Children who develop eczema or other skin and respiratory allergies early in life face a higher risk of developing a food allergy, according to a recent study published in JAMA Pediatrics.

Researchers analyzed 190 studies involving nearly 2.8 million children across 40 countries and estimated that 4.7% of children develop immunoglobulin E (IgE)-mediated food allergy by age 6 when clinicians confirm diagnosis through oral food challenge (95% CI, 3.2%-6.9%). 

“The study confirms that the risk of a child developing food allergy is relatively low — less than 5 out of 100 babies will develop food allergy,” Matthew Greenhawt, MD, professor in the Department of Pediatrics Section of Allergy and Immunology at Children’s Hospital Colorado and study co-author said in a written statement. “The study identified 342 risk factors, but the results tell us that there is no single cause or risk for food allergy.”

Early allergic disease, delayed introduction of allergenic foods, impaired skin barrier function, and select perinatal and demographic factors drove most of the excess risk, the study found. 

The analysis also found that vaccines are not associated with food allergy risk, a finding that Greenhawt, who is also chief medical officer of the Asthma and Allergy Foundation of America (AAFA), says counters persistent misinformation.

One of the strongest predictors was eczema during infancy. Children who developed atopic dermatitis in the first year of life had nearly four times the odds of developing food allergy compared with children without eczema (odds ratio [OR], 3.88; 95% CI, 3.03-4.97). 

The findings reinforce what many allergists already see in practice. 

“In eczema early on, especially with severe eczema and moderate eczema, they do seem to be a high-risk group for the development of food allergies,” said Michael Pistiner, MD, director of Food Allergy Advocacy, Education, and Prevention at the Food Allergy Center in Massachusetts General Hospital in Boston.

Indeed, each 5- to 10-point increase in SCORAD score (an atopic dermatitis severity scoring system) increased the odds of food allergy (OR, 1.22; 95% CI, 1.12–1.34), suggesting that the severity of eczema further raises risk. 

Pistiner said the findings are consistent with research supporting the “dual exposure hypothesis,” in which allergic sensitization occurs through the skin when the skin barrier is disrupted.

“The skin seems to be the sensitizing route,” said Pistiner, who was not involved in the study. He added that inflammation and barrier dysfunction may allow exposure to food proteins that trigger immune responses leading to allergy.

Other early allergic conditions also strongly predicted food allergy. Allergic rhinitis or conjunctivitis more than tripled the odds (OR, 3.39; 95% CI, 2.50-4.61; risk difference, 10.1%; 95% CI, 6.7%-14.4%), while early-life wheezing doubled the odds (OR, 2.11; 95% CI, 1.46-3.04; risk difference, 5.0%; 95% CI, 2.1%-8.8%).

The findings reinforce the concept of the atopic march, in which early skin and airway disease signal later food allergy risk.

Measures of skin barrier dysfunction showed similarly strong associations. Children with higher transepidermal water loss had more than three times the odds of developing food allergy (OR, 3.36; 95% CI, 2.41-4.68; risk difference, 10.0%; 95% CI, 6.3%-14.8%). Loss-of-function variants in the filaggrin gene nearly doubled the odds (OR, 1.93; 95% CI, 1.51-2.45; risk difference, 4.2%; 95% CI, 2.4%-6.4%).

The timing of food introduction stood out as a modifiable risk factor. Children who were introduced to peanut after 12 months of age had higher odds of food allergy than those who were introduced to peanut earlier (OR, 2.55; 95% CI, 1.40-4.64). Delayed introduction of egg, fish, fruit, and other foods showed similar but smaller effects.

The American Academy of Pediatrics (AAP) supports early introduction of allergenic foods, especially for infants at higher risk.

“Tolerance is thought to be achieved if the infant has exposure through the gut,” said Pistiner said. “That can help keep a child from developing allergy.”

But clinical implementation has not always been straightforward, Pistiner said. Some of the barriers include limited access to allergists, family concerns about allergic reactions, and time constraints in pediatric visits, he said. 

“A busy primary care clinician is talking about vaccines, trampolines, firearms,” said Pistiner. “They may not have the time to actually go into some of the early introduction guidance.”

Those barriers make shared decision-making important, he said, especially for infants with moderate or severe eczema who may benefit most from early allergen introduction.

“If we have a young child who has severe eczema, you want to have these conversations early on and potentially have a plan for how you want to approach the early introduction of allergens in the child,” Pistiner said.

Antibiotic exposure in infancy also increased risk. Infants who received systemic antibiotics in the first month of life had more than four times the odds of developing food allergy (OR, 4.11; 95% CI, 1.08-15.6), although estimates were imprecise (risk difference, 12.8%; 95% CI, 0.4%-40%). Antibiotic exposure later in the first year of life showed a more modest association (OR, 1.39; 95% CI, 1.14-1.70; risk difference, 1.8%; 95% CI, 0.8%-3.1%).

Several demographic and birth-related factors contributed to smaller but measurable increases in risk. Male children had slightly higher odds of food allergy than female children (OR, 1.24; 95% CI, 1.15-1.34; risk difference, 1.1%; 95% CI, 0.7%-1.6%), and firstborn children also faced higher risk (OR, 1.13; 95% CI, 1.06-1.20; risk difference, 0.6%; 95% CI, 0.3%-1.0%). Cesarean delivery modestly increased risk (OR, 1.16; 95% CI, 1.05-1.28; risk difference, 1.0%; 95% CI, 0.3%-1.2%).

Family history remained a strong predictor. Children with a sibling who had food allergy had more than double the odds of developing food allergy themselves (OR, 2.36; 95% CI, 1.86-2.99; risk difference, 6.0%; 95% CI, 4.4%-8.0%). Parental food allergy and other allergic diseases also increased risk, as did parental migration before birth (OR, 3.28; 95% CI, 1.92-5.60; risk difference, 9.7%; 95% CI, 4.9%-16.3%).

In contrast, the analysis found no meaningful increase in absolute risk associated with low birth weight, postterm birth, maternal diet during pregnancy, or maternal stress.

For clinicians, Pistiner said the findings reinforce the direction of current guidelines.

“We have good science here,” he said. “This recent publication actually supports the direction we’re already moving in.”

Pistiner has served as a consultant for AAFA, AAN, Kaleo Inc., DBV Technologies, Aimmune, Novartis, Bryn, Anjo, and FoodGraph; received funding from Kaleo Inc., DBV Technologies, ARS, Aquestive Therapeutics, Stallergenes Greer, National Peanut Board, Egg Nutrition Center, AAN and AAFA; and is co-founder of AllergyHome and Allergy Certified Training and FAMP-IT.org. Greenhawt reported having no relevant disclosures.

Lara Salahi is a health journalist based in Boston.