Correcting the bs on the abortion-breast cancer link
This is a response to Dienne whose terribly misinformed post on
abortion appeared on Eric Zorn's blog. The posting thread was whether
the lieutenant governor position should be retained; Eric was shocked
to agree with me that it should be. But commentator Dienne wanted to
change the subject to abortion. What irked Dienne was that someone had
the gall to lob a compliment my way: "Also, whether you agree with him
or not he's often backing himself up with a solid argument."
Dienne's response was:
Yeah,
like when he uses questionable studies quoted in other articles without
even looking at the study himself. For instance, Byrne likes to beat
the "abortion causes breast cancer" drum a lot, and his most recent
column on the subject involved a study that supposedly showed a 40%
increase in breast cancer among women who'd had abortions. Except that
Byrne neglected to mention that the particular form of breast cancer
referenced in the study is so rare that that 40% was actually
statistically insignificant. Byrne also neglected to mention a finding
in that same study that was statistically significant - that women who
gave birth before the age of 18 had a significantly lower incidence of
breast cancer than women who gave birth later. And the more children
that a woman had, the lower the incidence. So if Byrne is really
concerned about breast cancer, and not just seeking a red herring
against abortion, he should be encouraging young girls to start popping
out babies early and often.
Actually, Dienne, I did download the study and examined it closely.
The kind of "rare" cancer that you so dismissively wave away is Triple Negative Breast Cancer, which is harder to diagnose than others. It is a "clinically challenging" (harder to treat) type that more frequently occurs in younger and African-American women. More aggressive than other types, TNBC constitutes 10 percent to 20 percent of all breast cancers.
The authors call the sample they were using "large population-based." The finding that the odds ratio of (gravid) women having had an abortion and contracting breast cancer is 1.4 is not insignificant. I guess the fact that 187 of the 897 cancer patients studied had this form of cancer didn't concern you.
Dienne, you might not have noticed this, but the 2009 study in its text referred to abortion as a "known and suspected risk factor," like family history. The very table that you referred to ("Multivariate adjusted case-controlled odds ratios and 95% CIs for all breast cancers case, triple-negative and non-triple negative case, in relation to known and suspected risk factors....") lists abortion among those risk factors.
The study, published in the April, 2009 issue of the journal Cancer Epidemiology, Biomarkers and Prevention, was based on the same data set from an earlier study ("Risk of Breast Cancer Among Young Women: Relationship to Induced Abortion," Janet R. Daling, et al) in which the authors said: "Our data support the hypothesis that an induced abortion can adversely influence a woman's subsequent risk of breast cancer." It also states that abortion as a risk factor is "consistent with the effects observed in previous studies on younger women." It adds that the "results across all epidemiologic studies" are inconsistent and, as must studies do, called for more studies.
And there have been more, and the over-all result is not "settled science," even though your bullet points might say so. One of the sources that deniers of the possibility of an abortion-breast cancer (ABC) link often cite is a 2003 workshop on "early reproductive events and breast cancer" sponsored by the U.S. National Cancer Institute. One panel flatly stated that no ABC link has been established and dismissed studies that contradicted its conclusion as "unreliable." Since then, the NCI has followed that conclusion without qualification.
But, here's the interesting thing. Louise A. Briton of the NCI was the chief organizer of the 2003 workshop that spit out the unqualified denial of an ABC link. Yet, Louise A. Briton, a co-author of the latest study, reaffirms what she and her other co-authors stated in earlier papers: that the ABC link is "consistent" with other studies that listed
More from around the web
2 Comments
Wendy C said:
Interesting that the government webpage on this topic states no relationship between abortions or miscarriages exists.
"They concluded that having an abortion or miscarriage does not increase a woman’s subsequent risk of developing breast cancer."
http://www.cancer.gov/cancertopics/factsheet/risk/abortion-miscarriage
Karen Malec said:
The following letter to President Obama and party leaders in Congress shows how the National Cancer Institute conned women with its 2003 workshop which denied that abortion raises breast cancer risk:
http://www.abortionbreastcancer.com/download/LetterToObama.pdf
At the workshop, Louise Brinton and the National Cancer Institute said retrospective studies relying on women's reports of abortion histories, instead of medical records of abortions, were flawed because (they claimed) more breast cancer patients accurately report their abortions than do healthy women. Therefore, (they claimed) one would expect to find more abortions among breast cancer patients in a study relying on women's self-reports of abortion histories. This type of flaw is called reporting bias.
All but one of the studies reporting risk increases among women with abortions were retrospective. Two of the studies which Brinton and the NCI claimed were flawed because of the problem of reporting bias were studies conducted by the Daling team in 1994 and 1996. (Brinton was a co-author in the 1996 study.)
In 2007, triple-negative breast cancer was first described in medical literature. So the Daling team decided to use new technology to test for it when they conducted the 2009 study, Dolle et al. THEY USED THE ALLEGEDLY FLAWED 1990s DATA TO CONDUCT THEIR STUDY. They had saved 897 samples of cancerous breast tissue from study subjects participating in the 1990s studies. They tested each sample for triple-negative breast cancer and then matched the samples with reproductive history, including abortion histories. IN OTHER WORDS, THEY RELIED ON WOMEN'S SELF-REPORTS OF ABORTION HISTORIES.
Leave a Comment?
What your comment will look like:
said: