Preventive Medicine Column

Dr. David L. Katz

Once upon a time, the first human woman outlived the biological limits of her ovarian hormone production, and menopause was born. None can give the date for this event, but we can be confident it was a very, very, very long time ago.

Medical menopause, however, was born much more recently, arguably in the early 1940s when the FDA approved estrogen replacement for the treatment of menopause and its related symptoms.

Author Jared Diamond has shrewdly pointed out that invention can be the mother of necessity.  When we invented medical menopause, the necessity of treatments that ensued grew into a multi-billion-dollar industry. For a span of decades, science fostered this trend with observational studies suggesting the many benefits of hormone replacement.

The practice that came to prevail, even though many true authorities on the topic were dubious all the while, was the use of Prem/Pro, a combination of Premarin (conjugated equine estrogen, or, to spell it right out: estrogen derived from the urine of pregnant female horses), and Provera (medroxyprogesterone acetate, a synthetic progesterone many times more potent in its action than the native human variety).

The popular view for those recent decades, propagated in all the ways we might imagine by both our limited knowledge, and large sums of money changing hands, was that hormone replacement was good, and that Prem/Pro was good hormone replacement.

One of the dominant themes in modern medicine is “evidence-based” practice. As attention, over the past 30 years or so, focused more on standards of evidence, misgivings began to arise about the observational data on which enthusiasm for HRT was mostly based. Seeing X and Y happen together in a large group of people does not reliably mean that X is causing Y. It may be that everyone who does X happens to have Z, and Z causes Y. As an example, people who don’t have televisions are more prone to tuberculosis — but not because TVs prevent TB! Rather, not having a television is a proxy for poverty, and poverty in turn is associated with a number of the factors that truly do increase TB risk.

Early misgivings about the quality of evidence for HRT were greatly compounded by publication of the HERS trial findings in 1998.  HERS was the first significant randomized, placebo-controlled trial of hormone replacement therapy, and it suggested an overall lack of benefit for cardiovascular disease prevention. Nearly 3,000 women were enrolled.

Then came publication of the Women’s Health Initiative (WHI) results in 2002.  This randomized trial in nearly 20,000 women showed modest net harm from hormone replacement therapy, and advised directly against the practice.

Our latest news, again courtesy of the WHI, is a paper just published in JAMA, reporting the long-term experience of women in the trial who had undergone a hysterectomy and received only estrogen.  This treatment produced a clear reduction in breast cancer risk, without meaningful harms in other areas. The study is rekindling some of that love that once prevailed, and then was lost, for hormone replacement therapy.

Looking at all this in hindsight, we never had as much cause to love or hate HRT as our behavior during any interval suggested. Our enthusiasm for hormone replacement, and certainly for Prem/Pro, got well ahead of the data. Then, our renunciation of the practice was far more adamant and generalized than the data warranted. Both the HERS and WHI trials used the same kind of hormones, and both enrolled women a long time after menopause.   And, at worst, they suggested that benefits and harms were nearly a toss-up.

Studies since suggest that the benefits of hormone replacement appear to be most robust when the practice is applied as soon after menopause as possible. The benefits are greater when women are carefully selected; those with increased risk for blood clots should not participate, for instance. And there are many options — and most would say many far better options- than Prem/Pro, including other estrogen/progesterone combinations, estrogen alone, bio-identical hormones, and the synthetic estrogen-like drugs called SERMs, of which raloxifene (Evista) is an example.

There are two take-away messages. The first is that hormone replacement therapy at menopause is not good for all, nor bad for all. It clearly can confer benefit, but it depends how it is done, when, and for whom. When the right hormone preparations are thoughtfully applied in the right women for the right reasons at the right time, considerable benefit can come of it. There is potential for harm when the wrong preparations are thoughtlessly used for the wrong reasons in the wrong women. You and a doctor you can really talk to should sort this out together.

The other message is that science is incremental and evolutionary, rarely revolutionary. Evidence accumulates over time, and the weight of evidence tips slowly, but inexorably, toward the truth. When science becomes a teeter-totter of sequentially opposing truths, we have lost our balance, and our way.

That has been the history to date of hormone replacement therapy: over-reactions to the half-truths that any given trial provided.   We have the customary choice: repeat the follies of history, or learn from them.




Dr. David L. Katz;