Are modern children born too late?

Being a modern woman is unnatural. For that reason, deliberately unnatural childbirth might be the best way forward.

I live closer to nature than most Westerners. And still, I'm a persistent proponent of unnatural childbirth. I never have understood the ideal that childbirth should be as natural as possible. Why reject science and technology in an area where it is unusually useful?

In spite of this persistent bias in favor of the unnatural, I was surprised by the results of a large-scale randomized trial in 2018, called the Arrive Trial. It compared pregnant women who agreed to be randomized to either have their labor induced at 39 weeks and another group who were advised to wait for spontaneous start of labor until at least 40 weeks and 5 days (normal practice at the time). There was no significant difference in outcomes for the children. But there was a 16 percent reduction in C-sections in the group that were induced at 39 weeks.¹

The results were rather surprising, because induced labor is associated with a higher rate of C-sections when compared to labor starting spontaneously at the same time of pregnancy. The Arrive Trial indicates that just like so much else in obstetrics, that is a question of selection bias: The women who give birth spontaneously at term are simply better equipped to give birth in the first place. 

A cleverly made observational study of 55 000 births from 2018 aimed at getting around this selection bias. Instead of comparing labor inductions with spontaneous labor at the same time, it compared labor inductions with pregnancies that did not result in spontaneous labor at that time: Data from births that were induced in week 39 and 40 were compared with pregnancies that were ongoing on those dates and resulted in spontaneous och induced labor on a later date. That study came to the same conclusion as the Arrive Trial: For first-time mothers, inducing birth at 39 weeks reduced the risk of having a C-section later. For women who had given birth before, there was no significant difference.²

At 39 weeks, only about 20 percent of pregnant women will have given birth spontaneously.³

If all women who are expecting their first child would have been treated like the intervention group in the Arrive trial and had elected inductions during the first half of week 39, about 75 percent of births would start as inductions. That would in effect put an end to the ideal that nature should decide the timing of childbirth.

It is unlikely that 75 percent of today's pregnant women would agree to have their labor induced in week 39. Only 27 percent of the eligible participants for the Arrive trial agreed to participate in the study. We should expect that those 27 percent were self-selected according to some parameters. Maybe they were more tired of their pregnancies than the 73 percent who did not agree to participate. Still, the trial indicates that offering induction to women who want inductions is likely to have beneficial effects.

When digging through some data, I realize that I shouldn't be too surprised by the Arrive Trial. During the last 20 years, a number of randomized studies comparing induction of labor to expectant management have reached similar results. In those studies labor was most commonly induced after 40 weeks rather than after 39 weeks, but the results are in the same direction: The induction groups have better results in terms of survival of the babies, health problems of the babies and C-sections. Being pregnant too long simply seems to be no good.⁴⁵⁶

What's wrong with nature? 

Although I'm absolutely no proponent of natural birth for the sake of it, I was still surprised that nature can be this bad. There is a plausible explanation why the birth of humans is more difficult than the birth of most mammals (including our cousins the great apes) called The Obstetrical Dilemma: Since humans are large-brained primates that walk upright, fetuses need big heads to make room for a large brain and mothers need narrow pelvises in order to be able to walk upright. These two factors have caused a delicate balance between the need to have intelligent children and the need to move efficiently, resulting in a high rate of mortality for mothers and infants. 

That high rate of mortality over probably hundreds of thousands of years should have created an effective selection process for adaptations that mitigate the conflict between head size and pelvic size. And it has. Human fetuses are forced out into the world through a series of intricate rotations, something that is not the case for other great apes.

But why wouldn't evolution also have selected for the optimal timing of birth? Pregnancies actually do not last on average 40 weeks, as is commonly said. That suspiciously even number was more or less an educated guess made up in the 19th century, before large-scale statistics were invented. One study found out that the median pregnant woman gives birth at 40 weeks plus 5 days if it is her first pregnancy and at 40 weeks plus 3 days in subsequent pregnancies.⁷

When scientists try to estimate the ideal time to give birth, they tend to arrive at dates much earlier in pregnancy. "The existing evidence suggests that most complications of pregnancy are minimized at about 39 weeks of gestation", an article in Nature from 2006 states⁸. Children die the least, get the least breathing problems and emergency c-sections are the fewest. Longer-term studies haven't managed to pin-point a single ideal week to be born. But they indicate that being born too early or too late is associated with increases in behavioral problems, being born too early was associated with lower income and being born too late with higher rates of disability.⁹ ¹⁰

It is very difficult to find the reason why most children are born at some date between 40 and 41 weeks instead of at 39 weeks: More children die in the womb after 39 weeks, labor seems to get more difficult and longer-term prospects seem no better for children born after 39 weeks. The whole thing is puzzling. Among humans, pregnancy and labor are difficult enough as they are. Why would women have evolved to give birth too late?

Born too late?

I can think of one plausible explanation: The environment has changed compared to when humans evolved. What if most babies in history were actually born on an optimal date?

My evidence to support that idea? Very little, unfortunately.

To start with, no one knows why babies are born in the first place. What mechanism makes labor start? Nobody knows. Scientists have managed to detect hormonal changes that occur between two and four weeks before spontaneous labor starts. Those changes occurred both in pregnant women who gave birth prematurely and within the normal range.¹¹

That indicates that whatever makes labor start, it is not a sudden chock, but a longer-term process. As if a pregnant woman's body is making statistics on her fetus' level of maturity and her capacity to carry it, and then evaluates the data weeks before delivery actually occurs.

Pregnancy length is obviously also determined by genetic factors. In one study of data from Sweden, pregnant women who had had one too long pregnancy (more than 41 weeks) ran a 4.4 time greater risk than average of having another too long pregnancy, if her next child had the same father. If she was expecting a child with another father, her risk of overdue pregnancy was 3.4 times the average. A woman with a sister who had an overdue pregnancy runs 1.8 times the average risk of having one herself.¹²

Scientists know there are environmental factors too. Low socioeconomic status is a risk factor for premature labor, while the risk profile for overdue pregnancy has more to do with age. Physiological and psychological stress is a risk factor that labor starts prematurely. More premature births have been noticed among women with certain kinds of heavy jobs.¹³¹⁴¹⁵

Stressed to give birth? 

Scientists know that psychological stress and some kinds of hard work increases the rate of premature birth. From that information I'm prone to make the assumption that labor in general is caused by accumulated stress. Internal or external. For some pregnant women, very little stress is required to start labor. Those women are kept lying down for months, since the stress of standing upright would cause their children to be born prematurely. Other women require a lot of stress for labor to start. Their babies grow bigger and bigger and sometimes even die before labor starts. 

If stress is what triggers labor, then it would be logical if external stress is a variable factor that could make the average birth occur a bit earlier. Maybe even at those ideal 39 weeks on average?

Unfortunately, I have found almost no good data that supports this hypothesis. Only rather bad data that supports it very loosely.

To start with, it is clear that in premodern societies, pregnant women were forced to lead much more physically stressful lives than in present-day Western society. In numerous primitive societies, women were the main breadwinners. 19th century anthropologist Carl Lumholtz described the women of a northern Australian aboriginal tribe as providers and explained that having several wives who gathered for him was the only way a man could be rich. Men hunted mostly for sport, while the women were the serious food providers, Lumholz explained.¹⁶

There are many similar, if less extreme, reports from horticultural societies, where women performed most agricultural labor, firewood collection, cooking and childcare. The idea that war and hunting is what men should do and labor is what women should do occurs in primitive societies all over the world.

Not only were women the laboring sex. Their labor was also more individualistic than that of men. In Catching Fire, Richard Wrangham explains that the extensive sharing that hunter-gatherers are so famous for only really applies to the male side. For a man, getting married was as much about getting a stable supply of vegetable foods as getting sexual access to a woman. Despite extensive sharing among men, meat access was unstable, and men profited from monopolizing a share in the results of a woman's gathering.¹⁷

Napoleon Chagnon wrote that Yanomamö women almost always spent late afternoons gathering firewood. They weren't cooperative: If one woman found a good deposit of dry wood close to the village, she made sure to collect it all so no one else would get it.¹⁸

What possibilities does such a system leave for taking time off during pregnancy? Unfortunately I can't write an email to Napoleon Chagnon and ask. He died in 2019, at the age of 81. 

Ask the living

Being dead is more or less the rule among anthropologists who have seen almost uncontacted, primitive societies. But a few are also still alive. For example, Napoleon Chagnon's graduate student-turned-enemy, Kenneth Good. In the early 1990s, Kenneth Good got (in)famous for having married a very young Yanomamö woman called Yarima and brought her to the United States. The couple had three children before Yarima decided to stay in the jungle after a visit there, leaving behind their children in Kenneth's care under disorganized circumstances.

David, the oldest of those three children, was then five years old. He grew up shaped by the experience of suddenly losing his mother. At the age of 24, he managed to organize an expedition to the Amazon rainforest and reunited with her. He also encountered several half-siblings, most notably a younger brother who had taken the name Ricky Martin. A biologist by training, David Good subsequently came to study the gut flora of the Yanomamö.

I couldn't find any way to reach Kenneth Good, who is now 81 years old and seems to have retired. But David Good, now 36, published a book in 2015 and has appeared in newspapers and podcasts. The people behind one of those podcasts forwarded an email from me to David Good, but warned me that David is often on long trips into the deep Amazon rainforest.

That was as far as I came by now. I really hope to get in touch with David Good sooner or later, because there is a glaring lack of studies and anecdotes of pregnancy in primitive societies. Anthropologists recorded rules, beliefs and cultural practices around childbirth. But no one seems to have posed the question: To what degree do women keep up with their hard work while pregnant? How does that work affect their pregnancies? I assume that no anthropologists have been interested in that topic because they weren't mothers themselves. Most anthropologists were male. And among those who were female, I can't verify that any of those I have read were mothers. Presumably anthropologists seldom thought of combining pregnancy and work as a very big issue. 

Nonetheless, Kenneth and David Good's books give more anecdotes on childbirth under primitive conditions than most anthropology books, because they were both personally implicated in the process. In 1983, Kenneth Good's teenaged Yanomamö wife Yarima got pregnant. When she was about five months along and visibly pregnant, she joined her older brother, three bachelors and Kenneth on a trek to another village. After having walked for almost a day, carrying a basket with all her household stuff, Yarima gave birth to a very premature, lifeless fetus.¹⁹

In 1986, David Good was born in a hospital in Caracas: After the miscarriage a few years earlier, his father had resolved to get his wife the best possible maternal care. The birth was surprisingly quick and uncomplicated. Three years later his sister Vanessa was born two months earlier than expected in Yanomamö territory. It was planned that Vanessa would also be born in a hospital in the United States, but her early arrival canceled those plans²⁰. Babies born around the 31-32 week of pregnancy, like Vanessa, are normally treated in intensive care units for newborns. But since there weren't any such facilities in the jungle, Vanessa had to make do with standard Yanomamö care. David Good makes no mention of any health problems of his sister. To the contrary, through her brother's remarks she makes a both healthy and clever impression. From that information we can draw the conclusion that it is entirely possible to be born around week 31-32 in a Yanomamö village and still become entirely healthy.

Otherwise, I have found one single anthropological anecdote on childbirth, from Jean Brigg's Never in Anger: Portrait of an Netsilik Eskimo Family from 1971. For one and a half years, Jean lived with an Eskimo family with three daughters. The family lived in a traditional way from fishing and hunting. They only ventured to civilization to trade a few times every year. They entertained a nomadic lifestyle, moving every now and then. The mother, Allaq, was pregnant with a fourth child. When the family was moving one day, she ran beside the dog sled most of the six hour trip, as adults tended to do. When the family arrived at the new site, she helped build an igloo. Jean Briggs found it surprising that Allaq didn't give birth the same day, but a week later instead, quietly during the night after an evening of socializing. The infant was another healthy daughter.²¹

This is about all I have been able to find about pregnancy and childbirth in primitive societies in the anthropological literature I have at my disposal. I have read about a woman dying alone in childbirth among the Piraha, because the Piraha were such individualists that they had no routines for who should help a woman in difficult labor²². I have read about Nisa the !Kung woman who gave birth alone in the bush, according to local custom. I have read that !Kung women in general stick to their usual work of walking and carrying until labor begins, although women are expected to be emotionally labile during pregnancy.²³ But I haven't read any detailed observation by any Westerner commenting on the strain a primitive lifestyle puts on pregnant women and how they cope with that.

I can not compete with you

If we want to understand how much women in traditional societies exert themselves during pregnancy, two pieces of information are lacking:

1. How hard do premodern women work during pregnancy? 

2. How naturally well-equipped are they to handle that hard work? 

Also with the very limited data I have, I'm convinced that point 2 must differ between women in primitive societies and at least some women in modern society. However women in primitive societies are experiencing pregnancy, they need to be better equipped for it than me. At the very least, people who entertain a nomadic lifestyle must be able to walk. And I haven't really been walking since I got pregnant last winter; I have been slouching around instead, due to sheer exhaustion. I could never have kept up with a band of nomads (although I think I could have during normal times). Still, according to Western norms, my pregnancy is entirely healthy.

It's not because I was in a bad shape before I got pregnant. To the contrary, I was hardened by years of construction work and food cultivation for numerous hours every day. And still I collapse when I get pregnant. Admittedly, I'm old - almost 39 years. Pregnancy was a lot easier when I was younger: When I was pregnant at 21, I kept on gyming and jogging until the seventh month, and now I can barely walk. But 39 year old women have children in nomadic, primitive societies too. Somehow, those women need to be better constituted than me to make it. 

The only way I can explain the bad shape of me and many other modern pregnant women is that a new pregnancy phenotype has evolved among sedentary agricultural populations. Evolutionary competition is relentless: If pressure for fitness loosens in one place, it will tighten in another place. When people gave up nomadism and reliance on female agricultural labor, the competition between the children of different women could become more intense. There is always a trade-off between the interests of mother and child. When mothers are given more space, most of that is likely to be taken up by the pressure to make healthier children. In denser societies, the pressure to bring immunologically strong children to the world should be especially high. 

Mutations that make children stronger at the expense of mothers that would be impossible in nomadic societies were free to evolve in sedentary agricultural societies. Especially in socially stratified agricultural societies. A disproportionate number of people in highly stratified societies are descendants of upper class people. That is, from upper class women (class differences tend to develop later on the female side, but they do develop). And upper class women were a kind of breeding specialists. Subsistence work, childcare and sometimes also breastfeeding was outsourced to others. Pregnancies that were difficult for the mother, but produced healthier children, could be selected for more among upper class women than among working class women.

Through selecting for upper class individuals for many generations, evolution might have promoted numerous virtues. But it also promoted incubator mothers at the expense of working mothers. Civilization on all levels should have had that effect. The more sick leave pregnant mothers were allowed to take, the more sickening pregnancy could become. 

There is always a balance between maternal health and fetal health. That balance should shift when female living conditions change. But there is always a balance. And I find it plausible that during history, the timing of birth was one way to regulate that balance: Women remained pregnant roughly as long as seemed possible in their specific environments. 

How about mortality? 

There is another very loose indication that children of hard-working women are born on more optimal times than the children of women allowed to rest: the comparatively high maternal mortality rates among well-off women without access to modern medicine. 

The few studies that were performed before the 20th century showed significantly higher maternal mortality in higher social classes than in lower social classes. In 1781, a doctor estimated deaths in childbirth in the London slums and found them to be fewer than among his own patients. In 1851, an investigation found deaths in childbirth to be only half as common in a working-class community than in townships where higher classes lived. Such findings continued into the 1930s.²⁴ Physician Irvine Loudon, seemingly the foremost expert on historical maternal mortality, explained:

"The only plausible explanation for this social class difference is that the upper classes were more often delivered by physicians and, therefore, more likely to suffer unnecessary interference, whereas the lower classes were delivered by midwives"²⁵

Without absolving 19th century doctors from blame, I think there is an alternative explanation: Living conditions higher up on the social ladder did in itself cause more difficult deliveries. Mothers who got the opportunity to rest might have had longer pregnancies. And longer pregnancies tend to result in more difficult deliveries. 

Poor women should die more than richer women from certain obstetric causes, notably those caused by malnourishment, like labor obstruction from rickets. If richer women still died at much higher rates, they must have had some big disadvantage that offset the effects of poverty. It is possible, but far from certain, that bad doctors constituted that entire big disadvantage. 

Also in modern times, Western well-nourished women have a rather high tendency to die in childbirth if they don't get adequate medical care. For example, a Christian sect in the US in the 1980s, the Faith Assembly, rejected all kinds of professional medicine. They were well-nourished, white and had middle class jobs. But their maternal care consisted of prayer and assistance from family members and friends. Their maternal mortality was about 0.9 percent, comparable to countries with very little healthcare.²⁶

The only estimate for hunter-gatherers that I have found is lower. Kim Hill and Magdalena Hurtado estimated that between 1940 and 1970, when the Aché were still foragers, 4 deaths occurred out of 597 children born, which gives a maternal death rate of approximately 1 per 150 births, or 0.67 percent.²⁷ Otherwise it is difficult to find estimates of rates of maternal deaths among hunter-gatherers and horticulturalists.

How to be unnatural

So, it seems like all-in-all, the best time to give birth is at least a week earlier than when the average modern woman actually gives birth. What to make of that information?

In theory, telling pregnant women to become a bit more stressed could be one option. But I wouldn't recommend it to anyone. Premature birth is no good. Not even slightly premature birth, as the Danish population study indicated. If birth can in most cases actually be predicted two or four weeks in advance, that means that a woman who gives birth at 39 weeks has been destined to do so from 35 weeks. That means that she needs to start overstraining herself a number of weeks before her target delivery date. And then there is a very real risk that some of that strain will instead lead to premature birth. Although there is data saying that stress can lead to earlier labor, there certainly is not enough data to reliably calculate how to self-induce the right amount of strain and stress at the right time. 

Also, I wouldn't be surprised if the stress that (maybe) makes women who live under difficult conditions give birth earlier is to no small degree triggered by sheer physical pain. That is what I experience myself when I overstrain, so I guess that is what hardworking pregnant women in history and in present times experience: Persistently high levels of pain that results in stress that results in earlier delivery. 

The sad truth seems to be that the only rational things expectant mothers can do are to keep reasonably fit and use modern medicine wisely. Evolution has simply driven us into a number of impasses. First, it got us adapted for strenuous work. Then, social stratification created an upper class specialized in breeding. All in all, the combination of evolutionary pressures from different times seems to have created a toxic combination, consisting of physically taxing pregnancies that last too long. 

Those of us First World citizens who are pregnant today can only make the best of the situation. Next week is that ideal week 39 for me. Then I will go to the regional hospital for the induction I have persuaded the health care services to give me at some point. I will do my best to cooperate with the medicines and machines to bring a new human being into this world. Wish me good luck.

Domenico Ghirlandaio, detail from Birth of the Virgin, 1486-1490, Wikimedia Commons 

1. William A. Grobman, M.D., Madeline M. Rice et al. Labor Induction versus Expectant Management in Low-Risk Nulliparous Women, 2018

https://www.nejm.org/doi/full/10.1056/nejmoa1800566

2. Vivienne Souter, Ian Painter, Kristin Sitcov, Aaron B Caughey, Maternal and newborn outcomes with elective induction of labor at term, 2019 

https://pubmed.ncbi.nlm.nih.gov/30716284/

3. GC Smith: Use of time to event analysis to estimate the normal duration of human pregnancy, 2001

link: https://pubmed.ncbi.nlm.nih.gov/11425837/

Sci-hub link: https://sci-hub.se/https://pubmed.ncbi.nlm.nih.gov/11425837/} 

4. A B Caughey, What is the optimal gestational age for delivery?, 2006

https://www.nature.com/articles/7211536

5. Middleton P, Shepherd E, Morris J, Crowther CA, Gomersall JC, Induction of labour in women with normal pregnancies at or beyond 37 weeks, 2020

https://www.cochrane.org/CD004945/PREG_induction-labour-women-normal-pregnancies-or-beyond-37-weeks

6. Philippa Middleton, Emily Shepherd, Caroline A Crowther, Induction of labour for improving birth outcomes for women at or beyond term, 2018

https://www.cochranelibrary.com/web/cochrane/content?templateType=full&urlTitle=/cdsr/doi/10.1002/14651858.CD004945.pub4&doi=10.1002/14651858.CD004945.pub4&type=cdsr&contentLanguage=

7. G C Smith, Use of time to event analysis to estimate the normal duration of human pregnancy, 2001

link: https://pubmed.ncbi.nlm.nih.gov/11425837/

Sci-hub link: https://sci-hub.se/https://pubmed.ncbi.nlm.nih.gov/11425837/

8. A B Caughey, What is the optimal gestational age for delivery?, 2006

https://www.nature.com/articles/7211536

9. Hanan El Marroun, Mijke Zeegers, Eric AP Steegers, Jan van der Ende, Post-term birth and the risk of behavioural and emotional problems in early childhood, 2012

https://academic.oup.com/ije/article/41/3/773/835374}. In a big Danish study

10. Josephine Funck Bilsteen, MSc^1,2; David Taylor-Robinson, PhD³; Klaus Børch, PhD¹; et al, Gestational Age and Socioeconomic Achievements in Young Adulthood - A Danish Population-Based Study, 2018

https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2718099

11. Ina A Stelzer, Mohammad S Ghaemi, Xiaoyuan Han, et al. Integrated trajectories of the maternal metabolome, proteome, and immunome predict labor onset, 2021

https://pubmed.ncbi.nlm.nih.gov/33952678/

Article about the study

https://med.stanford.edu/news/all-news/2021/05/blood-markers-indicate-labor-approaching.html 

12. Anna S. Oberg, Thomas Frisell, Anna C. Svensson et al., Maternal and Fetal Genetic Contributions to Postterm Birth: Familial Clustering in a Population-Based Sample of 475,429 Swedish Births, 2013

https://academic.oup.com/aje/article/177/6/531/160108

13. Keith T Palmer, Matteo Bonzini, E Clare Harris, Cathy Linaker, and Jens Peter Bonde, Work activities and risk of prematurity, low birthweight and pre-eclampsia: an updated review with meta-analysis, 2013, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3653070/

14. J. L. Peacock, J. M. Bland, and H. R. Anderson

Preterm delivery: effects of socioeconomic factors, psychological stress, smoking, alcohol, and caffeine, 1995

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2550601/

15. Birth outcomes by neighbourhood income and recent immigration in Toronto

https://www150.statcan.gc.ca/n1/pub/82-003-x/2006010/article/birth-naissance/4060735-eng.htm

16. Carl Lumholtz, Among Cannibals, 1889

https://www.gutenberg.org/files/66299/66299-h/66299-h.htm#Page_160

17. Richard Wrangham, Catching Fire, 2009, chapter 7

18. Napoleon Chagnon, Yanomamö, Fourth Edition, 1992, page 124

19. Kenneth Good, David Chanoff, Into the Heart - One Man's Pursuit of Love and Knowledge Among the Yanomami, 1996, page 137-139

20. David Good, The Way Around, 2015, 23 percent

21. Jean Brigg's Never in Anger: Portrait of an Netsilik Eskimo Family, 1970, page 52-54

22. Daniel Everett, Don't Sleep,There are Snakes, 35 percent

23. Melvin Konner and Marjorie Shostak, Timing and Management of Birth among the !Kung: Biocultural Interaction in Reproductive Adaptation, 1987

Link: https://www.jstor.org/stable/656392

Sci-hub link:https://sci-hub.se/https://www.jstor.org/stable/656392

24. Irvine Loudon, Obstetric Care, Social Class and Maternal Mortality, 1986, https://www.jstor.org/stable/29524414

25. Irvine Loudon, Maternal mortality in the past and its relevance to developing countries today, 2000

https://www.sciencedirect.com/science/article/pii/S0002916523066856

26. Irvine Loudon, Maternal mortality in the past and its relevance to developing countries today, 2000

https://www.sciencedirect.com/science/article/pii/S0002916523066856

27. Kim Hill and Magdalena Hurtado, Ache Life History, 1996, 34 percent