Study finds no abortion link to breast cancer

By Shankar Vedantam
The Washington Post

E-mail this article Print this article Search archive

Having an abortion does not increase a woman's risk of contracting breast cancer later, according to the most comprehensive and definitive analysis conducted on the controversial issue, scientists said yesterday.

The conclusion was based on 53 studies involving 83,000 women in 16 countries. Researchers for the first time compared all studies according to the quality of their methodology: Better-designed studies found no link between abortion and breast cancer. Studies using weaker designs were inconsistent but on average found a link.

Several states tell women considering an abortion that it might increase their risk of breast cancer. Other states are debating whether to require doctors to warn women of a risk. There is no such requirement in Washington state, said Marcy Bloom, executive director of Aradia Women's Health Center in Seattle.

The U.S. government also had suggested that abortion might increase the risk, but it recently dropped that caution from an official cancer information Web site.

The governmental assertions all suggest that there is an ongoing scientific debate in which studies that found no link are balanced by others that did. The new analysis concludes that this approach distorts the truth. It is being published today in the Lancet, a respected British medical journal.

"We have demonstrated that a certain group of studies are unreliable and can't be trusted," said Valerie Beral, an Oxford University professor of epidemiology who coordinated the study. "You could say, if anything, these results suggest the possibility of a slightly reduced risk of breast cancer if you have had an abortion."

The studies deemed unreliable are those that anti-abortion activists long have championed. Beral, who conducted the analysis with an international group of scientists, said those studies were flawed because they asked women with breast cancer whether they had ever had an abortion.

Researchers think women with breast cancer are more likely to reveal an abortion because they are searching for explanations for why they got cancer, said David Grimes, former chief of the abortion surveillance branch of the Centers for Disease Control and Prevention. Healthy women have less incentive to volunteer such personal information, he said.

The better "prospective studies," Beral said, tracked women for years to see whether they had abortions and, later, whether they developed breast cancer. Other studies deemed reliable relied on official registries that kept track of abortions and breast-cancer cases in countries such as Sweden.
 
Both sides in the abortion debate predicted the issue would not go away. The most prominent advocate of the link between breast cancer and abortion accused the new analysis of bias, and said the new study was tainted by "political correctness."

Joel Brind, professor of biology and endocrinology at Baruch College of the City University of New York, said the Lancet paper excluded one well-conducted prospective study that had showed a link between abortion and breast cancer, while including three others that were flawed. He accused the peer-review process that approved the study for publication, and the Lancet, of being "corrupt."

"If they were not corrupt, they would never allow garbage studies like this to get published," Brind said. "These studies are junk science."

Beral defended the prospective studies that were included. She said two prospective studies had been excluded because their investigators no longer had the original data — one found a link between abortion and breast cancer, and one did not.

Because the studies were small, she said that they would not have changed the result even if they had been included.

Seattle Times staff reporter Julia Sommerfeld contributed to this report.

Copyright © 2004 The Seattle Times Company