Babies who cried inconsolably and also had eczema were more often born to mothers who ate little fish in pregnancy. Babies with colic alone showed no such link

Colic is one of the oldest puzzles in baby care. A healthy, well-fed infant cries for hours, draws up its legs, and nothing settles it. Doctors now describe it as a disorder of the conversation between gut and brain, which is a polite way of saying nobody is sure what causes it. Most of the diet research has looked at what happens after birth: breast milk, formula, the mother's diet while breastfeeding. A paper published online on 12 October 2026 in JAMA Network Open looks further back, at what mothers ate while they were pregnant.

The data come from Project Viva, a long-running study that enrolled pregnant women at practices in eastern Massachusetts between 1999 and 2002 and has followed their children since. The mothers were not picked for any family history of allergy. In early pregnancy, at about ten weeks, and again at about 28 weeks, they filled in a detailed food questionnaire. From those answers the researchers built two scores: a Mediterranean diet score from 0 to 9, and a dietary inflammatory index, where a higher number means a diet richer in foods linked with inflammation. They also counted weekly servings of the foods most often behind allergies: fish and shellfish, nuts and peanuts, dairy and eggs.

When the babies were about six months old, the mothers answered two questions. Had the baby ever had times when it seemed to be in agony, screaming, drawing its legs up and could not be calmed? If that had happened sometimes or often, the study counted it as colic. And had a health professional ever said the baby had eczema? Of the 1,324 mother and baby pairs with answers to both the diet and the baby questions, 273 babies had colic only, 152 had eczema only, 61 had both, and 838 had neither.

Here is the result that the headline numbers tend to blur. For colic on its own, neither diet score made any measurable difference. For eczema on its own, the same. The link appeared only in the 61 babies who had both. Each extra point on the Mediterranean score went with about 15% lower odds of a baby being in that group. Each extra point on the inflammatory index went with about 34% higher odds. Every weekly serving of seafood went with about 25% lower odds. These figures were adjusted for the mother's age, education, race and ethnicity, earlier births, smoking and her own history of allergy, asthma or eczema.

Dairy and eggs showed nothing either way. Nuts and peanuts gave a small surprise: a little more colic without eczema among mothers who ate more of them, about 5% higher odds per weekly serving, with the lower edge of the confidence range sitting exactly at no effect. The authors had expected the opposite and say so.

Why would diet matter for colic plus eczema but not for colic alone? The authors' suggestion is that colic is not one condition but several, which all look like the same inconsolable crying. In some babies it may be the first sign of an allergic, inflamed type of immune system, the kind that also shows up in the skin. If that is right, a mother's diet could plausibly nudge that type and leave the others alone. The omega-3 fats in fish are the obvious candidate, and trials of omega-3 supplements in pregnancy have hinted at less eczema and wheeze in the children. This study cannot test that idea. It only shows that the pattern fits it.

There are reasons for caution, and the paper names most of them. Both colic and eczema were reported by the mothers, not checked in medical records. The colic question is broad: one baby in four in this study met it, far more than would meet a strict clinical definition. Food questionnaires are rough tools. The families all had prenatal care and health insurance, and the pregnancies took place a quarter of a century ago, when advice on fish, peanuts and allergy was different. Many families dropped out before the six-month questions, and an observational study can never fully rule out that healthier eating simply travels with other things that protect babies.

One more detail is worth a careful reader's attention. The paper's key points box says the link between diet and colic was stronger in babies whose mothers had allergies themselves. Its results section, describing the same analyses in more detail, says the nut and seafood links with colic did not notably differ by the mother's allergy history. The breakdown sits in an online supplement. It is a small inconsistency, but it is the kind that gets repeated in summaries.

So what can a pregnant reader take from this? Nothing that changes current advice, and nothing that contradicts it either. Health bodies in the United States, the UK and Europe already encourage eating fish that is low in mercury, two to three servings a week in the US advice, during pregnancy, mostly for the baby's brain and eyes. This study adds a modest, unproven hint that the same habit might also go with fewer babies who have both colic and eczema. It gives no reason to cut out nuts, dairy or eggs, and no reason for a parent of a crying baby to look back at their pregnancy menu with guilt.