Hormone therapy: Not high risk, not high need

The news media last year were full of reports growing out of a major study of hormone therapy or women near menopause, called the Women’s Health Initiative. A part of the study was discontinued early because the evidence showed a small but definite extra risk of heart disease and cancer associated with one particular mix of hormones. These results came as a surprise to most physicians, who had assumed based on earlier trials that hormones would lower heart disease risk. The news reports sent many women running to their doctors worried about what to do and what this all meant.

Now that the dust has settled a bit, it’s easier to make sense of what we do and don’t know about hormones for post-menopausal women. In this column I’ll summarize why the actual risks of taking hormones is lower than most people thought when they first heard the news about the study. So if a woman had a good reason to wish to take hormones, the extra risk might actually not be so bad. Next time I’ll look at the alternatives to using the particular mix of hormones that was implicated in the study to show that even though the risk is low, many women will decide not to run that risk because they can do something else instead.

The portion of the study that was terminated in May 2002 looked at women taking a particular combination of the hormones estrogen and progesterone, 0.625 milligrams daily of conjugated estrogen and 2.5 milligrams daily of methoxyprogesterone (equivalent to a popular product, Prempro .625/2.5). The study found that compared to women receiving no hormones, these women had an increased risk of developing heart attack, stroke, or breast cancer. A significant fact overlooked by most news reports was that the actual death rate for women taking the hormones was no higher than for the non-hormone women, showing that these outcomes, while obviously bad, were at least not uniformly fatal.

When taking a medicine places us at higher risk for some bad thing happening, there are several ways we can report the identical statistics. The way the news reports tended report the risk was to compare the two groups of women (relative risk). So one might say, for instance, that women taking a certain medication had a 20 percent higher risk of stroke than women not taking the medicine.

Relative risk figures can be misleading because they tell us nothing about the actual size of the risk. Here are two cases in which the relative increased risk is 20 percent. In one case, women not taking the medicine have a 50 percent risk of having a stroke, and women who do take the medicine have a 60 percent risk. In the second case, women not taking the medicine have a 1 percent risk of stroke and those who do take it have a 1.2 percent risk. The average person, hearing these figures, would be much more impressed with the seriousness of the extra risk produced by the medicine in the first case compared to the second case.

Because the relative risks can be so misleading and uninformative, leaders in medical statistics today suggest that we should report these matters as follows: how many women would have to take this medicine, for how long, before one woman suffered the bad outcome?

If you look at the data that way, here is what you find: 286 women would have to take the estrogen/progestin combination for five years before one woman would have an extra heart attack from taking the medicine. For stroke, it would be 250 women for five years. For breast cancer, it would also be 250 women.

No one would want to discount how serious it is for you, if you are that one woman who ends up with a heart attack, stroke, or breast cancer. Yet many women, hearing these figures, wonder what all the fuss was about in the news media last summer. These women might well decide that if the hormone treatment did something good for them, the very slight extra risk was well within acceptable bounds. So next time we have to ask: What do hormones do for you that other treatment cannot do?

Note: In this and the next column I’m grateful to Dr. Mindy Smith, who teaches at the Sparrow/MSU Family Practice Residency. Smith prepared a presentation for a physicians’ course we both help to teach, which provides an excellent summary of hormone therapy and its pros and cons.

Dr. Howard Brody, M.D., teaches family practice and medical ethics at Michigan State University. You can reach him at brody@msu.edu.

Care to respond? Send letters to letters@lansingcitypulse.com. View our Letters policy.