The evolutionary causes of pregnancy nausea

Pregnancy nausea feels like a blood sugar police force. Maybe it actually is.

A traditional dessert of East Coast Malaysia. Wikimedia Commons

There are a few phenomena with the female body that lack a good evolutionary explanation. The existence of permanent female breasts is one of them. I launched a theory about that here. 

The existence of pregnancy nausea is another. I have a good reason to think about it now: I'm 9 weeks pregnant. Yes, I know, the phrase "I'm x weeks pregnant (x<12)" is not supposed to be uttered in public. The 12 week rule has a natural explanation: The miscarriage risk decreases dramatically at 12 weeks. Pregnancy is supposed to be a social happening, not a medical condition.

The problem is that for me, pregnancy really is a medical condition during the first 12 weeks. After that I'm my normal self, except that I look more and more like a seal. But for a few weeks, approximately between weeks 6 and 10, I'm so exhausted that I can barely move, I feel slightly sick and I like no food. Pregnancy is not a disease, they say, but the fatigue I experience for a few weeks in early pregnancy is worse than any disease I have ever experienced (admittedly, I haven't experienced many).

The lack of energy has only one advantage: At least I have a lot of time to read and think. And, naturally, one question I think about is: Why, why, why am I so sick and tired?

The anti-poison theory

The only existing explanation I have found was launched by biologists Samuel Flaxman and Paul Sherman in 2000 (sci-hub link). The explanation takes aim at the specific food preferences females tend to develop in early pregnancy. Women suffering from pregnancy nausea develop many aversions. A few of those tend to be against harmful substances like alcohol and cigarette smoke and against meats and strong-tasting vegetables that might contain poisons and pathogens. For that reason, the researchers theorize, nausea is an evolved mechanism to avoid infections and poisoning that could be dangerous to the fetus.

There are at least two problems with this theory. The first is that aversions and cravings during pregnancy are rather arbitrary. Flaxman and Sherman compiled findings from research of pregnant women's aversions and cravings and illustrated it with an interesting diagram. Meats, non-alcoholic beverages and vegetables were the most common aversions. Fruit-juices, sweets, dairy and ice-cream and meats were the most common cravings. Yes, that is the strange thing: Although meats were the most common aversion, they also were one of the most common cravings. Like if above all, nausea makes people form an emotional opinion of meat. For vegetables, the relationship between cravings and aversions was rather even. The rates of cravings and aversions to ethnic, strong and spicy foods were also almost identical.

This arbitrariness fits very well with my own experiences and with things people around me have told me. In my own experience and from what I have heard, cravings tend to be very particular: Exactly that kind of ice cream, exactly that vegetable, exactly that traditional dish. There seems to be rather little general logic in pregnant women's aversions and cravings, just like there seems to be little logic in the food preferences of people who experience nausea due to cancer treatment. I think the lack of consistency in cravings and aversions weakens Flaxman's and Sherman's hypothesis significantly. An adaptation that is supposed to make an organism abhor meats and still causes cravings for meats in a rather large minority is not a very efficient adaptation. 

Are meat pathogens really that dangerous in early pregnancy? 

Flaxman and Sherman believe pregnant women feel nauseous and often abhor meat because they are more under the threat of meat-borne pathogens than the populations at large. One weakness of this hypothesis is that infections are dangerous during pregnancy as a whole, while nausea and food aversions most often last only for the first three months of pregnancy. 

Flaxman and Sherman do not mention this problem. They only list evidence that infections are a significant risk to mother and fetus during pregnancy (which they clearly are). Flaxman and Sherman explicitly mention the parasite Toxoplasma gondii, which is often acquired by handling or eating raw or undercooked meat and has been linked to congenital neurological birth defects, spontaneous abortions, neonatal diseases, and ocular defects. However, they don't mention that fetal effects of toxoplasmosis are much more severe in late pregnancy than in early pregnancy. A study mentions that while infection in early pregnancy poses a less than 6 percent risk of  fetal transmission, rates of transmission range between 60% and 81% in the third trimester. That is, at a time when the vast majority of pregnant women do not feel sick and do not have significant food aversions. 

Another food-borne pathogen that Flaxman and Sherman do not mention but that Western pregnant women are much warned for is Listeria monocytogenes, a bacterium that can cause fetal death, preterm birth and neonatal sepsis. Also listeria has its worst effects in the third trimester of pregnancy. It is rare in the second trimester and in the first trimester only a handful of cases are known in the scientific literature. So pregnancy nausea can't possibly have evolved to protect pregnant women from listeria infection, since also that infection is the most dangerous when hardly anyone feels nauseous and suffers from food aversions. 

It could be that toxoplasmosis and listeria actually cause damage in the first trimester too through causing miscarriage. Miscarriages during the first trimester are very common: About 15 percent of all noticed pregnancies are estimated to end in miscarriage and 30-50 percent of pregnancies in total (of which a majority have chromosomal abnormities). But an early miscarriage is not especially costly to the mother. She can just go on and have another child and hope the meat will be less poisonous next time. A much greater cost to the mother is when food poisoning occurs at later stages in pregnancy. Then both mother and child are in mortal danger and the child can get brain damaged. Such brain damage is not necessarily visible at birth. If a woman who is pregnant in the third trimester catches listeria or toxoplasmosis, she might die herself from the disease or survive and raise a child with developmental retardation for years. Both are a great cost to the mother. 

If the function of pregnancy nausea were to avoid food poisoning, why does it then appear at the stages of pregnancy when the evolutionary costs of accidental food poisoning appear lower than during later stages of pregnancy? And why do so many pregnant females actually prefer "dangerous" foods anyway?

In evolutionary terms, pregnancy nausea is certainly a costly phenomenon. About 0.7 percent of pregnant women are so sick that they need to be hospitalized. Before modern medicine, women died from the condition. Its most famous victim is probably Charlotte Brontë. She became pregnant, vomited and vomited and died in the fourth month of pregnancy, on the 31 of March in 1855. She probably died from malnutrition and dehydration caused by being unable to eat and drink.

One could guess that such a phenomenon appears less in populations where women need to work hard to support their children. But actually, the opposite seems to be true. So called hyperemesis gravidarum, nausea so severe that it leads to nutritional deficiencies and dehydration, is much more common among women from South Asia and Africa than among women of European ancestry. 

The great Ramadan experiment 

For the above reasons, I couldn't really accept Flaxman's and Sherman's hypothesis. But I also didn't have a better explanation. That is, until I stumbled over a research article from 2008 about how Ramadan fasting among Muslims seems to increase long-term adverse outcomes. 

The researchers, economists Douglas Almond and Bhashkar Mazumder, used Ramadan as a natural experiment for evaluating effects of intermittent fasting during pregnancy. They studied data from Muslim populations in Uganda and Michigan. Some results were rather striking. For example, census data from Uganda showed that Muslims who were born nine months after Ramadan were 22 percent more likely to be disabled as adults compared to those who were never in utero during Ramadan. Effects were found for vision, hearing, and especially for mental (or learning) disabilities. The negative Ramadan effects were especially significant in early pregnancy. Almond and Mazumder made comparisons to nearby Christian populations and did not find any Ramadan effects among those. In Michigan they found that presumably Muslim Arab children who were in utero during Ramadan weighed less at birth compared to presumably Muslim Arab children who were not. The effect was stronger when Ramadan appeared in the summer and fasting hours were thereby longer. Ramadan happening just prior to conception reduced the fraction of male births by about 6 percentage points.

In 2009, economist Rein van Ewijk also found negative health effects from in utero Ramadan exposure from studying data from Indonesia. Van Ewijk mostly studied physical health parameters and found that having been exposed to Ramadan in utero was associated with worse physical health as an adult. For Christians in the same areas there were no detectable Ramadan effects at all.

A meta-study from 2018 observed that "some negative effects were particularly observed when Ramadan occurred during conception or the first trimester of the pregnancy". 

In many smaller studies of babies exposed to Ramadan fasting, doctors have often failed to identify any obviously clear negative effects of Ramadan fasting during pregnancy. Children exposed to Ramadan during the fetal stage often fulfill the proxies for newborn health as much as babies unexposed to Ramadan. A much clearer pattern emerged when economists, who are better equipped to study long-term effects, came into the picture. 

What if it's all about blood sugar?

When I saw those studies, the thought struck me: What if the function of pregnancy nausea is to keep blood sugar levels on an even level during the first part of pregnancy? It is common knowledge that eating often and not too much at the same time helps against pregnancy nausea. When pregnancy nausea hits, the (partial) cure is eating a little. The very reason pregnancy nausea is called "morning sickness" in English is that it is commonly experienced in mornings, before breakfast when blood sugar levels are low.

Personally I experience pregnancy nausea as an evil police officer who forces me to keep an even blood sugar level. Approximately between pregnancy week 4 and 11, nausea takes the place of hunger. I'm not hungry anymore and eating ceases to be pleasurable. Instead my eating becomes regulated by nausea. I eat only in order to control the attacks of sickness that always occur when I haven't eaten for a few hours. I actually don't like any food at all very much. Finding something I don't totally dislike can be a hassle. I often weigh the unpleasantness of feeling sick against the unpleasantness of eating something. The latter wins every few hours, but I only need, and want, to eat rather little every time. I also drink water much more often than usual.

For me, nausea and food aversions work in synchrony. Nausea forces me to eat regularly. Food aversions prevent me from eating very much at a time. Why eat more than necessary when no food is very inspiring anyway? I don't believe all aspects of my experience is universal for all women who experience pregnancy nausea. For example, I have heard about pregnant women who actually liked some food also when they felt nauseous. Still, I have the impression that pregnancy nausea in general suppresses both intermittent fasting and heavy eating. I also think it encourages hydration. Together, I think nausea and food aversions encourage an eating and drinking behavior that is more or less the total opposite of what is required for Muslims during Ramadan. Also professional advice to pregnant women say this: Women who feel high levels of nausea are advised to eat foods they feel they can tolerate every one to two hours, to only eat little every time and to drink water or other fluids between meals. That is as much anti-Ramadan as things can get. 

So when I saw the research studies of the long-term effects of Ramadan fasting, I thought there might be a connection: Maybe pregnancy nausea is a mechanism that strongly urges its sufferer to avoid tops and dips in blood sugar levels and/or dehydration. Ramadan fasting is a natural experiment that shows what happens when society encourages pregnant women to ignore those urges to guard their blood sugar and hydration levels. Muslim norms of Ramadan fasting is a social police that seriously challenges the blood sugar police.

A binge-eating problem? 

One fact that supports the blood sugar control theory is the well-known effects of diabetes in very early pregnancy. Children to mothers with pre-existing diabetes run a four to ten times higher risk of being born with congenital malformations. That is thought to be associated with too high blood sugar levels during the first trimester of pregnancy. 

This fact indicates that the negative Ramadan effect doesn't necessarily have to be due to starvation or dehydration. Part of the effects could also be due to spikes in blood sugar levels due to night time binge-eating. Muslims with diabetes are advised not to eat a lot of sugary food like other people when fast is broken at sunset because that could raise their blood sugar levels too much. The question is: Could there be a share of Muslim women who don't know that they are diabetics or prediabetics who get pregnant while doing like most people do - fasting and then eating large amounts of sugar? 

Type-2 diabetes is a sliding scale. A threshold value has been set at 7 mmol/L of fasting plasma-glucose. It is also not just a problem for old people. An age of onset of 30-39 years is not unusual. It wouldn't be entirely strange if women who suffer from undetected prediabetes or diabetes manage to get diabetes-style blood sugar fluctuations through alternating between fasting and social binge-eating. 

Flaxman and Sherman mention evidence of pregnancy nausea in three species of animals in the literature: Domestic dogs, captive rhesus macaques and captive chimpanzees. All three species have been observed eating less during early pregnancy. That observation indicates that high rather than low blood sugar might be the main problem. Also human females seem to eat less during the first stage of pregnancy compared to during latter stages: Pregnant women are infamous for gaining too much weight. But not during the first trimester. 

If high blood sugar in early pregnancy is a reason behind the worse outcomes for people who were made in Ramadan, the effects can only be expected to get worse and worse. The Arab world has high rates of diabetes and people originating from the Arab world have higher incidences of diabetes than people of European descent (study). If blood sugar spikes rather than starvation is the Ramadan problem for the next generation, more problems are to come. 

I have tried to find information about how women who suffer from pregnancy nausea actually manage to fast. Fasting and then eating a lot is against everything I know about how to manage pregnancy nausea. If I did that in early pregnancy, I would vomit all day long. But I haven't found any description of what fasting in spite of pregnancy nausea is like. I only found a question from a Muslim woman who vomited during pregnancy and wanted to know whether her fasting counts in spite of her vomiting of bile. Someone with religious authority reassured her that only voluntary vomiting of bile breaks the fast.

More research is needed

As usual, more research is needed. Not only to discover the evolutionary causes of pregnancy nausea, but above all to get any clarity over the effects of periodic fasting on fetuses. 

It is pretty strange that so few people care about this. It is well established that diabetes in early pregnancy is a risk factor for congenital malformations. Ramadan fasting is a great blood sugar experiment, in populations with high rates of diabetes and prediabetes. It should have attracted some interest since 2008 and 2009, when Almond and Mazumder and Van Ewijk made their studies, but I can't find any newer studies of similar design.

I think a sibling study would be ideal to understand Ramadan effects better. In general, more sibling studies should be made. But in this case it would be especially beneficial, since time of conception in relation to the Muslim calendar in general doesn't seem to be an active choice. Studies comparing siblings who were breastfed or not breastfed, exposed to alcohol or medications during the fetal stage or not have the weakness that they compare siblings subjected to different maternal choices: Why does a prospective mother choose to drink some alcohol during one pregnancy but not during another pregnancy? There is no guarantee that choice is unrelated to the psychosocial environment in which she raises the child, making claims that moderate alcohol use during pregnancy causes psychosocial effects dubious. 

We have the possibility that health-conscious mothers time their pregnancies to avoid Ramadan. Both Almond and Mazumder and Van Ewijk made efforts to show this wasn't the case in their study populations. In Almond and Mazumder's Michigan data, mothers who became pregnant right before Ramadan had a slight tendency to be better educated than those who got pregnant just after Ramadan. Midwives and doctors Van Ewijk interviewed had never heard of any couple trying to time conception in order to avoid Ramadan. To the contrary, the doctors and midwives said believers considered Ramadan beneficial and some even tried to time conception so birth would happen during Ramadan. Also Van Ewijk compared health and educational characteristics of mothers whose children were exposed to Ramadan in utero and of those who were not. He found the two groups to be largely similar. 

If getting pregnant less than nine months before Ramadan is not much of an active choice, then comparing siblings born in different months of the Muslim calendar should be highly interesting. I think Sweden could be a good place for such a study. Here many things get registered except, as usual, religion. However, immigrants self-segregate to very important extents. Christian Arabs live in one suburb, Somalis in another and so on. That way, religion can be rather accurately guessed from a person's address. Find a suburb where people actually fast during Ramadan and then compare siblings born there x months after Ramadan. That should be a doable project. Just give me some funding for independent research, that data the state keeps secret from mere mortals and a research partner who understands statistics, and I'm ready to go. I mean, I'm ready as soon as I get out of bed.