Preventive Medicine Column

Dr. David L. Katz

It may well be that from a cold, hard biological perspective, menopause- and, for that matter, the somewhat controversial male counterpart, andropause- are simply not supposed to happen.  The median age of menopause onset in the United States is approximately 50.  Until well into the Neolithic era- as recently as 5,000 years ago or so- average lifespan was less than 40.  Bluntly, virtually no one lived to experience menopause.

Average life expectancy among women in the U.S. today- somewhat higher than that of men- is inching ever closer to 80.  Menopause is thus an all but universal experience, and women spend only a little less than half their lives in the post-menopausal condition.

For a span of years up until the late 1990s, the prevailing view was that the marked decline in ovarian hormone (estrogen and progesterone) production that is the hallmark of menopause warranted replacement therapy.  This perspective was likely fostered to some extent by the notion that the levels of estrogen and progesterone in women of so-called “child bearing” age were normal, and data from observational studies clearly and consistently suggesting a reduced risk of chronic diseases with hormone therapy (HT).

But the operative word there proved to be “observational.”  In observational trials, participants choose their own interventions, and researchers simply monitor to see what happens to whom.  In such trials, women who chose HT had less premature death and chronic disease- cardiovascular disease in particular.

The liability of such studies is that people who choose A may differ in a whole variety of ways from people who choose B.  In the case of HT, it may have been, for instance, that the more health-conscious, or highly educated, or affluent women were more likely to make use of replacement hormones.

That, in fact, is just what we learned in the late 1990s when the first major randomized trial of HT, the HERS study (Heart and Estrogen/progestin Replacement Study), was published.  In a randomized trial, health-consciousness, education, and affluence are all neutralized, because luck of the draw, not personal preference, determines who gets what treatment.  In the HERS trial, overall mortality and chronic disease rates- especially certain cancers- were higher in the women using hormone replacement!

And then HT, already staggering from the blow delivered by the HERS data based on roughly 3,000 women, was laid low by results from the Women’s Health Initiative (WHI) Study in which over 15,000 women received both estrogen and progesterone replacement, and roughly 11,000 women who had undergone hysterectomy received replacement estrogen only.  In both cases, overall disease and death rates were slightly higher with hormone replacement than with placebo.

There were two major reactions to these trials.  One was that hormone replacement therapy at menopause went from hero to villain in short order.  Not only did the medical community do an about face, but spurred by alarming headlines that emphasized an increased risk of death, women, en masse, abandoned hormone therapy in fear.  Speaking as a clinician, I can attest that for some years now, it has been challenging to talk a woman into HT even when it was clearly warranted.

The second major impact of these trials was to accentuate the distinction between observational and randomized intervention study data.  The perception that took hold was that observational data could not be trusted, and data from randomized controlled trials were something akin to gospel.

But there has always been a very important limitation to even the largest and best-run clinical trials: they answer only the specific questions posed, and they provide answers for only people just like participants in the trial.  The HT trials used only one kind of hormone replacement (the estrogen was Premarin, and the progesterone was Provera); they enrolled only women willing to be randomly assigned to hormones or placebo; and they enrolled women who were ten years out from the onset of menopause on average.

These limitations are all addressed in a position statement on hormone replacement therapy just issued by the North American Menopause Society.  Also clarified in this detailed and thorough report is the fact that even when HT was meaningfully associated with increased risks, those risks were generally very small.  As an example, one of the biggest concerns about HT use for more than 5 years was increased breast cancer risk.  The data suggest HT use for 5 years by 10,000 women would result in a net increase of 8 breast cancers.  This is by no means trivial, especially if you or someone you love is one of those eight.  But it is a small absolute risk.

Among the more salient messages in the new position statement is that HT use right at menopause is very different from use following a ten-year delay.  HT at menopause decisively improves quality of life measures, forestalls osteoporosis, and appears to reduce risks of cardiovascular disease and diabetes.  Lost in the disappointment and drama of the HERS and WHI headlines was this: HT use at menopause is associated with a slight reduction in all-cause mortality.

Neither the North American Menopause Society, nor I, is advocating for routine use of HT at menopause.  Rather, we agree that HT was never the panacea it seemed before the HERS and WHI trials, nor the poison it may have seemed after.  As is true for almost all matters medical, there is potential here for both benefit, and harm, and thus the need for informed and individualized decisions.

Beware, in general, the drama of the medical headline that promises a breakthrough or warns of dire peril.  The truth is far more often on middle ground.

That is clearly where the truth about HT resides.  Used thoughtfully, it will help some of the people some of the time.  The new report can help inform the conversation you should have with your doctor to determine whether or not you are one of them.


Dr. David L. Katz;