Only Fuzzy Math Can Make the ABC Link Disappear
By Joel Brind, Ph.D.
To borrow from Yogi Berra, it looks like "déja-vu all over again." A supposedly definitive study of immense statistical power, published in a top medical journal, has once again allegedly proven there is no link between an induced abortion and an increase in breast cancer (the ABC link, for short).
This time, we're told, it was "a collaborative re-analysis of data from 53 epidemiological studies, including 83,000 women with breast cancer from 16 countries." It was authored by a prestigious group of Oxford researchers and published in March in the Lancet, one of the most prominent medical journals in the world.
In a pre-publication media blitz, lead author Valerie Beral wasted no time. She told the Associated Press, "The totality of the worldwide epidemiological evidence indicates that pregnancies ended by induced abortion do not have adverse effects on women's subsequent risk of developing breast cancer." Beral also told the Atlanta Journal-Constitution, "Scientifically, this really is a full analysis of the current data," suggesting a truly comprehensive review of the data.
In truth, the Beral study is seriously flawed. Its conclusions do not stand up to modest scrutiny, let alone close scrutiny. For starters, the claim that this is a "full analysis" is entirely false and seriously misleading.
With 41 studies published since 1957 that have data on the question of induced abortion and its relationship to breast cancer, one would think that Beral et al. started with these 41 studies, and then added 12 studies with previously unpublished data. Not so.
They arrived at the aforementioned figure of 53 studies mostly by deleting completely acceptable studies and then adding studies as yet unpublished and thus not peer-reviewed.
1) For acceptable reasons they threw out two studies: "specific information on whether pregnancies ended as spontaneous or induced abortions had not been recorded systematically for women with breast cancer and a comparison group."
2) But they also threw out 11 perfectly good studies for reasons that ranged from misleading to silly. For example, some were removed because the original authors couldn't be found. Following that logic, we ought to assume that the sun revolves around the earth. We've looked all around the University of Krakow in Poland, and since we cannot locate Dr. Copernicus, we should throw out his data which proved otherwise.
3) Finally, they did not even mention data from four studies, even though these studies had been previously published as abstracts or included in other reviews.
In summary, by deleting or ignoring 17 studies, the initial figure of 41 was reduced to only 24 studies. They then added data from 28 unpublished studies and, suddenly, there are now 52 studies (not 53, by the way).
The fact that the majority of these latter studies have not stood the test of peer review is troubling enough. But a closer look at the excluded studies is even more revealing.
Let's return to the 41 previously published studies. Twenty-nine show increased risk of breast cancer among women who had chosen abortion. (Epidemiologists call this a "positive association.")
Sixteen were "statistically significant," which means there is at least a 95% certainty that the results cannot be explained by chance. Beral excludes 10 of these for reasons that simply are not supportable.
In fact, if we average all of the 15 studies Beral excluded for unscientific reasons, they show an average breast cancer risk increase of 80% among women who had chosen abortion. By selectively eliminating studies that show an ABC link, Beral is able to find there is no significant effect of abortion on breast cancer risk.
Having thrown out studies that contradict her thesis, she then includes studies that are plagued with serious deficiencies.
Beral divided the included studies into two types: those which used retrospective methods of data collection (i.e., interviews of breast cancer patients versus women who had not had breast cancer), and those which used prospective methods (i.e., medical records taken long before breast cancer diagnosis).
Beral told the Washington Post that retrospective data-based studies are thought to be less reliable. Women with breast cancer "are more likely than healthy women to reveal they had an abortion, leading to the conclusion that there are more abortions among this group," she said.
This "reporting bias" or "response bias" is a key to Beral's argument. But something you'd never know from the universally glowing media portraits is that this hypothesis has been disproved over and over again in studies as far flung as Japan, the United States, and Greece! Moreover, the authors of the only study to claim direct evidence for such a reporting/response bias based that on an assumption which they have since publicly retracted. They had assumed that breast cancer patients "overreported" abortions (i.e., reported abortions that had never taken place).
That brings up another serious flaw in the Beral study: the exclusion of any published studies that critiqued the studies she found which supported her conclusion. To take just one example, when a group I was part of subjected a 1990 study Beral used to rigorous mathematical proof, we found that this study on Norwegian women actually did find a link between induced abortion and breast cancer, contrary to what the authors insisted.
It is understandable that many would place credence in studies that come from such high places as the Lancet or the New England Journal of Medicine or the National Cancer Institute.
However, as one who has been doing battle over the ABC link in medical and scientific journals and in other public settings for over a decade, nothing is clearer than that organized science and medicine have systematically denied the linkage.
Consider this: Even if one were to reject the abundant evidence that choosing to abort increases a woman's risk of breast cancer, it is undisputed - - even by Beral herself - - that a full-term pregnancy lowers a woman's long-term risk of breast cancer. Obviously, this protection is not afforded when a pregnancy ends in an induced abortion.
According to the journal Nature, "The authors hope to lay to rest a highly charged debate." It has, of course, done nothing of the sort. It may remain politically incorrect in many circles to warn women that abortion increases their chances of contracting breast cancer, but we must do so nonetheless.