By Joel Brind, Ph.D.
In Part II of this series of articles on the link between abortion and breast cancer, I described how the "reporting bias" or “response bias” argument has been systematically used by the most prominent medical research authorities in the western world to deny the reality of the ABC link. It is as hugely important to critics as it is demonstrably false.
The argument goes as follows:
When you compare the reproductive histories of breast cancer patients to healthy women (these are called "retrospective" data-based studies), the cancer patients will be more truthful about their past abortions than healthy women. If true, this would result in breast cancer falsely appearing to be more common among post-abortive than non-post-abortive women, thus incorrectly indicating an increased risk with abortion.
The argument continues:
To prove whether there is a reporting bias that results in an ABC link, studies are needed which do not rely on women's recall and reporting of their reproductive histories. Rather those studies which rely on medical records of abortion and/or which follow cohorts of women after their abortions are sure to be free of possible of reporting bias.
Epidemiologists refer to this latter type of data as prospective, i.e., data on abortion exposure are gathered before it is known who will end up being a breast cancer patient and who will not. Such studies as these, it is contended, are far superior to those which rely on retrospective data.
During the period of 1997 to 2008 about a dozen prospective data-based studies emerged from some of the most high profile institutions and journals. All of them showed no ABC link, fueling the official denial of the link far and wide, a wall of denial still essentially in place.
But just because a study is based on prospective data does not necessarily make it scientifically sound. Those dozen studies are largely provably false. In this article, I will provide a taste of what has passed for science in official circles -- scientific merit being adjudged strictly according to whether a study supports the "safe abortion" narrative.
It is more than interesting that the first solid prospective data-based study had actually been published back in 1989. Based on fetal death certificates filed for abortions done in New York State, the study found the politically incorrect results that women who'd had an abortion nearly doubled their breast cancer risk.
Since this does not fit the "safe abortion" narrative, the authors of the recent spate of prospective studies have systematically misrepresented the New York study. Some ignore it altogether, some mischaracterize it as being subject to reporting bias, and some flat-out claim no such study exists!
In 1997, a prospective study on Danish women (funded by the U.S. Department of Defense) was published in the top American medical journal, the New England Journal of Medicine, and found precisely zero effect of abortion on breast cancer risk. Its timing is significant because that was exactly three months after my group's "comprehensive review and meta-analysis" gained substantial media traction. We found a 30 percent overall increased breast cancer risk among post-abortive women worldwide.
The Danish study was widely touted as definitive because it was so big: based on 1.5 million women, representing an aggregate of over 370,000 abortions and over 10,000 cases of breast cancer.
But upon closer examination, the Danish study reads like a manual of how to distort the data in order to achieve a desired outcome. Among the more egregious examples was the use of breast cancer records dating back to 1968, but including abortions only from 1973 onwards. This is a violation of the most basic scientific rule: cause must precede effect! But apparently, not when you are trying to prove abortion is safe for women.
Yet more effective at covering up the truth was the omission of legal abortion records dating back to 1940, when abortion was legalized in Denmark, and then writing as if the date of legalization were in 1973.
Then there is the trick that has been used in most of the prospective studies purporting to show that there is no ABC link: limiting the follow-up period after abortion. You see, one of the strengths of retrospective studies is that, typically, older women are interviewed about their reproductive histories years or even decades after their abortions.
Why is that hugely important? Because it often takes 10-20 years for cancer to develop after an abortion.
But in the 1997 Danish study, fully one fourth of the subjects were still under the age of 25 at the end of the study! Of course, these young women had lots of abortions, but almost no breast cancer.
Not only was this lack of follow-up time used as a device to eliminate the evidence of increased breast cancer risk with abortion in prospective study after study, a very diverse bag of tricks can also be found at work.
For example, one of the many ABC cover-up studies published by the Oxford group of epidemiologists, published in 2001, showed no ABC link. But the medical records they used had clearly missed over 90 percent of the abortions performed on women during the study period. Buried deep in the paper's discussion section was the following curious revelation: "Our data on abortions are substantially incomplete ..." That alone should have convinced the journal to discard the study on the spot!
Then there is the "dagger under the table" technique. The "dagger" is the printed dagger that signifies a footnote. There was a pivotal footnote that appeared under the key data table in the prestigious Harvard Nurses Study II, published in 2007.
The daggered footnote signifies, in that case, that a key statistical adjustment to the data on abortion was omitted from the overall result. Hence, the authors could (misleadingly) report no significant ABC link in the overall conclusion, rather than the significant risk increase of at least 10 percent that they had actually observed.
Finally, another key trick now used by just about all ABC-link deniers is the complete omission of cases of breast cancer in situ, the type of breast cancer which appears earliest -- the incidence of which type has also dramatically risen in the US by some 400 percent in recent decades!
At this point one may seriously ask: How on earth can the medical research establishment get away with such a widespread, systematic cover-up of a serious health risk from an elective procedure?
Of course there is the media complicity, but there is also the ability of a small cohort of well-placed individuals in key positions at agencies such as the U.S. National Cancer Institute (NCI), which funds most of the cancer research in the U.S. Scientists who don't toe the party line can lose their research grants.
In the final installment of this series, I'll describe the biology that underlies the ABC-link.
Joel Brind, Ph.D. is a professor of biology and endocrinology at Baruch College of the City University of New York and co-founder of the Breast Cancer Prevention Institute. This article is reprinted with permission from NRL News Today.