“Hormone replacement linked to dementia” dominated health news coverage this week. More than just the most recent scary headline, the story is yet another bizarre attempt to discredit hormone replacement therapy (search) (HRT) for post-menopausal women.

Women given an estrogen-progestin form of HRT reportedly had twice the rate of “probable dementia” — mostly Alzheimer’s disease (search) — as women given placebos, according to a study in the Journal of the American Medical Association (May 28).

“It’s another nail in the coffin” for the use of hormones during and after menopause (search), a St. Louis gynecologist told The Associated Press.

A look at the study, though, reveals a much less scary story. After interviewing the lead researcher, I wonder why these unexpected results were even published.

Assuming for argument’s sake that the results are valid and generally applicable to post-menopausal women — big assumptions — the purported increase in risk of dementia is small — only an additional 23 cases of dementia per 10,000 HRT users aged 65 years or older per year.

This is probably not a meaningful — or even measurable — increase in risk as the prevalence of dementia is estimated to be more than 10 percent for 65-year-olds, increasing to 50 percent for those over 85.

The alleged small increase in risk also means HRT wouldn’t cause dementia in the vast majority of users. Since dementia is unpredictable, bypassing the proven benefits of HRT based on a slight-to-imaginary reduction in risk doesn’t make sense.

Regardless of whether the study’s results are correct, their publication and accompanying publicity campaign are premature. Lead author, Wake Forest University’s Sally Shumaker, reluctantly acknowledged during an interview that the study is incomplete.

There are more than 30 metabolic, structural and infectious conditions that may lead to dementia. The potential genetic and lifestyle factors that might cause those conditions are many more and largely unknown.

But none of these factors or conditions were explicitly considered by the researchers in their study. HRT can’t be blamed for causing dementia when other causes haven’t been ruled out.

Shumaker said the study design — a randomized, double-blind, placebo-controlled clinical trial — made consideration of other risk factors unnecessary.

Random assignment of treatment and placebo among study subjects, Shumaker said, likely distributed risk factors evenly among the two study groups, thus canceling out the possibility that factors other than HRT are responsible for the observed differences between the groups.

But rigorous research practices — even for a randomized, double-blind, placebo-controlled clinical trial — require consideration of potential confounding factors. This is especially true for a study producing novel results.

Shumaker’s randomization argument is further weakened by the fact that study subjects were volunteers, not a representative group of women. Volunteers may differ in significant ways from nonparticipants. In Shumaker’s study, for example, less than one percent of volunteers were diagnosed with Alzheimer’s. The rate is about 10 percent among the general population of 65-year-olds, though.

Shumaker acknowledged that further study of potential confounding factors will continue and the results will be published in the future.

Shumaker also didn’t provide a reasonably certain biological explanation for why HRT would increase the risk of dementia. A future study will address this issue also.

So what’s the big rush to publish uncertain, incomplete and even trivial results that only scare and confuse women and their physicians?

Media coverage was also disturbing in that headlines typically referred to “hormone replacement.” But the study only examined one form of HRT, a specific estrogen-progestin regimen. The estrogen alone and even other estrogen-progestin forms of HRT weren’t studied.

HRT has been the target of much bad press over recent years. That may be because it received undeserved good press for several decades.

HRT clearly helps relieve symptoms of menopause, such as “hot flashes.” For many years though, drug companies urged doctors to recommend HRT to post-menopausal women to reduce heart disease and osteoporosis. But the studies supporting these recommendations are under fire.

Studies indicated, for example, that HRT-users had less heart disease than non-users. It turns out the studies’ data were subtly, but significantly, biased. HRT-users tended to be better-educated and wealthier and less likely to have heart disease in the first place.

The junk science pendulum on HRT has now swung the other way with researchers seemingly stumbling over themselves to link HRT with various health problems, such as breast cancer.

Since the HRT controversy can be confusing, women should consult their physicians on the benefits and risks of HRT. Just make sure your physician isn't as confused as that St. Louis gynecologist.

Steven Milloy is the publisher of JunkScience.com, an adjunct scholar at the Cato Institute and the author of Junk Science Judo: Self-defense Against Health Scares and Scams (Cato Institute, 2001).