Quoted from Living Downstream:

From its peak in the 1990s, the breast cancer rate began to ebb. It stabilized in 2001-2003 and then dropped noticeably in 2003 and stabilized again in 2004. While this disease remains, by far, the most prevalent cancer among U.S. women, and while U.S. women still have the highest rates in the world, we can now say with some conviction that the constellation of factors contributing to breast cancer—whatever they are—seem to be receding. The serial killer known as breast cancer is claiming proportionally fewer victims than a decade ago.

But why?

There are four possible explanations, and they all have some merit, and they all have some defects. About them all, I remain agnostic. At this writing, the most popular hypothesis is that the ongoing decline in breast cancer is attributable to the widespread collective decision by postmenopausal women to stop taking hormone-replacement drugs in 2002. In that year, the New England Journal of Medicine reported trial results from the Women’s Health Initiative that indicated likely excesses of breast cancer and undeniable excesses of heart problems among women taking estrogen and progestin hormones to ease the side effects of menopause. These findings made headlines around the nation and triggered a repudiation of pharmaceutical approaches to menopause. Among California women alone, the use of estrogen-progestin replacement drugs plunged by 68 percent between 2001 and 2003.


The second hypothesis—which is not mutually exclusive with the first—is that declining rates of mammography screening explain the declining numbers of breast cancer diagnoses.


The third possible explanation for falling rates of breast cancer is disproportional underreporting. It’s possible that the reports of recent cases, still dribbling in, are more likely than reports of older cases to be missing from the analysis.


The fourth possibility is that declining breast cancer rates are caused by declining exposures to causative agents other than (or in addition to) pharmaceutical estrogen. Indeed, biomonitoring data show blood levels of some hormonally active suspected breast carcinogens, such as DDT and PCBs, are now finally falling, many years after their ban. (More on this in Chapters Five and Twelve.) This hypothesis would help explain why the beginning of the recent recession of breast cancer predates the swift and dramatic change in hormone replacement drug use. Supporting this idea is a recent Spanish study that demonstrated an association between breast cancer risk and body burden of all estrogenic chemical contaminants, excluding natural hormones.


For my readers in the present epoch, I urge the following approach: the evidence today says that lowering population-wide exposures to estrogen prevents breast cancer. The abandonment of hormone replacement therapy as a treatment for menopause and the attendant drop in breast cancer incidence is a natural human experiment that serves as Exhibit A. On the grounds that we are morally obligated to act on the basis of the best information available to us now, let’s initiate another natural human experiment: screen chemicals in commerce for their ability to act like estrogen and systematically phase out the estrogen mimics. Then see what happens to the breast cancer rate. Go on to publish papers with titles such as, “The Impact on Breast Cancer Incidence of Eliminating Hormonally Active Agents from Agriculture and Consumer Products.”

Steingraber, Sandra (2010-03-23). Living Downstream: An Ecologist’s Personal Investigation of Cancer and the Environment (p. 40-42). Da Capo Press. Kindle Edition.