Light Drinking During Pregnancy: 7 Things You Need to Know

Last year, the CDC ignited a firestorm of criticism by stating that women should “stop drinking alcohol if they are trying to get pregnant or could get pregnant”, and–because an estimated 50% of pregnancies in the U.S. are unplanned–any woman who drinks alcohol should use birth control.

“Its [the CDC’s] underlying message was unmistakable: Women should consider themselves first a vessel for human life and make decisions about their health and behavior based on that possibility,” Rebecca Ruiz wrote at Mashable, in a typical reaction.

The (completely understandable) outrage at the CDC’s tone-deaf and condescending messaging has, unfortunately, drowned out information on the key question for many pregnant women: Is any amount of alcohol during pregnancy okay? And are there times when it should be absolutely avoided?

We all know that heavy drinking and binge drinking are harmful during pregnancy. Fetal alcohol syndrome, caused by heavy drinking during pregnancy, affects an estimated 2 to 7 children out of 1000. Another 2 to 7% of U.S. children are thought to suffer milder forms of cognitive impairment due to alcohol exposure in the womb.

But what about light drinking, a champagne toast or an occasional glass of wine at dinner? Women–pregnant, pre-pregnant, and otherwise–receive conflicting advice about the safety of light drinking.

Economist Emily Oster, in her bestselling pregnancy advice book Expecting Better, says pregnant women can be comfortable with “1 to 2 drinks a week in the first trimester” and one drink daily afterward, a stance she continues to stand behind.

Many doctors also greenlight the occasional drink, as Ruth Graham of Slate notes: “Many doctors seem perfectly comfortable with moderate alcohol consumption in the late stages of pregnancy. When I told my doctor that I was enjoying a glass of wine per week in my third trimester, she didn’t bat an eye.”

The CDC, on the other hand, maintains that no amount of drinking is safe.

“Any drinking is going to put your child at risk,” according to Clark Denny, a CDC epidemiologist. “You should not drink if you are pregnant, are considering getting pregnant or even if you could possibly get pregnant.”

Other countries, like the U.K. and France, that once gave less stringent advice–pregnant women should not exceed 1-2 drinks per week–now state no amount of drinking is safe, and that women who are pregnant or trying to conceive should abstain entirely.

At the heart of this debate is the research itself. So, what do we actually know about light drinking during pregnancy?

Let’s start with alcohol and the risk of miscarriage.

  1. Light drinking, the equivalent of 1-2 drinks per week, during the first trimester, boosts the odds of a first trimester miscarriage by about 30%, and the odds of a early second trimester miscarriage (between 13-16 weeks) by about 70%, according to a large study of over 90,000 pregnancies in Denmark. The risk rose with greater intakes. Drinking 4 or more drinks per week during the first trimester more than doubled the odds of a miscarriage. A U.S. based study found that women who drink 2 times per week in their first trimesters had a 25% chance of miscarriage, compared to a 14% chance for those who abstained. Other studies also find a higher risk of miscarriage for light drinkers (see here and here).

Does this mean that light drinking early in pregnancy causes miscarriage? Unfortunately, we cannot say for sure. Experimental studies in humans are impossible, so there are a lot of unknowns.

For one, some women may understate how much they actually drank during pregnancy, so the apparent rise in miscarriage could actually stem from moderate to heavy, rather than light, drinking.

Women who drink heavily during pregnancy often also smoke or have partners who smoke, and are more likely use drugs, receive worse prenatal care, live in poverty and experience extreme chronic stress–all of which can raise the risk of miscarriage and cognitive problems in their children, and any and all of these factors could exacerbate the toxic effects of alcohol on the fetus.

Poor nutrition and smoking in particular seem to worsen the effects of alcohol; alcohol lowers the amount of nourishment reaching the fetus.

Women who drink alcohol during their first trimester may have less nausea. (I for one could not get near alcohol in my first trimester, as much as I might, after a long day of nausea and chasing after a toddler, desire a drink.) Although about 25% of women do not experience nausea during pregnancy, an absence of nausea is linked with an increased risk of miscarriage, probably because poorly developing pregnancies produce fewer symptoms.

And finally, the problem may not arise from drinking during early pregnancy but from drinking before pregnancy.

2. Drinking alcohol may lower the chances of pregnancy, and increase the chances of miscarriage, by causing chromosomal abnormalities in the egg before ovulation. Alcohol has been shown to impair meiosis, the critical two-step cell division in maturing egg follicles, leading to chromosomally abnormal eggs. Chromosomal abnormalities account for over half of first trimester miscarriages. Worse, because eggs take several months to fully mature, even drinking in the months before conception could be harmful.

How much alcohol do you have to drink to harm your eggs? Again, we don’t have a good answer. The degree of harm likely depends on a lot of other things, like your age, your overall fecundity, your alcohol tolerance, how much alcohol you drink, and when you drink relative to critical phases in the egg’s development.

Clearly, even formerly heavy drinkers go on to have chromosomally normal and perfectly healthy pregnancies, so the effect is not absolute. The increase in chromosomal abnormalities is probably most problematic for couples already suffering from fertility issues. Among couples undergoing IVF, for instance, drinking appears to lower their chances of pregnancy.

Because the human data are limited, we have to turn to animal models. In one study using monkeys, the equivalent of binge drinking (4-5 drinks at a sitting) twice a week lowered the number of chromosomally normal eggs and increased the chances of miscarriage.

3. Alcohol does not reach the developing embryo until the 3rd week after fertilization, or pregnancy week 4, right after most pregnancy tests turn positive. In other words, even if you got pregnant on your honeymoon while more than a little tipsy and drank cocktails on the beach for the rest of the week, you have nothing to worry about. That alcohol did not reach your embryo.

What about harm to the fetus’s developing brain? Here’s the problem: No one knows at what threshold drinking alcohol becomes harmful, and that threshold may vary from person to person, just like alcohol tolerance and metabolism varies from person to person.

Heavy drinking and frequent binge drinking are clearly bad, but what about that occasional glass of wine or cocktail?

Emily Oster finds the not one drop rule propounded by the CDC and others absurd, and it’s easy to see where she is coming from. Lots of chemicals known to be harmful in large quantities are completely safe in small amounts. As she puts it,

“If you have too many bananas (and I mean a LOT of bananas), the excess of potassium can be a real problem, but no doctor is going around saying “No amount of bananas have been proven safe!” He’d be laughed out of a medical conference.”

But this argument, a version of the Paracelsus principle–the dose makes the poison–depends on the poison in question. Some toxins, like lead, are considered unsafe at any amount.

We don’t have great information about the actual threshold at which alcohol causes harm, or when harm is most likely to occur. But let’s sift through what we do know.

4. By the third week after conception (the 5th week of pregnancy), alcohol and its byproducts cross the placenta. Based on animal research, the fetus is believed to experience the same blood alcohol level as its mother.

5. Alcohol is a known neurotoxin. Although how alcohol causes damage is not entirely clear, neuronal loss with heavy or binge drinking is evident in animals and in humans. Harmful effects may be especially pronounced in the developing brain, particularly during the first trimester, when many of the changes in facial morphology in FAS appear to originate. Damage to slow growing brain structures, like the cerebellum, likely occurs throughout pregnancy.

6. Large epidemiological studies find no evidence of cognitive impairment with light drinking (1-2 drinks per week) in the second and third trimesters and less than a drink a week, on average, in the first trimester. This research is the basis of Emily Oster’s claim that a drink a day in the second and third trimesters is okay.

One of the largest of these studies is the U.K. Millennium cohort, which has followed a nationally representative sample of 11,000 children born betwen 2000-2002. At age 3, age 5, and age 7, both boys and girls whose mothers drank about 1-2 drinks per week actually had better overall cognitive performance and fewer behavioral and attention problems than children whose mothers abstained from drinking during pregnancy. Heavier drinking, on the other hand, was associated with worse cognitive performance and more behavioral problems.

(The higher test scores among children of light drinkers was almost surely NOT because alcohol benefited their development, but because women who drank lightly were on average more educated and of a higher socioeconomic status than women who abstained. This confounding of light drinking with education and socioeconomic status is actually a huge problem for interpreting this study’s results. Parental education, income, and social class all predict better cognitive performance and fewer behavior problems. So, who is to say that these children would not have been more advantaged had their mothers abstained throughout pregnancy?)

Other studies have found no effect on test scores or mental health at age 11 among children of mothers who drank less than 1 glass per week during their first trimester; no increase in mental health or behavior problems among children of light drinkers at age 2, 5, and 8 (but worse mental health seen among moderate, binge, and heavy drinkers); and no impairment in cognition, learning, or attention among 14 year olds whose mothers drank an average of less than a glass a day early or late in pregnancy; and no reduction in IQ, attention, or executive function at age 5 in women who drank up to an average of 5 drinks per week.

In short, light drinking during pregnancy, less than a drink a day on average, and no more than 1 drink at a sitting, has not been shown to cause detectable harm.

Because light drinking is more common among highly educated, high income women, the advantages of which could mask any subtle impairments caused by small doses of alcohol, we need to take these findings with a grain of salt.

7. Individual differences in alcohol metabolism and clearance probably determine the threshold at which alcohol causes harm. This is clear even among heavy drinkers: Only about 5% of babies born to women who abuse alcohol during pregnancy suffer from FAS.

But a recent study looking at IQ at age 8 found that among women who were genetically poor alcohol metabolizers, moderate drinking (1-6 drinks/week) was associated with lower IQ. Children of faster metabolizers, on the other hand, did not have lower IQs on average, nor did children of among poor metabolizers who abstained from drinking during pregnancy.

The Bottom Line

Drinking more than 2 drinks a week in the first trimester appears to substantially increase the risk of miscarriage. That said, we do not know whether alcohol causes miscarriage. A lack of nausea, or other problems that often go along with alcohol, like smoking, may contribute to or fully explain this effect.

One possible reason is that alcohol can cause chromosomal abnormalities in the egg prior to conception. Alcohol use before conception does NOT preclude a healthy pregnancy (clearly!). Still, couples who are struggling to conceive may boost their chances by abstaining from alcohol.

What about children’s cognitive abilities and emotional and mental health? Here, very light drinking, less than a drink per week on average, has not been not been shown to be harmful. And many studies find no harmful effects of 1-2 drinks per week.

However, women vary considerably in their ability to metabolize alcohol, based on their body size, liver size, and genetics. Other factors, like whether you drink with food and how quickly you drink (no tequila shots, please!), also affect the amount of alcohol reaching the fetus. So setting a “safe” threshold is basically impossible.

Given this variability, here’s my personal take: It seems wise to largely avoid alcohol during pregnancy, especially during the first trimester. An occasional glass of wine drunk slowly with dinner is probably fine, but I personally don’t see the upside to pushing the limits.

At the same time, women who drank a glass of wine here and there have no cause for alarm. The CDC’s draconian, not-one-drop stance is probably based on a slippery slope argument. If they say an occasional glass of wine or beer is fine, then some women might read this as license to overindulge.

On the other hand, Emily Oster’s advice to drink “up to 1 drink a day in the second and third trimesters, and 1 to 2 drinks a week in the first trimester” feels too risky for my blood. Most studies define light drinking as 1-2 drinks per week, much lower than Oster’s recommended maximum for the second and third trimesters.

At the end of the day, every woman is going to weigh the risks and benefits of light drinking a little differently. A reasonable woman could see the available evidence and feel comfortable with 1 to 2 drinks a week; another reasonable woman could see the available evidence and decide to abstain entirely.

What about drinking after pregnancy? Check out my post on drinking while breastfeeding.


Introducing Solids: Is It Best to Wait Until 6 Months?

In 2005, the American Academy of Pediatrics (AAP) changed their recommendations from gradually introducing solids between 4 and 6 months of age to exclusively breastfeeding for “about the first 6 months of life. For the first 6 months “your baby”, they assert, “needs no additional foods (except Vitamin D) or fluids unless medically indicated.”

My daughter at 9 months, eating peas and Cheerios

You might think that these recommendations were updated because of a new evidence showing substantial benefits for the baby of delaying the introduction of solids.

But, nope, this is not the case. Instead, these changes were motivated by studies showing an absence of harm to the baby and some minor potential benefits for the mother.

The World Health Organization was the first major medical organizations to recommend exclusive breastfeeding for 6 months, for reasons they outline in a 2002 research review (updated in 2007):

  1. No evidence exclusive breastfeeding for 6 months results in poor growth
  2. No apparent increase in the risk of allergies*
  3. A longer period of lactation-induced infertility, about an extra month on average
  4. Greater postpartum weight loss between 4-6 months, about a pound on average
  5. Lower risk of illness for the first six months, and possibly throughout the first year

* (Since this review was published, new evidence suggests delaying solids until 6 months may increase the risk of food allergies. More on this complex topic below.)

When weighing the pros and cons of exclusive breastfeeding for 6 months, the social and environmental context matters. The WHO crafted its guidelines to maximize health outcomes in both developing and developed nations. But benefits that have a large public health impact in the developing world, where access to basic medical care, birth control, clean water, and nutritious food is limited and inconsistent, can have little or no impact in the developed world.

Benefits of Exclusive Breastfeeding for Six Months

For mothers living in the developed world, the benefits of exclusive breastfeeding for 6 months are small, arguably negligible.

We have been blessed with access to birth control, so lactation-induced infertility is not a critical means of pregnancy prevention and birth spacing.

And, when it comes to weight loss, although almost every mother I know would gladly shed an extra pound or 2 of baby weight, the benefit is really small, just an extra a pound on average. Surely there are other activities that do not require 8-hours a day, round-the-clock efforts that would be just as effective forms of weight loss.

Certainly neither of these two benefits warrant the stress placed on introducing solids “at the right time.”

Of the reasons given by the WHO, the only truly persuasive one is a reduced risk of illness for the baby during its first 6 to 12 months of life. Here again, though, there are important differences between the developing and the developed world.

In the developing world, exclusive breastfeeding for 6 months is literally life saving. Within the 42 nations with the highest child mortality rates, widespread adoption of exclusive breastfeeding for 6 months would prevent an estimated 13% of the deaths of children under the age of 5. This is a larger percentage than would be saved with improved access to antimalarial medication or antibiotics.

Within the developed world, the picture is more nuanced. Here, completely weaning from breast milk before 6 months, or introducing solids before 4 months, or introducing formula before 6 months substantially increases the risk ear infections, diarrhea, pneumonia, and bronchitis.

But gradually introducing solids between 4 to 6 months while continuing to breastfeed does not. Studies that compare the outcomes of infants who were exclusively breastfed until 6 months (EBF or exclusively breastfed) with those who were exclusively breastfed for at least 4 months and then partially breastfed thereafter (MBF or mostly breastfed) find that 6 months of exclusive breastfeeding neither harms nor benefits growth and development.

The only apparent benefit of delaying until 6 months was a 25-50% lower risk of diarrhea in the first year of life. The effects on ear infections, pneumonia, and bronchitis were unclear: A couple of observational studies found a lower risk, but the majority of studies, including the only two randomized controlled trials, found no effect. And longitudinal study conducted in Belarus found no differences in weight, height, BMI, asthma, allergies, or other illnesses by age 6.

“Exclusive breastfeeding for six months does not seem to confer any long-term (at least to early school age) protection against obesity or allergic disease, nor any benefits in cognitive ability or behaviour, compared with exclusive breastfeeding for three to four months with continued partial breastfeeding to six months.”  – Kramer & Kakuma, 2012

Costs of Exclusive Breastfeeding

Lower Iron Status. EBF infants tend to have lower iron levels than mixed fed or formula-fed infants. Two randomized controlled trials, one conducted in Honduras and the other in Iceland, found significantly higher iron levels among MBF than EBF infants. Reassuringly, at 6 months of age, iron levels in both groups remained within the normal range, and the EBF infants were not more likely to be iron deficient or to develop anemia.

Still, iron needs rise sharply in the second half of the first year when newborn iron stores have been depleted, and lower iron status in the first six months could place EBF babies at higher risk of iron deficiency as they age. This is why the CDC considers “breastfed babies who after age 6 months are not being given plain, iron-fortified cereals or another good source of iron from other foods” to be at high risk for iron deficiency and anemia.

Increased Risk of Food Allergies. On this point, the advice given to parents has seesawed from it is best to delay solid foods, especially highly allergenic ones, to it is best to introduce these foods early, between 4 to 6 months. This is because the evidence behind the advice is of poor quality and inconsistent. A handful of observational studies find that early introduction increases the risk of allergies. A comparable number find that late introduction increases the risk. And not all of the studies find when solids are introduced matters at all.

Some researchers to speculate that there is something of a sweet spot in infancy, between 4-6 months of age, when introducing solids lowers the risk of allergies. This protective effect may be strongest for peanuts and cooked whole eggs. But until proper clinical trials have been done, it seems wise not to make too much of these findings. (UPDATE: A randomized controlled trial has now shown that introducing peanuts before 1 year of age lowers the risk of a peanut allergy substantially. More on this study here.)

The AAP no longer recommends waiting to introduce solids until 6 months or later to reduce the risk of allergies, and instead advises waiting until at least 4 months. And of course, this statement on solids and food allergies conflicts with guidance the AAP gives elsewhere to exclusively breastfeed for 6 months.


Like so many parenting decisions, the actual evidence suggests that we can all relax a bit. For those of us lucky enough to live in the developed world, introducing a small amount of solids between 4 to 6 months is unlikely to harm our babies, and equally unlikely to benefit them.

That said, when giving your baby solids before 6 months, it makes sense to provide small amounts. The immune benefits of breast milk are dose-dependent: the more milk the baby takes in, the greater the protection from illness. This is why solids during the first year should not be used to replace but to supplement breastfeeding. And when introducing solids, especially to exclusively breastfed babies, it makes sense to emphasize iron rich foods like broccoli, meat, eggs, fish, and fortified cereals.

Here is most sensible statement I found on the 6-month rule, from a review article in The BMJ, the flagship journal of the British Medical Association:

“From a biological perspective, the point when breast milk ceases to be an adequate sole source of nutrition would not be expected to be fixed, but to vary according to the infant’s size, activity, growth rate, and sex, and the quality and volume of the breast milk supply.”  – Fewtrell, Booth, & Lucas, 2010

When did you first introduce solids? Did the risk of allergies or infections play a role in your decision?