Study Adds to Discourse on Cancer, Abortion
For several years, some researchers and anti-abortion advocates have argued that having an abortion increases a woman’s risk of developing breast cancer.
Most scientists thought that the matter had been largely settled earlier this year when a National Cancer Institute panel examined the small group of studies investigating the subject and concluded that the risk was insignificant.
But the question was reopened again today, in the middle of a presidential election in which abortion rights have played a major role, by a new study suggesting that an abortion raises the risk of breast cancer 30%.
A team from Pennsylvania State University and Baruch Medical College in New York City reanalyzed the same studies considered by the NCI team. In their study in the October Journal of Epidemiology and Community Health, the researchers conclude that a short-term disruption of hormonal balance produced by abortions causes an estimated 5,000 cases of breast cancer each year.
“The evidence is overwhelming,” said co-author Vernon Chinchilli of Penn State.
Few others seem to agree, however.
*
“We don’t dismiss this out of hand, but we sure look at it with a slightly jaundiced eye,” said Dr. Michael Burnhill, vice president for medical affairs of Planned Parenthood Federation of America.
“This man [Joel Brind, the lead author] is an avowed anti-abortionist,” Burnhill said. “This material was used to start a real scare campaign in Washington, with posters in subways, handouts and lobbying for a bill to make doctors warn about the risk. I have to wonder if we are dealing with real, unadulterated science.”
“The study is invalid,” said Boston University epidemiologist Lynn Rosenberg, who has analyzed the individual studies. “It means nothing.”
“This won’t change our position,” said an NCI spokeswoman.
And even if the results are confirmed, said Stanley Henshaw, deputy director of research for the Alan Guttmacher Institute, “the cancer risk caused by abortion would still be less than the risk taken by every woman who postpones childbirth until her 30s.” It is also smaller, he noted, than the increased statistical risk associated with living in a city, graduating from college or having a high income.
“I’m just amazed by the interest in an article which, if entirely accurate and true, would not show much of a risk of breast cancer,” Burnhill said.
Speaking at a news conference Friday in Hershey, Pa., the four researchers who prepared the study denied that they were politically motivated and indicated that they had divided opinions on the subject of abortion.
The team made a decision that they would publish the results, no matter how they turned out, said Penn State’s Walter B. Severs. “Our job as scientists was to do the job and report the facts. The research speaks for itself.”
*
Researchers have published more than 30 studies examining the possibility of a link between abortion and breast cancer. Some have shown a slight increase in risk, some have shown no effect at all, and some have shown a slight protective effect.
“There couldn’t be a very large relationship, or they would all go in the same direction,” Burnhill said. Despite the lack of firm evidence, three states--Louisiana, Mississippi and Montana--require physicians to warn women seeking abortions about the presumed risk.
One criticism of these earlier studies is that none is large enough to yield definitive results if there is a small increase in risk. Brind, a professor of endocrinology at Baruch, and his colleagues attempted to get around this problem by using a controversial technique called meta-analysis to combine data from 23 of the small studies into one larger study involving 25,967 women with breast cancer and 34,977 control patients without cancer.
The 30% increase in risk they observed in their analysis, when projected to the 800,000 first-time abortions performed annually in the United States, would give rise to 25,000 additional cases of breast cancer each year by the mid-21st century as the initial group of women exposed to legal induced abortion in the 1970s ages.
Brind believes that the increased risk of breast cancer results from a sharp rise in a woman’s estrogen levels when a pregnancy is terminated. “Excess exposure to estrogen is involved in most known breast cancer risk factors,” Brind said. If the pregnancy goes to term, hormonal levels are restored naturally and a protective effect eventually occurs.
But, critics say, Brind’s study--and meta-analysis in general--suffers from the same problems afflicting the individual studies: questions about comparability among the study groups, lack of knowledge of other pertinent lifestyle factors that may have a greater influence on breast cancer, and an inability to distinguish between induced and spontaneous abortions.
Brind’s group included studies from Asia, for example, where the risk of breast cancer is four to five times as high as in the United States.
*
Perhaps the most serious potential weakness of the studies, according to NCI researchers, is a phenomenon called recall bias. It involves the accuracy of reports gathered from women 20 years or more after the events in question. That is a problem under the best of circumstances, and more so when the study involves an illegal act.
“People have to understand that if you are studying older women, they would have had these induced abortions before they were legal,” said Dr. Malcolm Pike of USC. “If you are a breast cancer patient trying to figure out why you got it, you might be willing to tell the truth about an abortion. But if you speak to controls, they might not tell you about them because they were illegal.”
If researchers are studying younger women, he added, they are likely to find a risk because both abortions and pregnancies increase the risk of breast cancer for about five years after the event.
The bottom line, most experts agreed, is that Brind’s study does little to move the debate forward. As NCI concluded in February, “The available studies do not permit definite conclusions about the relationship between breast cancer and either spontaneous or induced abortions.”
(BEGIN TEXT OF INFOBOX / INFOGRAPHIC)
Cancer Risk Factors
A new study claims that an abortion increases a woman’s risk of breast cancer by 30%. Many other lifestyle factors have also been shown to increase the risk of breast cancer.
FACTORS INFLUENCING RISK / ESTIMATED INCREASED RISK
According to new study:
* Abortion: 30%
According to National Cancer Institute:
* Older age (65-69 vs. 30-34): 1,600%
* Residence in North America vs. Asia: 300%-400%
* Residence in urban areas: 50%
* Higher educational status or family income: 50%
* Mother or sister with breast cancer: 100%-200%
* Obesity: 100%
* Tallness: 50%-100%
* Dense breast tissue: 100%-300%
* No children or first birth past age 30: 100%-200%
Sources: Louise A. Brinton and Susan S. Devesa, National Cancer Institute
