In sharp contrast to prevailing hypotheses, taking combination hormone replacement actually doubles a woman’s risk of developing dementia, including Alzheimer's disease, when the therapy is begun at age 65 or older, according to a study in today’s Journal of the American Medical Association. Hormone replacement also did not prevent mild cognitive impairment in trial participants.

Evidence from earlier studies suggested that hormone replacement would decrease the risk of dementia. Although prospective observational trials have not shown any benefit, a meta-analysis of 14 epidemiologic studies suggested that estrogen could cut AD risk by half.

The Women's Health Initiative Memory Study assessed combination therapy's effect on risk in a large randomized, double-blind, placebo-controlled trial with nearly 4,500 participants aged 65 or older. Sally A. Shumaker, Wake Forest University Health Sciences, Winston-Salem, North Carolina, and colleagues report that overall, 61 trial participants were diagnosed with probable dementia. Of those, 40 (66 percent) were receiving estrogen and progestin, while 21 (34 percent) were taking placebo. About half of the cases in each group were classified as AD.

The results did not change significantly even after the investigators controlled for baseline dementia rates, adherence, use of statins and nonsteroidal antiinflammatory drugs, socioeconomic and educational status, and prior hormone use.

The study is an ancillary to the larger hormone therapy trials within the Women’s Health Initiative (WHI) that were halted last summer because the participants were found to be at greater risk for heart disease, stroke, pulmonary embolism, and breast cancer (see ARF related news story). It is possible that small strokes contributed to the increased risk of AD in this trial, the researchers write. The present study also comes hard on the heels of an analysis of the WHI study published earlier this month, which found that, contrary to prior studies and general expectation, combined estrogen-progestin therapy does not improve health-related quality of life (Hays et al., 2003).

The absolute risk of dementia is still small, the researchers point out. The increased risk of HRT would increase the number of cases by 23 for every 10,000 women. Even so, the authors conclude: "Estrogen plus progestin should not be prescribed with the expectation that it will enhance cognitive performance in postmenopausal women. When considered in conjunction with the WHI results reported earlier, the WHIMS estrogen plus progestin data reinforce the conclusion that the risks of estrogen plus progestin outweigh the benefits."

Participants in the trial took a combination of estrogen and progestin. The effect on risk may be different in women taking only estrogen, rather than the combination, the researchers suggest. That trial is still in progress.—Apoorva Mandavilli


  1. Basically, the results from over 4,500 women studied in the WHI and WHIMS studies conclude independently that combination hormone replacement therapy (CHRT) holds no efficacy in protecting women from Alzheimer's when the hormones are begun at age 65 or older. Taken together with last year's report that CHRT does not protect against atherosclerosis, there is now no indication for CHRT as prevention of any illness. Indeed, in the NHLBI studies and in one of the new WHI studies, side effects (clotting and Alzheimer's, respectively) were increased in the group of women taking CHRT.

    More than 80 percent of women on CHRT are under 65, however, and it remains unclear whether hormones begun perimenopausally might be more effective than those begun many years later: Certainly this result emerged from the Cache County study (Zandi et al., 2002). More importantly, it is worth noting that virtually all of the basic and early clinical literature suggesting that hormones protected against Alzheimer's utilized unopposed, estrogen-only formulations. The estrogen-only protocol is now only available to women who have had hysterectomies, since uterine cancer rates can increase. An arm of the WHI testing unopposed estrogen replacement continues in women who have had hysterectomies, and results are expected in 2005. Biologically, progestins antagonize many of the effects of estrogens, giving plausibility to the hypothesis that progestins may be "undoing" any good effects of estrogens.

    If the estrogen-only study shows benefit, the effort to identify brain-specific SERMs (selective estrogen receptor modulators) will be bolstered, in an effort to provide the protective effects without the side effects on reproductive tissues. If unopposed estrogens are not protective against Alzheimer's, or if a protective window is not identified, enthusiasm for this proposed strategy will be very difficult to sustain.


    . Hormone replacement therapy and incidence of Alzheimer disease in older women: the Cache County Study. JAMA. 2002 Nov 6;288(17):2123-9. PubMed.

  2. I'd like to follow up on Sam Gandy's comment regarding the antagonistic effect of progestins on the estrogenic effect on memory. While the results from the WHI trial on unopposed estrogen will report more definitive answers, we published in 2000 observational results from the Kame Project showing mean change scores in performance on the Cognitive Abilities Screening Instrument (CASI) over two years among 837 women of +0.79 (sem=.19) for women who were never on hormone replacement therapy; +1.68 (sem=.36) for unopposed estrogen users (p=.04) and -0.41 (sem=.50) for current estrogen-progestin users (p=.02). These figures were adjusted for age, education, language spoken at the interview (Japanese/English), surgical menopause and baseline CASI score. These earlier results support a modest beneficial effect of current unopposed estrogen use on rate of cognitive change, and a modest detrimental effect of combined estrogen-progestin use relative to women who have never taken hormone replacement. This was the first observational study to examine potential differences in combined therapy versus unopposed estrogen only. In this study, all women receiving combined therapy were on medroxyprogesterone acetate. In addition to studies of unopposed estrogen, other formulations of combined therapy should be examined.


    . Postmenopausal estrogen and estrogen-progestin use and 2-year rate of cognitive change in a cohort of older Japanese American women: The Kame Project. Arch Intern Med. 2000 Jun 12;160(11):1641-9. PubMed.

    . Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA. 2003 May 28;289(20):2651-62. PubMed.

  3. Reply by Alzforum Editors to Comment by Mikko Laakso

    The election of this paper by Alzforum readers represents an example of where the Milestone status reflects not so much an endorsement of a study, but an acknowledgment that the paper has changed the field in practical terms. In this case, the paper was a setback in the U.S. in that it effectively ended clinical research on HRT and made it more difficult for scientists to obtain funding for clinical, animal, or cell-based research on the effect of various preparations of estrogen or estrogen-like compounds on the aging brain. The Alzforum continues to cover this topic in Live Discussions, news, and commentary. Perhaps this study is more of a “Gravestone” to ongoing research than a true “Milestone.” It did, however, prompt a reorientation toward different compounds and different times of dosing.

    View all comments by Gabrielle Strobel
  4. I don't understand how this can be a milestone paper. Where is the criticism of the scientific community? This paper is likely to cause great damage among the female population in terms of increased AD in the future. The study setting is questionable. The results only apply to the horse urine estrogen, which seems to have adverse effects not only in AD, but in cardiovascular diseases and stroke. This paper ignores the positive effects of all of the estrogens that are available, which do exert a protective effect against AD (and cardiovascular diseases).

    The message this paper gives is only that the horse urine estrogen should not be used under any conditions. Period. Any conclusions about other estrogen preparations should not be drawn based on this study.

    View all comments by Mikko Laakso
  5. Reply to comment above by Gabrielle Strobel.

    That this paper may well be a gravestone is an apt description. I think this paper and related papers have and will continue to cause untold damage. The WHI studies were designed to address a very specific aim: whether or not the widely used drugs Premarin (pregnant mare urine) and Provera (the synthetic progestin and anti-estrogen) would be beneficial for the heart health of post-postmenopausal women. The studies were a success in discrediting this form of treatment, but it is unfortunate that few have looked beyond the effects of estrogen on the heart. The negative results of the WHI studies should be interpreted within the context of an extensive literature on the biology of the gonadal steroid hormones and their receptors, as well as their mechanisms of action. This has rarely been done.

    Most people, including physicians and journalists, continue to ignore that estrogen and progesterone have beneficial functions for the brain and are not just a source of dementia. The brain is a major target of estradiol and progesterone throughout life. In the brain, estradiol influences higher cognitive functions, mood, pain mechanisms, motor skills, susceptibility to seizures, hot flashes and sleep. As with estradiol, progesterone and its metabolites also protect the brain and have been shown to have sedative, hypnotic, anesthetic, anxiolytic, sleep-modulating, anticonvulsant, antidepressant and antipsychotic effects in animals and humans. In contrast, not only is the synthetic progestin Provera an ineffective neuroprotectant, but it actually antagonizes estrogen-induced neuroprotection and exhibits none of progesterone’s beneficial properties.

    The therapeutic potential of estradiol and progesterone has not yet received a proper scientific evaluation. Too much emphasis has been placed on the negative effects of estrogen on the heart and blood vessels to the exclusion of all other estrogen targets, particularly the brain.

    View all comments by Dominique Toran-Allerand

Make a Comment

To make a comment you must login or register.


News Citations

  1. Safety Concerns Derail Estrogen/Progestin Trial

Paper Citations

  1. . Effects of estrogen plus progestin on health-related quality of life. N Engl J Med. 2003 May 8;348(19):1839-54. PubMed.

Further Reading


  1. . Estrogen replacement and risk of Alzheimer disease. JAMA. 2003 Mar 5;289(9):1100-2. PubMed.

Primary Papers

  1. . Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA. 2003 May 28;289(20):2651-62. PubMed.