Search the Site

Episode Transcript

When Jessica Cohen was pregnant with her first child, her blood pressure started to rise. She knew that could be a sign of preeclampsia, a serious condition that obstetricians are always on the watch for. But her doctor suggested she come in for some extra blood pressure checks during her pregnancy. That was it.

When she was pregnant with her second child, the same thing happened again. But this time… her OB had a very different response.

COHEN: In addition to bringing me in, um, multiple times for blood pressure checks multiple times a week they were doing detailed ultrasounds and doing what are called anti-natal fetal surveillance checks and checking the health of the placenta and keeping very careful track of how the pregnancy was going.

The response was so different the second time around because of one simple fact: Jessica was now over age 35. That meant she was labeled as “A.M.A.” – advanced maternal age. In other words, she was high risk. And it made a huge difference, not only during her pregnancy, but also her delivery.

COHEN: I had many machines that beeped. I was definitely connected to many wires. I probably had a pretty, pretty medicalized birth, but it all ended okay.

Jessica isn’t complaining. She was happy to have the extra precautions. And she welcomed a healthy baby boy into the world. But the experience left her wondering about why age 35 set off so many alarm bells. And more importantly, did it make a difference? What was the benefit of labeling her and other women as high risk? And what were the downsides?

As luck would have it, Jessica is an associate professor at the Harvard School of Public Health. And so, as she was caring for her newborn, she was also planning a study. Birth of a baby, birth of an idea.

*          *          *

From the Freakonomics Radio Network, this is Freakonomics, M.D.

*          *          *

I’m Bapu Jena. I’m an economist and I’m also a medical doctor. Each episode, I dissect an interesting question at the sweet spot between health and economics. Today — when it comes to giving birth, the United States can be a dangerous place. The number of women who die during childbirth is extremely high compared to other developed nations, and so is the number of women who experience complications during pregnancy and birth. Why is it so dangerous to give birth in America? And for some women who are pregnant, could being labeled “high risk” actually make things better?

Let’s talk about age 35. What is it about turning 35 that makes pregnancy riskier for a mother and her baby? The answer, you might guess, is nothing. Sure, it’s true that older women are more likely to experience complications during pregnancy and in childbirth. The risk of maternal death, birth defects, miscarriage, premature births and other issues – they increase with age. But there’s nothing special about age 35. Turning 35 doesn’t automatically make this experience more dangerous. You don’t blow out the birthday candles and suddenly become more susceptible to preeclampsia. This marking point exists simply so that obstetricians everywhere follow the same standard. It gives doctors a really rough estimate of when the process of childbirth is more likely to get complicated, and to require additional medical care. But for all intents and purposes, 35 is really an arbitrary cutoff point. It’s a midway point between 25, when complication risks are relatively low, and 45, when the risks are higher.

The question that Jessica had was: Does the extra care matter? Should all women be receiving the same care given to women who are labeled as high risk?

COHEN: Trying to disentangle the relationship between providing extra services and ultimate outcomes is really tricky because for the most part older women get more services, but older women also have worse outcomes.

As complicated as it is to unravel the link between the level of care pregnant women receive and their outcomes, it’s crucial work. Because, despite standards like the advanced age cutoff, women experience troubling disparities during pregnancy. For example, infant and maternal mortality rates are three times higher among Black women compared to white women. And that’s just one example.

COHEN: How much prenatal care someone gets has a lot to do with their socioeconomic status, their race, ethnicity, and a bunch of other factors that are also correlated with the mom’s and babie’s outcomes. We think that some women get “ too much” prenatal care like way more ultrasounds than they might need and some definitely don’t get enough. we know for sure that some women who should be seeing specialists, maternal fetal medicine specialists are not.

But is the gulf in outcomes a direct result of the gulf in prenatal care? If all women received the same care, would it erase the disparities?

COHEN: On the one hand, paying closer attention could get women diagnosed, for example, with preeclampsia or conditions earlier in pregnancy. It could avert stillbirth, which is part of the objective. On the other hand, of course, paying much closer attention, can also lead to additional intervention that may not have been needed or warranted. It’s really hard to weigh those trade-offs.

Jessica’s study was a brilliant way to sort whether more medical care for pregnant women could, on average, improve outcomes, specifically among women who were older. Her idea was simple. Take a set of women who were just a few months shy of age 35 during their pregnancy and a set of women who had just crossed that threshold, compare the kind of care they received and then look at their outcomes. She and her collaborators obtained data on more than 50,000 women who fit the bill.

COHEN: The first thing that we looked at is an indicator called severe maternal morbidity. This is an index that combines a range of severe outcomes for the mom. These are things like needing a blood transfusion or having, um, eclampsia which can trigger seizures.

They looked at the rate of stillbirths and also perinatal mortality. That’s an important term here — it means a fetal death anytime from 28 weeks of pregnancy through seven days after delivery — in other words, a late stillbirth or an early newborn death. And they also looked at premature births and low birth weight, which can indicate the overall health of an infant. The researchers compared all of these outcomes among the two groups of women on either side of age 35.

So, what did they find? Well for one thing, they confirmed that women just above age 35 received substantially more intensive prenatal care than women just a bit below age 35.

COHEN: Detailed ultrasounds increased by almost 16 percentage points at age 3., and antepartum fetal surveillance increased by almost five percentage points. These are pretty, um, pretty big changes for these tasks for women who are just a few months apart from one another.

So, women just above age 35 got more medical care. But, did these women do any better? The researchers found that there was no difference in maternal morbidity. Then they looked at the infants. The number of infants born prematurely or with low birth weight weren’t higher among women just under age 35. So, maybe that extra care wasn’t really having any benefit. But then the researchers spotted something.

COHEN: What we see alongside the jumps in prenatal care intensity and monitoring just above age 35 is a steep decline in perinatal mortality for women who are just above age 35, relative to women who are just below it. Perinatal mortality was almost eliminated completely for women who are just above age 35. It was like 80 percent or more decline.

Because the only difference between the two groups of women was the amount of care they received, Jessica reasoned that this was the cause for the decline in perinatal deaths among pregnancies that were labeled “advanced maternal age”.

COHEN: I think it is logical to say that the additional attention, the additional services that women just above age 35 are receiving, are leading to demonstrable benefits, in terms of the survival of the baby.

Now, it’s important to note the researchers couldn’t say which of the extra intervention mattered. Was it more prenatal visits that made the difference? More ultrasounds? More specialists on the case? A combination of some — or all — of the above?

COHEN: It could be that during the delivery the doctor is on higher alert or the nurse is on higher alert. Or one thing I think about a lot, um, is how seriously are patients taken when they raise a concern? So, one thing you hear a lot in the messaging around trying to avert stillbirth is paying attention to fetal movement. So, asking women to notice i less movement feeling fewer kicks, that kind of thing. So, you could imagine a scenario where a woman calls her doctor and says, “I’m feeling a little less movement.” Maybe when you’re over age 35, that triggers a sort of visit right away. And when you’re younger, a little younger, or you don’t have that flag on your chart, they tell you to, drink a sugary beverage, lie down, and count the kicks.

The study was published at the end of 2021 and Jessica is now trying to look further into that last question. Are some women’s concerns taken less seriously than others during pregnancy? If so, are there racial and socioeconomic differences between the two groups that would explain why some women are treated differently than others?

COHEN: I really think that the evidence from this study suggests that many more of these deaths are avertible than, um, than we realize. Um, and you know, the key is figuring out how to do it.

We started this episode with some stark data about the state of childbirth in the United States. Jessica’s study showed that being classified as high-risk could actually make pregnancy and delivery safer.

Her findings aren’t alone. There’s another study, from back in 2010, that gets at this issue from a different angle. Columbia University economist Doug Almond and his coauthors looked at the care given to newborns with low birth weight. Just like women are considered high risk once the clock hits 35 years, infants are considered high risk when they weigh less than three and a quarter pounds. And that label garners them extra care. So, just like Jessica’s study, Doug and his colleagues looked at whether that extra care made a difference. They found that newborns who weighed less than three and a quarter pounds had a lower one-year mortality rate than newborns just above this cutoff point — even though, in general, the more a baby weighs, the more likely it is to survive.

In both this study and Jessica’s, being labeled high risk meant more care and better outcomes. But is that the final word? What about when the problem isn’t too little care but too much? That’s after the break.

*          *          *

Generally speaking, we tend to think it’s a bad sign when someone gets labeled as a high-risk patient. By definition, it means that the risk of bad outcomes is higher, and that’s not a good thing. But Jessica’s study pointed out an interesting twist. That label actually carries some power. It gets us more care. In the case of childbirth, for some babies, that label could make the difference between life and death. So, based on this research, you might be tempted to conclude that all pregnant women should be labeled as high risk, just so they all get more medical care.

Well, as you can probably guess by the fact that we’re only halfway through the episode, that’s not the end of the story. The question is: Are there times when extra care makes pregnant women worse off, maybe because it leads to interventions that they don’t need?

Take, for example, cesarean deliveries. The current c-section rate in the U.S. is about 32 percent. In other words, a third of all babies are delivered this way. That’s — well, a lot. But it wasn’t always like that.

SHAH: we intervene in childbirth in the United States with major abdominal surgery 500 percent more than we did in the early 1970s. And yet even though it’s a surgery that’s designed to rescue people, term infants are 0 percent better off, and an American today is 50 percent more likely to die in childbirth than her own mother. And then the story just gets even more bizarre from there when you try and explain or understand why.

That’s Neel Shah. He’s an obstetrician at Harvard Medical School. He’s also the chief medical officer at the Maven Clinic, a virtual health clinic for fertility and pregnancy care.

SHAH: And as an obstetrician myself, I had a hard time reconciling the fact that when I do a C-section personally, I’m always right. But we can’t all be right if we’re doing that much surgery. And it’s not just the fact that we do a lot of it.

Neel started investigating the high rate of C-sections in the U.S. back in 2013 and uncovered a peculiar statistic.

SHAH: When you look, not just across time, but across geography, there’s almost no other health care service that varies more in terms of utilization. So, C-section rates at some hospitals are 7 percent, and at other hospitals, it’s 70 percent. And then it’s the only health care service where, after you account for the risk of the population that you’re taking care of, you see more variation, not less.

In other words, even among the subset of pregnant women who could be at risk of needing a C-section, there’s a huge amount of variation in whether or not they actually get one. So, what’s going on? Do a third of all childbirths actually require cesarean delivery? If not, why are there so many of them?

Now, there are lots of medical reasons to do a C-section. Doctors may opt to do a C-section if labor stops progressing, or the fetus is in a difficult position for delivery, if the baby or mother is in distress, or several other medical reasons. But the decision is still up to the O.B. It’s a judgment call. It’s not like driving directions telling us to turn right or left. There is always an element of guesswork — of discretion. And figuring out why the rate is so high — and so much higher than it used to be — is important because the fact is, it’s a risky procedure.

SHAH: The odds of hemorrhage, infection, organ injury, they’re three times higher with a C-section than with a normal delivery. It’s a longer recovery, which is really, really hard to recover from a surgery while taking care of a newborn infant. And then the other thing is that obstetricians are the only surgeons that cut on the same scar over and over again, because most moms have more than one baby. So, if a vascular surgeon has to go back and operate where they operated before, that’s a bad day in their work week, but for an obstetrician, that’s a Tuesday. And the first time you do a C-section, it’s a pretty straight-forward surgery, but the second or third time, you have to cut through all of that scar tissue. And sometimes it’s like a melted box of crayons. And the placenta, which is an organ that only exists in pregnancy, is a big bag of blood vessels that gets 25 percent of everything that the heart is pumping. And if the placenta gets caught up in that scar tissue, people can bleed to death. It’s called placenta accreta. And that condition has become 800 percent more common today compared to the 1970s. So, that’s the consequence that we’re talking about.

Neel wanted to see if he could figure out what had driven the C-section rate in the U.S. so high. He had his theories — and maybe you do too, or maybe you’ve heard some. One theory is that hospitals get paid more by insurance companies for C-sections than for vaginal deliveries. While that’s true, Neel didn’t find much evidence that money was a key reason that O.B.s chose do C-sections, in part because the payments that O.B.s get are much smaller than the payments that hospitals get and for any given woman, the O.B. is the one making the decision to do a C-section or not.

But some research suggests financial incentives could play a role. One research paper calculated that O.B.s do often earn a few hundred dollars more for a C-section than for a vaginal delivery. Back in 2013, a group of researchers looked at whether that incentive plays a role. They compared the rate at which pregnant women who are also physicians are ushered into surgery versus pregnant women who weren’t physicians. They found that physician-mothers were less likely to undergo C-sections than non–physician mothers. The researchers interpreted that data as an indication that at least some doctors were incentivized to perform C-sections for reasons that they couldn’t fully justify when the patient was also a doctor — for example, money.

But even if financial incentives do play a role, these almost certainly wouldn’t account for all of the 3,000 or so C-sections performed every day in the U.S. Neel and others have also wondered about other explanations. Maybe the high rate of C-sections is because doctors are worried about being sued if a vaginal delivery goes poorly. Some evidence suggests this might be true. Maybe doctors decide to do C-sections for reasons having nothing to do with the health of the mom and the baby. For example, maybe a delivery is taking a long time and the hospital needs the bed. Or maybe OBs turn to surgery when they want to get home; that would explain why C-section rates spike at the end of the day. But even putting all these together, Neel felt that these couldn’t explain the enormous growth we’ve seen in C-sections.

There’s another theory that I think a lot of people hold — maybe it’s the mothers. Women are having babies at older ages today, and so the concern about potential complications is more common, leading to more elective C-sections. There’s also more obesity, more hypertension, more diabetes, more twin births due to I.V.F.  — all of these could contribute to the decision to do a C-section. And on top of that, a small percentage of women seem to request C-sections instead of delivering naturally., Neel looked into all of this. Here’s what he found.

SHAH: All of the conventional wisdom doesn’t bear out in trying to explain this. It’s true that some people want a C-section, but they are the vast minority. Less than a half percent of people electively request a C-section. So, that’s very, very few people and certainly not what’s driving the, um, statistics that I described. And it turns out that even a young, healthy 25-year-old today is many times more likely to get a C-section than in the 1970s. And so, fundamentally, what is absolutely true is that you cannot blame women or moms. You can’t blame the C-section rate on them. It’s not them.

Okay. So, with all of the conventional explanations ruled out, what was left? Well, Neel traveled around the country looking at hospitals, trying to figure out the difference between a hospital with a 7 percent C-section rate and one with a 70-percent C-section rate. He did a bunch of studies, unpacking the problem one question at a time. And there’s one that especially caught my eye. At some point in his quest, Neel started wondering about nurses. During most hospital births, nurses are the ones who are closest to the moms. They are the monitors, the coaches, the primary caregivers. Across all areas of medicine, nurses really are the unsung heroes.

But as Neel started trying to uncover why some hospitals have high C-section rates and others have low ones, something surprised him. It looked like some labor and delivery nurses took care of a lot of moms that ended up giving birth by C-sections and other nurses took care of very few. And, by the way, the nurses — whose patients ended up being more likely to have C-sections — weren’t specialized in any particular way and their patients weren’t otherwise at higher risk of needing a C-section. So, Neel wondered whether the way that a given nurse manages and supports a person in labor could be contributing to the wide variation in C-section rates across the country. So, he and several colleagues took data from more than 3,000 births attended by 72 different nurses. And they looked to see if the rate of C-sections varied according to the nurse who was attending the birth. And, well —

SHAH: We found that the nurse that gets assigned to the patient can influence the odds of getting a C-section fivefold.

Among all those births, 26 percent were cesarean deliveries. But here’s the thing — the range of C-sections per nurse stretched from 8.3 percent to 48 percent. For at least one of those 72 nurses, 48 of every 100 childbirths they attended ended with C-sections. For another nurse, that was true for fewer than ten out of every 100 births they attended.

SHAH: I was not surprised to find that the nurse that gets assigned to you has a role in whether or not you get a C-section. I was surprised to find how much of a role they can potentially have. And what that really means is that childbirth is a team sport, and the odds of getting a C-section are not just based on the person giving birth or the obstetrician or the nurse. It’s all three together.

Neel and his colleagues wondered if it could be that some nurses were simply assigned the more complicated labors. That would explain the variation. But a further analysis ruled out that possibility. So, Neel’s conclusion was that some nurses have higher C-section rates, and some have lower C-section rates.

Why? Well, There’s a host of issues at play here. To start, it may be helpful to go back to the 1970s and the introduction of electronic fetal monitoring, which watches the baby’s heartbeat during labor. These devices are great for telling us when the baby is completely fine, or something is definitely wrong. But anything in between is ambiguous. There isn’t clear evidence on how to interpret the information. Combine that with the extremely low tolerance for risk during childbirth and you create the potential for a lot of overreaction based on this device.

That’s not all. Neel also found that the way hospitals are designed often blends labor and delivery with surgery right from the get-go. In particular, operating rooms are typically right next to the labor rooms. Also, there aren’t any guidelines on how many labor and delivery rooms a hospital needs to have related to the number of mothers who give birth there each year. A hospital with a hundred births per year could have the same number of rooms as a hospital with a thousand births a year. And, as you can probably guess, hospitals with a lot of deliveries tend to have higher C-section rates, possibly because they don’t have enough rooms.,

But there’s one other important factor and it’s more about mindset than anything physical. It’s about choice and risk. See, the thing about C-sections is that they can never be proven to be the wrong decision.

SHAH: You’re always incentivized to do the C-section. Whenever you do a C-section, the reason that I believed I was always right, is because if the baby comes out looking perfect, I would think, “It’s a good thing I did a C-section. I just saved the day.” And if the baby comes out looking lackluster with low Apgar scores, I think, you know, it’s confirmation bias. “That was a sick baby. It’s a good thing I did a C-section.” So, either way, it’s pretty good to be me. I’m always right.

And interestingly, nobody ever gets labeled “low risk” when it comes to childbirth. It’s either high-risk or nothing.

And that’s what so much research surrounding childbirth comes back to — how we think about risk. There’s more to it than just deciding on the riskiness of any specific birth. It’s about how we move between individuals and population-level data. It’s easy to look at the high C-section rate in the U.S. and think, “Ok, there must be some unnecessary C-sections going on out there.” But it’s another thing to decide, when a 38-year-old woman has been in labor for 24 hours, whether she does or doesn’t need a C-section. It’s so much easier to just go for the operation.

We see this dance between over- and under-reacting all the time in childbirth. There’s a study by Vini Singh, a health economist at UMass-Amherst that illustrates it so perfectly. She looked at electronic health records from 86,000 deliveries to try to spot what influences O.B.s to choose surgery over vaginal birth. She found that when a delivery had complications, doctors were far more likely to switch to the other delivery mode on the very next birth they attended. So, a complicated vaginal birth led to a cesarean on the next one, and vice versa.

That kind of thinking is what we call a heuristic — a shortcut, basically. In addition to thinking through each case, the doctor also — maybe inappropriately — relies on recent prior experience to tell him or her what to do. Vini looked at a bunch of other possibilities that could have shaped the decision and none of them changed her conclusion. And unfortunately, this approach — which is really more of a reaction than a decision —— resulted in slightly more problems during childbirth.

One of the reasons we want to understand the dynamics at play in the medical decisions surrounding childbirth is cost. The total spending on healthcare in the U.S. exceeded four trillion dollars in 2020, almost 20 percent of G.D.P. And the medicalization of childbirth is a meaningful part of that. Here’s Neel Shah again:

SHAH: Of our entire G.D.P., just the cost of hospitalizing moms and babies for a birth is 0.6 percent of it. If you were to take our whole G.D.P. of our country — and spread it out across the table, you can see the cost of hospitalizing moms and babies to give birth with your naked eye.

And yet maternity leave, paternity leave, and everything else following birth is woefully underfunded.

So, on one hand, providing more care to older pregnant women could, on average, improve their child’s outcomes, just like Jessica Cohen saw with her study on maternal age. But on the other hand, providing that advanced care to every mom may not make sense and for some moms, may do more harm than good. So, how should doctors balance evidence from big statistics with the fear of making the wrong decision for the patient right in front of them? Both Neel and Jessica spoke about the need to listen more carefully to mothers.

COHEN: Pregnancy and especially delivery is an evolving situation. Someone can go from low risk to high risk pretty quickly. It’s really something that needs to evolve with the pregnancy and with the situation. It’s hard to stick up for yourself in the doctor-patient dynamic and making sure that your concerns are heard your voice is heard and that, if you have a concern that it’s taken seriously is really important.

Jessica’s observations as a mother and a researcher who has studied these issues are so important. But they’re not just specific to the care of pregnant women.

SHAH: Maternal health really is a bellwether for the wellbeing of society as a whole. And the outcomes that we see are at the intersection of gender inequity, racial inequity, geographic inequity, and honestly, like, generational inequity. It is harder to start and grow a family today. It is more dangerous from a maternal mortality standpoint, and it’s really connected to, uh, the broader narrative in our country right now about hope and opportunity, feeling like it’s eroding for for young families in particular, because it’s scarier to have a baby and start a family right now. And all these things are related.

I want to end today’s show on a little bit of a lighter note, with some of my own observations. You know, having a child is a strange thing. It’s one of the most exciting things that can happen to a person but it’s also one of the scariest, even when everything goes well. From before a child is born to every day after, every decision that we make feels like a life-or-death event. Did my pregnant wife’s grilled cheese sandwich touch my cold cut sandwich? I can’t remember. Can I warm up this baby bottle in the microwave? Is that safe? Is this swaddle too tight? Is it true that breast milk is more valuable than bitcoin? And why does Google keep taking me to these Mommy blogs?

Well, in all seriousness, it turns out that for many of us, we aren’t equipped to make the real life or death choices on our own. We need help from the healthcare system. And we should expect better.

Anyway, that’s it for Freakonomics, M.D. this week. I hope you enjoyed our discussion today. You can find links to all the studies we mentioned at

As always, I want to thank you for listening. It’d be great if you could give us a review on Apple Podcasts or wherever you’re listening. It helps new people discover the show.

And as I’ve said before, if you have any thoughts on the show, follow Freakonomics, M.D. on Twitter at DrBapuPod — that’s D-R B-A-P-U P-O-D — and share your thoughts! You can also still shoot me an email at

*          *          *

Freakonomics, M.D. is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and People I (Mostly) Admire. All our shows are produced by Stitcher and Renbud Radio. You can find us on Twitter and Instagram at @drbapupod. Original music composed by Luis Guerra. This episode was produced by Jessica Wapner and mixed by Eleanor Osborne. The supervising producer was Tracey Samuelson. Our staff also includes Alison Craiglow, Greg Rippin, Rebecca Lee Douglas, Morgan Levey, Zack Lapinski, Ryan Kelley, Mary Diduch, Jasmin Klinger, Emma Tyrrell, Lyric Bowditch, Jacob Clemente, Alina Kulman, and Stephen Dubner. If you like this show, or any other show in the Freakonomics Radio Network, please recommend it to your family and friends. That’s the best way to support the podcasts you love. As always, thanks for listening.

JENA: This is Bapu. I’m here at a dinner party with some friends, trying to get some information about today’s show. Hey, Kent, have you ever spilled breast milk on the counter?

KENT: Oh, I’m not going out like that.

Read full Transcript


  • Jessica Cohen, professor of health and behavioral economics at the Harvard School of Public Health.
  • Neel Shah, professor of obstetrics, gynecology and reproductive biology at Harvard Medical School and chief medical officer of Maven Clinic.