Is It Something I Ate? Baby Wind, Colic and Elimination Diets While Breastfeeding

It is three in the morning, the baby has been crying for an hour, and you are mentally rewinding through everything you ate yesterday looking for the thing you did wrong.

Almost everyone does this. The answer is usually that you did nothing, and the crying is not a message about your dinner. But there is a genuine exception worth knowing properly, because the people it applies to often spend months being told it is just colic.

Most Wind Is Not About Your Diet

Start with the reassuring part, which is also the evidenced part.

The foods that produce gas in an adult do it through fibre being fermented in your own large intestine. Those gas-producing compounds are not what crosses into breast milk. So cabbage, broccoli, beans, onions and cauliflower do not make a baby windy, however persistent the belief is.

Spicy food is fine too. Flavours do reach your milk, and that turns out to be useful rather than a problem, because early exposure to varied flavours is linked to better acceptance of varied food later in weaning.

Most newborn wind is swallowed air and an immature digestive system. Babies gulp, they feed fast, they have an underdeveloped gut, and they have not yet worked out how to move gas through it. That improves with time, and almost nothing you eat changes its course.

What Normal Crying Actually Looks Like

Useful to have a benchmark, because without one every hard evening feels like evidence of a problem.

Crying in healthy babies follows a predictable curve. It increases from around two weeks, peaks somewhere around six to eight weeks, and settles down by three to four months. Late afternoon and evening are the worst windows, which is exactly when you are most tired and least able to be philosophical about it.

Colic is the label for the top end of that curve rather than a disease in itself. The traditional definition is crying for more than three hours a day, more than three days a week, for more than three weeks, in a baby who is otherwise healthy and growing well. Roughly one in five babies meets it.

The detail that matters most in that definition is otherwise healthy and growing well. A baby who is crying a lot but feeding, gaining and developing normally is a different situation from one who is crying and also not thriving.

Rule These Out First

Before touching your diet, there are four things that cause far more windy, unsettled babies than food ever does.

Latch and positioning. A shallow latch means more swallowed air at every feed. This is the single most common fixable cause, and a feeding assessment costs you one appointment.

Fast letdown or oversupply. If your milk comes fast, babies gulp to keep up, swallow air, pull off, splutter and get frustrated. It looks exactly like a reaction to something you ate. Feeding more upright, or leaning back so they work against gravity, often transforms it.

Tongue tie. Can restrict the seal and lead to the same air swallowing, and it is often missed at first.

Overtiredness. An overtired newborn cries in a way that looks like pain. Evening crying and evening tiredness overlapping is not a coincidence.

One mother described working through it the other way round. She cut dairy, then soya, then eggs, then wheat over about six weeks, until she was eating from a list of four foods and had lost more weight than she was comfortable with. The baby cried exactly as much. A feeding assessment eventually found a shallow latch, and the crying improved within a fortnight while she was eating normally again.

That sequence is common, and it runs in the wrong order. Diet is the thing to check last, not first.

The Exception That Is Real

Cow's milk protein allergy is the one dietary cause with genuine evidence behind it.

It affects somewhere around 2 to 3 percent of infants, and unlike the gas-producing compounds, cow's milk protein from your diet does pass into breast milk in small amounts. In a sensitised baby, that is enough to produce symptoms.

It comes in two forms, and telling them apart matters because they look nothing alike.

	Immediate type	Delayed type
Timing	Minutes to about 2 hours	2 to 72 hours later
Skin	Hives, swelling of lips or eyes	Worsening eczema, often stubborn
Gut	Vomiting soon after a feed	Reflux, diarrhoea or constipation, blood or mucus in the nappy
Other	Wheeze, breathing changes	Persistent distress, poor weight gain, feed refusal
Recognised?	Usually, quickly	Often missed for months

The delayed type is the one that gets labelled colic, because it has no dramatic moment to point at. The pattern that should prompt a conversation is crying plus something else: eczema that will not settle, blood or mucus in the stool, or weight that is not tracking as it should. Crying on its own, in a thriving baby, is much more likely to be the normal curve.

Immediate symptoms involving breathing or swelling need urgent medical attention, not a diet experiment.

How an Elimination Trial Should Actually Be Run

If the picture genuinely fits, an exclusion trial is reasonable. It should be done with a health visitor, GP or dietitian rather than alone, and it has a specific shape.

One food group, not several. Cow's milk protein first, since it is by far the most likely. Cutting four things at once tells you nothing about which mattered.

Properly strict, or it does not count. That means milk, butter, cheese, yoghurt, cream and the milk hidden in bread, biscuits, sauces and processed food. A partial exclusion produces a partial answer, which is worse than no answer.

For two to four weeks. Delayed reactions take time to clear, so a three day trial proves nothing. Equally, going beyond four weeks with no change means the answer is no.

Write down what you see, not what you remember. Sleep deprivation makes memory unreliable, and a simple daily note on crying, nappies and skin is the difference between a real result and a hunch.

The Step Everyone Skips

Reintroduction is what turns a trial into a diagnosis, and it is the part most people never do.

If symptoms improve, you bring the food back and watch. If they return, you have an answer worth acting on. If nothing happens, the improvement was probably the normal crying curve flattening out at the same time, which is extremely common given that most trials start around the peak.

Without that step you can end up avoiding dairy for a year on the strength of a coincidence. Most babies with cow's milk protein allergy grow out of it during the first few years anyway, so even a confirmed diagnosis is usually temporary and needs reviewing.

What It Costs You Nutritionally

This is the part that gets skipped in the panic, and it is the reason not to cut things speculatively.

Breastfeeding raises your calcium requirement substantially, and dairy is where most people get theirs. Take it out without replacing it and you are running a deficit at a point when your body is already being asked for a lot.

Non-dairy calcium comes from fortified plant milks, which is the easiest swap, along with tinned sardines and salmon with the bones, tofu set with calcium, almonds, sesame and tahini, kale and other dark leafy greens, and dried figs. Choose fortified plant milks deliberately, because the unfortified organic versions often contain almost none.

Vitamin D matters alongside it, since calcium absorption depends on it and postpartum levels are frequently low. Our dairy-free nutrition guide has the full replacement list, and it works just as well postpartum as it does in pregnancy.

Ask for a dietitian referral if dairy is coming out for more than a few weeks. It is a reasonable request, and getting it right is much easier with someone checking the numbers.

Things Not Worth Cutting

For completeness, since all of these get suggested.

Gassy vegetables. No transfer, no mechanism, no evidence. Spicy food. Fine. Common allergens in general. Guidance changed years ago, and avoiding peanuts, eggs and dairy without a diagnosed reason is not protective. Our guide to introducing allergens covers the baby's side of that.

Lactose. Worth naming because lactose-free products get recommended constantly and usually miss the point. When there is a genuine cow's milk problem it is almost always the protein, not the lactose, and lactose-free milk still contains the protein. Primary lactose intolerance in babies is very rare.

Caffeine is the one small exception. A little reaches your milk and newborns clear it slowly, so if your baby is unusually wired, trialling it down for a few days is harmless. Our breastfeeding guide has the numbers.

When to Get Seen Quickly

Do not wait out a trial if you see blood or mucus in the nappy, faltering growth or dropping centiles, vomiting after most feeds rather than ordinary posseting, severe eczema that treatment is not touching, feed refusal, or any swelling, wheeze or breathing change.

And one for you rather than the baby. Relentless crying is genuinely hard on a parent, and feeling desperate about it is not a character flaw. If it is affecting your mood or you are running out of coping room, tell your health visitor or GP. Our guide to eating in the first six weeks deals with the practical side of keeping yourself fed through it, which is the thing that usually goes first.

This article is general information and does not replace medical advice. Do not start an exclusion diet while breastfeeding without speaking to your health visitor, GP or a dietitian first, and seek urgent help for any swelling, wheeze or breathing difficulty.