From winter 2003 - Abortion and breast cancer - Is there a link?
(Right click and choose 'save as...' to download a printable version of this article)
If a scientist discovered a risk factor that increases the chance of breast cancer by 30%, you'd have thought it would have spurred huge headlines and impassioned demands for action. With the exception of AIDS, no other health issue has been as politicised as breast cancer. Yet as scientists zero in on what one called the single most avoidable risk factor for breast cancer, barely a peep has been heard for more research, more funds or more information. That's because the risk is abortion.  (D Byrne)
For several decades now, the incidence of breast cancer in the Western World has been increasing. The lifetime risk of breast cancer in women is now around twelve percent. There are several possible explanations in the literature but one in particular is a recurring theme: the connection between abortion and breast cancer.
Finding a link between abortion and breast cancer is of limited significance if there is no plausible biological explanation. A 1980 study looked at the effects of a known carcinogen on rats of differing parity. Rats allowed to become pregnant and go to term before exposure to the carcinogen had a mammary carcinoma rate of six percent. The rates in virgin rats and rats with induced abortions were 68% and 78% respectively.
The authors suggested that abortion can increase the risk of breast cancer in two ways: firstly by abolishing the protective effect of a first full term pregnancy and secondly by superimposing its own independent risk. This second proposed factor is related to oestrogen exposure: oestrogen promotes both normal and abnormal breast tissue growth and has a specific effect on undifferentiated breast tissue. It is the undifferentiated cells that are particularly vulnerable to malignant change later in life. Early pregnancy is characterised by high mitotic activity and proliferation within the breast. A woman who goes through several weeks of a normal pregnancy and then has an abortion is left with more potentially malignant cells. In contrast, the well-known risk reduction of a full term pregnancy results from differentiation and fewer potentially malignant cells. By eight weeks gestation, oestrogen concentration is typically six times higher than at conception. In contrast, first trimester miscarriages usually do not generate oestradiol in quantities greater than the non-pregnant state.
The earliest study linking induced abortion with later development of breast cancer was published in Japan in 1957. Women who had had abortions carried a 2.6 relative or 160% increased risk compared with women who had not had abortions. In addition, this study did not find any association between miscarriage and breast cancer. Generally, this has been confirmed in numerous studies since. A 1981 Californian study concluded that, in the absence of a full-term pregnancy, young women were 2.4 times more likely to develop breast cancer if they had had an abortion. A 1989 prospective study showed a 90% increased risk. An alarming 1994 paper compared approximately 900 breast cancer patients against a control group. In women with no relevant family history there was a relative risk of breast cancer after abortion of 1.4; those with a positive family history had risks of 1.8. A small subgroup of women with first abortions below the age of 18 and a positive family history had incalculably high risks: every one of these women developed breast cancer by the age of 45.
By 1996 there were 23 studies, 18 of them showing increased risk. A meta-analysis revealed at least 30% increased risk of breast cancer after abortion over and above loss of the protective effect of first full term pregnancy. In the UK Patrick Carroll of the Pension and Population Research Institute published a major analysis of our own data. Using figures from the Office of National Statistics he discovered a high positive correlation (0.84) between cumulated breast cancer incidence and cumulated abortion rates in women aged 45-49. Projection of the trends of increased breast cancer incidence to the year 2023 suggests a 60 percent increase in this age group. The total number of cases in women of all ages is expected to more than double to around 77,000 per year.
There has been a reluctance to inform the public about the ABC (Abortion Breast Cancer) link. Some of the more egregious examples include the following:
A 1982 paper concluded, 'The results are entirely reassuring'  However, a closer look at the results section shows that their data included 'only a handful' of women who had had an abortion; a figure is not given, therefore no valid calculation can be made. Furthermore, figures for abortion and miscarriage were combined, masking any hope of finding the answer. These results were not 'entirely reassuring': they were completely irrelevant.
An Australian study looked at risk factors, including abortion, for breast cancer but failed to report on the abortion factor. The data only came to light in another small 1995 metaanalysis by a French group; they looked at the Australian figures and calculated a 160% increased risk of breast cancer after abortion. In fact abortion was the strongest risk factor but the original researchers had concealed their findings. 'It was the strongest risk factor they found. It was the only one that was clearly statistically significant. And this you don't do. This is not what you see in scientific research, ever. I've never seen it before, where the most significant finding in a study is specifically left out of a research paper...and we hypothesise that there is more of it.'
In an accompanying editorial to Daling's 1994 study, an official of The National Cancer Institute tried to explain away the 50% increased risk as response bias. This is a theory that women who had abortions and later develop breast cancer are more likely to admit to their abortion than women who had abortions and don't develop breast cancer. In 1991 Swedish researchers first suggested it as seven women who had reported abortions weren't on the national computerised abortion registry and were therefore thought to be making it up. In 1998 they finally admitted that this registry is incomplete but the response bias theory is still used as a way of 'explaining' research that suggests the ABC link.[16,17] Abortion is associated with significant amounts of denial so it seems at least equally plausible that women who develop breast cancer could be more likely to deny their abortions. Whatever this theory's merits, Daling's researchers had tested specifically for response bias and found no evidence of it.
A 1997 Danish record linkage study claimed to show no evidence of increased risk. It contains a catalogue of errors. The researchers counted abortions from 1973 when the law was liberalised and a computerised registry begun; however, Danish abortions had been hand recorded since 1940. Using data on the median age of abortion, it was possible to calculate an exclusion of around 60,000 women who had had abortions but were not included in the study. Another major error includes logging breast cancer cases from 1968 but abortions since only 1973. This breaks one of the most basic rules of epidemiological research: exposure must precede outcome.[19,20]
If the lifetime risk of breast cancer is close to twelve percent and abortion introduces an extra 30% risk, this 1.3 relative risk increase means a four percent increased risk in absolute terms. Using the figure of a 25% mortality rate from breast cancer, this works out as at least one extra breast cancer death per 100 abortions or 1,000 deaths per 100,000 abortions. As there are 190,000 UK abortions each year, there will be roughly 1,900 extra breast cancer deaths over the course of a lifetime for each yearly cohort of aborted women.
It is no longer possible to argue that abortion is safe: abortion providers and their apologists should be held to account.