
The trial tested one specific combination, conjugated equine estrogens taken orally with medroxyprogesterone acetate, at one dose.
The estrogen plus progestin arm was stopped at an average of 5.2 years of follow-up.
When the principal results were published in JAMA in 2002, the reported hazard ratio for invasive breast cancer was 1.26, with a nominal confidence interval of 1.00 to 1.59. In absolute terms, that was 38 cases per 10,000 women per year in the hormone group compared with 30 per 10,000 in the placebo group, a difference of 8 additional cases per 10,000 women per year.
That is the number. A 26 percent relative increase became the headline. Eight additional cases per 10,000 women per year, in a population averaging 63 years of age, taking one formulation, did not.
Five things the headline left out
There were two trials, not one. Women who had undergone hysterectomy received estrogen alone rather than estrogen plus a progestogen. In that trial of 10,739 women, breast cancer incidence went down, with a hazard ratio of 0.79 and 7 fewer cases per 10,000 person-years, and that reduction persisted through 20 years of follow-up, as summarized in the 2024 JAMA review of the WHI trials. The estrogen-alone finding was almost entirely absent from public coverage.
Relative risk is not absolute risk. A 26 percent increase sounds enormous. Eight events per 10,000 women per year sounds like what it is, which is a small absolute change that a woman is entitled to weigh against her own symptoms and her own risk profile.
Age at initiation was collapsed into a single average. The same 2024 review reports that the composite measure of harms and benefits showed 12 excess events per 10,000 person-years in women aged 50 to 59, compared with 38 excess events per 10,000 person-years in women aged 70 to 79.
The trial did not test what most symptomatic women need. Women with moderate to severe hot flashes were largely excluded because giving them placebo for years was not workable. The trial that reshaped symptom care was not a symptom trial.
One formulation is not every formulation. Oral conjugated equine estrogens with medroxyprogesterone acetate is not interchangeable with transdermal estradiol and micronized progesterone, which is what most of my patients who use hormone therapy are actually using today.
What twenty years of follow-up has shown
The follow-up data are more nuanced than either the 2002 headlines or the current wave of enthusiastic marketing would suggest. From the long-term WHI analyses:
Neither trial showed an increase in all-cause mortality. The hazard ratio was 0.97 for estrogen plus progestin and 1.03 for estrogen alone.
Both regimens reduced hip fracture, with a hazard ratio of 0.67 in each trial during the treatment phase.
Stroke risk was elevated in both trials during treatment, with hazard ratios of 1.37 and 1.35, which is a real finding and one I discuss directly with every patient.
Absolute risks attributable to hormone therapy in women in early menopause are small. The 2024 review characterizes them as generally fewer than one additional adverse event per 1,000 women per year.
The reviewing authors, several of whom led the original trials, conclude that hormone therapy is appropriate to consider for women in early menopause without contraindications who have bothersome vasomotor symptoms, with individualized care and shared decision-making. That is a materially different message from the one that reached the public in 2002.
Where the guidelines stand today
The Menopause Society, the professional body that grants the MSCP credential I hold, states the current position plainly in its 2022 hormone therapy position statement:
The benefits of hormone therapy outweigh the risks for most healthy symptomatic women who are aged younger than 60 years and within 10 years of menopause onset.
The same statement holds that therapy does not need to be discontinued automatically at an arbitrary age, and that continuation past 65 can be considered after appropriate evaluation for persistent symptoms, quality of life, or bone health.
Regulatory language has moved as well. On November 10, 2025, the FDA announced the removal of the boxed warnings covering cardiovascular disease, breast cancer, and probable dementia from menopausal hormone therapy products, while retaining the endometrial cancer warning on estrogen-alone products. Reasonable clinicians have debated whether that change went too far. What is not debatable is that the warning language a generation of physicians was trained on no longer stands as written.
Bioidentical is a marketing word before it is a medical one
This is where I spend a good deal of visit time, because the confusion is understandable and it is being actively exploited.
Estradiol and micronized progesterone are structurally identical to the hormones the human body produces, and both are available in FDA-approved, regulated products. That is bioidentical in the meaningful sense of the word, and it is ordinary prescribing.
What is different is custom compounded hormone therapy sold through wellness clinics and pellet programs. The Menopause Society's position statement is direct about the concerns, citing minimal government regulation and monitoring, the potential for overdosing or underdosing, impurities or lack of sterility, and a lack of scientific efficacy and safety data. If a program will not tell you what is in the preparation, who made it, and what evidence supports it, that is your answer.
What an individualized evaluation looks like in my office
I will not print a protocol here, because there is not one that applies to everybody. What I can describe is the structure of the conversation, which is the part most women never get:
Where you actually are. Perimenopause and postmenopause are not the same clinical situation, and the timing of your final menstrual period matters more than your birthday does. This is also where I separate a menopause story from a thyroid or metabolic one, which is not always obvious from a symptom list.
What the symptoms are costing you. Sleep, cognition, mood, joint pain, intimacy, and work performance are clinical information, not complaints. I ask about the ones women tend not to volunteer.
Your personal risk picture. Cardiovascular history, blood pressure, lipids, clotting history, migraine pattern, breast cancer history and family history, and your bone health all belong in the decision. So does the interaction between sleep, stress, and hormones, which changes how I interpret what you are describing.
Route and formulation, discussed as tradeoffs. Oral and transdermal delivery do not carry identical risk profiles, and that difference is part of an honest conversation rather than a detail to settle at the pharmacy.
A decision you are allowed to revisit. This is the part that a fifteen-minute visit structurally cannot support. We can start, stop, adjust, or decide against hormone therapy entirely, and reassess as your health changes. Nothing about this has to be settled in one appointment, which is one of the reasons visit length is a clinical variable rather than a comfort feature.
I manage hormone therapy inside the primary care relationship rather than sending it out as a referral. That means the physician evaluating your cardiovascular risk is the same physician discussing your hormones, reading your labs, and following you over time.
If You Were Told No Without a Conversation
Being told no is not the same as being evaluated. There are women for whom hormone therapy is not appropriate, and identifying them is part of the work. What should not happen is a decision reached without your history, your risk profile, and your symptoms ever entering the room.
If you are trying to work out whether the advice you were given is current, three questions are worth asking at your next visit:
Was my individual cardiovascular, clotting, and breast cancer risk actually assessed, or was this a general policy?
Does the recommendation account for how far I am from my final menstrual period?
Is this decision something we can revisit, or is it final?
If the answers are unsatisfying, you are entitled to a second opinion from a clinician who is credentialed in this specific area. You can look up Menopause Society Certified Practitioners through the Society's practitioner directory.
At Asklia Concierge & Metabolic Medicine in Cave Spring, this conversation is part of primary care, not an add-on. You can review what is included in the Women's Health membership, read more about my training and the MSCP credential, or schedule a Meet + Greet at no cost to ask questions before deciding anything. You can also reach the office at 540-410-9275.
This article is for general educational purposes and is not medical advice. Hormone therapy decisions should be made with your own physician based on your individual health history.
