Many doctors agree with the Food and Drug Administration’s recent decision to remove the black box warnings on at least some forms of hormone therapy used to treat the symptoms of menopause. But in making the announcement, health officials misleadingly suggested that women could take the drugs for long-term benefits to the heart and brain. Menopausal hormone therapy is not currently recommended for those uses.
On Nov. 10, the FDA announced it would be removing the black box warnings for increased risk of cardiovascular disease, breast cancer and probable dementia on all hormone treatments for menopause. Such products include a variety of formulations of estrogen, often with another hormone, that work to alleviate symptoms such as hot flashes, night sweats and vaginal dryness, which can be severe in some women.
The warnings were first applied in 2003, following the results of a large randomized controlled trial that was part of the Women’s Health Initiative. This study, funded by the National Institutes of Health to better understand how to prevent disease in older women, was widely misinterpreted as finding that risks of menopausal hormone therapy, or MHT, outweighed benefits for managing menopause symptoms. Use of MHT plummeted.
In the intervening years, it has become clear that certain forms of MHT, such as oral pills that act systemically, or throughout the body, do have some risks, but can be safely used to treat symptoms in most women without risk factors when started within 10 years of menopause. This is when symptoms are most likely anyway. Other forms of MHT, such as vaginal estrogen, pose little if any risk. Menopause is when a woman stops menstruating; a woman is considered postmenopausal once she has gone a year without a period.
Medical groups such as the American College of Obstetricians and Gynecologists have long recommended MHT after considering an individual’s risks and benefits and have petitioned the FDA to remove the black box warnings for low-dose vaginal estrogen. They do not, however, advocate the use of hormones for overall health when a woman does not have any symptoms, due to a lack of convincing evidence.
In making their black box announcement, FDA Commissioner Dr. Marty Makary and Health and Human Services Secretary Robert F. Kennedy Jr. misleadingly extolled the benefits of MHT in preventing cardiovascular disease and dementia.
“Hormone replacement therapy has been found to reduce the risks of cardiovascular disease and mortality by as much as 50%, Alzheimer’s disease by 35%,” Kennedy said near the beginning of the Nov. 10 press conference announcing the change.
“By a large body of evidence, there are now recognized to be profound long-term health benefits that few people, even physicians, know about,” Makary later said of hormone therapy, calling it “life-saving.”
“What are we doing not offering women this potentially powerful treatment? With few exceptions, there may be no other medication in the modern era that can improve the health outcomes of women on a population level than hormone replacement therapy,” he added.
Kennedy and Makary made similar claims in a variety of subsequent appearances, including Kennedy’s citation of the same figures in a Dec. 2 Cabinet meeting.
Experts told us their statements don’t reflect the full scientific literature on the topic.
“The totality of the data do not support a preventive benefit of hormone therapy for these diseases,” Dr. Nanette Santoro, a menopause researcher at the University of Colorado Anschutz, told us. “One must selectively pull out specific numbers from specific studies to come up with a miraculous reduction in risks.”
No guidelines, neither from ACOG, the Menopause Society, the International Menopause Society nor the Endocrine Society, she noted, recommend hormone therapy to prevent cardiovascular disease or dementia.
Makary and Kennedy also downplayed risks of systemic MHT, implying that breast cancer risks are not borne out by the data. This is misleading and does not tell the full story.
“They’ve systematically cherry-picked things that show benefit and that diminish risk,” Dr. Martha Hickey, a menopause clinician and researcher at the University of Melbourne, told us.
“Providers should always have a nuanced conversation about the risks and benefits of systemic hormone therapy with their patients. Each woman is different, and this conversation needs to be tailored to the needs and priorities of each woman,” Rebecca Thurston, a menopause researcher at the University of Pittsburgh who focuses on cardiovascular and neurocognitive health, told us. She emphasized that MHT is highly effective for hot flashes and night sweats and can also prevent menopause-associated bone density declines but is not recommended for the prevention of heart disease and dementia “based on the highest quality science.”
The FDA did not respond to a request for comment.
Cherry-Picked Dementia Benefits
In touting the benefits of MHT, Makary and Kennedy repeatedly claimed that hormone therapy has been found to lower the risk of Alzheimer’s disease by 35% and reduce cognitive decline by 64%. The numbers were also cited in an HHS fact sheet.
Both reflect findings in individual scientific studies, but they are cherry-picked and not representative of the overall literature.
Pauline Maki, a menopause and cognition researcher at the University of Illinois College of Medicine, explained in a series of social media posts that the 35% figure is “misleading and inaccurate in light of what we now understand.” It comes from a 1996 case control study of a single retirement community in Laguna Hills, California.
The study looked at how often women with an Alzheimer’s disease or dementia diagnosis on their death certificate had self-reported using hormone therapy, compared with those who did not have such diagnoses noted. It found that women with Alzheimer’s were less likely to have said they used hormones, with hormone use associated with a 35% lower risk of the disease.
While that might sound pretty good, Maki said, this doesn’t prove that it’s the hormones that made the difference, as other factors associated with hormone use could be driving the result instead.
There are also much larger and more recent case control studies that Makary and Kennedy did not mention. While the cited Laguna Hills study drew from a population of around 9,000 and involved fewer than 1,500 women who died 30 or more years ago, three population-wide case control studies have been published in the past six years, each with 60,000 or more participants.
“When you look at those studies, a very, very different result emerges,” Maki said in a video post. “Far from reducing the risk of Alzheimer’s disease, those studies found zero evidence for a reduction, and in fact, found an elevated risk.”
Maki said that women should not be concerned about increasing their dementia risk if they are using hormones to alleviate menopause symptoms. “But if you use it for primary prevention, particularly long term, there’s no evidence of benefit,” and there could be potential harm, she said.
The second study Kennedy and Makary have cited is a 2005 Danish follow-up study of 343 women who had previously been in a randomized controlled trial and had received hormone therapy or placebo years earlier. Researchers administered a cognitive screening test and found that while there was no difference in the average scores among those who ever took hormones versus those who never did, the subset of women who took hormone therapy for two to three years had a 64% lower risk of testing positive for cognitive impairment.
Maki told us that the study had a better design in terms of exposure to hormone therapy, but was “weak in its measure of cognitive impairment and is weak methodologically as only a subset of women in the original study participated in this follow-up.” She said it was odd to highlight either this paper or the Laguna Hills paper given all the other studies that have been conducted on the subject over the past 20 years.
When randomized controlled studies have been done, they have failed to find that hormone therapy has cognitive benefits in recently postmenopausal women. Four such trials have found no effects — good or bad — on cognition.
One randomized controlled trial — the WHI Memory Study — found that at 18 years of follow-up, women were 26% less likely to die of dementia if they had taken one type of an oral estrogen-only therapy. These women had all had hysterectomies, allowing them to take systemic estrogen without progesterone. (In women with intact uteruses, progesterone or a synthetic form called progestin needs to be added because estrogen on its own can increase the risk of uterine cancer.)
But the absolute benefit was very small, and the authors said the result should be interpreted with “particular caution.” The same trial found earlier that a combined estrogen and progestin therapy more than doubled the risk of probable dementia in postmenopausal women 65 years and older.
Some scientists have hypothesized that the timing of MHT matters, such that hormone therapy might benefit the brain if given early in the menopause transition or before disease sets in — and then shift to being neutral or harmful later. But as the Menopause Society’s 2022 position statement on MHT explains, the concept has not “been definitively supported” in randomized controlled trials.
Based on “good and consistent scientific evidence” — the highest graded recommendations — the society concluded that “[i]n the absence of more definitive findings, hormone therapy is not recommended at any age to prevent or treat a decline in cognitive function or dementia.”
Possible Cardiovascular Benefits in Some Women Uncertain
Makary and Kennedy have frequently claimed that hormone therapy dramatically cuts the risk of heart disease, casting it as a clear and definitive finding.
While there is some evidence of a cardiovascular benefit when MHT is started within 10 years of menopause or before the age of 60, this remains a hypothesis that has not yet been confirmed in trials.
“Until we have that next trial that would answer these questions, we can’t make these very large sweeping claims,” Dr. Chrisandra Shufelt, a women’s health internist at the Mayo Clinic in Florida who has studied the impact of hormones on cardiovascular disease, told us.
Both health officials have specifically said that hormone therapy slashes cardiovascular disease risk by 50%. HHS has used the same statistic in its press release and fact sheet.
The study often cited for this, however, is a 1991 review that specified that those findings pertained to oral estrogen-only hormone therapy — the kind given to women without uteruses — and needed to be confirmed in a randomized controlled trial.
Indeed, the review predates findings from the 2002 and 2004 WHI studies, which were specifically done to test whether hormone therapy actually did prevent cardiovascular disease in primarily healthy postmenopausal women.
“There’s observational data and there’s lots of it,” Marcia Stefanick, a WHI investigator and chronic disease prevention researcher at Stanford University, told us, noting that’s why the WHI study was done in the first place. “We know that the women who are using menopausal hormones were very different from the women who weren’t,” she explained, adding that they were “less obese, less likely to smoke, more physically active, more highly educated, less likely to eat high fat diets and have salt in their diet.”
At the time, hormones were commonly prescribed for prevention and were recommended by medical societies for that purpose.
“What piqued our interest in the late 80s and early 90s was that hormone therapy was increasingly being used for other indications — that is, prevention of cardiovascular disease in particular,” Dr. Jacques Rossouw, the project officer for the study from its inception until his retirement in 2014, told us. He emphasized that he was speaking to us in his personal capacity and not on behalf of the WHI investigators nor his former employer, the NIH.
Surprising many, when the WHI study testing MHT with oral estrogen and progestin was stopped after five years in 2002, it did not identify cardioprotective effects — and in fact the medications appeared to increase risk for the group as a whole. Nor did the oral estrogen-only part of the study find heart benefits when it halted in 2004. (A separate trial known as HERS similarly found no cardiovascular benefit with MHT in women with existing coronary disease, with an increased risk of blood clots, contrary to earlier observational studies.)
When later stratified by age, and after longer follow-up, the WHI results suggested that the cardiovascular risks of hormone therapy are generally lower in younger women closer to menopause and higher in older women starting therapy 10 years or more after menopause. These risks dissipated over time, and were more evident in women taking combined therapy. For younger women 50 through 59 years of age taking estrogen alone after 13 years of follow-up, the study pointed to a reduced risk of heart attack and coronary heart disease.
“If you look at the estrogen and progestin trial, there’s no benefit for younger women at any point for anything,” Stefanick noted. “It’s only in the estrogen-only trial, and those are women who had a hysterectomy.”
In trying to understand why the observational studies diverged from the trial data, these subanalyses, combined with other studies, including in monkeys, led to the so-called “timing hypothesis.” It proposes that estrogen has beneficial effects on the heart if given early and before plaques have formed in arteries, but is neutral or can be harmful later in life when women have established plaques.
As several guidelines document, there is wide agreement that timing affects the cardiovascular risk of hormone therapy — and that oral MHT is safe in most younger postmenopausal women who have bothersome symptoms. But there is far less agreement that this means MHT is preventing cardiovascular events in these women and that hormones should be given for the purpose of preventing heart disease.
Yet another WHI subanalysis published this year found that among the subset of younger women with moderate or severe hot flashes and night sweats in the trials, both estrogen-alone and combined MHT reduced symptoms without changing atherosclerotic cardiovascular disease risk. But for postmenopausal women age 70 and older with these symptoms, that risk was increased in both cases.
The WHI studies tested just one dosage of a particular oral estrogen and progestin, and it is unclear if the results apply to other dosages, types or modes of delivery. While Makary and others have emphasized that newer products may have reduced risk, by the same token, it hasn’t been shown that they would necessarily show any cardiovascular disease benefit.
“Most of us do believe it’s safer, but being safe is not the same as being beneficial,” Stefanick said, speaking of transdermal estrogen, a systemic product that is widely considered safer for the heart than oral formulations since it bypasses the liver. She added that she was not aware of any evidence of transdermal formulations showing heart disease benefits.
Makary has also referenced a 48% decline in fatal heart attacks with MHT, pointing in the press conference to a review published in Circulation. The 2023 review, which Shufelt co-wrote, was describing a subgroup finding in a 2015 Cochrane systematic review. It notably did not back the use of hormones to prevent heart attacks, and stated that it is “appropriate that no medical societies” recommend MHT for that purpose.
While the overall finding of the Cochrane review was that MHT did not protect against mortality or cardiovascular disease, when broken down by timing of therapy, a subanalysis found there was a 30% risk reduction of death from any cause and a 48% reduced risk of coronary heart disease (defined as cardiovascular deaths and nonfatal heart attacks) in women starting hormone therapy less than 10 years after menopause.
These results were largely driven by the three largest trials, Shufelt told us, and were only possible because the analysis pooled results from studies using different types of MHT, including both oral estrogen alone and combined therapy. One of the included trials is a 2012 Danish study of around 1,000 participants that was not blinded and did not use a placebo. The Cochrane review itself notes that if the Danish study is not included, the findings would not be statistically significant (see table 3). Moreover, the review found an increased risk of blood clots in veins in the younger subgroup — a fact Makary and Kennedy never mention.
The authors of the Cochrane review said in a commentary that their analysis of younger participants “should be interpreted with caution due to its post hoc nature and is therefore not adequate to make recommendations for its use to prevent cardiovascular disease in this population.”
In addition to studies looking at health outcomes, there are two smaller randomized controlled trials that have assessed whether there might be cardiovascular benefits of MHT when given to younger postmenopausal women. A 2014 study using a lower dose of estrogen failed to find any beneficial effect on atherosclerosis progression after four years.
A 2016 trial of around 650 participants found that in women within six years of menopause, MHT did appear to stall the thickening of the carotid artery. However, there was no impact on two other secondary measures of coronary atherosclerosis.
Shufelt said that it is not clear that the observed slowing of artery thickening is indicative of a cardioprotective effect. Typically, the thickness is only concerning for preclinical heart disease if it reaches a certain level, and the women still had normal thicknesses, she said. She likened the study results to slightly lowering a person’s blood pressure from an already normal pressure.
Shufelt said the studies were important because they showed that hormone therapy doesn’t increase risk in these populations, but “we’re not there yet to say that it prevents disease.”
Incomplete Information on Breast Cancer Risk
Kennedy and Makary both have emphasized that the WHI results were misinterpreted, and use of MHT subsequently fell. This narrative is broadly correct, but the officials misled on exactly how the study was misinterpreted, downplaying statistically significant findings on breast cancer risk.
The 2002 WHI study was “not statistically significant, but it triggered a media frenzy and led to the FDA applying unscientific black box warnings to all hormone replacement therapy products in 2003,” Kennedy said during the Nov. 10 press conference.
Makary also emphasized that the WHI breast cancer results were not statistically significant. “If we don’t have statistics, then we don’t have science,” he said.
Rossouw, the former WHI project officer, called this portrayal of breast cancer risk and WHI “just dead wrong.”
The WHI was stopped around three years early in 2002 after an independent board of researchers saw that breast cancer risk had reached a level that they had decided in advance would be unacceptable. Clinical trials are required to have such a board of safety monitors. The data also showed risks of blood clots and strokes, while failing, as we’ve said, to substantiate the idea that hormone therapy could prevent cardiovascular disease. The WHI researchers concluded that, given its overall balance of risks and benefits, MHT could not be recommended to prevent chronic disease and that it would not be ethical to continue the trial.
Rossouw was the first author of the 2002 WHI paper released in conjunction with the termination of the study. This is the publication Makary and Kennedy were referring to when they cited breast cancer results that were not statistically significant. But Rossouw explained that these data were preliminary. The women stopped their treatments under the study protocol after a median of 5.6 years. The researchers published their 2002 paper based on the 5.2 years of data available at the time. “There’s a lag,” he explained. “When you stop a trial early, there’s some data that are in process that you have not yet analyzed.”
The researchers were subsequently able to analyze the data collected in the final months of the trial. “When we did the full analysis and published those data, the breast cancer risk in the estrogen-progestin trial was statistically significant,” he said. The researchers wrote in a 2003 study that they had found a 24% increase in breast cancer cases in women who were randomly assigned to receive the combined hormonal therapy.
Furthermore, as the researchers continued to follow the women for years after the trial ended, they repeatedly found a statistically significant increased risk of breast cancer for combined MHT, both at 13 years and 20 years of follow-up. The WHI ultimately showed a decreased risk of breast cancer in women who took estrogen-only therapy.
The WHI investigators wrote in a Nov. 12 response to the FDA that “combination estrogen plus progestin increased the risk of breast cancer irrespective of age and this risk became more significant over time in the trial and with longer duration of follow-up.”
Some observational studies also have found an increased risk of breast cancer associated with use of MHT. One meta-analysis published in 2019 found that taking MHT was associated with increased risk of breast cancer even among people who took it between one and four years, with increasing risk the longer people took the therapy.
Hickey called the evidence showing an increased risk for breast cancer from combined MHT “strong and consistent.” She added that it makes mechanistic sense that hormone therapy might have some impact on breast cancer. “I think with breast cancer, there’s a very strong biological mechanism by which menopausal hormone therapy increases the growth of existing breast cancers and possibly the initiation of new breast cancers,” she said.
In general, practice guidelines support the use of MHT for troublesome menopausal symptoms while urging doctors to inform patients about possible risks and benefits.
“Women should be counseled about the risk of breast cancer with hormone therapy, putting the data into perspective, with risk similar to that of modifiable risk factors such as two daily alcoholic beverages, obesity, and low physical activity,” the Menopause Society advises doctors, for example.
“Hormone therapy does confer some risk, and so what we don’t want to do is overgeneralize or underestimate that risk, because it potentially could be harmful to some people,” Dr. Monica Christmas, director of the menopause program at University of Chicago Medicine and associate medical director of the Menopause Society, told us.
People who have already been diagnosed with breast cancer are typically recommended to avoid systemic MHT, according to the group’s guidelines, although it may be an option in specific cases. Vaginal estrogen, which is applied topically, is an option for those experiencing vaginal symptoms.
What some people arguably misinterpreted were the implications of the WHI results for treating menopausal symptoms.
The 2002 study notably did not say that the trial showed that the risks of MHT outweighed its benefits when used to treat troublesome hot flashes and night sweats that come with menopause. Investigators, including Rossouw, stated this at the time.
However, this nuance sometimes was lost as study authors, communications officers, journalists, doctors and the lay public interpreted and communicated the WHI results. Many left with the impression that the researchers had found that the risks of MHT outweighed the benefits in general, and not just for preventing chronic disease. In 2001-2002, nearly 39% of women between ages 52 and 65 used MHT. As of 2017-2020, fewer than 5% did.
Makary also dismissed the WHI data by saying that it only applied to an older progestin. “Today, hormone therapy is available in formulations that do not appear to carry the same increased risk of blood clots or breast cancer that was seen in earlier studies,” he wrote in a Nov. 10 viewpoint published in the Journal of the American Medical Association. There is some support for this idea, but this conclusion is not certain.
The Menopause Society guidelines say that some “but not all” of the available data suggest a lower breast cancer risk from certain hormones compared with the type tested in the WHI study. There are no data from randomized trials to tease apart whether breast cancer risks vary by hormone type.
Editor’s note: FactCheck.org does not accept advertising. We rely on grants and individual donations from people like you. Please consider a donation. Credit card donations may be made through our “Donate” page. If you prefer to give by check, send to: FactCheck.org, Annenberg Public Policy Center, P.O. Box 58100, Philadelphia, PA 19102.