“The incidence of rape resulting in pregnancy are very low.”
— Rep. Trent Franks (R-Ariz.), June 12, 2013
This column has been significantly updated.
Rep. Franks made this comment during a House Judiciary Committee debate over a bill that would ban abortions after 20 weeks of pregnancy, in which he opposed a Democratic amendment to make exceptions for rape and incest.
After a firestorm erupted, Franks later sought to clarify his somewhat ungrammatical comments, claiming he was referring to women seeking abortions in the sixth month. “Pregnancies from rape that result in abortion after the beginning of the sixth month are very rare,” he said. “This bill does not address unborn children in earlier gestations. Indeed, the bill does nothing to restrict abortions performed before the beginning of the 6th month.”
We’re not sure his clarification really tracks with the comment he originally made after this statement: “But when you make that exception, there’s usually a requirement to report the rape within 48 hours. And in this case that’s impossible because this is in the sixth month of gestation.” After all, how many women know they are pregnant after being raped?
Readers can listen to the audio recording above and judge for themselves.
In any case, Franks raises an interesting issue: What is the incidence of pregnancy after a rape? And is it much lower than rate of pregnancy after consensual sex?
Because of the violence and stigma associated with rape — as well as different definitions — there are a wide range of statistics concerning rape. The Centers for Disease Control and Prevention, for instance, estimates that nearly 1.3 million American women were victims of rape or attempted rape in 2010. (About half were actual rapes.) But RAINN, the Rape Abuse and Incest National Network, says about 64,000 women were raped between 2004-2005, citing Justice Department data.
Obviously, the number of rapes will make a huge difference in the number of rapes that result in pregnancy. But Franks spoke of “incidence,” which suggests he is speaking about the rate of pregnancy.
The most widely cited study of this question (embedded below) was published in 1996 by the American Journal of Obstetrics and Gynecology, which determined a national rape-related pregnancy rate of 5.0 percent per rape among victims between the ages of 12 and 45. The study was based on a survey of 4008 adult women, over a three-year period, which covered a range of questions on drug and alcohol abuse but also included questions intended to draw out information on sexual assaults.
All told, there were 315 rape victims of reproductive age in this sample, resulting in 20 rape-related pregnancies. In an interview, Dean. G. Kilpatrick, one of the researchers and director of the National Crime Victims Research and Treatment Center at the Medical University of South Carolina, acknowledged that, given the sample size, it might have been appropriate to provide a confidence interval. Still, he added, “most people who have looked at the methodology were pretty satisfied that this is a reasonable way to go about it.”
The study concluded that in one year, this rate would result in more than 32,000 rape-related pregnancies, which Kilpatrick said would be about 50,000 a year when adjusted to today’s population. RAINN, using the same incidence rate, calculates about 3,200 pregnancies.
Though the study is nearly two decades old, it has not been updated. The 5 percent figure is a little higher than a separate study that estimates the chance of getting pregnant from a single act of unprotected sex was 3.1 percent; a European study pegged the chances of getting pregnant as 25 percent at two days before ovulation but a 5 percent average over the rest of the cycle, though the possibilities rapidly dwindled within days of ovulation.
Indeed, a 2002 study speculated that the incidence of pregnancy from rape could be even higher — 6.4 percent — in part because women have no choice in refusing sex, whereas in consensual sex a woman may refuse if she thinks the chances of pregnancy are higher.
Franks’ spokesman did not respond to a request for comment.
Update: Some readers have suggested we misinterpreted Franks’ comments, with one pointing us to liberal and conservative commentary arguing that Franks was not repeating the controversial statement made by Todd Akin last year regarding “legitimate rape.” We agree that Franks’ comments were not like Akin’s, which is why we never mentioned Akin.
But it is worth noting that some readers believe Franks, in using the word “incidence,” was referring to either the number of abortions in the case of rape or the number of rape-induced pregnancies. The Guttmacher Institute says that one percent of women who have abortions say they are victims of rape and less than a half percent say they became pregnant because of incest.
We stand by our interpretation, but it demonstrates why it is risky for a politician to utter imprecise language on a subject such as abortion. The Pinocchio rating might be different depending on the interpretation.
The Pinocchio Test
To some extent, one could argue that whether an incidence rate of one out of 20 is “very low” or not is a matter of opinion. But, by inference, Franks’ comment only makes sense if he is comparing the rate from rapes to rates of pregnancy through consensual intercourse.
The available research suggests that not only is there little or no difference, but the rate of pregnancy after rape may actually be slightly higher. Under any definition, that does not qualify as “very low.” There is no reason to suggest any distinction.
Update, June 14: Ben Carnes, communications director for Franks, says that Franks misspoke and intended to refer to the number of abortions due to rape. We had originally awarded Four Pinocchios for Franks’ statement, but in light of the clarification, we have removed the rating from this column. We don’t try to play gotcha here, and might not have written a column if Carnes had responded to our initial inquiry. In any case, given that some readers interpreted Franks’s statement differently, it no longer appears appropriate to have a rating.
Check out our candidate Pinocchio Tracker