WHI, One Year Later — WHY?

It has been a year since the Women’s Health Initiative Estrogen plus Progestin (WHI) arm of this large trial was stopped early because it caused harm (1 [1]). But people are still talking about Hormone Replacement Therapy and estrogen deficiency. Why?

Little has changed because the fundamental, negative ideas about women and about menopause have not changed. Universal menopausal hormone therapy and condemnation to disease and deficiency after menopause are based on a cultural belief that women are inferior and need fixing. The purpose of this article is to outline the prejudice against women on which these concepts of “estrogen deficiency” and “replacement” are based, to identify the ways these ideas are unscientific and to propose some things we can do to change these harmful ideas.

First, a personal word. I am a specialist physician for whom studying, practicing, teaching and creating new science about women’s reproduction is my life work. For over twenty years I have been saying that menopause is a natural part of women’s life cycle, the low estrogen levels after menopause are healthy and not abnormal. This made me unpopular—I have been labeled me as “way-out” or worse. Therefore the WHI results for me were a vindication.

The apparently positive responses of menopausal women to estrogen therapy were the keystone in the arch of reproduction-based prejudice against women. With WHI’s destruction of that keystone, the arch should fall. Right? Not right because the strong cultural negative ideas about women remain. For me, the Women’s Health Initiative estrogen plus progestin trial results provide evidence-based proof of misogyny.

What do I mean by that statement? Let’s start with the tough one—the word “misogyny” is harsh. In Norwegian it roughly translates to “woman hating.” Misogyny means that unintentionally or intentionally women are being treated badly—this is commonly supported by a negative view of women’s contribution to the society and different reproductive biology compared with men. However, women are normally different. Women’s estrogen levels in our normal life cycle do decrease to low after menopause. By contrast, men’s testosterone levels, although they decline somewhat in older ages, on average continue to be high throughout a man’s lives. Thus the concept of “estrogen deficiency” arises. And from that follows the prescription of estrogen to all menopausal women to prevent the supposed negative effects of low estrogen levels. But, as Dr. Susan Love said, “If estrogen deficiency is a disease, all men have it!” (2 [1]).

The strongest evidence for the belief that menopausal women are estrogen deficient has been that estrogen treatment apparently decreased heart attacks and dementia and was beneficial to women’s health and well being. In the early 1990s being filmed for “The Best Years” for the Canadian Broadcasting Corporation, I was asked by Joyce Resin to respond to Dr. M. Gelfand’s statement that estrogen improved women’s quality of life. My immediate response, “Better quality of life with estrogen is a smoke-screen for keeping women diseased and dependent.” (I realized how radical that sounded as I was saying it! I unconsciously ended the sentence with a slight questioning upturn in my voice.) WHI results show that estrogen does not improve the quality of life for women (3 [1]).

The Women’s Health Initiative, the largest controlled trial of ovarian hormone therapy ever performed, provides the strongest proof that Medicine (which is part of Culture and has been strongly influenced by the Pharmaceutical Industry) causes harm for women. Further a meta-analysis of controlled trials shows the same thing (4 [1]). This requires that the concept that “menopause-means-estrogen-deficiency-needs-replacement” must be replaced with a scientifically accurate and non-prejudicial idea about menopause.

To explain further—WHI estrogen plus progestin results are “evidence-based.” What does that mean? Modern medicine, to ensure scientific support for health care decisions and therapies, demands that what I, as a physician do, be based on good evidence, ideally from randomized controlled trials. It requires that I test my ideas (for example, about the use of cyclic progesterone for heavy periods or progesterone daily for treatment of hot flushes) in randomized controlled trials. It also demands that I abandon any therapy ideas that haven’t been proven both effective and safe in controlled studies. As a 23-year veteran of research in women’s health, I heartily agree with the goals of evidence-based medicine.

Evidence-based medicine is also founded on that fact that different kinds of studies have different value from clinical impressions, to observational studies to controlled trials to meta-analysis of controlled trials. At the bottom of the list is a doctor’s or many doctors’ impressions: “X is beneficial because my patients feel better.” That is biased because the women a given doctor treats are not representative of all women and came with a problem expecting help. It is also biased by the relatively small number of women one physician cares for (because some important harmful effects are relatively rare and therefore would not be expected to occur in the practice lifetime of one physician). And finally because in general happy patients come back and unhappy or harmed ones may disappear, or if they return be considered anxious or their side-effects discounted. The opinion of doctors has been that estrogen therapy is good for women. The company making estrogen has played a very strong role in the education of today’s physician leaders, especially in the field of gynecology. The idea of treatment with estrogen rested on the culturally accepted and fundamentally prejudicial idea that menopausal women are ill, or becoming that way, because of estrogen deficiency.

Did doctors and scientists have evidence-based data that estrogen therapy was beneficial for menopausal women? Sort of. All kinds of observational studies showed it. Observational studies enroll volunteer women who, for example, are all asked whether or not they have ever used Therapy Y. The differences in outcomes (let’s say Disease D) between those using and not using Therapy Y are all credited to or blamed on Therapy Y. But there are many unmeasured differences that may be present between groups using or not using Therapy Y. These may or may not be directly related to Therapy Y, and may or may not directly influence Disease D. Observational studies in the dozens involving hundreds and thousands of women have almost consistently shown that menopausal women taking estrogen were less likely to have heart attacks.

However, in observational studies the women who asked for estrogen, or whose doctors insisted they take it, tended to be healthier. In the United States of America estrogen-requesting women tended also to be of higher socioeconomic status and education (and would therefore have health insurance that over 35 million in the USA don’t). In real life as well as in studies, women who were prescribed estrogen were healthier (the healthy cohort bias) (5 [1]). Doctors also followed women taking hormone therapy more closely so they would have likely have had any illnesses detected and treated earlier (ascertainment bias). And women who kept taking estrogen were healthier simply because they were good pill takers—the compliance bias occurs because good takers of placebos have better health than erratic pill-takers (6 [1]). Therefore, estrogen was considered to prevent heart attacks and the consistency of the evidence from many studies was used as proof. So strong was this (biased) belief that a medical Professor, and leader in a large and reputable disease-focused organization in Canada demanded that I, as an invited speaker to a forum on Menopause and Osteoporosis in Toronto, say that estrogen prevented heart disease. When I protested that there was no randomized, controlled trial evidence for that statement, I was told I “had” to say it. I ended up saying that most physicians believed that estrogen prevented heart disease followed quickly by the statement that I thought that was not true because only lower quality, biased studies had shown it.

The strongest kind of study in evidence-based medicine is a double-blind randomized placebo-controlled trial. Double blind means no one conducting the study, participating in the study or making judgments about lab tests knows the real nature of treatment each participant is receiving. Randomized means that there is a rolling of the dice kind of assignment to different therapies. Placebo-controlled means that one of the therapies is inactive or a placebo. In the case of estrogen therapy for heart disease prevention, a randomized placebo-controlled trial in men without heart disease was prematurely discontinued in the 1970s because it showed that estrogen treatment caused heart disease as well as other serious risks in men (7 [1]). Results from that study should have demanded that a similar high-quality study be performed in women before prescribing it for heart disease prevention. Instead, those data were virtually ignored and hundreds of observational studies (remember, a less strong design) were funded, conducted and published. Meanwhile, millions of healthy menopausal women around the world, especially in North America, have been taking harmful hormone “replacement” therapy for disease prevention.

About two decades passed after that trial of estrogen in men before science-based feminist groups (such as the National Women’s Health Foundation in the USA) and the collective writings of skeptical feminists (2 [1];8 [1];9 [1]) had an effect. A randomized controlled trial of heart disease prevention with estrogen, the Women’s Health Initiative (WHI), was begun. Although the estrogen plus progestin arm was stopped, WHI has many other arms that are still ongoing including one testing only estrogen in women who have had a hysterectomy. That study arm, which has about 5,000 fewer women enrolled, has less statistical power to show potential negative effects on heart disease. It may also not show the breast cancer risk of the study in women without hysterectomy because women who have had a hysterectomy, even if ovaries are conserved, have lower basic risks for breast cancer (10 [1]). In addition, other ongoing arms of WHI are studying calcium and vitamin D for osteoporosis prevention and documenting whether a low fat diet compared with a conventional one prevents breast cancer.

The WHI Estrogen plus Progestin arm (E plus P) tested therapy with conjugated equine estrogen (estrogen from pregnant mare’s urine, Premarin®) plus very low dose medroxyprogesterone (a created kind of progesterone called a progestin, Provera®) both taken daily. Women enrolled did not have severe hot flushes and night sweats, heart disease, osteoporosis or breast cancer because the idea of the study was to test the popular concept that E plus P would prevent the chronic serious diseases that menopausal estrogen deficiency was supposed to cause for all women. The WHI showed the opposite of what was expected—E plus P caused heart disease and breast cancer as well as blood clots and strokes. E plus P also prevented osteoporotic fractures including hip fracture and bowel cancers but these good things did not begin to counterbalance the bad (1 [1]).

The ultimate scientific proof is present when several double-blind randomized placebo-controlled trial results are analyzed together to provide what is called a “meta-analysis.” It should rather be called a “mega” analysis because it involves thousands of participants and carries great scientific weight. A recent meta-analysis of long-term ovarian hormone therapy that includes the WHI E plus P results shows no benefit in heart disease prevention and further documents that estrogen causes stroke, blood clots and breast cancer (4 [1]).

Any logical person much less a scientist would now say that the WHI results require re-thinking of the concepts on which menopausal “replacement” and “estrogen deficiency” were based. The arch of prejudice against menopausal women should crumble. Right? If the arch itself still stands, though perhaps it has become a bit wobbly, this says something. I think it is further proof that the menopause-equals-estrogen-deficiency idea is a cultural prejudice against women. In being asked about the WHI’s discontinuation Dr. Wulf Utian, executive director of the North American Menopause Society, said, “It’s not just a matter of what the data says—truth is opinion” (11 [1]). He is counting on societal prejudice to spin WHI results and to maintain the status quo.

Another aspect of the response to the WHI that bothers me is that women were forgotten in all the hype about the study’s abrupt termination. The website for the WHI study said, “Women with a uterus should stop their study drugs immediately.” Following that order led to hot flushes in many women who were on hormone therapy. And women who heard the news of the study in the media also often stopped their therapy abruptly (12 [1]). That led to severe hot flushes in hundreds of thousands of women and in virtually all women who had previously experienced them. Because it is well known that rapid withdrawal from estrogen causes night sweats and hot flushes, all physicians should have been ready with advice for women wanting to stop estrogen. However, the few resources available are those provided by feminist physicians (www.drsusanlove.com [2]; www.cemcor.ubc.ca [3] ). Without helpful information, and because night sweats/hot flushes are so miserable, many women, in desperation, have reluctantly restarted unwanted estrogen treatment.

Back to evidence-based medicine. Although, both as a scientist and a woman, I strongly support evidence-based medicine the process through which our society creates science is complex. The process begins when, as a clinician, I feel a need for new therapy. Let me be specific—about 25% of midlife women have heavy vaginal bleeding (13 [1]) because of high estrogen levels and low progesterone levels during early perimenopause (14 [1];15 [1]). A physiology-based therapy would right this hormonal imbalance by giving a high dose of natural progesterone for at least 14 days at the end of the cycle. Progesterone’s job is to counterbalance estrogen’s effects everywhere in the body but especially to cause thinning of the endometrial lining of the uterus and to decrease heavy flow.

Given this identification of an important problem for women, and after many years of successfully using this new therapy in clinical practice, colleagues and I at the Centre for Menstrual Cycle and Ovulation Research decided to apply for funds to do a controlled trial. We then spent several months working through nights and weekends writing and assembling the complex data needed for a grant application. This proposal was to test the use of high dose cyclic progesterone in a randomized double blind, placebo-controlled trial and compared with birth control pills, the standard treatment doctors currently use for heavy bleeding in midlife women. We sent this application to Canada’s publicly funded scientific granting agency, the Canadian Institutes for Health Research.

Response—this application is funded because it is important and because perimenopausal women currently make up a large portion of the population of women in Canada. It is funded because it could potentially decrease women’s anguish at flooding menstruation, women’s time lost from work, could prevent most hysterectomies and is a women-centred and physiology-based therapy. Wrong. It is three times rejected and each revision is criticized in a different way. Why? Perhaps because it is not about molecular genetics (science has its fashions) or perhaps women’s bleeding, because of society’s taboo (16 [1]) triggers a negative response in the largely men on the “peer review” committee. Or perhaps this was rejected because our outcome measure, vaginal flow, was assessed using two kinds of women’s self-report data from a standardized diary form (Daily Perimenopause Diary (17 [1])) and was not a “hard” endpoint. Whatever the reason, it means that as a physician I am left to prescribe and teach about a therapy that is not evidence-based. And that every day hundreds of midlife women have their lives disrupted by heavy flow or undergo unnecessary surgeries.

Another part of the process of making science is that the results of a study must be published. In the realm of science, something that is unpublished doesn’t exist. I once got funding for and successfully completed a trial of cyclic medroxyprogesterone treatment for bone loss in healthy, active young women whose menstrual cycles or ovulation were disturbed. That study used a randomized, double blind design and was placebo-controlled. Further it showed the highly positive results that progestin therapy significantly increased bone density while the placebo-treated women lost bone. When I submitted the manuscript of these results it was rejected nine times over three years before it was eventually published (18 [1]). During those frustrating three years, I nearly lost my application for tenure at the University of British Columbia. This means I was almost fired from my job. The process of creating evidence-based medicine is fraught with hazard for a feminist scientist.

So, let’s get back to WHI, menopause, evidence-based medicine and prejudice. I believe that we have a unique opportunity, right now, as women at this anniversary of the initial WHI results. We must call attention to the negative ideas about women that are part of our culture and that WHI so clearly illustrates. Destroying the concept of estrogen deficiency will be hard. Don’t forget that we, as women, are also members of this culture and at least in part adopt much of what culture believes about women. It is now time to declare that menopause is a normal part of every woman’s life cycle. I call menopause “graduation.” As a menopausal woman I truly feel freed—I’ve survived the premenopausal demands of cyclic estrogen and the chaos of perimenopause which is estrogen’s storm season. As women it is now time to declare that the estrogen deficiency concept of menopause is wrong, based on prejudice and not on science.

What can we do? First change the language. I and others (19 [1]) have been trying for many years to persuade physicians, women and organizations to stop using the term “hormone replacement therapy” or “HRT.” Those terms imply that menopausal women are deficient and need fixing. Now the Federal Drug Association in the USA also says menopause treatment should be termed “hormone therapy.” (However, that term is vague and could apply to testosterone for “andropause” in men, DHEAS purchased under the counter, or ground up calf adrenal glands!) Instead “Ovarian Hormone Therapy” is a good descriptive term and rolls off the tongue in its abbreviation as “OHT.” Use it! And demand that others also say OHT.

Most thought leaders and women’s health organizations are now either directly or indirectly advocating abandoning any hormone therapy for menopausal women. But OHT continues to be an important and physiological treatment for women whose menopause came early (before age 40 or perhaps 45), for women with disturbing hot flushes especially causing chronic sleep disruptions, and for menopausal women with osteoporosis plus hot flushes (20 [1]). However, my concept of OHT is different from what many imagine. To me it means “bio-identical” or “natural” kinds of estrogen and progesterone both given in physiological doses daily unless a woman wishes flow. (If a woman desires flow I would give estrogen for days 1-25 a month and full dose progesterone days 14 through 27 of the month.) On doses of estrogen and progesterone that are equally balanced, menopausal women have no further bleeding based on my extensive clinical experience. (Note that this hasn’t been proven in a controlled trial). However, with the full dose estrogen and very low dose progestin treatment tested in the WHI, 41% of women in the hormone arm had irregular, abnormal bleeding (requiring that the randomization code be broken) as well as increased rates of hysterectomy over placebo-treated women (1 [1]). This unpredictable, abnormal bleeding, which planners knew would occur based on evidence from an earlier randomized, controlled study (21 [1]), is further evidence of medicine’s misogyny. (Would an abnormal and unpredictable penile discharge be considered a trivial complaint of a man?)

Have I changed my thoughts and practice since WHI? I have always believed that menopause is normal with naturally low estrogen and progesterone levels. And my teaching, writing and prescribing practices have always been consistent with that concept. However, I will never again prescribe estrogen as a pill. There is now overwhelming controlled trial evidence that pill forms of estrogen cause unacceptably high rates of clots and perhaps, through clotting, also increased strokes and heart disease. There are not yet strong data but it is likely that estrogen given through the skin (as a gel, crème or patch) will be less likely to activate liver clotting factors and thus safer.

In summary, the prejudicial, negative idea that menopause-means-estrogen-deficiency is the arch for which hormone “replacement” therapy was the keystone. With the negative results of the WHI the prejudice against menopause and women must stop. We now have evidence-based proof that “replacement” is harmful to women. Each of us must practice living with these important ideas—menopause is normal and low estrogen levels are natural.

What can we do? We must pressure the Canadian Institutes for Health Research to fund important research for women. CIHR must ensure that taboos and prejudice against women are eliminated from the process of allocating Canadian’s tax dollars to fund research. Until research studies rising from women’s needs are funded equitably and published fairly, the prejudice will continue. Medical journals must be pressured to review articles in a blinded fashion, without knowledge of the identity of the authors. More broadly and practically, if we hear doctors, reporters, other women, our partners, our kids or anyone saying “hormone replacement therapy,” “HRT” or “estrogen deficiency” we must hold up our hands, interrupt. “That feels to me like prejudice.” Say it strongly. “That feels like prejudice.” Further, say that these prejudicial ideas are unscientific. We have evidence-based proof from WHI. It is also time to be humble and ask why so many of us, who want to help women, previously “bought” the menopause-means-estrogen-deficiency-requires-replacement concept. Finally, although we will eventually be successful in eliminating menopause-related prejudices, all of us must be alert to the new mutations of misogyny our culture will undoubtedly create.

Reference List

1. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in health postmenopausal women: principal results from the Women's Health Initiative Randomized Control trial. JAMA 2002; 288:321-333.
2. Love S. Doctor Susan Love's Hormone Book. San Francisco: Random House 1997.
3. Hays J, Ockene JK, Brunner RL, Kotchen JM, Manson JE, Patterson RE et al. Effects of estrogen plus progestin on health-related quality of life. N Engl J Med 2003; 348(19):1839-1852.
4. Beral V, Banks E, Reeves G. Evidence from randomised trials on the long-term effects of hormone replacement therapy. Lancet 2002; 360(9337):942-944.
5. Barrett-Connor E. Postmenopausal estrogen and prevention bias. Ann Int Med 1991; 115:455-456.
6. Petitti DB. Hormone replacement therapy and heart disease prevention. Experimentation Trumps Observation (editorial). Journal of the American Medical Association 1998; 280:650-652.
7. Coronary Drug Project Research Group. Coronary drug project: findings leading to the discontinuation of the 2.5 mg/day estrogen group. Journal of the American Medical Association 1973; 226:652-657.
8. Prior JC. Critique of estrogen treatment for heart attack prevention: the nurses' health study. A Friend Indeed: for Women in the Prime of Life 1992; VII:3-4.
9. Prior JC. One voice on menopause. J Am Med Women Assoc 1994; 49:27-29.
10. Kreiger N, Sloan M, Cotterchio M, Kirsh V. The risk of breast cancer following reproductive surgery. Eur J Cancer 1999; 35:97-101.
11. Kolata G, Petersen M. Hormone Replacement Study a Shock to the Medical System. July 10, 2002, 1-5. 2002. New York, New York Times.
12. Grady D. A 60-year-old woman trying to discontinue hormone replacement therapy. JAMA 2002; 287:2130-2137.
13. Kaufert PA. The perimenopausal woman and her use of the health services. Maturitas 1980; 2:191-205.
14. Santoro N, Rosenberg J, Adel T, Skurnick JH. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab 1996; 81:4,1495-1501.
15. Prior JC. Perimenopause: The complex endocrinology of the menopausal transition. Endocr Rev 1998; 19:397-428.
16. Martin E. Medical metaphors of women's bodies: menstruation and menopause. Int J Health Serv 1988; 18:237-254.
17. Hale GE, Hitchcock CL, Williams LA, Vigna YM, Prior JC. Cyclicity of breast tenderness and nighttime vasomotor symptoms in midlife women: information collected using the daily perimenopause diary. Climacteric. 2003.
18. Prior JC, Vigna YM, Barr SI, Rexworthy C, Lentle BC. Cyclic medroxyprogesterone treatment increases bone density: a controlled trial in active women with menstrual cycle disturbances. Am J Med 1994; 96:521-530.
19. Speroff L. It's time to stop using the word 'replacement'. Maturitas 2000; 34:1-3.
20. Prior JC. Menopause. In: Gray J, Johnson G, editors. Therapeutic Choices. Ottawa, Ontario, Canada: C.K. Productions, 1995: 468-477.
21. Lindenfeld EA, Langer RD. Bleeding patterns of the hormone replacement therapies in the postmenopausal estrogen and progestin interventions trial. Obstet Gynecol 2002; 100(5 Pt 1):853-863