The Atlantic— by Jean Twenge
Deep anxiety about the ability to have children later in life plagues many women. But the decline in fertility over the course of a woman’s 30s has been oversold. Here’s what the statistics really tell us—and what they don’t.
In the tentative, post-9/11 spring of 2002, I was, at 30, in the midst of extricating myself from my first marriage. My husband and I had met in graduate school but couldn’t find two academic jobs in the same place, so we spent the three years of our marriage living in different states. After I accepted a tenure-track position in California and he turned down a postdoctoral research position nearby—the job wasn’t good enough, he said—it seemed clear that our living situation was not going to change.
I put off telling my parents about the split for weeks, hesitant to disappoint them. When I finally broke the news, they were, to my relief, supportive and understanding. Then my mother said, “Have you read Time magazine this week? I know you want to have kids.”
Time’s cover that week had a baby on it. “Listen to a successful woman discuss her failure to bear a child, and the grief comes in layers of bitterness and regret,” the story inside began. A generation of women who had waited to start a family was beginning to grapple with that decision, and one media outlet after another was wringing its hands about the steep decline in women’s fertility with age: “When It’s Too Late to Have a Baby,” lamented the U.K.’s Observer; “Baby Panic,” New York magazine announced on its cover.
The panic stemmed from the April 2002 publication of Sylvia Ann Hewlett’s headline-grabbing book, Creating a Life, which counseled that women should have their children while they’re young or risk having none at all. Within corporate America, 42 percent of the professional women interviewed by Hewlett had no children at age 40, and most said they deeply regretted it. Just as you plan for a corner office, Hewlett advised her readers, you should plan for grandchildren.
The previous fall, an ad campaign sponsored by the American Society for Reproductive Medicine (ASRM) had warned, “Advancing age decreases your ability to have children.” One ad was illustrated with a baby bottle shaped like an hourglass that was—just to make the point glaringly obvious—running out of milk. Female fertility, the group announced, begins to decline at 27. “Should you have your baby now?” asked Newsweek in response.
For me, that was no longer a viable option.
I had always wanted children. Even when I was busy with my postdoctoral research, I volunteered to babysit a friend’s preschooler. I frequently passed the time in airports by chatting up frazzled mothers and babbling toddlers—a 2-year-old, quite to my surprise, once crawled into my lap. At a wedding I attended in my late 20s, I played with the groom’s preschool-age nephews, often on the floor, during the entire rehearsal and most of the reception. (“Do you fart?” one of them asked me in an overly loud voice during the rehearsal. “Everyone does,” I replied solemnly, as his grandfather laughed quietly in the next pew.)
But, suddenly single at 30, I seemed destined to remain childless until at least my mid-30s, and perhaps always. Flying to a friend’s wedding in May 2002, I finally forced myself to read the Time article. It upset me so much that I began doubting my divorce for the first time. “And God, what if I want to have two?,” I wrote in my journal as the cold plane sped over the Rockies. “First at 35, and if you wait until the kid is 2 to try, more than likely you have the second at 38 or 39. If at all.” To reassure myself about the divorce, I wrote, “Nothing I did would have changed the situation.” I underlined that.
I was lucky: within a few years, I married again, and this time the match was much better. But my new husband and I seemed to face frightening odds against having children. Most books and Web sites I read said that one in three women ages 35 to 39 would not get pregnant within a year of starting to try. The first page of the ASRM’s 2003 guide for patients noted that women in their late 30s had a 30 percent chance of remaining childless altogether. The guide also included statistics that I’d seen repeated in many other places: a woman’s chance of pregnancy was 20 percent each month at age 30, dwindling to 5 percent by age 40.
Every time I read these statistics, my stomach dropped like a stone, heavy and foreboding. Had I already missed my chance to be a mother?
As a psychology researcher who’d published articles in scientific journals, some covered in the popular press, I knew that many scientific findings differ significantly from what the public hears about them. Soon after my second wedding, I decided to go to the source: I scoured medical-research databases, and quickly learned that the statistics on women’s age and fertility—used by many to make decisions about relationships, careers, and when to have children—were one of the more spectacular examples of the mainstream media’s failure to correctly report on and interpret scientific research.
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, millions of women are being told when to get pregnant based on statistics from a time before electricity, antibiotics, or fertility treatment. Most people assume these numbers are based on large, well-conducted studies of modern women, but they are not. When I mention this to friends and associates, by far the most common reaction is: “No … No way. Really?”
Surprisingly few well-designed studies of female age and natural fertility include women born in the 20th century—but those that do tend to paint a more optimistic picture. One study, published in Obstetrics & Gynecology in 2004 and headed by 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.) Another study, released this March in Fertility and Sterility and led by Kenneth Rothman of Boston University, followed 2,820 Danish women as they tried to get pregnant. Among women having sex during their fertile times, 78 percent of 35-to-40-year-olds got pregnant within a year, compared with 84 percent of 20-to-34-year-olds. A study headed by Anne Steiner, an associate professor at the University of North Carolina School of Medicine, the results of which were presented in June, found that among 38- and 39-year-olds who had been pregnant before, 80 percent of white women of normal weight got pregnant naturally within six months (although that percentage was lower among other races and among the overweight). “In our data, we’re not seeing huge drops until age 40,” she told me.
Even some studies based on historical birth records are more optimistic than what the press normally reports: One found that, in the days before birth control, 89 percent of 38-year-old women were still fertile. Another concluded that the typical woman was able to get pregnant until somewhere between ages 40 and 45. Yet these more encouraging numbers are rarely mentioned—none of these figures appear in the American Society for Reproductive Medicine’s 2008 committee opinion on female age and fertility, which instead relies on the most-ominous historical data.
In short, the “baby panic”—which has by no means abated since it hit me personally—is based largely on questionable data. We’ve rearranged our lives, worried endlessly, and forgone countless career opportunities based on a few statistics about women who resided in thatched-roof huts and never saw a lightbulb. In Dunson’s study of modern women, the difference in pregnancy rates at age 28 versus 37 is only about 4 percentage points. Fertility does decrease with age, but the decline is not steep enough to keep the vast majority of women in their late 30s from having a child. And that, after all, is the whole point.
I am now the mother of three children, all born after I turned 35. My oldest started kindergarten on my 40th birthday; my youngest was born five months later. All were conceived naturally within a few months. The toddler in my lap at the airport is now mine.
Instead of worrying about my fertility, I now worry about paying for child care and getting three children to bed on time. These are good problems to have.
Yet the memory of my abject terror about age-related infertility still lingers. Every time I tried to get pregnant, I was consumed by anxiety that my age meant doom. I was not alone. Women on Internet message boards write of scaling back their careers or having fewer children than they’d like to, because they can’t bear the thought of trying to get pregnant after 35. Those who have already passed the dreaded birthday ask for tips on how to stay calm when trying to get pregnant, constantly worrying—just as I did—that they will never have a child. “I’m scared because I am 35 and everyone keeps reminding me that my ‘clock is ticking.’ My grandmother even reminded me of this at my wedding reception,” one newly married woman wrote to me after reading my 2012 advice book, The Impatient Woman’s Guide to Getting Pregnant, based in part on my own experience. It’s not just grandmothers sounding this note. “What science tells us about the aging parental body should alarm us more than it does,” wrote the journalist Judith Shulevitz in a New Republic cover story late last year that focused, laser-like, on the downsides of delayed parenthood.
How did the baby panic happen in the first place? And why hasn’t there been more public pushback from fertility experts?
One possibility is the “availability heuristic”: when making judgments, people rely on what’s right in front of them. Fertility doctors see the effects of age on the success rate of fertility treatment every day. That’s particularly true for in vitro fertilization, which relies on the extraction of a large number of eggs from the ovaries, because some eggs are lost at every stage of the difficult process. Younger women’s ovaries respond better to the drugs used to extract the eggs, and younger women’s eggs are more likely to be chromosomally normal. As a result, younger women’s IVF success rates are indeed much higher—about 42 percent of those younger than 35 will give birth to a live baby after one IVF cycle, versus 27 percent for those ages 35 to 40, and just 12 percent for those ages 41 to 42. Many studies have examined how IVF success declines with age, and these statistics are cited in many research articles and online forums.
Yet only about 1 percent of babies born each year in the U.S. are a result of IVF, and most of their mothers used the technique not because of their age, but to overcome blocked fallopian tubes, male infertility, or other issues: about 80 percent of IVF patients are 40 or younger. And the IVF statistics tell us very little about natural conception, which requires just one egg rather than a dozen or more, among other differences.
Studies of natural conception are surprisingly difficult to conduct—that’s one reason both IVF statistics and historical records play an outsize role in fertility reporting. Modern birth records are uninformative, because most women have their children in their 20s and then use birth control or sterilization surgery to prevent pregnancy during their 30s and 40s. Studies asking couples how long it took them to conceive or how long they have been trying to get pregnant are as unreliable as human memory. And finding and studying women who are trying to get pregnant is challenging, as there’s such a narrow window between when they start trying and when some will succeed.
Another problem looms even larger: women who are actively trying to get pregnant at age 35 or later might be less fertile than the average over-35 woman. Some highly fertile women will get pregnant accidentally when they are younger, and others will get pregnant quickly whenever they try, completing their families at a younger age. Those who are left are, disproportionately, the less fertile. Thus, “the observed lower fertility rates among older women presumably overestimate the effect of biological aging,” says Dr. Allen Wilcox, who leads the Reproductive Epidemiology Group at the National Institute of Environmental Health Sciences. “If we’re overestimating the biological decline of fertility with age, this will only be good news to women who have been most fastidious in their birth-control use, and may be more fertile at older ages, on average, than our data would lead them to expect.”
These modern-day research problems help explain why historical data from an age before birth control are so tempting. However, the downsides of a historical approach are numerous. Advanced medical care, antibiotics, and even a reliable food supply were unavailable hundreds of years ago. And the decline in fertility in the historical data may also stem from older couples’ having sex less often than younger ones. Less-frequent sex might have been especially likely if couples had been married for a long time, or had many children, or both. (Having more children of course makes it more difficult to fit in sex, and some couples surely realized—eureka!—that they could avoid having another mouth to feed by scaling back their nocturnal activities.) Some historical studies try to control for these problems in various ways—such as looking only at just-married couples—but many of the same issues remain.
The best way to assess fertility might be to measure “cycle viability,” or the chance of getting pregnant if a couple has sex on the most fertile day of the woman’s cycle. Studies based on cycle viability use a prospective rather than retrospective design—monitoring couples as they attempt to get pregnant instead of asking couples to recall how long it took them to get pregnant or how long they tried. Cycle-viability studies also eliminate the need to account for older couples’ less active sex lives. David Dunson’s analysis revealed that intercourse two days before ovulation resulted in pregnancy 29 percent of the time for 35-to-39-year-old women, compared with about 42 percent for 27-to-29-year-olds. So, by this measure, fertility falls by about a third from a woman’s late 20s to her late 30s. However, a 35-to-39-year-old’s fertility two days before ovulation was the same as a 19-to-26-year-old’s fertility three days before ovulation: according to Dunson’s data, older couples who time sex just one day better than younger ones will effectively eliminate the age difference.
Don’t these numbers contradict the statistics you sometimes see in the popular press that only 20 percent of 30-year-old women and 5 percent of 40-year-old women get pregnant per cycle? They do, but no journal article I could locate contained these numbers, and none of the experts I contacted could tell me what data set they were based on. The American Society for Reproductive Medicine’s guide provides no citation for these statistics; when I contacted the association’s press office asking where they came from, a representative said they were simplified for a popular audience, and did not provide a specific citation.
Dunson, a biostatistics professor, thought the lower numbers might be averages across many cycles rather than the chances of getting pregnant during the first cycle of trying. More women will get pregnant during the first cycle than in each subsequent one because the most fertile will conceive quickly, and those left will have lower fertility on average.
Most fertility problems are not the result of female age. Blocked tubes and endometriosis (a condition in which the cells lining the uterus also grow outside it) strike both younger and older women. Almost half of infertility problems trace back to the man, and these seem to be more common among older men, although research suggests that men’s fertility declines only gradually with age.
Fertility problems unrelated to female age may also explain why, in many studies, fertility at older ages is considerably higher among women who have been pregnant before. Among couples who haven’t had an accidental pregnancy—who, as Dr. Steiner put it, “have never had an ‘oops’ ”—sperm issues and blocked tubes may be more likely. Thus, the data from women who already have a child may give a more accurate picture of the fertility decline due to “ovarian aging.” In Kenneth Rothman’s study of the Danish women, among those who’d given birth at least once previously, the chance of getting pregnant at age 40 was similar to that at age 20.
Older women’s fears, of course, extend beyond the ability to get pregnant. The rates of miscarriages and birth defects rise with age, and worries over both have been well ventilated in the popular press. But how much do these risks actually rise? Many miscarriage statistics come from—you guessed it—women who undergo IVF or other fertility treatment, who may have a higher miscarriage risk regardless of age. Nonetheless, the National Vital Statistics Reports, which draw data from the general population, find that 15 percent of women ages 20 to 34, 27 percent of women 35 to 39, and 26 percent of women 40 to 44 report having had a miscarriage. These increases are hardly insignificant, and the true rate of miscarriages is higher, since many miscarriages occur extremely early in a pregnancy—before a missed period or pregnancy test. Yet it should be noted that even for older women, the likelihood of a pregnancy’s continuing is nearly three times that of having a known miscarriage.
What about birth defects? The risk of chromosomal abnormalities such as Down syndrome does rise with a woman’s age—such abnormalities are the source of many of those very early, undetected miscarriages. However, the probability of having a child with a chromosomal abnormality remains extremely low. Even at early fetal testing (known as chorionic villus sampling), 99 percent of fetuses are chromosomally normal among 35-year-old pregnant women, and 97 percent among 40-year-olds. At 45, when most women can no longer get pregnant, 87 percent of fetuses are still normal. (Many of those that are not will later be miscarried.) In the near future, fetal genetic testing will be done with a simple blood test, making it even easier than it is today for women to get early information about possible genetic issues.
What does all this mean for a woman trying to decide when to have children? More specifically, how long can she safely wait?
This question can’t be answered with absolutely certainty, for two big reasons. First, while the data on natural fertility among modern women are proliferating, they are still sparse. Collectively, the three modern studies by Dunson, Rothman, and Steiner included only about 400 women 35 or older, and they might not be representative of all such women trying to conceive.
Second, statistics, of course, can tell us only about probabilities and averages—they offer no guarantees to any particular person. “Even if we had good estimates for the average biological decline in fertility with age, that is still of relatively limited use to individuals, given the large range of fertility found in healthy women,” says Allen Wilcox of the NIH.
So what is a woman—and her partner—to do?
The data, imperfect as they are, suggest two conclusions. No. 1: fertility declines with age. No. 2, and much more relevant: the vast majority of women in their late 30s will be able to get pregnant on their own. The bottom line for women, in my view, is: plan to have your last child by the time you turn 40. Beyond that, you’re rolling the dice, though they may still come up in your favor. “Fertility is relatively stable until the late 30s, with the inflection point somewhere around 38 or 39,” Steiner told me. “Women in their early 30s can think about years, but in their late 30s, they need to be thinking about months.” That’s also why many experts advise that women older than 35 should see a fertility specialist if they haven’t conceived after six months—particularly if it’s been six months of sex during fertile times.
There is no single best time to have a child. Some women and couples will find that starting—and finishing—their families in their 20s is what’s best for them, all things considered. They just shouldn’t let alarmist rhetoric push them to become parents before they’re ready. Having children at a young age slightly lowers the risks of infertility and chromosomal abnormalities, and moderately lowers the risk of miscarriage. But it also carries costs for relationships and careers. Literally: an analysis by one economist found that, on average, every year a woman postpones having children leads to a 10 percent increase in career earnings.
For women who aren’t ready for children in their early 30s but are still worried about waiting, new technologies—albeit imperfect ones—offer a third option. Some women choose to freeze their eggs, having a fertility doctor extract eggs when they are still young (say, early 30s) and cryogenically preserve them. Then, if they haven’t had children by their self-imposed deadline, they can thaw the eggs, fertilize them, and implant the embryos using IVF. Because the eggs will be younger, success rates are theoretically higher. The downsides are the expense—perhaps $10,000 for the egg freezing and an average of more than $12,000 per cycle for IVF—and having to use IVF to get pregnant. Women who already have a partner can, alternatively, freeze embryos, a more common procedure that also uses IVF technology.
At home, couples should recognize that having sex at the most fertile time of the cycle matters enormously, potentially making the difference between an easy conception in the bedroom and expensive fertility treatment in a clinic. Rothman’s study found that timing sex around ovulation narrowed the fertility gap between younger and older women. Women older than 35 who want to get pregnant should consider recapturing the glory of their 20‑something sex lives, or learning to predict ovulation by charting their cycles or using a fertility monitor.
I wish I had known all this back in the spring of 2002, when the media coverage of age and infertility was deafening. I did, though, find some relief from the smart women of Saturday Night Live.
“According to author Sylvia Hewlett, career women shouldn’t wait to have babies, because our fertility takes a steep drop-off after age 27,” Tina Fey said during a “Weekend Update” sketch. “And Sylvia’s right; I definitely should have had a baby when I was 27, living in Chicago over a biker bar, pulling down a cool $12,000 a year. That would have worked out great.” Rachel Dratch said, “Yeah. Sylvia, um, thanks for reminding me that I have to hurry up and have a baby. Uh, me and my four cats will get right on that.”
“My neighbor has this adorable, cute little Chinese baby that speaks Italian,” noted Amy Poehler. “So, you know, I’ll just buy one of those.” Maya Rudolph rounded out the rant: “Yeah, Sylvia, maybe your next book should tell men our age to stop playing Grand Theft Auto III and holding out for the chick from Alias.” (“You’re not gonna get the chick from Alias,” Fey advised.)
Eleven years later, these four women have eight children among them, all but one born when they were older than 35. It’s good to be right.