More debate on the abortion/breast cancer link

The letter below was published in the Tribune's Voice of the People in response to my Jan. 31 column in which I deplored the politicalization of science, including the political attacks made on scientists who see a connection between breast cancer and inducted abortions. I had mentioned one study that discovered a statistical link between the two, but the c0-author of the study took umbrage. Here is her letter, and below it is a response from Joel Brind, Professor of biology and endocrinology at Baruch College of the City University of NY and a leading advocate for women who may have suffered from breast cancer because of an induced abortion.

Science and politics

I appreciated columnist Dennis Byrne's "Save us from the politics of science; When ideology drives discussion, we're in trouble" (Commentary, Jan. 31), and his commitment to uncovering the tangled web of science and politics.

While I wholeheartedly endorse Byrne's aims, I am saddened and dismayed to see that he committed precisely the same error of tying together science and politics to serve a specific agenda by perpetuating the misrepresentation of my American University of Armenia student's work in the following statement: "The group (Coalition on Abortion/BreastCancer) recently pointed to a study co-authored by researchers from the Johns Hopkins School of Public Health and the University of Pennsylvania School of Nursing that reported a nearly tripled breast cancer risk for post-abortive women."

The research was not aimed at linking elective abortion to the incidence of breast cancer, and we specifically note that our finding of an association between breast cancer and induced abortion was not robust.

The paper, in fact, reports a study of possible epidemiological connections between breast cancer and Type 2 diabetes in the Armenian population.

Discussing the work I sponsored and supported a student to complete and publish, without reading and understanding the report, only adds to myth and misinformation. I hope that, in the future, Byrne will consider reading research on which he wishes to comment and contacting the authors of studies like this one before entangling them in a debate that besmirches both science and scientists with the worst of political motives.

— Sarah H. Kagan, professor of gerontological nursing, School of Nursing, clinical nurse specialist, Abramson Cancer Center, University of Pennsylvania,Philadelphia

Now for Brind's response:

Here are a few specific criticisms of her letter:
1) " the research was NOT aimed at linking elective abortion to the incidence of breast cancer". This is entirely irrelevant. Of course the study was primarily interested in finding out if Type II diabetes was a risk factor, but it looked at all the other likely risk factors, and abortion did indeed come out with almost a 3-fold risk increase.
2) " we specifically note that our finding of an association between breast cancer and induced abortion was not robust." Indeed, this is exactly the sort of gratuitous, self-negating statement that is sorely out of place in a scientific paper, especially since an objective description of this finding would be "statistically significant and moderately strong".  Also interesting that this odd statement was not in the original, MPH thesis version of Khachatryan's study (available online via google scholar).

3) In fact, in the original thesis version of the paper, on her short list of 4 conclusions, conclusion #3 reads: “Induced abortions are positively associated with development of breast cancer.” The published paper of which Kagan is a co-author merely calls for "further research", with induced abortion not mentioned at all in the context of conclusions or recommendations.



Leave a comment
  • fb_avatar

    FETUS IS NOT A BABY (GOOGLE THE HUMAN DEVELOPMENT CHART), but a parasite because the classification of the biological relationship that is based on the behavior one organism (fetus) and how it relates to the woman's body:

    as a zygote, it invaded the woman's uterus using its TROPHOBLAST cells, hijacked her immune system by using NEUROKININ B and HCG--- so her body doesn't kill it, steals her nutrients to survive, and causes her harm or potential death.

    "The placenta functions as an immunological barrier between the mother and the fetus, creating an immunologically privileged site. For this purpose, it uses several mechanisms:
    It secretes Neurokinin B containing phosphocholine molecules. This is the same mechanism used by parasitic nematodes to avoid detection by the immune system of their host.[2]"

    " Due to its highly-negative charge, hCG may repel the immune cells of the mother, protecting the fetus during the first trimester. It has also been hypothesized that hCG may be a placental link for the development of local maternal immunotolerance."

    "It is also possible for a symbiotic relationship to exist between two organisms of the same species." -- Gale's Science of Everyday Things.

    just like a parasitic twin ---

    "an animal or plant that lives in or on another (the host) from which it obtains nourishment. The host does not benefit from the association and is often harmed by it"

    pregnancy CAUSES HARM:


  • I tried to look up the paper and can't do so until next month without paying for a subscription to the journal. But I did look at the abstract and noticed that the study's sample size was 302, which is pretty low. This leaves me wondering exactly how many women in the study actually had abortions.

    Sample size matters. If one study of 3000 people shows no link, and a study of 300 shows a link, the study of 3000 people takes precedence. I don't know the literature in this abortion-breast cancer issue, but I'd be willing to wager that there are very many studies out there with a larger sample size. It is important to set the context of this study as one of many other studies. We should pay more attention to the bigger studies, and this is not one of them.

    False positive associations in medical research are very common. "Statistically significant" can be misleading--If one does 100 studies to check for a correlation between two factors, 5 of those studies will show a statistically significant link even when in reality there is none.

Leave a comment

  • Advertisement:
  • Advertisement:
  • ChicagoNow is full of win

    Welcome to ChicagoNow.

    Meet our bloggers,
    post comments, or
    pitch your blog idea.

  • Visit my new website

    I'm a freelance writer, editor and author. I can help you with a wide variety of projects. Check out my new website at

  • Subscribe to The Barbershop

    Enter your email address:

    Delivered by FeedBurner

  • Dennis Byrne’s Facebook Fan Page

  • Like me on Facebook

  • Our National Debt

  • Twitter

  • Tags

  • Recent Comments

  • /Users/dennisby/Desktop/trailer.mp4
  • Advertisement: