People often ask me about the frustrations we run up against in trying to make the average person aware that there really is a link between abortion and breast cancer (ABC link).

We do have the evidence, after all. Going back to 1957, an overwhelming majority of studies (25 out of 31 epidemiological studies worldwide, including 11 out of 12 in the U.S.) have shown a higher risk of breast cancer in women who have chosen abortion.

In response to such inquiries, I never hesitate to point out the bright side. It would be impossible to make the case at all were it not for the fact that there has been at least enough scientific integrity to allow so much evidence to find its way into the published medical literature. This is no small triumph.

Unfortunately, of late, there have been troubling signs that more evidence of the ABC link has managed somehow not to appear in published studies on breast cancer risk factors. This new absence is very significant, for it includes studies coming from prominent researchers publishing in prominent medical journals.

In 1996, I and my colleagues at the Penn State College of Medicine summarized all the ABC link data to date in a " Comprehensive review and meta-analysis," which was published in the British Medical Association's Journal of Epidemiology and Community Health. For purposes of this article, it's worth highlighting that we acknowledged the well known tendency to keep non-significant data in the proverbial "file drawer."

That is, studies which show no significant trends are less likely to be published than studies which do show significant trends. How does that relate to our meta-analysis which summed up data across published studies?

It raises the possibility that most studies might actually not show the ABC link, but they simply have not been published due to the "file drawer" effect. Following this argument, the emergence of a significant ABC link from a meta-analysis of published studies might not, in fact, present a true picture of what research has shown cumulatively.

What is our response? As we also pointed out in our meta- analysis, the real problem is exactly the opposite in studies involving abortion: "there is indirect evidence to suggest the opposite trend in bias" - - that is, the bias is to not publish data which do show the ABC link. I call this the "reverse file drawer effect." As it happened, direct evidence of this bias surfaced shortly after we had submitted our meta-analysis for publication.

Back in 1988, Dr. Tom Rohan et al.published a study which focused on dietary risk factors and breast cancer in Australian women which was published in the American Journal of Epidemiology. As was appropriate, the study also looked at all the "classical" risk factors for breast cancer, including reproductive history. But curiously, no data were shown for abortion.

Fast forward to 1995. A research team headed by French epidemiologist Nadine Andrieu published a paper in the British Journal of Cancer, combining data from several earlier studies, including the Rohan study. Intriguingly, in presenting Rohan et al.'s earlier data, the Andrieu paper revealed for the first time the Australian data on induced abortion.

We learn from Andrieu's team that Rohan et al. had found that the strongest and most significant risk factor for breast cancer in Australian women (stronger even than family history of breast cancer) was induced abortion (as opposed to spontaneous abortion, or miscarriage): a whopping 160% increased risk! For a research team to omit its most significant finding is unheard of-but alas, not when the subject is abortion.

And sad to say, this disturbing trend of keeping evidence of the ABC link out of the published scientific literature is gaining in popularity; witness three recent studies on ethnic women.

The first of these studies, by Le Ming Bu et al., dealt with women in mainland China. It was actually published in abstract form in 1995 in the American Journal of Epidemiology. Consequently, only its most basic findings were presented.

Even so, the findings are rather striking (and statistically significant). If a woman had one abortion, it raised her breast cancer risk by 190%. Two or more abortions raised breast cancer risk by 260%!

Over three years have passed since that abstract appeared. The full paper with far more detail and likely to attract far more attention than a mere abstracts has yet to appear.

But that is certainly not for lack of competence or reputation on the part of the authors or their lack of trying to get it published. It was co-authored by Janet Daling et al. of the top- notch Fred Hutchinson Cancer Research Center in Seattle, Washington (who told me it has already been rejected by three major medical journals!).

The second study, published in 1996 in the Proceedings of the National Science Council, ROC, was written by Dr. Fu-Ming Lai et al. and concerns women in Taiwan. The Lai study focuses almost entirely on reproductive factors and their impact on breast cancer risk.

In fact, the list of reproductive variables examined including abortion -- is impressively thorough. The data tables are not quite as thorough, however. Data on abortion are con conspicuously absent.

That didn't stop the authors from categorically declaring the non-existence of any ABC link, as in: "the number of spontaneous or artificial abortions were not found to be related to an increased risk of breast cancer." To draw such a conclusion without even showing the data -- especially when the data for every other variable are shown is entirely improper.

Not being one to withhold the benefit of the doubt, I sent a polite request to Dr. Lai for the missing data. Seven months later, Dr. Lai has yet to reply.

Closer to home, also in 1996, Dr. A.H. Wu of the University of Southern California and colleagues published the paper "Menstrual and reproductive factors and the risk of breast cancer in - Americans" in the British Journal of Cancer. Considerably larger than the Lai study, it encompassed 492 breast cancer patients of Chinese-, Japanese-, and Filipino-American ancestry who lived in California and Hawaii. Like the Lai study, it also dealt with a very comprehensive list of reproductive factors, including induced and spontaneous abortion.

Concerning these last two variables, the data were not entirely absent as was the case with the Lai study but, rather, selective. In particular, data were given separately for induced and spontaneous abortion but only among the childless women in the study. Since only a small portion of the patients were childless (24%), the 92% increased risk of breast cancer among childless women could be written off as not statistically significant.

As in most study populations, the majority of the patients (76%) did have children. However, for them, the authors lumped together induced and spontaneous abortions, then spin-doctored the following conclusion: "The present study suggests a small increased risk with both spontaneous and induced abortion; neither increase was statistically significant." It is now over seven months since I requested the missing data from Dr. Wu. He also has yet to reply.

For all its selectivity and inadequacies, the Wu study still provided enough data to constitute the 11th study out of 12 in the U.S. (25th out of 31 worldwide) to evidence the ABC link. And now the list includes separate studies on three major American racial groups.

World Conference on Breast Cancer

Abortion and Breast Cancer
Positive Response To Latest Developments Presented At World Conference On Breast Cancer

by Joel Brind, Ph.D.

It was encouraging and somewhat surprising to have an update of our "Comprehensive review and meta-analysis" on induced abortion and breast cancer accepted for presentation at the July [1997] "World Conference on Breast Cancer" in Kingston, Ontario. The conference was organized by veteran pro-abortion activist Bella Abzug in her capacity as president of Women's Environment and Development Organization.

The original paper, which I had written in collaboration with colleagues from Penn State College of Medicine in Hershey, was published in the British Medical Association's Journal of Epidemiology and Community Health last October. My presentation in Ontario centered on several additional studies essentially confirming our original finding of 30% increased risk of breast cancer attributable to abortion. Nothing is more an article of faith for radical feminists such as Bella Abzug than that abortion is indispensable to women's equality.

Therefore I expected that the room would be largely empty -- especially given that there were five other concurrent sessions -- for such an against-the-grain presentation. But to my surprise, not only was the room full, but the attentive audience included Ms. Abzug herself.

As readers may recall, our meta-analysis found that 24 out of the 30 epidemiological studies published worldwide since 1957 have confirmed the abortion-breast cancer (ABC) link. More recent publications supporting and challenging the ABC link were discussed in my presentation in Kingston.

For one, there is the continuing heated debate about the Melbye study from Denmark published in this January's New England Journal of Medicine (NEJM). The U.S. National Cancer Institute and others continue to maintain that it disproved the ABC link. The World Conference provided me the opportunity to point out the serious flaws in the Melbye study, as had been done in an article published in National Right to Life News (5/23/97), and in a letter to the editor, which appeared in the June 19 edition of NEJM. Among other egregious flaws, Melbye et al. had misclassified some 60,000 women in the study as not having had any abortions who actually did have legal abortions on record!

For another, there was the Rookus study on Dutch women published last December in the Journal of the National Cancer Institute (JNCI). Readers may remember how the authors of this well-designed study found a 90% risk increase with abortion, but then proceeded to explain it away on the basis of something called "reporting bias".

This curious argument holds that the ABC link only appears to exist because women with breast cancer are more likely to admit having had abortions than are healthy women. As a result, they say, studies which depend on women reporting their reproductive histories (as most studies do) will show an artificially elevated risk.

The authors of the Dutch study claimed to have evidence of reporting bias. But as was demonstrated both to readers of NRL News (12/10/96) and readers of the April 12 edition of JNCI, by way of another letter to the editor, it was this so-called evidence which had been artificially manufactured.

It was the JNCI editorial that accompanied the Dutch study which the journal's editors (who are also high-ranking NCI officials) used to attack our meta-analysis by name. They claimed we had made "a leap beyond the bounds of inference" for inferring a causal link between abortion and breast cancer.

In our letter rebutting these charges, we pointed out how, rather than relying solely on the statistical link, a causal association was also supported by the biological facts, namely:  

1. "Estrogens are strong growth promoters of normal and most cancerous breast tissue.
2. Most known risk factors for breast cancer are attributable to some form of estrogen overexposure.
3. Maternal estradiol (estrogen) rises 20-fold (2,000%) during the first trimester of a normal pregnancy. [Estrogen makes breast tissue grow, including potentially cancerous tissue. If the pregnancy is completed, other hormones cause the breast tissue to differentiate into milk-producing tissue. If pregnancy isn't completed, the excess estrogen leaves the tissue free to grow into abnormal or even cancerous cells.]
4. But "pregnancies which abort spontaneously (miscarry) usually generate subnormal amounts of estradiol; no increased risk of breast cancer is seen.
5. The incidence of breast cancer is dramatically increased in rats whose pregnancies are aborted."

Most newsworthy in this regard, as I stated in Kingston, was the fact that in the editors' response to our letter in the JNCI not one of these biological facts was challenged!

Then there was the intriguing news about the mysterious appearance of the ABC link in Australian women. This evidence had been suppressed for seven years.

Back in 1988, a study on dietary risk factors for breast cancer in Australian women was published in the American Journal of Epidemiology. However, the study, principally authored by Dr. T.E. Rohan (now in Toronto with the National Cancer Institute of Canada), also listed standard risk factors -- including reproductive history and family history. None of these factors had a statistically significant impact on Australian women in the study (except for removal of the ovaries). Abortion was never mentioned in the 1988 paper.

It was a 1995 paper, principally authored by a French researcher Nadine Andrieu, where the Rohan data on abortion made its first appearance. And it showed a statistically significant 160% breast cancer risk increase among women who had chosen abortion -- much stronger even than family history in these Australian women.

Why were the abortion data suppressed for seven years? Dr Rohan acknowledged, in a phone call I placed to him, that with the politically sensitive status of abortion, he had never even tried to include the results on abortion -- the strongest connection he found -- in the original 1988 paper.

"Canada's national newspaper", the Globe and Mail (July 15) called my presentation "one of the most controversial" of the Kingston conference. Indeed, it provoked many questions from the audience, the first from Ms. Abzug herself.

She wanted to know, if induced abortion increases breast cancer risk, how could Japan, which has had a high abortion rate for decades, have the world's lowest breast cancer rate? The answer which was explained in detail to the Kingston audience (as well as NRL News readers [4/6/95]) is simple.

Four epidemiological studies on Japanese women have been published since 1957. All show increased risk, with an average risk increase of about 100%. All these studies were scientifically correct: they compared Japanese women who had undergone an abortion to Japanese women who had not -- not to women from the U.S. or anywhere else. And in Japan, where breast cancer risk in the absence of abortion is about 2%, abortion doubles it to about 4% -- still much lower than the U.S. breast cancer rate.

Other provocative questions followed from this audience, well represented by breast cancer survivors, most notably concerning survival among women diagnosed with breast cancer while pregnant, and those who got pregnant after treatment for breast cancer. I reported that, contrary to commonly given medical advice, survival was markedly better in both cases among women who chose childbirth instead of abortion. Ironically, the best research in the world on these subgroups of women was performed at the Princess Margaret Hospital in Toronto, only 150 miles from where we sat in Kingston.

As the Globe and Mail article also reported, my world conference presentation "received applause from the audience, which included many pro-choice advocates." Thankfully, some organization concerned with women's health and women's rights has finally taken a good look at the ABC link. After all, abortion is the single most avoidable risk factor for breast cancer, and women everywhere have the right to know that!

by Joel Brind, Ph.D., Professor of Endocrinology, Department of Natural Sciences, Baruch College of the City University of New York, New York, NY 10010

In the 3-1/2 years since I sent Commissioner Kessler a detailed letter summarizing the research literature on abortion and breast cancer, considerable additional data have been gathered, bringing the issue into much sharper focus. To date, a total of 30 published reports describe 24 separate epidemiological studies which give specific data on induced abortion and breast cancer incidence. 19 of the 24 report overall increased breast cancer risk, 12 with statistical significance.Several important conclusions can be clearly drawn based on this substantial body of worldwide knowledge dating back to 1957:

1) Only induced abortion--not spontaneous abortion--is consistently linked to the incidence of breast cancer. The biological basis of this difference is also clear: Most spontaneous abortions are characterized by subnormal ovarian estradiol secretion. It is the surge of estradiol early in a normal pregnancy which provides an estrogen overexposure by which most known risk factors increase breast cancer risk.

2) Induced abortion increases breast cancer risk independently of its effect in delaying first full term pregnancy. An early full-term pregnancy decreases breast cancer risk. Since induced abortion also abrogates this protective effect, it raises breast cancer risk in two ways for young nulliparous women.

3) The increased breast cancer risk attributable to induced abortion cannot be explained by response bias in case-control studies. The only study claiming to provide direct evidence of response bias relies on the specious conclusion that breast cancer patients report having had abortions that never took place, and the only other study using prospective data found a statistically significant 90% risk increase.

4) There is now evidence of a particularly strong interaction between induced abortion and family history of breast cancer, shown by two studies published in 1994.

5) There is no basis for assuming that the somewhat younger average gestational age of medically induced abortions will confer any less of a breast cancer risk increase than surgical abortion: Neither of the two studies which looked at the timing of first trimester induced abortions found a significant difference between abortions before versus after 9 weeks. Endocrinological evidence backs this up: Estradiol begins to surge measurably within a few days after conception. Unfortunately, the short time alotted today does not permit me to report specific data. However, along with colleagues at the Penn State Hershey Medical Center, I have completed a "Comprehensive review and meta-analysis" on the subject, which is now in press for this October's Journal of Epidemiology and Community Health. Although subject to embargo, I can provide the FDA a copy if you are interested.

In the drug approval process to date for mifepristone/misoprostol, has breast cancer, even as a potential risk factor, ever come up? Indeed, the overall, highly significant positive association between induced abortion and breast cancer, which we have documented in the meta-analysis, demands that women be warned at the very least. Such warnings are already mandated to be given to any women considering induced abortion by law in Louisiana, Montana and Mississippi, with more such laws in the pipeline.

by Dennis Byrne [ Chicago-area writer and public affairs consultant.]

How long will this nation sit by as a powerful, well-funded industry continues to expose women to the No. 1 preventable risk of breast cancer?

How long will the industry's political flunkies, who receive millions in campaign funds from this special interest, be allowed to turn a blind eye to a danger that kills thousands of women every year?

How long will a biased media keep silent in the face of a hazard that directly imperils more than 1 million women a year?

No, I'm not talking about the chemical industry, daily poisoning the environment with its toxins. Nor the producers of fatty food or alcohol, also factors suspected of increasing breast cancer.

The industry I'm talking about is the abortion business--consisting of abortion "providers," their clinics, ideological supporters, grant-giving foundations and the rest of the political power structure that refuses to even admit that a scientific debate, let along scientific evidence, exists about the dangers of induced abortions. They--despite their claims of superior benevolence and compassion--are threatening thousands of women's lives with an unspeakably painful disease.

Yet in the month of May, a time of renewal, promise, new life and marches throughout the country against breast cancer, millions of women are being deceived about this risk, or denied the knowledge of important studies.

Twenty-seven out of 34 independent studies conducted throughout the world (including 13 out of 14 conducted in the United States) have linked abortion and breast cancer. Seventeen of these studies show a statistically significant relationship. Five show more than a two-fold elevation of risk. In turn, the abortion industry says all those studies are trumped by one study, whose methodology, critics say, is seriously flawed.

The biological hypothesis is that during pregnancy, a woman's breasts begin developing a hormone that causes cells--both normal and pre-cancerous--to multiply dramatically. If the pregnancy is carried to term, those undifferentiated cells are shaped into milk ducts and a naturally occurring process shuts off the rapid cell multiplication. An induced abortion leaves a women with more undifferentiated cells, and so, more cancer-vulnerable cells.

When I first wrote about this issue in 1997, the scorn and name-calling flowed in. Anti-choice fanatic. Ignorant bozo. Misogynist. Since then, much has happened. The United Kingdom's Royal College of Obstetricians and Gynecologists became the first medical organization to warn its abortion practitioners that the abortion-breast cancer link "could not be disregarded." It said that the methodology of the principal ABC (abortion-breast cancer) researcher, Joel Brind, was sound.

John Kindley, an attorney, warned in a 1999 Wisconsin Law Review article that physicians who do not inform their patients of the ABC link expose themselves to medical malpractice suits. He concluded that about 1 out of 100 women who have had an induced abortion die from breast cancer attributable to the abortion.

The American Cancer Society Web page lists induced abortions (along with pesticides, chemical exposures, weight gain and other factors) among elements that may be related to breast cancer, and that the relationship is being studied.

Earlier, Dr. Janet Darling and colleagues at the Fred Hutchinson Cancer Research Center, in a study commissioned by the National Cancer Institute, found that "among women who had been pregnant at least once, the risk of breast cancer in those who had . . . an induced abortion was 50 percent higher than among other women." The risk of breast cancer for women under 18 or over 29 who had induced abortions was more than twofold. Women who abort and have a family history of breast cancer increase their risk 80 percent.  The increased risk of women under 18 with that family history was incalculably high.

Being pro-choice didn't shield Darling from the usual attacks. She fought back. "If politics gets involved in science," she then told the Los Angeles Daily News, "it will really hold back the progress that we make. I have three sisters with breast cancer, and I resent people messing with the scientific data to further their own agenda, be they pro-choice or pro-life. I would have loved to have found no association between breast cancer and abortion, but our research is rock solid, and our data is accurate. It's not a matter of believing, it's a matter of what is."

Yet the Web site of the Y-ME National Breast Cancer Organization, sponsor of many marches, fails to mention even the possibility of the ABC connection in its list of risk factors. Not even under its list of fuzzy, not "clear-cut" factors. Not even the existence of a scientific debate over induced abortion is worth a mention.

As if women had no right to know.

If you want to know more, look in on the Web page of the Palos Heights-based Coalition on Abortion/Breast Cancer (http://www.abortionbreastcancer.com/). You may not agree with everything there but at least you'll be respected for your intellectual ability to make an informed choice.