Has ‘mainstream ’medicine's interest in preserving the notion of 'safe abortion ’marginalized a significant and statistically provable danger to women's health?

What do you suppose would happen if a reputable team of American cancer researchers, funded by the National Cancer Institute (NCI), the federal agency that pays for most cancer research in the United States, conducted a study that produced convincing evidence that American women who have had an induced abortion end up with a significant (50%) higher risk of acquiring breast cancer than those who have never had an abortion? And what if these scientists then published the results in the Journal of the National Cancer Institute (JNCI)?

Reasonable people would expect the JNCI would also publish an editorial in the same issue highlighting the study as important for American women to know. After all, induced abortion is a matter of choice, and breast cancer is the leading killer of middle-aged women. With well over a million abortions performed in the United States each year, and more than 180,000 cases of breast cancer resulting in some 40,000 breast cancer deaths, one would expect the NCI to sound the alarm.

In fact, this very scenario occurred in late 1994, with a slight difference: The JNCI editorial called the reported increased risk “small in epidemiologic terms,” and concluded that “the overall results as well as the particulars are far from conclusive, and it is difficult to see how they will be informative to the public.” This dismissive tone was adopted indiscriminately by the media and the possibility of a causal link between abortion and breast cancer (ABC link) faded quickly from the public mind.

Considering the investment of “mainstream” medicine and public health in the reputation of “safe abortion,” a reluctance to allow abortion to cross the usually low threshold of inclusion on a list of cancer risk factors is understandable, if regrettable. But what is particularly significant and troubling about this episode is that it is neither the first nor the last chapter in the story of a death-dealing cover-up that has spanned the globe for decades, and which continues in full force, from the highest places.

Early Research

Long before Roe v. Wade, a nationwide study on cancer in Japan, published in that nation's most prominent (English language) cancer research journal, found a 150% to 200% increased incidence of breast cancer among women who had had abortions. These were cancer patients diagnosed between 1948 and 1952, and the study was published in 1957.

With abortion being uncommon in most of the world at the time, few studies appeared in the next several years. The most prominent was a multi-center study conducted by the World Health Organization (WHO) in places as far-flung as Taipei, Taiwan; Boston, Mass.; and Sao Pauolo, Brazil. Although this study did not distinguish between induced abortion and spontaneous abortion (miscarriage), it revealed a disturbing trend “in the direction which suggested increased risk associated with abortion — contrary to the reduction in risk associated with full-term births.” That was 1970, when abortion was first legalized in several U.S. states, and the landmark study was published in the WHO's own Bulletin.

Millions of “safe, legal” abortions later, in 1981, the first study on the ABC link in American women was published in the British Journal of Cancer. In the study, prominent University of Southern California epidemiologist Malcolm Pike and colleagues reported that women were 140% more likely to be afflicted with breast cancer by the age of 33 if they had aborted their first (or only) pregnancy.

Some public health authorities took Pike's findings seriously and responded appropriately. For example, Willard Cates of the Centers for Disease Control and Prevention (CDC), writing in the journal Science in 1982, expressed “concern about ... possibly higher risks of breast cancer in certain women.” National Institutes of Health scientist Bruce Stadel and colleagues at the CDC, in a letter published in the prominent British journal The Lancet, stated unequivocally, “Induced abortion before first-term pregnancy increases the risk of breast cancer.”

Such warnings were all the more credible as experimental verification in laboratory animal experiments were also appearing, most notably in the work of Jose and Irma Russo of the Michigan Cancer Foundation. Their experiments, still viewed today as landmarks, were published in the American Journal of Pathology in 1980.

Protecting Abortion

But signs of trouble in the medical research community emerged as well. A very prestigious group of Oxford epidemiologists, headed by Martin Vessey, reported in the British Journal of Cancer the results of a study they conducted on almost 1,200 English breast cancer patients. In the first line of their abstract, Vessey and companions noted the Pike study by name, and proclaimed that, on the contrary, the Oxford results were “entirely reassuring.” In other words, there was no trend toward increased risk in women who had had an abortion.

The trouble was the number of subjects in the Vessey study who had actually chosen abortion: It was “only a handful,” admitted the researchers (deep in the body of the text). In truth, the Vessey study was entirely irrelevant regarding any effects of induced abortion, yet it was put up as an authoritative study to protect the reputation of “safe” abortion. It seemed to do the trick, however, with the dreaded term “breast cancer” being absent from the ongoing public debate on abortion for more than another decade.

There was, however, a stream of studies quietly being published in the medical literature in the United States, Europe, and Japan that continued to verify the ABC link. But even in 1992, the American Medical Association's report of the safety of abortion, published in its own journal (JAMA), made no mention of breast cancer whatsoever; nor did the New England Journal of Medicine (NEJM) in its review of potential as well as established breast cancer risks, mention abortion at all.

Late in 1992, having been engaged in breast cancer research among other projects, I came across evidence of the ABC link buried in the medical literature. In October 1994, the NCI study (mentioned above) by Janet Daling et al. of the Fred Hutchinson Cancer Research Center was published in the JNCI, with its gainsaying editorial. Two years later, my own “comprehensive review and meta-analysis” of the ABC link, written with colleagues from the Penn State College of Medicine, was published in the British Medical Association's Journal of Epidemiology and Community Health (JECH). In our study, we combined data from every available published study, using the most conservative statistical model. Of the 23 studies analyzed, 18 showed increased risk of breast cancer after induced abortion (now 25 of 31 studies; 11 of 12 on American women), and the average risk increase was — remains — a statistically significant 30%.

Bias, But Whose?

One of the reasons we opted for publishing the study in a British journal was to avoid the kind of editorial response Daling had received from the JNCI. However, two months later, the JNCI (Dec. 4, 1996 issue) published an editorial that attacked our study, calling our inference of a cause-and-effect relationship “a leap beyond the bounds of inference.” They resorted to ABC link critics' favorite four-letter word: bias. It was not an accusation of bias on our part against abortion (our research team represented both “pro-life” and “prochoice” points of view), but rather to the claim that actual findings of increased breast cancer risk in women who have had abortions are likely due to bias on the part of the study subjects.

This sort of bias is variously referred to as “response bias,” “reporting bias,” or “recall bias.” The theory is that in epidemiological “case-control” studies (most are of this type), women who have breast cancer may be more forthcoming and honest about reporting their abortions than the healthy “control” subjects to whom they are compared. If this occurs, the cancer patients will appear to have had a greater frequency of abortion than the healthy women, and will translate statistically to an increased risk of breast cancer among women who have aborted.

A team of Swedish researchers, headed by Olav Meirik of the World Health Organization, claimed to have direct, significant evidence of response bias in Swedish women. Specifically, they compared interview-based data to computerized medical records on the same population of Swedish women and found that healthy women underreported their induced abortion history “relative to over-reporting among cases.” In other words, their “evidence” of response bias depended on the assumption that abortions reported by some breast cancer patients, of which abortions the computer had no record, had not taken place at all — they had been “over-reported.” (This claim was retracted by Meirik et al. in March 1998.)

Importantly, every other attempt to produce evidence of response bias has come up empty. Further, the only American study based on computerized medical records rather than interviews (therefore immune to response bias) found a 90% increased risk of breast cancer with induced abortion.

None of this has managed to deter those bent on denial of the ABC link. The NCI has effectively adopted the position that a woman need not worry about the risk of breast cancer when considering abortion, based on a single study of Danish women published in the Jan. 9, 1997 NEJM. The study concluded that “Induced abortions have no overall effect on the risk of breast cancer,” but was a broad analysis of all 1.5 million Danish women born between 1935 and 1978; more than 300,000 abortions, and some 10,000 cases of breast cancer. The New York Times did not hesitate to jump on the NCI bandwagon, claiming that the Danish study “largely disproved the notion that abortions cause breast cancer.”

Flawed Study

Now, a year-and-a-half later, the NCI and its followers show no indication of modifying their position. On closer inspection, the Danish study (funded in part by our U.S. Department of Defense) falls apart. To detail but one of its fatal flaws, the study was based on abortion records dating back only to 1973, whereas, contrary to the study's claim, abortion had been legalized in Denmark, for reasons other than medical necessity, in 1939, and some 60,000 women recorded in the study as having had no abortions, actually had legal abortions on record. Although such flaws have been enumerated in the NEJM and other medical journals, such statements are as easily buried as the 40- plus years of published research that documents the ABC link.

The NCI and other federal agencies are subject to oversight by Congress, however, and some Washington officials are fulfilling the job they are constitutionally mandated to do. In a recent letter to the chairman of the Commerce Committee's Subcommittee on Oversight and Investigation, Reps. Christopher Smith (R-N.J.) and Henry Hyde (R-Ill.) said: “It is clear that Congress must take the NCI to task for its alarming attempts to bury the ABC link.” The letter included a formal request for public hearings to be held by the Subcommittee. A similar letter was sent by Reps. Dave Weldon (R Fla.) and Tom Coburn (R-Okla.), who are physicians.

It is comforting to know that Lincoln was right: “You can't fool all the people all the time.” Especially when it concerns the most avoidable risk for breast cancer.

Joel Brind PhD is a professor of biology and endocrinology at Baruch College of the City University of New York, and editor-publisher of the Abortion-Breast Cancer Quarterly Update.