Within a decade of Roe vs. Wade, the Supreme Court opinion that legalized abortion in 1973, animal research showed a correlation between abortion and breast cancer. Then, in 1981, a much-publicized study of 163 Los Angeles women, appearing in the British Journal of Cancer, found that women who had an abortion were more than twice as likely to develop breast cancer. But later research wasn't consistent with the British journal's findings, and many scientists began to suspect flaws in the methodology of the British study.
One problem: Almost all of the studies done on the topic were observational "case-controlled" studies, meaning the medical histories of women who were diagnosed with breast cancer were retrospectively tracked and compared with a similar group of women who did not have breast cancer. The drawback to observational studies is that they are based on patients' self-reported recollections of their medical histories -- not medical records or exams. Researchers say that this approach can lead to problems in getting accurate information, especially if the study involves a medical situation that may have some social stigma, such as abortion in the United States.
Scientists suspect that some of the women who had abortions, especially before the Supreme Court made them legal in 1973, might not accurately disclose that history -- a phenomenon in medical research known as "reporting bias." Some of the studies also failed to account for separate breast-cancer factors, such as family history and diet.
Because of reporting and other biases involved in abortion-breast cancer research, "most of the conclusions from [these] case-controlled studies aren't worthwhile," says David Grimes, a clinical professor of obstetrics and gynecology at the University of North Carolina at Chapel Hill, who studied the abortion-breast cancer issue while working as an epidemiologist at the U.S. Centers for Disease Control and Prevention.
In 1996, Brind, of Baruch College, and researchers at Pennsylvania State University revisited the abortion-breast cancer theory by conducting a comprehensive review of all 23 known studies on the issue. Together, the research suggested a 30% increase in breast cancer risk for women who have had at least one abortion. Anti-abortion groups touted the analysis, appearing in the British Medical Assn.'s Journal of Epidemiology and Community Health, as conclusive proof of a link.
Some scientists, however, questioned Brind's research method, arguing that an examination of a lot of small, poorly designed studies might only lead to a larger, flawed conclusion. Others pointed out that Brind is not an epidemiologist, which is more common in research of this kind, and noted his association with some anti-abortion groups. Brind says any implication his political views influenced his research is "ridiculous."
Overall, the U.S. abortion rate is at its lowest level since 1974, according to the Alan Guttmacher Institute, down to 1.3 million from a peak of 1.6 million in 1991. Breast cancer, though, is a different story. Rates of breast cancer in the U.S. have been climbing steadily for more than two decades. Many scientists say better detection and the increase in other known risk factors, such as obesity and environmental causes, may explain part, but not all, of the rise in breast cancer rates.
One woman's story
Charnette Messe, a mother of two from Groton, Conn., was diagnosed with breast cancer last spring. At first, her doctors weren't optimistic about her prognosis, but after she recently underwent chemotherapy and radiation, her doctors believe she may be in remission.
Messe is convinced that an abortion she had 11 years ago is to blame for her cancer. After all, she points out, she doesn't fit into any of the major risk categories: She eats well, exercises often and has no family history of cancer. Messe adds that politics is not playing a role here either. Although she and her husband now oppose abortion, she says she used to support abortion rights until having her children just a few years ago.
"I'm not saying [breast cancer] is my punishment," says Messe. "I just feel in my heart that this illness is because of my regrettable decision."
To escape the epidemiological bias of the previous studies on the topic, from self-reporting or other sources, researchers finally turned toward another study method, known as a cohort study. In this kind of research, women are tracked prospectively, meaning from the time they have an abortion onward to see if they contract breast cancer. But this kind of study takes time, usually decades. Considering the stigma around abortion, there was a second problem: Where could they find a population of women to follow? The answer came from Scandinavia.