Junk science is finding more victims on the Prairies. A cottage industry—the harvesting of urine from pregnant mares, used to manufacture hormones for therapeutic use by women—has lost two-fifths of its suppliers. A fear campaign based less on facts than perceptions has scared away many customers.
This misuse of statistical research confers no kindness on the women who need the estrogen. The miseries of menopause cover a range from extreme flushes and sweats to a high risk of osteoporosis and resultant bone fractures which often lead to death. Besides moderating these problems, hormone therapy gave women back their self-esteem, reduced the risk of heart attacks in younger women, countered depression and improved sexuality. Because of this, those who use Hormone Replacement Therapy (HRT) consider it a Godsend. They feel better and enjoy life more.
A primary source for these replacement hormones, pregnant mares urine (PMU), opened up a rural opportunity and farms quickly sprang up everywhere to supply the product. This started in western Canada around 1968 and peaked a few years later, at just over 400 ranches specializing in PMU. But recents events have tarnished the gratitude of the agricultural economy and of women for this benevolent gift.
As night follows day, this niche market became the target of scaremongers, fueled by the high priests of the post-modernist, non-science called “epidemiology.” The first blow fell in July 2002 with the release of a “study” conducted by the Women’s Health Initiative which was published in the Journal of the American Medical Association. It ignited a firestorm of scary headlines around the world claiming that HRT was linked to breast cancer, strokes and heart attacks.
A look at the details of the research reveals the sloppiness. The study composed two control groups of 8,304 each, one taking estrogen and one not. The numbers in each that contract these other conditions is then compared, and relative risk factors calculated. In this case, preliminary data showed the relative risk factor for breast cancer, for example, was in the range of 1.00, which equals no risk at all, to 1.59. That is not in itself cause for alarm. You need to see a risk factor of at least 3.0 to 4.0 before you should even worry that you might be on to something.
But the percentage for the slight increase in relative risk, the one the media prefer to use, is then multiplied across the population to create what are called “Trojan” numbers, after the legendary Trojan horse. The one number smuggled into a larger one makes an inconsequential result look significant. So we hear that out of 50 million menopausal women, six million take estrogen, and that they therefore face a 41 percent higher risk for breast cancer, 29 percent for heart attacks and 22 percent for cardiovascular disease.
That gives them “virtual,” as opposed to real, body counts. In reality, the supposed excess number of women in the estrogen group that had heart attacks was all of seven out of 8,304. That’s not very frightening, so neither the scientists nor the media transmission belts bother to discuss that number. Furthermore, the data suggested a slight protective effect when all forms of cancer were combined; they don’t talk much about that.
The study also achieved a new low in honesty. Supposed to be “double-blind,” so that neither the researchers nor the participants knew who was taking the treatment and who was taking the placebo, the research abandoned that anonymity. Claiming that the statistics for breast cancer were so high that they could not morally continue, the scientists unblinded the study and put a stop to it early. That, of course, generated more scary headlines, even though the action genuinely outraged the scientific community at large.
The blood was in the water now. Junk scientists had a new headline-generating tool, which they applied ruthlessly, drawing on long experience which teaches that media fads produce more grant money than a reputation for rigorous testing. This latest tactic had the added benefit of reducing the workload in the drive for truth. Soon all sorts of other studies involving HRT stopped early on similar pointless grounds. The National Heart, Lung and Blood institute cancelled one, a Scandinavian one was axed and the next one abandoned because of the so-called risk of stroke. On and on it went.
By the beginning of 2004, more than half of the woman previously taking the therapy abandoned it on the strength of these “studies.” Ironically in April of that year the farce came full circle when another bunch of epidemiologists completely exonerated HRT. That, of course, generated fewer headlines, mostly of the “who-can-you-believe?” variety.
The damage was done. At the beginning of May, 2005, a large number of PMU operations under the Wyeth Organics banner were mothballed. In Manitoba, 24 out of 79 where shut down. In total Wyeth closed 49 of its 122 operations, including all 19 in Alberta. All thanks to sloppy statistics masquerading as science.
The farmers affected now must scramble in tough times to replace that revenue and the well-being of millions of women has been compromised.. To quote Ernest Rutherford, a Nobel Prize winner in Chemistry, “If your experiment needs statistics, you ought to have done a better experiment.” We need to do some serious thinking about our current use of epidemiology.