The incidence of immunoglobulin E–mediated food allergy diagnosed by food challenge is likely about 5% by age 6 years, according to a systematic review and meta-analysis of 190 studies including approximately 2.8 million participants published in JAMA Pediatrics. The strongest and most credible risk factors included early allergic disease, delayed allergen introduction, genetic susceptibility, antibiotic exposure, demographic factors, and birth-related variables.
The researchers synthesized data from studies conducted across 40 countries. A total of 16 studies (n = 18,279 participants) assessed the incidence of immunoglobulin E–mediated food allergy confirmed by food challenge.
Pooled analysis showed that the baseline incidence of food allergy was 4.7%. Regional variation was observed, with higher estimated incidence in Australia (10%) and the US (7%) and lower incidence in the Middle East (2%) and Africa (2%). Certainty of evidence for incidence was rated moderate.
For risk factors, 176 studies identified 342 potential predictors. Thirty-eight factors were supported by high-certainty evidence, 69 by moderate-certainty evidence, 120 by low-certainty evidence, and 115 by very low-certainty evidence.
Early Allergic Disease, Skin Barrier, and Genetics
The strongest associations were observed with early allergic disease. Atopic dermatitis during the first year of life was associated with 3.88 times the odds of developing food allergy. Allergic rhinitis was associated with 3.39 times the odds, and wheeze with 2.11 times the odds.
Disease severity was also associated with increased risk. Increasing atopic dermatitis severity, measured by Scoring Atopic Dermatitis, was associated with 1.22 times the odds of food allergy for each 5- to 10-point increase in score.
Markers of skin barrier dysfunction were also associated with food allergy. Higher transepidermal water loss was associated with 3.36 times the odds of food allergy, and filaggrin gene loss-of-function variants were associated with 1.93 times the odds.
Delayed Food Introduction and Antibiotics
Delayed introduction of allergenic foods was also associated with increased risk. Introduction of peanut after 12 months of age was associated with 2.55 times the odds of food allergy, and similar patterns were observed for delayed introduction of fish, egg, and fruit.
Early-life antibiotic exposure was also associated with increased risk. Systemic antibiotic exposure during the first month of life was associated with 4.11 times the odds of food allergy. Antibiotic exposure within the first year of life was associated with 1.39 times the odds, and antibiotic exposure during pregnancy was associated with 1.32 times the odds.
Demographic, Familial, and Birth-Related Factors
Several demographic and familial factors were also associated with increased risk. Male sex was associated with 1.24 times the odds of food allergy, and being firstborn was associated with 1.13 times the odds.
Family history of food allergy was also associated with increased risk. Maternal history of food allergy was associated with 1.98 times the odds, paternal history with 1.69 times the odds, both parents affected with 2.07 times the odds, and affected siblings with 2.36 times the odds.
Parental migration before birth was associated with 3.28 times the odds of food allergy. Self-identification as Black vs White was associated with 3.93 times the odds, and vs non-Hispanic White with 2.23 times the odds.
Birth-related factors showed smaller associations. Cesarean delivery was associated with 1.16 times the odds of food allergy, and increasing maternal age was also associated with increased odds.
Several factors showed no meaningful association with food allergy, including low birth weight, postterm birth, partial breastfeeding, maternal consumption of fish or cheese during pregnancy, maternal stress during pregnancy, and high household income.
Subgroup analyses found no credible subgroup differences across risk of bias, food allergy definitions, study year (pre- vs post-2015), or the number of allergenic foods assessed (single vs multiple).
Overall, 66% of included studies were rated as having high or probably high risk of bias in at least one domain, most commonly related to prognostic factor measurement.
Limitations
The researchers noted that many reported risk factors were supported by low- or very low-certainty evidence, often derived from single small studies. Studies also adjusted for different sets of confounders and rarely accounted for all relevant predictors, making it difficult to determine how multiple risk factors interact or whether effects are additive.
In addition, observational associations cannot establish causality. Most evidence came from higher-income countries, which may limit generalizability.
Risk factor analyses also included studies with and without food challenge confirmation, even though incidence estimates were restricted to challenge-confirmed diagnoses. The researchers noted that the lack of modern US birth cohorts using protocolized food challenges remains a gap that the ongoing SunBEAm birth cohort may help address.
The researchers concluded that food allergy likely arises from the interaction of genetic, microbial, environmental, social, and comorbid mechanisms combined with allergen exposure, often involving one or more major risk factors alongside additional contributing factors.
Disclosures: Several researchers reported advisory roles, consultancy work, research funding, and other relationships with pharmaceutical companies and advocacy organizations. Full disclosures are provided in the published article.
Source: JAMA Pediatrics