In 2002 Sylvia Ann Hewlett terrified career-oriented women when she hit them with a cold truth: Regardless of advances in medicine, it’s difficult for women to get pregnant beyond their early 30s. If women wanted to have children, her book Creating a Life urged, they better leave their desks and start planning. Now.
The book sparked a media flurry and spawned a decade’s worth of articles spouting largely the same statistics chronicling women’s waning fertility. “Fifty percent of women over 35 will fail to get pregnant over the course of eight months,” warned New York Magazine in 2010, “and after that the odds keep dropping.” In June, the Wall Street Journal warned readers that fertility rates “gradually begin to decline around age 32 and then rapidly decline after age 37.” The articles have become so frequent they have birthed a new genre: lady panic.
But in this month’s Atlantic cover story, Jean Twenge goes on a factchecking deep-dive through the most commonly bandied about statistics and comes to a pretty shocking conclusion: Everything you’ve ever heard about the biological clock has been wildly distorted. Take this report, for example:
The widely cited statistic that one in three women ages 35 to 39 will not be pregnant after a year of trying, for instance, is based on an article published in 2004 in the journal Human Reproduction. Rarely mentioned is the source of the data: French birth records from 1670 to 1830. The chance of remaining childless—30 percent—was also calculated based on historical populations.
“In other words,” Twenge writes, “millions of women are being told when to get pregnant based on statistics from a time before electricity, antibiotics, or fertility treatment.”
And, in other words, no journalist bothered to read (or report on) the methodology of the study.
Failing to mention the century of your statistics is pretty bad, but it gets worse. Twenge traced a number of the most incendiary facts, including the oft-spouted statistic that “20 percent of 30-year-old women and 5 percent of 40-year-old women get pregnant per cycle,” to pamphlets released from the American Society for Reproductive Medicine, which she called asking for citations. Apparently she’s the only journalist in a decade to probe the information: not only could they not provide corroborating studies, they admitted to massaging the statistics for simplicity.
Now, the onus for the botched facts doesn’t fall entirely on lazy reporting. Part of the reason that journalists are spouting historical figures is that scientists, too, have been relying on ancient data sets. Pregnancy, it turns out, is a difficult thing to evaluate. There’s only a short period to find couples trying to get pregnant before some of them become pregnant and drop out, throwing off the data pool. And by the time prospective parents reach their late-30s the pool is already corrupted by the super-fertile twenty-somethings who get knocked up early and don’t, but probably could, continue procreating into their late-30s and 40s. (Twenge recommends looking at studies of cycle viability, which monitors couples day-by-day as they attempt get pregnant.)
But as she deftly reports, had journalists bothered to review the research holistically, even the small body of contemporary research paints a more optimistic picture of the biological clock:
In one study… David Dunson (now of Duke University), examined the chances of pregnancy among 770 European women. It found that with sex at least twice a week, 82 percent of 35-to-39-year-old women conceive within a year, compared with 86 percent of 27-to-34-year-olds. (The fertility of women in their late 20s and early 30s was almost identical—news in and of itself.)
In CJR’s January cover story, “Survival of the Wrongest,” writer (and Atlantic contributor) David H. Freedman recounts the chronic inconsistency of reporting of health findings. Freedman blames this on the breadth and variation of scientific inquiry—where many incremental findings amount over time and frequency to a solid fact—allowing journalists hand-select the evidence that fits their final argument.
Twenge botches her reading of Dunson so freakishly that one has to assume that anything else she says about what she read is similarly untrustworthy.
She claims that Dunson shows age differences don't matter because the chances of a 35-39 couple conceiving at 2 days prior to ovulation are the same as a young couple at 3 days prior. In fact, Dunson's data shows that on any given day, the fertility rate of the younger couples is much higher than that of the older couples. The overall likelihood of pregnancy in a cycle is much greater for younger couples (so much for the suggestion that looking at cycle viability produces different results). Yes, it's trivially true that the fertility rate for 35-39 couples at Day Ovulation-2 is the same as for 19-26 couples on Day Ov-3. But that doesn't erase the age difference; it highlights it. It's saying that at the very best, the older couple can only achieve a level of fertility that younger couples achieve and surpass earlier in the cycle.
For CJR, the lesson is that just because someone is contrarian doesn't necessarily mean that they're accurate. I'm guessing that Fitts might have been better served by taking her own advice and actually reading the sources that Twenge purports to describe.
#1 Posted by Tom T., CJR on Mon 15 Jul 2013 at 12:50 AM
In fact we do not include full citations in our patient education materials.
However, we have a number of reports from our Practice Committee which include a host of references and are available on our web site.
http://asrm.org/uploadedFiles/ASRM_Content/News_and_Publications/Practice_Guidelines/Committee_Opinions/Age-related_fertility.pdf
http://asrm.org/uploadedFiles/ASRM_Content/News_and_Publications/Practice_Guidelines/Committee_Opinions/Testing_and_interpreting_measures_of_ovarian_reserve-noprint.pdf
Sean Tipton
Director, Public Affairs
American Society for Reproductive Medicine
#2 Posted by Sean Tipton, CJR on Mon 15 Jul 2013 at 10:12 AM
Tom, Twenge doesn't use the Dunson study to prove that both age cohorts are identical; she uses it to show that the data behind conception is much more complicated than often portrayed. You can extrapolate from the study that while older couples may have to pay closer attention to ovulation cycles, it is not completely impossible for them to reproduce. Clearly there is a difference, it's just not a completely dire difference.
#3 Posted by Alexis Sobel Fitts, CJR on Mon 15 Jul 2013 at 10:12 AM
This kind of super-skewed reporting on women's fertility is skewered in Susan Faludi's Backlash - written in 1990.
#4 Posted by fleurdamour, CJR on Mon 15 Jul 2013 at 02:52 PM
Interesting piece. Thanks for writing.
#5 Posted by Karina Ioffee, CJR on Wed 31 Jul 2013 at 12:57 PM