How Has The publication of the Women’s Health Initiative (WHI) Study in July 2002 changed the approach taken to the management of menopause?

Dr Philip Thomas, Essay for Masters of Reproductive Medicine (UNSW) 2011


In addressing this topic it is prudent first to examine the design of the WHI, its chief findings and the flaws in the initial presentation of the results due to the influence of the media.

The WHI study was a large, well designed, randomised controlled trial (seeking level 2 evidence), which ran from 1993 to 1998, recruiting 161,809 women aged 50 to 79, across 40 centres in the USA (1). Older women were chosen, as they had to be largely free from vasomotor symptoms, in order to be successfully blinded. The components were firstly, two randomised trials of postmenopausal hormone replacement therapy (HRT) and secondly a dietary modification trial examining low fat diet, normal diet and calcium/vitamin d supplementation. The HRT trial examined combined HRT (cHRT, 0.625mg of conjugated equine oestrogen (CEE) and 2.5 mg medroxyprogesterone acetate) versus placebo and randomised 16,608 women. The other arm randomised 10,739 hysterectomised women to either 0.625 mg CEE (eHRT) or placebo. The cHRT arm was terminated early (in 2002) because of a statistically significant increase in invasive breast cancers and cardiovascular events (2), which exceeded pre determined benefit/harm parameters. The oestrogen only arm was terminated in 2004 because of an increased risk of stroke and a trend towards increased dementia/cognitive impairment (3). There was however no change in incidence of coronary heart disease, a reduction in hip fracture, and, importantly, no increased risk of breast cancer. There has been criticism of the statistical treatment of the data in view of the trial’s multiple endpoints (4).

Importantly, the two HRT arms of the trial were randomised against their own placebo groups and not each other. The groups were dissimilar (5). Those in the oestrogen only arm were more likely to be obese, less active, had more pre-existing coronary artery disease (CAD); and had more early births, longer duration of previous HRT use, and a higher chance of previous bilateral oophorectomy (as they had had hysterectomies). Therefore, although the WHI was an RCT the two HRT groups could not be directly compared.

The status of HRT prior to WHI

This was the first major RCT examining these issues. Prior to 2002 most data arose from case-control and observational studies (level 3 evidence) (6). These studies suggested that long-term HRT users had reduced cardiovascular risk, but increased risk of breast cancer and thromboembolism. The WHI trial however showed that in a relatively asymptomatic older population commencing cHRT on average 13 years after menopause, one could expect a reduction in fractures, an increased risk of breast cancer and cardiovascular events and no overall cardiovascular benefit (7). Note that the published results of the WHI therefore agreed with the previous 30 years of case-control and cohort data, with the exception of cardiovascular results (8).

Problems with initial data presentation

There was however a 10% rate of disagreement of the diagnosis of myocardial infarction and 3% for death due to coronary artery disease when analysed by the adjudicating panel. This small difference meant that the increased risk of cardiovascular events no longer reached statistical significance (9). In addition, 40% of the cHRT arm was unblinded due to vaginal bleeding, leading to a possible impact on doctors’ management and diagnoses (10) with respect to cardiovascular and breast disease.

How were the results of the WHI study released?

The WHI trial results were summarised to the press in Australia, by Dr Andrew Penman, then the CEO of the NSW State Cancer Council, as follows: “The research found that combined HRT led to a 26% increase in breast cancer… the link was so significant that the arm of the study using combined HRT had to be stopped”. He was referring to a relative risk increase of 0.26, and the popular press, many health practitioners and most patients interpreted this as an absolute risk increase of 26%. The popular press in this country and others ran stories to that effect. As always, bad news travels fast and it makes good reading. People also remember bad news.

What was the initial effect on health consumers/patients?

Some interpreted this to mean that 1:4 women on HRT would get breast cancer. Overnight up to two thirds of women ceased their HRT, many of whom experienced severe oestrogen withdrawal symptoms.

How did the medical profession react?

The mood was extremely cautious, with many advisory bodies issuing strongly worded guidance that HRT should only be used in the lowest possible dose, for the shortest possible time and only then in women who had the most severe symptoms (6). Most physicians even if well informed, felt that the extra risk was not worth it in view of the negative media and public sentiment. If a patient was to develop breast cancer, and HRT had been used, then there was real potential for litigation. Even today, women who develop breast cancer who have been previous users of HRT are led to think that the HRT had a large part in its aetiology, regardless of her specific circumstances. It also became very difficult to recommend HRT for the prevention of cardiovascular disease despite the fact that there is a probable “window of opportunity” when HRT is initiated under the age of 60, when the risk of cardiovascular disease is actually reduced. This is due to persistence of estrogen receptors on vascular endothelium (6).

What were the conclusions of the post WHI analyses?

Subsequent subgroup analysis showed that only the women in the cHRT arm who were 20 or more years post menopause had a statistically significant increase in CAD. Subtracting this group from the rest of the trial participants meant that there was no longer a difference between treatment and placebo groups. Therefore based on this, one can only conclude that the WHI findings of increased risk of CAD only apply to a particular subgroup of older women. The data from the eHRT group indicated that younger women experienced a decreased risk (7). Subsequent re analysis of the cHRT arm data in 2007, supported these findings (11).

In addition, subsequent analyses and data from RCT’s and cohort studies have changed the conclusion for those who commence HRT around the time of menopause.

The critical window hypothesis: this states that estrogen can reduce the risk of CAD in women who commence HRT at a relatively young age, before atherosclerotic plaque (with necrosis and inflammatory change) has set in. It is worth emphasising that in subsequent subgroup analysis of the WHI data, only the risk of CAD in women who were 20 years post menopause, reached statistical significance. There was no increase in risk for those aged 50-59; indeed there was a reduced overall mortality (5).

The WISDOM trial was a further long-term level 1 trial examining the effects of HRT in women 10-13 years post menopause (31). Combining this data with that of thaw WHI now supports cardio protection in women starting HRT near menopause, especially when an estrogen only regimen is used. Unfortunately the WHI trial did not study the appropriate population for the right length of time to establish that long term HRT does not exert a primary preventing effect on the risk of CAD (10).

The question of secondary prevention: However, pooled data and experimental models indicate that healthy endothelium is needed to respond to estrogen and the beneficial effects of estrogen are diminished with progressive atherosclerosis (Clarkson’s monkey model). In addition the vasodilatory effect of estrogen diminishes with age. There have been several randomised trials of HRT in the secondary prevention of CAD, all suggest no beneficial effects of HRT, regardless of the type of estrogen or progestogen used (12).

Venous thromboembolism, HRT and the WHI

Most studies including the WHI indicate that the risk of venous thromboembolism increase in the first two years, and is highest in those with thrombophilia or who are obese (6). The baseline risk is about 1/10,000 at the age of 50 and so a doubling of risk still represent a very low overall incidence (10).

Ovarian cancer

The cancelled cHRT arm of the WHI reported a non-significant increase in the risk of ovarian cancer. Other case control studies have concurred with this and the risk may be confined to endometrioid cancers (15). However individual studies lack numbers and the risk is likely to be very small.

Cognitive function and dementia

The WHI studied only women commencing HRT over 65 and showed a slight detriment, but smaller observational studies have shown a small therapeutic window whereby HRT may be helpful. Results are inconsistent (6)

WHI and stroke

The WHI reported an overall increase in the risk of strokes in both HRT arms, but no increase in fatal strokes. (9, 16) the findings of the Nurses Health Study were similarly although the latter failed to show an increase in the first ten years of therapy, when those older than 60 were excluded. An increase in TIA’s and strokes must be assumed in those commencing HRT many years after menopause (6).

Bowel and uterine cancers

A small non-significant decrease in these cancers was seen in the WHI cHRT arm 21%, (17) eHRT had no effect on bowel cancer. If the cHRT arm had continued it is thought that a larger reduction would have been apparent. These principles are consistent with previous concepts of progestogen protection of the endometrium.

Breast cancer

Prior to the WHI trial observational (or cohort, level 2) studies had suggested increased risk of breast cancer for medium term users of cHRT of around 1.53 after an average 8 years (6). The WHI actually reported a relative risk increase of only 1.26 after 5.6 years, but the media expressed this as a 26% risk of breast cancer. Systematic reviews of all level 1 data now suggest an increased risk of 4/10,000 woman years for cHRT or 2/1000 women after 5 years. This risk is best contextualised by comparing to the elevated risk associated with late menopause (55 years, RR 1.22), moderate alcohol consumption (three drinks per day, RR 1.4), or nulliparity (RR 1.67). In addition, the eHRT arm of the WHI showed a non-significant reduction in breast cancer (18), whereas long-term observational data (level 2) suggests that more than 20 years of eHRT may increase breast cancer risk again (6).

Menopausal vasomotor symptoms and quality of life

The older population of women in the WHI were selected for their lack of vasomotor symptoms, however the WHI did show that joint pains were significantly reduced by HRT and these recurred upon cessation (6). This is consistent with previous data.

What has changed in terms of doctors’ prescribing habits?

The mood of caution has given way to real attempts to assess and use lower doses of estrogen. Although half the standard dose of CEE has been shown to prevent bone loss and treat vasomotor symptoms, this dose not have the same effect on circulating lipids and lipoproteins. This was elegantly demonstrated by Clarkson’s monkey model (19), however Clarkson’s monkeys still enjoyed a decrease in atheroma formation despite a lesser numeric benefit on lipid levels.

Immediately following the release of the WHI, many women were mistakenly advised against HRT by their doctors or pharmacists on the basis of it being “too risky” and were steered in the direction of many zero-evidence alternative and complimentary therapies. This meant that many women, after 2002, missed the window of opportunity for cardio-protection, possible cognitive benefit, and prevention of urogenital atrophy (6).

The rise of other pharmacological treatments for menopause

As a corollary of physician uncertainty and risk minimisation described above, there has been renewed attention to drugs such as tibolone. Tibolone was first described in the 1960’s, is a derivative of wild yam and is structurally related to the 19-nortestosterone progestins used in oral contraceptives. I like to think of it as a pro-drug, as it is metabolised into three different drugs in the liver and intestine (20), which exert different actions in different tissues. It is as effective as standard HRT regimens for treatment of vasomotor symptoms (21) provides an estrogenic effect on the vagina (22) benefits libido (23) may benefit cognition (24) and benefits bone density. Effects on the cardiovascular system are likely to be neutral (25). It does not increase breast density or risk of de novo breast cancer and may in fact decrease it due to its inhibitory action on estrone sulfatase (26). However according to the LIBERATE trial (27) tibolone is still contraindicated after breast cancer, especially in those who are estrogen receptor positive.

The rise of complimentary medicine treatments for menopause

The WHI trial resulted in renewed interest in complimentary medicines for menopausal symptoms. The sale of alternative therapies is now a worldwide multi-billion dollar industry (10) and these varied compounds are described as dietary supplements so are not required to demonstrate safety or efficacy. Middle aged, well-educated women are the most frequent users of alternative therapies (28) however the evidence for their use is poor. Very few have an effect better than placebo (an effect which can in itself be as high as 50%) and none have any benefit on bone density (29). Calcium and vitamin D supplementation remain proven options for preventing bone loss.

Conclusion and summary

Clinical trials make a strong argument that the best approach is to initiate HRT in the early post-menopausal years, before there has been a significant drop in estrogen and commencement of atherosclerotic plaque. This means then that appropriately timed HRT can exert a primary preventive effect. The overall impression is that the latest data on HRT does not warrant the panic that gripped the world in 2002. Longer-term therapy is appropriate for women who are aware of the potential side effects and small potential for harm (6). When HRT is initiated soon after menopause for symptoms control and to improve quality of life there are likely to be additional bone, heart and possibly cognitive benefits that outweigh the risks which are not significantly elevated under the age of 60. As for the question of how long a woman should continue HRT for, simply put a woman should continue HRT for as long as she wants the benefits. Although some estrogen effects are long lasting, the full impact is lost soon after cessation and the reduced risk of cardiovascular mortality is lost by the fifth year (30).

In summary, I do not believe the WHI trial “changed everything forever”. What it did achieve is to raise awareness of the power of the media in influencing public opinion. We as a profession need to be extremely vigilant in how data of this importance is expressed to the media and the public. WHI triggered a critical reappraisal of how we see HRT, the doses used and how we counsel patients. As such, I believe it was in the end a victory for patient advocacy, and all future users may be more fully informed of the risks and benefits of the therapy undertaken, although the conclusions had been largely reached before by cohort and observational studies.


  1. The Women’s Health Initiative Study Group, Design of the Women’s Health Initiative clinical trial and observational study, Controlled Clin Trials 19:61, 1998.
  2. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg CL, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J, Writing Group for the Women’s Health Initiative Investigators, Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women’s Health Initiative randomised controlled trial, JAMA 288:321, 2002.
  3. The Women’s Health Initiative Steering Committee, Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. The Women’s Health Initiative Randomized Controlled Trial, JAMA 291:1707, 2004.
  4. Babushka Patel, Robyn Norton and Stephen McMahon MJA 2002 177 (7): 345-346
  5. Stefani ML, Cochrane BB, Hsia J, Bared DH, Liu JH, Johnson SR, The Women’s Health Initiative postmenopausal hormone trials: overview and baseline characteristics of participants, Ann Epidemiol 13:S78, 2003.
  6. MacLennan A. Hormone therapy: a 2008 perspective. Obs Gyn and Repro Med 19:1, 2008
  7. Writing group for the WHI Investigators. Risks and benefits of estrogen plus progestin in healthy post menopausal women. Principle results from the WHI randomised controlled trial. JAMA 2002; 288: 321-333
  8. Manson JE, Hsia J, Johnson KC, Rossouw JE, Assaf AR, Lasser NL, Trevisan M, Black HR, Heckbert SR, Detrano R, Strickland OL, Wong ND, Crouse JR, Stein E, Cushman M, for the Women’s Health Initiative Investigators, Estrogen plus progestin and the risk of coronary heart disease, New Engl J Med 349:523, 2003.
  9. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg CL, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J, Writing Group for the Women’s Health Initiative Investigators, Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women’s Health Initiative randomised controlled trial, JAMA 288:321, 2002.
  10. Post menopausal hormone therapy. In: Speroff L Clinical gynecologic endocrinology and infertility. Lippincott 2011.
  11. Toh S, Hernández-Diaz S, Logan RF, Rossouw JE, Hernán MA, Coronary heart disease in postmenopausal recipients of estrogen plus progestin therapy: does the increased risk ever disappear? A randomized trial, Ann Intern Med 152:211, 2010.
  12. Clarke SC, Kelleher J, Lloyd-Jones H, Slack M, Schofield PM, A study of hormone replacement therapy in postmenopausal women with ischaemic heart disease: the Papworth HRT Atherosclerosis Study, Br J Obstet Gynaecol 109:1056, 2002.
  13. Waters DD, Alderman EL, Hsia J, Howard BV, Cobb FR, Rogers WJ, Ouyang P, Thompson P, Tardif JC, Higginson L, Bittner V, Steffes M, Gordon DJ, Proschan M, Younes N, Verter JI, Effects of hormone replacement therapy and antioxidant vita- mins supplements on coronary atherosclerosis in postmenopausal women. A randomized controlled trial, JAMA 288:2432, 2002.
  14. The ESPRIT Team, Oestrogen therapy for prevention of reinfarc- tion in postmenopausal women: a randomised placebo controlled trial, Lancet 360:2001, 2002.
  15. Purdie DW, Bain CJ, Siskind V, Russell P, Hacker NF, Ward BG, Quinn MA, Green AC, Hormone replacement therapy and risk of epithelial ovarian cancer, Br J Cancer 81:559, 1999.
  16. Margolis DJ, Knauss J, Bilker W, Hormone replacement therapy and prevention of pressure ulcers and venous leg ulcers, Lancet 359:675, 2002.
  17. AndersonGL,JuddHL,KaunitzAM,BaradDH,BeresfordSAA, Pettinger M, Liu J, McNeeley SG, Lopez AM, for the Women’s Health Initiative Investigators, Effects of estrogen plus progestin on gynecologic cancers and associated diagnostic procedures. The Women’s Health Initiative Randomized Trial, JAMA 290:1739, 2003.
  18. Womens health initiative steering committee. Effects of conjugated equine estrogen in post menopausal women with hysterectomy. The WHI randomised controlled trial. JAMA 2004; 291:1701-1712
  19. Appt S, Clarkson TB, Lees CJ, Anthony MS, Low dose estrogens inhibit coronary artery atherosclerosis in postmenopausal monkeys, Maturitas 55:187, 2006.
  20. de Gooyer ME, Deckers GH, Schoonen WGEJ, Verheul HAM, Kloosterboer HJ, Receptor profiling and endocrine interactions of tibolone, Steroids 68:21, 2003.
  21. Trevoux R, Dieulangard P, Blum A, Efficacy and safety of Org OD 14 in the treatment of climacteric complaints, Maturitas 5:89, 1983.
  22. Siseles NO, Halperin H, Benencia HJ, Berg G, Pilnik S, Mesch V, Arrighi B, Wikinski RW, A comparative study of two hormone replacement therapy regimens on safety and efficacy variables, Maturitas 21:201, 1995.
  23. Nathorst-Böös J, Hammar M, Effect on sexual life — a com- parison between tibolone and a continuous estradiol-norethisterone acetate regimen, Maturitas 26:15, 1997.
  24. Fluck E, File SE, Rymer J, Cognitive effects of 10 years of hormone-replacement therapy with tibolone, J Clin Psychopharmacol 22:62, 200
  25. Nathorst-Böös J, Hammar M, Effect on sexual life — a com- parison between tibolone and a continuous estradiol-norethisterone acetate regimen, Maturitas 26:15, 1997.
  26. Chetrite G, Kloosterboer HJ, Pasqualini JR, Effect of tibolone (Org OD14) and its metabolites on estrone sulphatase activity in MCF-7 and T-47D mammary cancer cells, Anticancer Res 17:135, 1997
  27. Kenemans P, Bundred NJ, Foidart J-M, Kubista E, Von Schoultz B, Sismondi P, Vassilopoulou-Sellin R, Yip CH, Egberts J, Mol-Arts M, Mulder R, van Os S, Beckmann MW, on behalf of the LIBERATE Study Group, Safety and efficacy of tibolone in breast-cancer patients with vasomotor symptoms: a double-blind, randomised, non-inferiority trial, Lancet Oncol 10:135, 2009.
  28. Adams J et al. Med J Aust 2003; 179:297-300
  29. Davis S. Finding the balance: the who and WHI of hormone replacement therapy. NPS news, 32: 2004
  30. Grodstein F, Stampfer MJ, Colditz GA, Willett WC, Manson JE, Joffe M, Rosner B, Fuchs C, Hankinson SE, Hunter DJ, Hennekens CH, Speizer FE, Postmenopausal hormone therapy and mortality, New Engl J Med 336:1769, 1997
  31. Vickers M, Martin J, Meade T. The Women’s international study of long-duration oestrogen after menopause (WISDOM): a randomised controlled trial. BMC Women’s Health 2007.