* Reprint, in Forum Gynécologie Suisse 1/2014*
In 2002, a premature press release of preliminary results from the Women's Health Initiative (WHI) started an international hormone hysteria with profound consequences for women's health worldwide. Decades of observational and cohort study data went out the window in one fell swoop, catapulted by the world press. (1)
The early termination of the WHI had been prompted primarily by a marginally significant increase of 0.25% in cardiovascular events per year. (2)
The media, however, almost exclusively headlined the breast cancer risk and terrified a generation of women the world over. The popular press translated a relative risk increase of 26% (or 8 more breast cancers per 10'000 women over a baseline of 30 breast cancers per 10'000 women per year) into an absolute risk of 26%. Women were told that 1 in 4 would get breast cancer if they used HRT.
Within less than a month malpractice law suits started pouring in accusing doctors of having killed their patients with hormones, and millions of women flushed their hormone pills down the toilet. The stock value of pharmaceutical companies producing sex steroids plummeted and funding for HRT research and development was reassigned.
Even scientific meetings that at least tried to deal with the cardiovascular results were not much better. I heard a keynote speaker express from the podium of an national menopause meeting her own feelings of guilt over having killed the brain of her aging mother by causing vascular dementia with Prempro®(the study drug of the WHI's estrogen/progestin arm) that she had prescribed for her. Few in the audience were aware of the fact that the speaker had recently quit her academic position in favor of employment by the then makers of Raloxifene, who were trying to capture the market void left by HRT.
Both the American College of Obstetricians and Gynecologists and the North American Menopause Society rapidly caved in under the general pressure, decided to err on the side of caution and, while still not having access to actual WHI data, published the now infamous recommendation to use HRT “in the lowest effective dose for the shortest possible time”, and only for severely symptomatic women. Other specialty societies worldwide followed suit with the laudable exception of the International Menopause Society. Its much more measured menopausal hormone therapy statement of the time has been largely adopted by the rest of the world a decade later and clearly states in its current version that for most menopausal women HRT carries more benefits than risks. “New data and re-analyses of older studies by women's age show that, for most women, the potential benefits of HRT given for a clear indication are many and the risks are few when initiated within a few years of menopause.” (3)
In 2004, Anderson, et al. showed an actual 23% reduction
of the relative risk of breast cancer over placebo in the estrogen-only arm of the WHI (i.e. 6 fewer breast cancers per 10,000 women over a baseline of 30 breast cancers per 10,000 women per year). (4)
At the time this reduction was not yet statistically significant, but subsequently reached significance with longer follow up.(5
) While this was commented on in the Washington Post and the New York Times, these favorable results were widely ignored by the popular press and certainly never reached the larger public.
Also completely ignored were the well documented benefits demonstrated by both arms of the WHI, such as the protection from osteoporosis and that HRT shows one of the best reduction in hip fractures of any drug ever studied. And of course, nobody talked about the fact that only about 4% of women even die of breast cancer while about 50% die from cardiovascular disease.
Confused by the intense anti-HRT-terror campaign, women felt betrayed and misinformed by their gynecologists, turned to alternative health products if not outright quackery, and abandoned HRT in droves. Sprague, et al. report that by 2004 overall HRT use had already dropped by almost 50% compared to 2000, before the WHI publication. (5)
By the year 2010 it had dropped even further by almost 80% compared to a decade earlier, so that only 1.7% of postmenopausal US women still used estrogen/progestin formulations and 2.7% estrogen-only. The percent of hormone users is slightly higher for hysterectomized women, because until recently women in the US routinely had their ovaries removed at hysterectomy, and were thus more likely to be symptomatic than women undergoing a more gradual, natural menopause.
Of course, right from the beginning many critical voices questioning the design and methodology of the WHI were raised, but were universally not heard by the wider public. The principal objection to WHI was and remains that it did not study the population that traditionally is prescribed HRT: namely healthy women with menopausal symptoms. Women with hot flashes were in fact actively discouraged from participating in both arms of the WHI in order to avoid high dropout rates in the placebo group. This led to a WHI study population with an average age of over 63 and a BMI of 28.5 in the E/P arm and even a BMI of 30.1 in the estrogen-only arm, making it even less applicable to European women, who on average are more slender than their US counterparts. (2,4).
Some of the - in retrospect- incomprehensible flaws in the WHI study design can be explained by remembering that it was originally conceived as study to prove, once and for all, HRT as a means of primary cardiovascular prevention in postmenopausal women. In fact, throughout the 1990's US National guidelines (ATP II) still recommended HRT as a first line treatment for postmenopausal women with elevated serum lipids along with diet and exercise. It had been shown time and again in multiple trials that HRT lowered total cholesterol and LDL, and increased HDL. Cardiovascular survival benefits were supported by numerous observational data, including the Nurse's Health Study. The significance of the fact that HRT also increased inflammatory markers and triglycerides was then not yet perceived. In the last decade of the last century doctors also began to realize that heart attacks were not an exclusive male privilege and that the vast majority of women - about half - actually do die of cardiovascular disease- over ten times more than the 4% who die of breast cancer. This contributed to the interest by non-gynecologists in the outcome of the WHI trial.
As one of my colleges once remarked, “In medicine it takes a decade to even begin to undo the damage caused by a bad study once it has been generally adopted”. This decade has now passed. Due to the large number of participants in both arms of the WHI, over 16,000 and 10,000 women respectively, an enormous data set of useful information was accumulated. Because of the massive size of the study population of the original prospective randomized WHI trial and its follow-up and observational studies, it was possible to analyze various subgroups and still have high enough numbers to get valid data. One such subgroup that has been subsequently analyzed consists of the women in the WHI under 59. Another subgroup analysis entailed the separation of the women who had been takin hormones prior to WHI enrollment from the women who started taking hormones for the first time when they entered the WHI at an average age of 63.
In 2011, LaCroix, et al reported on the follow-up study of the estrogen-only arm of the WHI. Of particular interest were again our “typical menopausal women” from 50 to 59. (6)
The report indicated an excess mortality of 13/10,000 women/year in women in this age group for women who did not take estrogen, i.e. received placebo. Twelve of the 13 deaths were due to cardiovascular disease. An increase in deaths from invasive breast cancer was also seen in the women who had received placebo, with the women on estrogen alone having a statistically significant 23% lower incidence of invasive breast cancer. In fact, the lower incidence of breast cancer was the only major outcome that persisted after the active treatment phase, while the decrease in hip fracture was no longer statistically significant 5 years after discontinuation of estrogen.
In 2013, prompted by La Croix's publication, a group of researchers from Yale University calculated the best point estimates for the total excess mortality toll due to avoiding estrogen for hysterectomized women aged 50-59 in the United States from 2002 and 2011. (7)
They concluded that more than 50'000 deaths could have been avoided in menopausal women without a uterus, if the WHI had not stampeded US women into abandoning estrogen. In the past, more than 90% of these women would have been using ET when they were in their fifties. Now, the current ET use rate is as low as 7.5%. One reason for this tremendous decline is that less than 40% of women are being offered ET at the time of hysterectomy and oophorectomy, with 25% or less using ET 10 months later.
As a result of the insights gained from these later analyses, which only arose when the WHI data was finally made available to a larger circle of researchers than the original investigators, the pendulum finally began to swing the other way. The concept of a “window of opportunity” for menopausal hormone treatment was developed: Women who actually started on HRT within the first few years after their last menstrual period, before their arteries had been clogged with plaques due to rising lipid levels and increasing prevalence of metabolic syndrome, were shown to have exactly the cardiovascular benefits that all the observational and cohort studies from the decades prior to 2000 had shown. Moreover, there was no increase in breast cancer in younger menopausal women taking HRT. In other words, if the typical women who always had taken HRT- healthy peri-menopausal women with hot flashes- begin to take HRT, the possible harm done is clearly outweighed by the benefits. It is only the bulk of the WHI population, which is not representative of typical menopausal women in real life and clinical practice, to which HRT conveys a significant risk.
Unfortunately, for many women this renewed swing in favor of menopausal hormone therapy comes too late. The damage has already been done and for many can no longer be reversed. For the lay public the new recommendations are perceived as a not-to-be-trusted and confusing reversal and about face. It will be an uphill battle to convince women that science is more likely correct than the boulevard press and the sales clerk in the health food store. The same is true for many family doctors and non-women's health specialists, who do not have the time to sort out the intricacies of the confusing WHI data and the newer HRT studies finally reaching the scientific literature.
The net result of the severe confusion among doctors and the lay public wrought by the Women's Health Initiative study and the curious mode of releasing its results was that the generation of women who turned menopausal around the turn of the last century is very proud to have mastered menopause on their own without resorting to the evil hormones that the even more evil pharmaceutical companies, who had so successfully duped their gullible gynecologists, had been trying to force down their throats. But as a consequence of having successfully avoided HRT, many of these same women also have by now well documented osteoporosis and a massively increased risk for hip fractures. (8)
Hormone deprivation and urogenital atrophy also did not help their marriages. Now heart disease, excess death rates and, for those without a uterus, breast cancer (!) are being added to the health risks incurred by their misplaced “heroism”.
At the same time, almost three quarters of menopausal female gynecologists in Scandinavia use HRT themselves and among menopausal spouses of male gynecologists, 68-72% are current users, according to a 2007 survey by Pedersenet al. , essentially unchanged from surveys before publication of the WHI. (9)
But among their patients HRT is substantially lower. The same is true in Israel and Germany. In 2012 Buhling et al. reported their 2010-2011 survey, which shows that 97% of female gynecologists or the female partners of male German gynecologists used or would use HRT for the management of vasomotor symptoms. But only 37% of German women 45 to 60 years of age used HRT according to a 2005 report. Even in the US, where fewer than 3% of menopausal women were using HRT by 2010 (5)
, about half of menopausal female gynecologists used hormones themselves according to a Gallup 2003 survey. A 2013 survey from New York puts that number at 74%. A case of “Quod licet iovi non licet bovi.”
?! Or just not enough time and energy in our busy practices to be more convincing than the patients’ hair dressers and the boulevard press…
I wholeheartedly concur with Andrew Kaunitz's 2012 Editorial in the Journal Menopause that we should take the time to talk to our patients and help them make an informed decision about this admittedly very confusing issue. (10)
1. Shock, terror and controversy: how the media reacted to the Women's Health Initiative. Brown S. Climacteric 2012;15:275-80
2. Principal results from the Women's Health Initiative randomized controlled trial. Writing Group for the Women's Health Initiative Investigators. JAMA 2002;288:321-33.
3. Updated IMS recommendations on postmenopausal hormone therapy and preventive strategies for midlife health. D. W. Sturdee and A. Pines on behalf of the International Menopause Society Writing Group CLIMACTERIC 2011;14:302-320
4. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial.
Anderson GL, et al. JAMA. 2004 Apr 14;291(14):1701-12
5. A sustained decline in postmenopausal hormone use: results from the National Health and Nutrition Examination Survey, 1999-2010. Sprague BL, et al. Obstet Gynecol 2012;120:595-603.
6. Health outcomes after stopping conjugated equine estrogens among postmenopausal women with prior hysterectomy: a randomized controlled trial. LaCroix AZ, et al. JAMA 2011;305:1305-14
7. The mortality toll of estrogen avoidance: an analysis of excess deaths among hysterectomized women age 50 to 59. Sarrel PM, et al. Am J Pub Health 2013 July 18.
8. Hip fracture in postmenopausal women after cessation of hormone therapy: results from a prospective study in a large health management organization. Karim R, et al. Menopause. 2011 Nov;18(11):1172-7.
9. Impact of recent studies on attitudes and use of hormone therapy among Scandinavian gynaecologists. Pedersen AT, et.al. Acta Obstet Gynecol Scand. 2007;86(12):1490-5.
10. Disparity in Menopausal Hormone Therapy Use Between Women Obstetrician Gynecologists and Women Overall. Are Obstetrician Gynecologists Underserving Their Patients? Kaunitz, AM. Menopause. 2012;19(10):1070-1071.