When my wife became pregnant with our first child in early 2007, we discovered the joys—and the anxieties—that accompany parenthood even before a baby is born. For better or worse, modern medicine has taken some of the guesswork out of preparing for a child—you can determine gender (pink or blue crib sheets?), resolve questions of paternity, and test for the many things that can and occasionally do go wrong in a fetus's development. Even if prenatal testing doesn't allow the parents-in-waiting to sleep much better at night, it affords them the opportunity to plan for what's in store and to make better-informed choices on the most fraught of decisions: whether to continue the pregnancy to term.
But the testing regimen can itself be a source of anxiety, with some procedures presenting a risk to the fetus. And for such a high-stakes decision, medical science's assessment of the trade-offs involved can be rather unscientific. For one common prenatal test, amniocentesis, the common recommendation that women over the age of 35 should have the test may not only be unscientific, but exactly backward. The 35-year rule, it turns out, was based on a fallacy in reasoning that might have you questioning other treatment protocols on your next visit to the doctor's office.
Amniocentesis, a common screen at the end of the first trimester, involves the insertion of a needle into the uterus to extract a couple of tablespoons' worth of the amniotic fluid that surrounds the developing fetus. Lab tests on the extracted fluid provide a clear indication of whether the unborn child has Down syndrome or other genetic abnormalities. But the procedure carries with it an increased probability of miscarriage in the weeks that follow.
Is it worth risking a carefully nurtured fetus for information on an unborn child's genetic destiny? At the time my wife and I began to ask such questions, standard obstetric practice was that amnio or its alternative, chorionic villus sampling, be offered automatically to any woman over 35, but not to younger women without specific risk factors like a family history of genetic disorders. On the face of it, an age cutoff makes sense—fetuses of older women are at higher risk of genetic defects—and one would expect that the 35-year cutoff was developed on the basis of sound scientific evaluation.
It stands to reason that a woman should get amnio if the benefits of amnio exceed its costs. The clinical doctrine of testing after 35 would imply that the costs of amnio outweigh the benefits for younger women, but around age 35, the benefits begin to exceed the costs.
What are these costs and benefits? The cost is the increased probability of miscarriage from fluid extraction. (This higher risk is thought to be fairly similar for women of all ages.) The benefit of amnio is discovering genetic defects, the chances of which accelerate with age—the odds that a 20-year-old woman will give birth to a Down syndrome baby is a little under 1 in 2,000; this doubles by the time she's 32, and it's nearly 10 times higher by 38. There is similarly a steep increase in prevalence of other genetic abnormalities. So, for two otherwise comparable women—with likeminded attitudes toward abortion, comparable incomes, and similar family sizes—it seems reasonable to suggest that the one with a tenfold higher risk would get a bigger benefit from testing.
But why 35? The cutoff was first developed in the mid-1970s, based on estimates that the miscarriage risk from amnio was 1 in 200. For women under 35, this was greater than the odds of giving birth to a child with Down syndrome. For women older than 35, the ever-increasing probability of Down and other genetic defects was such that it exceeded this 1-in-200 threshold. The 35+ rule became firmly entrenched clinical gospel—in 2007 (just as new rules on genetic screening were announced), the New York Times quoted Dr. Deborah Driscoll, chairwoman of the obstetrics department at the University of Pennsylvania, as saying that "[i]t's been pretty much ingrained in obstetricians' minds that 35 is the cutoff age."