Cribsheet

by Emily Oster

  For most women, even those whose babies latch well, breastfeeding is at least somewhat painful early on. Any pain should be mostly gone after the first minute or two of nursing, not continue. Certain conditions can cause ongoing pain—for example, nipple yeast infections—but are treatable. It would be a shame not to figure that out, so if your pain persists, ask for help.

  Nipples can become cracked and sore, or bleed. There is no magic solution to fix this problem. Many women swear by lanolin cream or various gel packs and pads, but there is no randomized evidence suggesting that any of these things are successful.11 The only thing with any support in randomized trials is the practice of rubbing breast milk on your nipples regularly. I will caution, however, that this data comes from just one trial, and it is small.12

  Of course, there is no reason not to use lanolin or to rub breast milk on your nipples, so if you feel like that works, or you want to try it, awesome. My friend Hilary, when I asked her about this, wrote me: “MOISTURIZE THE NIPS EVERY TIME.”

  The very good news is that for most women, regardless of what actions they take, nipple pain does resolve, or at least lessen to manageable levels, after a couple of weeks. This is based on evidence from trials where women had reasonably severe nipple trauma—bleeding, open sores—so even if things look very grim, remember that in most cases, they will resolve themselves.13

  This evidence also says that still having agonizing pain after two weeks is not typical, nor is it something that should be dismissed with “Oh, it will get better if you keep trying.” If you’re experiencing this, get help. Many states have breastfeeding hotlines, and La Leche League can often connect you with a lactation specialist over the phone if you do not want to go as far as seeing someone in person.

  Nipple pain is different from mastitis, an infection you can get at any time during nursing. Some things will increase the risk of mastitis—including not fully emptying the breasts with each feeding, having an oversupply, or not emptying the breasts frequently enough—but its onset is largely random. It is not hard to diagnose—the symptoms are a red, painful, swollen breast and a high fever—and may need to be treated with antibiotics. Mastitis can be extremely painful and is not something to ignore.

  NIPPLE CONFUSION

  If you are considering breastfeeding, you will have heard about the dreaded nipple confusion. Many sources will tell you to be very careful about using artificial nipples—on a bottle or a pacifier—since babies will become confused and decide not to latch on to the breast.

  In this discussion, it seems important to separate bottle-feeding—where the baby is learning that food can come from another source—and pacifiers, which do not produce food.

  Despite the warnings, there is simply no evidence that the use of pacifiers impacts breastfeeding success. This has been shown by more than one randomized trial,14 including trials that start infants on a pacifier at birth. At least one of these trials gives a sense of why someone might have (incorrectly) concluded that pacifier use matters for breastfeeding. This trial enrolled 281 women and counseled them either in favor of or against pacifier use. Pacifier use was less in the group that was discouraged from it.15 The main analysis in the paper—which is shown in the first two bars of the graph on this page—compared breastfeeding rates at three months for women in the pacifier encouragement group with the pacifier discouragement group. This analysis showed no impact of the intervention on breastfeeding rates. In both groups, about 80 percent of moms were nursing at three months, even though one group was much more likely to also use a pacifier with their babies.

  The authors then do something clever, which is to compare breastfeeding rates at three months for moms who chose to use a pacifier versus not, without using the randomization. Basically, they treat the data as if they didn’t have a randomized trial at all, and just saw breastfeeding and pacifier-use rates for mothers.

  The results from this analysis are in the second set of bars in the graph below. Here we see that moms who use a pacifier are less likely to be breastfeeding at three months. The researchers’ conclusion—comparing the two sets of results—is that some other factor causes both the pacifier use and the early cessation of breastfeeding. For example, given the rhetoric around pacifiers, it is easy to believe that women who choose to use pacifiers may have a less intense desire to breastfeed.

  We should base our conclusions on the randomized data, which tells us that pacifier use doesn’t affect breastfeeding success. But since much of the rest of the evidence in the literature is based on these observational correlations, it is not surprising that people have bought into the myth of pacifiers causing nipple confusion.

  Evaluating the role of bottle-feeding nipple confusion is more complicated because there are two factors: the role of supplementing with formula, and the role of nipple confusion. Imagine that breastfeeding success is associated with supplementation—say, because women who have a harder time nursing are more likely to supplement. You will then find that infants who are fed by a bottle early on are less likely to be breastfed in the long run, but this could have nothing to do with the nipple.

  A very nice randomized trial addresses this issue using a simple design.16 Infants needing supplementation are randomized into either supplementation with a bottle or supplementation with a cup, where nipple confusion is not thought to be an issue.17 These authors found that overall, the method of supplementation did not matter. Both groups had breastfeeding durations of around four months, and exclusive breastfeeding for two to three weeks. Bottle or cup, the results were the same, suggesting that nipple confusion was not an issue.

  MILK SUPPLY

  My mother had her trusty Dr. Spock book in the 1980s. My mormor (grandmother) had her own guide: a set of six little books called The Mother’s Encyclopedia, first published in 1933. The book makes for great reading. It covers everything from measles to appendicitis to summer camp. Even better, it’s in alphabetical order, so you get a discussion of caesarean section followed immediately by a section on competitive sports.

  The discussion of breastfeeding in this book spends a large amount of time on the question of supply and, in particular, notes that many “modern” women have trouble producing enough milk. The book blames the recommendation that women nurse only every four hours, and only from one breast. Perhaps the best part is the discussion of “primitive” mothers (their words, not mine) who “nurse their babies when they cry—on any and all occasions!”

  The authors note that this “primitive” method is very good for milk supply, although they caution that no one would ever recommend that modern parents return to this approach. It’s a good lesson in how things change; generally, the recommendation now is to nurse on demand, at least early on, since this establishes a plentiful milk supply. Schedules, to the extent we get them, come later.

  A biological mechanism links the frequency of feeding to milk supply. The system is designed to have a feedback loop where you produce more milk when the baby needs more. The existence of this loop is why, for example, people who are looking to increase their supply will sometimes pump after feedings to trick the body into thinking demand is greater than it is.

  Despite a basically reasonable evolutionary design, this doesn’t always work quite as planned. First, it can take a lot of time for your milk to start flowing. Second, even once there is milk, you can have an undersupply. And third, on the opposite end, you can have an oversupply.

  When your baby is first born, you’ll produce a small amount of colostrum, an antibody-rich substance. (You’ll actually start producing this in late pregnancy.) Over the first few days, as you nurse, your body will eventually (in theory) switch over from producing colostrum to producing milk in more copious amounts. The expectation is that this switch to more full milk production—scientifically termed lactogenesis II, and sometimes referred to as your milk “coming in”—will occur within the first seventy-two hours after you’ve given
birth. If this doesn’t happen, you will be deemed as having “delayed lactogenesis.”

  In fact, it does take longer than that for many women. The graph on this page—from a study of 2,500 women—shows the distribution of days from baby’s birth to milk production. Almost a quarter of women have milk production delayed beyond three days. This is even higher—about 35 percent—for first-time mothers.18

  Delayed onset of milk does—in the data—correlate with a higher likelihood of earlier breastfeeding cessation.19 This may be because delayed onset of milk leads to excess infant weight loss, which makes it harder to get breastfeeding going. It may also be that if you are not especially committed to breastfeeding in the first place, this setback is enough to turn you off from it altogether.

  Regardless of whether it is causal, delayed milk onset can be extremely frustrating. There are a few factors that correlate with it.20

  Smoking during pregnancy slows down milk production, as does obesity. Women who have a caesarean section are more likely to have later onset, as are those who have an epidural during labor. In terms of post-birth modifiable behaviors, both feeding on demand and initiating breastfeeding within an hour of birth are associated with a lower likelihood of delayed milk onset. It is worth emphasizing that these are correlations, not necessarily causal links, and for something like the epidural there may be good reasons to do it anyway. And even if you do everything as suggested, your milk may still be delayed.

  Once the milk has arrived, there still may not be enough of it—or there may be too much.

  For women who do not have enough supply, a first-line suggestion is generally to try to use the “demand-driven” feedback loop to increase supply. Doctors may recommend that you pump after each feeding, or at least after some of them, to try to convince your body that you need more milk. Our general knowledge of the biology of lactation suggests this could be helpful, although I can find no research that gives any helpful guidance on how to do this most successfully.

  You’ll also find a variety of suggestions on the internet about how to increase your milk supply. These include herbal remedies—fenugreek is the most common, although others, like nettle tea, do come up—as well as particular foods (dark beer, for example) and a suggestion that you stay hydrated.

  It is always good to stay hydrated, but there is no reliable evidence that it promotes milk production.21 Beer actually makes things worse (more on this on this page).

  The evidence on herbal remedies is mixed.22 To take fenugreek as an example, a 2016 review article covered two small randomized studies of the effect of fenugreek consumption on breast milk. In one study, milk production was increased. In the other, it was not. Evidence on other herbal remedies (shatavari, malunggay) shows similarly mixed results. None of these herbs shows any side effects at the recommended doses, so it will not hurt you to try them, but they are not magic bullets.

  There is more positive evidence on pharmaceutical remedies. In particular, the drug domperidone has been shown in a variety of randomized studies to increase milk production.23 (Unfortunately, it is not available in the US, so this may be somewhat unhelpful to point out. Readers in the UK can get it there, and it is also available in Canada.)

  It is possible that no matter what you do, you will have little or no milk—this isn’t common, but it does happen, and it’s often a surprise when it does, since it is not frequently discussed. This is typically diagnosed as insufficient glandular tissue (IGT), which simply means a lack of sufficient milk glands. For some women, this is a congenital condition—if this is you, you’ll likely have to supplement, at least to some extent.

  Women who have had a breast reduction may also have a limited milk supply, depending on the method of reduction. Again, some degree of supplementation may be necessary.24

  On the other side, you can have too much milk. This can happen naturally, or it can result from an overenthusiastic attempt to avoid the too-little-milk problem. The recommendation of adding pumping sessions after nursing sessions early on to increase milk supply can overcompensate—I know a few women who were zealous pumpers early on and then found themselves with liters of extra milk and very uncomfortable breasts.

  The main problems with an oversupply are that it can be very uncomfortable and can increase your risk of mastitis. Your breasts become engorged with milk; they are hard and hot, and they ache. Pumping can relieve the discomfort, but it contributes to the feedback loop and prolongs the issue. If you want to calm down the supply, you have to deal with the engorgement problem.

  There are a variety of recommended techniques to do this—acupuncture, acupressure, particular kinds of massage, cold packs, hot packs, breast-shaped hot packs, cabbage leaves, and so on.25 The evidence on these is spotty—there are a few randomized trials, most of which are small and subject to some bias. Cold and hot packs do seem to provide some relief, as do cold or room-temperature cabbage leaves. (Yes, you read that right: cabbage. You keep the leaves in the fridge and wrap your breasts in them. No one said being a mom was glamorous.)

  One trial shows some benefit of something called gua sha therapy, which involves scraping the skin to produce light bruising. Gwyneth Paltrow swears by this, so take what you will from that.

  In addition to pain, an issue with oversupply is that when the baby does start to nurse, the milk may come very fast and overwhelm him, making it hard to actually eat. Basically, it is like you trying to drink from a firehose. Pumping for a couple of minutes—or hand-expressing milk—right before you nurse can help with this problem. It will also improve as the baby gets bigger and the oversupply problems calm down.

  THE BREASTFEEDING DIET

  “Hi Emily!” Humphrey wrote. “The baby is doing great. But Maggie’s parents say she can’t eat cauliflower or drink tap water because she is breastfeeding. They said the baby will cry more. Could this be right?”

  After nine months of careful food avoidance, it adds insult to injury to think that breastfeeding will introduce a similar set of restrictions. Can you return to your rare steak? Those unpasteurized cheeses you’ve been craving—are they still off-limits? And what about a glass of wine—or even a couple of glasses? Is that okay?

  Good news: mostly, breastfeeding moms have no dietary restrictions.

  Let’s start with the food part. The only food women are medically advised to avoid during breastfeeding is high-mercury fish.26 That’s it! No swordfish, king mackerel, tuna. But other fish are fine, as are unpasteurized cheeses, sushi, rare steak, deli meats, and on and on.

  If your baby is suffering from colic—excessive crying as an infant—there is some evidence that avoiding common dietary allergens could help. For more on this, see this page.

  What about cauliflower?

  There is something of an old wives’ tale that gassy foods (cauliflower, broccoli, beans) lead to a gassy baby, and can make colic worse. I can find only one paper on this, and it is based on a mail survey that asked parents about many foods and compared the food consumption for babies with colic to those without.27 Although this study did claim to find some minimal evidence that cauliflower and broccoli lead to more colic, the problems with the data collection and analysis are so significant (use of mail survey with poor response rate, excessive response among people who were hyperconcerned about breastfeeding, problems with statistical precision) that I think it is safe to ignore it.

  Eat what you want.

  What about alcohol? Many women hear—from the internet, typically not from their doctors—that they should avoid alcohol altogether, or that if they drink at all, they should “pump and dump.” On the other side, some people will tell you that having alcohol (beer, specifically) will increase your milk supply. So you should have more! Are either of these true?

  No, not really.28

  When you drink, the alcohol level in your milk is about the same as your blood alcohol level. The baby consumes the milk, not the alcoh
ol directly, so the level of alcohol they are exposed to is extremely low. One paper carefully calculates that even if you had four drinks very quickly and then breastfed at the maximum blood alcohol level, the baby would still be exposed to only a very, very low concentration of alcohol, one that is extremely unlikely to have any negative effects.29 And this is in a kind of “worst-case scenario.” This paper cautions that drinking four drinks quickly will impair your ability to parent and is not healthy, so it should be avoided, but the issue isn’t alcohol in your breast milk. Therefore, there is no need to pump and dump. The milk has the same alcohol concentration as your blood. As that goes down, so does the milk alcohol level. It isn’t stored in the milk.

  Given this, it is not surprising that we do not find much evidence of the impact of a mother’s alcohol consumption on her infant. There are some reports that babies sleep in shorter intervals when they consume milk after their mom has been drinking, but this isn’t supported in all studies. And no long-term impacts have been identified.

  What if you want to be super, super cautious and not expose your baby to alcohol at all? No problem. You can have a drink, but you need to wait for two hours afterward to let the alcohol metabolize before breastfeeding. For two drinks, that increases to four hours.30

  These studies all caution—correctly—that we do not know much about binge drinking, or frequent heavy drinking (three or more drinks every day). Many women who binge drink frequently also did so during pregnancy, and they differ in other ways from women who do not binge drink. Even if you are not pregnant or nursing, binge drinking isn’t good for your health. Binge drinking during pregnancy is very dangerous for your baby, and after birth, it will impair your ability to parent.