Quotes of the Day

Abortions increase the risk of low birth weight in future pregnancies by a factor of three, and of premature birth by a factor of two, according to the largest U.S. study of its kind. The study is hardly perfect; the data is more than 40 years old and doesn't distinguish between medical abortions and "spontaneous abortions," better known as miscarriages. Yet the report, published today in the Journal of Epidemiology and Community Health (JECH), shows one of the strongest links yet between miscarriage or abortion on premature birth and low birth weight — major risk factors for infant death or sickness.

What makes report significant is the size and detail of data. Some previous, smaller studies on abortion and future birth weight have suffered because researchers were unable to untangle the effects of abortion from, say, the effects of being poor (which also happens to increase a woman's odds of having an abortion). But the researchers behind the JECH study, which evaluated just over 45,000 single-child live births from 1959 to 1966, were able to adjust for an impressive array of confounding variables, including race, age, weight, height, marital status, occupation, the number of prenatal visits, the number of previous children, smoking and drinking habits, drug habits, infant gender and both parents' education levels.

That kind of rigor makes the new findings particularly important. The study not only found a link between abortion or miscarriage and low birth weight, but it also found that the risk appears to increase with every subsequent miscarriage or abortion. The accruing risk, says co-author Tilahun Adera at Virginia Commonwealth University, suggests that termination of pregnancy is a true cause of low birth weight and preterm birth rather than a variable associated with such conditions. "It's not just an association," he says. "The risk of premature birth increases with the increasing number of abortions."

Women who had had one, two or three prior abortions or miscarriages were three, five and nine times more likely, respectively, to have a low-birth-weight child, the data showed. Though it's still not clear why that's so, doctors theorize that the cervix may be weakened by miscarriage or abortion, increasing the risk of preterm birth later on. Or, it could be that uterine adhesions or infections from the terminated pregnancy slow the growth of the fetus in subsequent pregnancies.

Recent major studies from Australia and Canada have also concluded that miscarriages and induced abortions raise the odds of premature birth and low birth weight — but only modestly. (Those studies were able to distinguish women who had miscarried from women who had intentionally ended their pregnancies.) Many other studies have found no clear link at all. Perhaps that's because different study populations, taken from all over the world, involve different risk factors for premature birth; or it may be simply that the sample sizes in some studies were too small to pick up relatively small differences between women who had had abortions and those who had not.

The big question, however, is how well data from the 1960s really represents American women today. Back in the '60s, induced abortions were illegal in the U.S. It's possible that some women in the study had abortions but denied it — even to their doctors — or claimed to have miscarried. That makes the data harder to interpret. Illegal abortion techniques of the day, moreover, were no doubt cruder than abortion procedures today, and they may have caused more permanent damage to the reproductive system.

Indeed, the public-health implications of the JECH study may be more suitable for developing countries, says Adera — places where abortion is still illegal, and where prenatal care may be similar to what was offered in the U.S. half a century ago. Still, he says, all over the world, "Women need to be informed about these risks."

• Laura Blue
• A comprehensive study finds that abortion and miscarriage lead to low birth weight and preterm birth. But is 40-year-old data still relevant today?