Rachel M. MacNair
More than two dozen studies suggest the possibility that the
abortion of a first pregnancy may cause an increased risk in breast
cancer. Connection theorists who argue this position explain the
phenomena as resulting from the hormonal changes and carcinogenic
responses that occur when an established pregnancy is suddenly
terminated. Epidemiological studies provide some support for the
hypothesis. An upsurge in breast cancer in women, especially pre-menopausal
women, render explanations that derive from sharp expansions in
certain conditions--such as the concurrent upsurge in abortions--plausible.
If the evidence continues to mount, then the question arises as
to whether there are predictive models that can suggest the reactional
behavior of abortion providers.
Biological Explanation and Suggestive Studies
In 1989, Dr. Nancy Krieger published a paper that argued there
might be some sort of link between the abortion of a first pregnancy
and the incidence of breast cancer. In the journal Breast Cancer
Research and Treatment, she wrote:
"According to this hypothesis, an early first full-term pregnancy would provide the greatest protection against breast cancer by drastically reducing, early on, the presence of undifferentiated and hence vulnerable breast cells, thereby decreasing the risk of subsequent transformation . . . Other types of pregnancies, however, might increase risk of breast cancer. If a woman's first pregnancy resulted in a first trimester abortion, the dramatic rise in undifferentiated cells that takes place during the first trimester would not be followed by the marked differentiation occurring during the second and third trimesters. The consequent sharp increase in the number of vulnerable cells would thus elevate breast cancer risk . . . abortions occurring after a woman's [first full-term pregnancy] or any subsequent [full-term pregnancy] . . . might have little effect, because the number of undifferentiated cells eligible to proliferate would be markedly smaller, due to the prior pronounced breast development induced by each [full-term pregnancy]." (1)
Before a woman's first pregnancy, her breasts consist mostly of connective tissue surrounding a branching network of ducts, with relatively few milk-producing cells. When a first child is conceived, progestational hormones flood the mother's system. Under the influence of these hormones, her breast cells undergo massive growth. The network of mild ducts begins to bud and branch, developing more ducts and new structures called "terminal end buds." These end buds begin to form "alveolar buds", which will develop into the actual milk-producing glands called "acini." This period of rapid growth toward maturity is when breast cells are most likely to be affected by carcinogens. Quite a bit of research shows that when a woman completes her first full pregnancy, hormonal changes occur which permanently alter the structure of her breasts in a way that reduces her risk of breast cancer. Conversely, a premature termination of a first pregnancy interrupts this process, circumventing the protective effects of a full-term pregnancy and possibly leaving millions of breast cells in transitional states.
Breast cancer is rising sharply among women (though not among men) worldwide, and alarm bells have rung in the medical research community and governmental health agencies about the need to find out why. There are probably many reasons, but the possible connection to the abortion of a first pregnancy does show some promise among the epidemiological factors, enough promise to warrant more definitive study.
Early studies on breast cancer which included socioeconomic status usually found that rich women have a higher rate of the disease than poor women. Dr. Krieger showed, by studying women by census block, that this was in fact a difference in socioeconomic status rather than race (2). Before 1969, a legal abortion in a hospital was expensive, which meant that women of higher economic status were much more likely than other women to obtain abortions (3). If there is indeed a link between abortion and breast cancer, that may help account for the link between breast cancer and socioeconomic status.
In a study that seems to confirm this, researchers found that women in Taiwan, which until recently had little experience with Western-style abortion, have no significant difference in breast cancer rates based on socioeconomic status (4).
Inexpensive or free abortions would change this pattern, and studies in states which provide free abortions indicate the pattern has in fact changed. Washington State legalized abortion in 1970, three years before the Roe v. Wade decision. Affluent women in Washington have had little difficulty securing abortions. In the early 1970s, Washington began to publicly fund abortions for the indigent. The breast cancer rate among poor women rose by 53% in the period from 1974 to 1984, while it actually dropped 1% among wealthy women (5). Affluent women, who supposedly have always had access to abortion, experienced no increase in the rate of breast cancer, while the rate among low-income women rose substantially.
A similar study in California, which also funds abortions for the indigent, found that by 1990, among young white women, there was no difference in their rate of breast cancer between rich and poor (6). Washington and California have, to some extent, equalized poor and rich women's access to abortion, and appear to have simultaneously equalized their risk of breast cancer.
One factor often cited as a risk factor for breast cancer is occupation. White females in professional, managerial, clerical and teaching environments suffer statistically higher instances of breast cancer death than women in lower socioeconomic positions (7). Some, but not all, of this increased risk can be attributed to well-established risk factors like age at first birth. As the United States Supreme Court has noted, however, abortion has become a way of life for women who hold these kinds of jobs (8). Recognizing that, it would seem at least plausible that a higher rate of breast cancer among these women could be attributed to their higher rate of abortion.
While an increase in risk cannot be conclusively proven through epidemiological studies alone, the demographic effects of such an increase would be impossible to miss. If abortion causes even a slight increase in the risk of cancer, the staggering abortion rate of our society must eventually make that risk evident. Tragically, population-based studies seem to suggest just that. Other causes such as diet, genetics, radiation, miscarriage, and environmental pollution have always existed. They can be responsible for the relatively stable base level of breast cancer, but the recent global surge requires more explanation.
Many advocates of legal abortion have been aware of the evidence
suggesting the possibility of a breast cancer/abortion link since
at least 1982. In March of that year Willard Cates, Jr. wrote
an article in Science which discussed the evidence of a
link (9). His manuscript was reviewed by D.A. Grimes, C. Tietze,
R.W. Rochat, and C.W. Tyler, authors who have contributed many
articles to publications like Family Planning Perspectives,
Studies in Family Planning, American Journal of Obstetric Gynecology,
and Obstetric Gynecology.
The Response of Abortion Providers
The Planned Parenthood Federation of America, which operates
the largest chain of abortion clinics in the country and is currently
the major source of expansion in the industry, issued the following
statement in November of 1993:
"There is not one study in mainstream medical literature that proves a cause and effect relationship between abortion and breast cancer. In fact, there are many studies that show no relationship.
The largest -- and most comprehensive -- study to date of a possible link between abortion and breast cancer was done in Sweden and reported in the British Medical Journal on December 9, 1989. It followed, for as long as 20 years, 49,000 women who had received abortions before the age of 30. Not only did the study show no indication of an overall risk of breast cancer after an induced abortion, it suggested there could well be a lightly reduced risk. Many other studies have shown no relationship.
The two studies cited by those promoting the alleged link are of limited value because they don't account for the rest of the woman's pregnancy history. These studies are not considered conclusive from a medical point of view.
Those who allege the breast cancer/abortion link rely on a self-published booklet by an author with no known expertise in the field. This document, and a shorter brochure, are distributed by an organization in Virginia ("AIM") which is unrecognized in the medical field and fails to describe its mission or source of support. Studies establishing this alleged link have not appeared in accepted medical journals or undergone mainstream peer review.
Researchers don't know what causes breast cancer. In looking at possible risk factors, it is known that having a full-term pregnancy before age 35 is protective against breast cancer. Studies have shown that nuns, for example, have a much higher rate of breast cancer than the general population. If a woman were to have an abortion and no full-term pregnancy before 35 years of age, she would not have the protection offered by a full-term pregnancy. But neither would a woman who gets pregnant after age 35, or a woman who never has a child. In any event, the relationship between pregnancy and breast cancer is a meaningful element of a woman's medical history, but hardly a reason in and of itself to make decisions about childbearing. . . or becoming a nun.
Since no reliable, accepted study shows a link between abortion and breast cancer this is not information that should be conveyed to clients. In fact, to do so would be irresponsible. Bogus medical arguments and flawed conclusions serve only to create unwarranted fear in women; in no way do they contribute to informed consent.
Recognizing that breast cancer takes the life of five women every hour, that one woman in nine will get breast cancer during her lifetime, and that there has been no real progress in curing breast cancer in the last 40 years, we support greatly increased efforts to determine the causes of breast cancer, expand education and screening, and improve treatment."
The fact that there are more than the two studies claimed in this statement is shown in the accompanying list. The non-expert they cite is Dr. Joel Brind, professor of endocrinology at Baruch College, City University of New York, who is also a breast cancer researcher on staff at Beth Israel Hospital and Mount Sinai Hospital in New York City, and his work is submitted for peer review. An argument could be made about whether he is correct or whether his research is flawed or tainted by bias, but he is not without credentials; neither is Dr. Nancy Krieger, whose work was cited at the beginning of the article. While it is true that the information is being propounded by groups such as "AIM", and other organizations with a very definite taint of anti-abortion bias, that only means that their obvious predilection should be a cause of skepticism, not complete dismissal. After all, Planned Parenthood's unmistakable financial interest in the outcome of this debate would taint their pronouncements as well.
The Swedish study (10) which they claim shows abortion to cause no increase in "overall" risk of breast cancer, does not give them the data support they seek. "Overall" risk was not at issue. Connection theorists have never claimed that every woman who loses a pregnancy has a greater risk of breast cancer, but that women who have their first pregnancy terminated are at greater risk. The Swedish researchers made no effort to identify a control group, nor did they focus on women who aborted their first pregnancy. In Sweden, unlike America, most women who get legal abortions have already had one or more children, and thus most women in this study have the lower risk of breast cancer associated with the protective effect of the first full pregnancy. The Swedish study included women who already had a child with women who aborted their first pregnancy. They then compared the combined results to the total population, which included a high number of women who had had abortions, rather than to women who had not had abortions. Because of these methodological errors, the Swedish researchers effectively masked any possible link between first pregnancy abortion and breast cancer.
The only time first-pregnancy abortions were distinguished from the whole, the statistics showed that women who had an abortion after a live birth had a breast cancer risk of 58% of the "average" risk in the study, while women who had an abortion before a live birth had a risk of 109% of "average". The "average" risk in the Swedish study is based on the population at large. According to the authors, the "average" risk of breast cancer is 40% higher than it had been prior to the legalization of abortion. If one converts the risk factors to reflect this 40% rise, one sees that women who have an abortion after a live birth have an adjusted risk factor of 81% while women who have the abortion first have an adjusted risk of 153% that of Swedish women before the legalization of abortion.
In short, even though this study is cited by Planned Parenthood to refute the connection theory, if one simply counts the women who aborted before their first live birth, the figures actually support the theory.
Most interesting of all is the statement that "no study has proven a cause and effect relationship between abortion and breast cancer." That is accurate. It is also true that no study has proven a cause and effect relationship between smoking and lung cancer, as the Tobacco Institute constantly reminds the public. As every medical professional knows, scientific proof of causation would require ethically prohibited direct experimentation on living human beings. No study ever will "prove" the link. This is how the tobacco industry perpetuates the idea that there is a question about whether smoking causes lung cancer, defying almost the entire remainder of the scientific community.
The analogy between the two industries -- tobacco and abortion -- goes farther than that. Starting with the initial public relations campaigns in the 1920s which succeeded in inducing more women to smoke, cigarettes have been presented as the "torch of freedom" for women. Advertising campaigns have advised women that "they've come a long way" by virtue of having the freedom to smoke in public. Associating the image of "liberation" to the purchase of products and services has been used by the abortion industry as well, to similarly potent effect.
Of course, women have known for some time that smoking does cause lung cancer and many other diseases, and many have continued to smoke in spite of this knowledge. Surely, if the first pregnancy abortion connection to increased breast cancer risk were proven, then many women would still have abortions. It is, after all, only an increased risk, not a sure thing; the crisis of the moment may be more pressing on the mind that the exigencies of the future.
Nevertheless, there are many women who feel ambivalent about obtaining an abortion and would decide against it with this additional information. Few people could be found outside the abortion industry, or perhaps among abortion defenders in their social movement, who would argue that the matter should not be more thoroughly researched, and that women are entitled to know this information. The information on the bad health effects of cigarettes has led to reductions in tobacco consumption among women, but has not eliminated it entirely.
Still, abortion defenders in their social movement have pervasive
influence in society. The tobacco industry might be green with
envy concerning some of the tactics of abortion defenders. The
media, for instance, has for the most part abandoned the cause
of the tobacco industry; although there are still magazines with
large tobacco advertising accounts that will neglect to discuss
the health hazards, the electronic media will allow full coverage
of anti-tobacco activity. Virtually no one discusses the virtues
of tobacco in the media, and very few take the protestations of
the tobacco industry seriously. Abortion providers currently enjoy
a much better reputation among the media, at least for the time
The Response of Abortion-sympathetic Health Groups
Former United States Surgeon General Joycelyn Elders had a packet of several of the studies put into her hands by representatives of Feminists for Life of America at a public meeting in Washington, D.C. in October, 1993 (12); she promised to read the material. Her response to it is still awaited as of this writing. FFLA also distributed the material to every member of Congress and much of the press, and they were not the only organization to be disseminating this information. The politics of the abortion debate seems to preclude the kind of response to this information that would be expected, were this potential cause a relatively non-controversial item.
The National Breast Cancer Coalition (NBCC), an American grassroots
advocacy effort, issued a Position Statement on Abortion and Breast
Cancer on March 24, 1994. This statement is quoted in its entirety:
"Recently some organizations that are part of the anti-choice movement have begun a campaign publicizing that abortion increases the risk of breast cancer. This assumption is based on the selective use of some epidemiologic studies which show a slight increase in breast cancer risk after abortion. It ignores the numerous studies which have shown no relationship or a reduced risk of breast cancer.
Breast cancer issues have been receiving more national attention due largely to the efforts of the NBCC to educate the public and government of the need for more research to end this devastating epidemic. It is not surprising that organizations that have no interest or concern for women with breast cancer are using this increased awareness to further their own agendas. Any organization's use of misleading information regarding the link between breast cancer and abortion to promote its own political gain rather than a concern about breast cancer should be condemned as a deception to the American public.
The NBCC supports more research to answer questions regarding any possible link between breast cancer and abortion. In particular, the Coalition calls for an international collaboration to evaluate all of the existing data. Until such time that conclusive scientific evidence exists, women should not feel the pressure of misleading propaganda intended to influence their decisions."
Accusing all organizations that publicize the possibility of a connection of having "no interest in or concern for women with breast cancer" is particularly ungracious to groups such as Feminists for Life of America, which participated in the NBCC's petition drive calling for greater governmental attention to breast cancer research. The assertion is not otherwise documented.
Still,, the NCBB calls for more research on this possible link, which is more than Planned Parenthood's statement suggests. It is true that those who oppose abortion would be convinced more quickly, while those who favor abortion availability can reasonably be expected to be more reluctant and demand greater evidence. Nevertheless, both sides can agree on the need for greater research so that more definitive and unassailable answers can be found. Yet a distinction may need to be made between those who have a strong philosophical identification with abortion availability and those who actually have a financial stake in abortion provision.
The National Women's Health Network, the only national United
States public-interest membership organization devoted solely
to women and health, also emphasizes the unproven nature of the
possible link, yet takes the prudent and reasonable position that
further research is needed and important, that the evidence so
far is sufficient to merit further resources into research. Its
Fact Sheet, entitled "Abortion and Breast Cancer:
The Unproven Link.", was issued in January 1994 . It states,
Approximately seventy five studies have been conducted world wide that have collected information on breast cancer and reproductive factors such as abortion, childbirth, menstrual cycles, and birth control pills. Unfortunately, data on abortion and breast cancer has been published in only about twenty of the studies. Much vital information is not publicly available. Those studies that have been published show conflicting results. Some studies report a strong association between abortion and subsequent diagnosis of breast cancer; other studies find no association at all.
An oft-cited was reported by Pike et al. in 1981. It found that the risk of breast cancer was increased 2.4 times in women who had either a spontaneous or induced abortion before a first full term pregnancy (FFTP). Hadjimichael found that breast cancer risk was increased 3.5 times among women who had experienced an abortion before a FFTP (Hadjimichael, 1986). A Danish study found a smaller increase in risk (1.4 times) in a group of women whose first pregnancies terminated before 28 weeks (Ewertz, 1987). Howe found that the risk of breast cancer was increased by 1.9 times among women who had undergone an induced abortion (Howe, 1989).
On the other hand, several studies have reported that abortion does not increase breast cancer risk. In 1981 Vessey reported no indication of any association between breast cancer and abortion (Vessey, 1981). LaVecchia and Parazzini also studied the relationship between breast cancer and abortions/miscarriages. In the 1987 study, the researchers reported "little relation of breast cancer risk with abortions or miscarriages." (La Vecchia, 1987). In a recent study the same researchers again found no consistent relationship (Parazzini, 1991). Data by Rosenberg et al. "Suggest(s) that the risk of breast cancer is not materially affected by abortion, regardless of whether it occurs before or after the first term birth" (Rosenberg, 1988). The risks were equal for women who had experienced an abortion and those who had not. In summary, the evidence for the link between abortion and breast cancer is completely inconclusive at present.
Why is the evidence currently inconclusive? Although there have been many studies that gathered information on the possible relationship of abortion to breast cancer, there are many questions which haven't been researched fully.
For example, some researchers speculate that if abortion is associated with breast cancer, it may increase the risk only for women who have abortions when they are teenagers. New studies need to be conducted now, because larger numbers of women who had abortions as teenagers are approaching the age when breast cancer is more common.
Does it make a difference if abortion occurs before or after childbirth? Some of the studies mentioned in this paper looked at timing of childbirth in relation to abortion, but much more data is needed.
Is there any difference between induced abortion and spontaneous abortion (miscarriage) in later risk of breast cancer? Some of the published studies have combined women who had abortions and miscarriages into one group when risks were analyzed. Studies need to be designed to look at these two events separately.
Many other similar questions haven't been fully researched. Until these issues are explored, it will not be possible to conclusively answer the question of whether or not abortion increases the risk of breast cancer. However, researchers in another field have conducted interesting experiments that raise questions about how the breast might be affected by abortion.
How might the physiology of the breast be affected by abortion? Before a first pregnancy, a woman's breasts consist of a network of ducts and terminal end buds, immature cells that have the potential to become specialized milk-producing cells. When a woman conceives, hormones involved in pregnancy cause these terminal end buds to differentiate into alveolar buds and lobules in preparation for the breast feeding. The terminal end buds grow and proliferate rapidly as they differentiate. Researchers believe that after the cells have differentiated, they are not as susceptible to carcinogens (cancer-causing agents) because the proliferative ability of the cells is reduced.
Biological evidence from animal studies demonstrates that there is a plausible explanation for an association between breast cancer and abortion. When pregnancy is interrupted, as in abortion, the mammary glands contain some areas with completely differentiated structures and other areas of immature cells (Russo, 1981). Thus, the breast is more susceptible to the initiation or promotion of cancer. Studies of this effect in rats are useful since rats' breast tissue is similar to humans. In experiments, Russo has found that animals who had an induced abortion were at the same risk of breast cancer as animals treated with a chemical typically used to induce cancer in rats. This is the only study reporting this finding. More work needs to be done to confirm the effect of abortion in rats and to determine whether the developmental mechanisms are the same in women.
When will we have the answer? An international collaboration to analyze all seventy five studies with data on abortion and breast cancer has begun. Results of this summary will be available in about two years. This will nearly triple the amount of data currently available and has the potential to address some of the questions which are currently unanswered.
Two of the studies cited as showing no link are in fact included on the accompanying list as ones that do show such evidence; again, their data tend to support the theory when abortion of the first pregnancy is looked at specifically. The Vessey study (12) looked at women in ten-year age blocks, so that any increased risk in 38 year old women would be diluted by a lack of increase in 31 year old women. Similarly, the Rosenberg study (13) did not match the ages in its groups; while the women with breast cancer averaged 52 years of age, the control group averaged only 40 years old. Since age has long been recognized as a major risk factor, this lack of corresponding ages taints the study substantially. Still, an approach of treating studies not showing a link and studies which do equally, and mentioning them both, is more even-handed and reasonable than the approach taken by Planned Parenthood.
It would stand to reason that those who oppose abortion would
become convinced with less evidence, and groups that favor abortion
accessibility such as the National Women's Health Network would
be much more cautious and skeptical and demand greater evidence.
Nevertheless, all those who have women's interests at heart in
both viewpoints come to the same conclusion: the evidence thus
far clearly indicates the need for further research, because women
are entitled to have answers.
A Possible Predictive Model
Let us suppose, for the sake of argument, that the increasingly accumulating peer-reviewed studies continue to grow in size and number as time goes on. If this is not the case, then the following discussion will be irrelevant, and the viewpoints expressed by groups such as Planned Parenthood will be confirmed. If it is the case, however, then the public policy implications deserve exploration.
What would be the response from those who hold to medical ethics? The right of patients to have information is basic. Any possible complications from a surgical procedure (or a drug regimen) are unambiguously information that any patient has a right to know beforehand.
Additionally, any advocacy position stating that women have the right to control their own bodies, to have sovereignty to make decisions for their own lives, would obligate the dissemination of this information. The consciousness of women cannot be raised if they are kept in the dark on any subject. This is all the more the case on anything involving this magnitude of impact upon a woman's life.
There is another enterprise which has been faced with the necessity of defending its product against studies suggesting an increase in cancer due to its use: the tobacco industry. Its reactional behavior could provide a predictive model for abortion providers as well.
The January 1967 issue of True magazine featured an article that said there is no evidence to show that smoking causes cancer. That issue was widely advertised, and free copies of the article were sent to tens of thousands of influential Americans with a note from True. However, the organization Action on Smoking and Health filed a complaint to the Federal Trade Commission (FTC), and established that the article had actually been written by an employee of the industry's public relations firm. The ads had been placed and the mailings had been sent surreptitiously by the industry, not by the magazine (14).
The tobacco companies know full well that the fact that women (and men) have knowledge of the health impacts of their products decreases their business. They therefore continually deny this health impact and do so in the face of mounting evidence. They have countered with questionable studies of their own, and used ad hominem attacks, using such terms as "fumophobe" to describe opponents (15).
Clearly, a major distinction between the two businesses exists. The tobacco industry has a product it wishes to sell, and it makes no pretense otherwise. Abortion clinics, on the other hand, insist that they provide a service only because it is necessary, and their spokespeople will occasionally make remarks on the merits of reducing the need for those services. Therefore, the use of the model of the tobacco industry's response has predictive value only inasmuch as the those in abortion provision in reality view themselves as having a product to sell. Will the financial incentives outweigh the established medical ethics of providing complete informed consent for patients? Will they continually deny the evidence as it mounts, counter with questionable studies of their own, and employ ad hominem attacks?
A glimpse of what may be forthcoming is provided in statements made during a televised interview of Pamela Maraldo, then president of Planned Parenthood Federation of America. Discussing the most well publicized study to date (16) Maraldo was asked, "If, indeed, your panel of medical experts studies this study by Dr. Daling, and you find it to be solid, good science, what are the chances you will begin warning women about this possible link?" Maraldo responded, "Even if it's solid, good science, then to begin to warn women and upset women on the basis of one study is clearly irresponsible. One study is not adequate evidence to change a policy or to upset a frightened woman."(17)
Based on this model, further predictions would include the public, the media, members of Congress, and the breast cancer research community finding this response reasonable and acceptable initially, while becoming increasingly impatient as evidence of a link continues to mount, assuming it does so. Any action exposing a disregard for mounting evidence--especially where a clear financial interest can be demonstrated--and callous indifference for a women's right to informed consent, are ultimately not likely to be widely appealing.
Taking the road of medical ethics would not only be the morally
right thing for abortion providers to do, it would also be the
most astute strategy: abortion providers very much need an image
of careful professionalism in medical settings. Squandering that
image would likely be far more damaging to them than any revelations
about possible cancer links could ever be. The predictive value
of the model of the tobacco industry depends entirely on how abortion
providers actually view their own practice.
Impact on the Individual
Women ambivalent about having an abortion may find information suggesting a breast cancer link ultimately dissuasive. Women being pressured to have an abortion may find the information a useful tool in resisting the interests of others. Some may be more circumspect with regard to contraception, or even elect to forego sexual opportunities that could lead to pregnancy. Others, determined to obtain an abortion, are unlikely to be swayed by what they will probably view as just another risk.
What about those women for whom all these options are foreclosed by virtue of their having had an abortion? Researchers like Daling and Krieger insist that there is no cause for panic, (18) and the science appears to affirm this. Even if the evidence were to become conclusive, an increased risk only means the presence of another factor which should be considered as part of a woman's medical history and in consultation with her physician.
Yet it must be said that while the increase in risk appears to
pose little cause for alarm among individual women, the
increase also suggests a profound impact on women as a group,
considering the current high abortion rate. If the evidence for
an abortion/breast cancer link eventually becomes unassailable,
then we will know that thousands of women have died of breast
cancer who might otherwise have lived if not for the prevalence