An abortion or a miscarriage does not increase a woman's risk of breast cancer, according to results released Monday from a decade-long study of more than 100,000 women.
The findings are the latest, and perhaps most convincing, in a series of studies that have discredited a concern cited by antiabortion activists to dissuade women from having the procedure.
"It's important for women to have the facts," said Karin B. Michels of Brigham and Women's Hospital in Boston. Michels is lead author of the study.
She said her group's study was "very much in line" with a 2003 expert panel convened by the National Cancer Institute that concluded no evidence supported a link between abortion and breast cancer. The institute funded Michels' study as well.
Karen Malec, president of the Coalition on Abortion/Breast Cancer, took issue with the findings. Her group uses the purported link as an argument against abortion.
"Clearly [the cancer institute] must suspect a link, or else they know that a link really exists," Malec said. "Why else would they continue to pay for these studies?"
Texas, Minnesota and Mississippi require physicians to warn women seeking an abortion about the supposed cancer risk. Several other states considered similar laws but rejected them in light of the 2003 consensus report.
The new results, reported in the Archives of Internal Medicine, are from an arm of the Nurses' Health Study that involved 105,716 women.
Beginning in 1993, Michels and her colleagues biennially questioned the women -- who were ages 29 to 46 in 1993 -- about abortions, miscarriages and breast cancer.
They found that through 2003, a total of 16,118 had had at least one induced abortion and 21,753 had had at least one spontaneous abortion (miscarriage). The team found 1,458 new cases of breast cancer -- an incidence that was the same among women who had had an abortion (induced or spontaneous) and those who had not.
The design of the study, Michels said, was much more reliable than previous studies, which started with women who'd had breast cancer and asked them if they had undergone an abortion. Such studies introduce "recall bias," she said, because women with breast cancer are more likely to acknowledge an abortion, thinking it might be related to their condition.
"There will always be some underreporting of abortion [among healthy women] because it is such a sensitive issue," she said.
Malec criticized the design of the study, which allowed reports of abortions that had occurred through 2003. Malec said that would tend to "dilute" cancer links because cancer in those women would not have had time to develop. "That's like asking, 'If you smoke cigarettes today, would you develop lung cancer tomorrow?' " she said.
Michels said 90% of the abortions and miscarriages reported in the study occurred before 1993.