Antibiotic Prescribing during Infancy and Risk of Treated Obstructive Airway Diseases during Early Childhood: A Registry-Based Nationwide Cohort Study in Denmark

Widespread antibiotic use leads to bacterial resistance, and antibiotic use in early life may be associated with asthma in childhood. To date, studies of this association have led to inconsistent findings. Additionally, data are limited regarding cohort effects on antibiotic use in children, which may impact underlying susceptibility to adverse effects. Using nationwide registry data on all children born in Denmark during 2004-2012, our objectives were to (1) examine birth-season and birth-year cohort effects on antibiotic prescribing during the first year of life (henceforth, ‘infancy’), and (2) to estimate 1-, 2-, and 3-year risk differences (RD) for the association between antibiotic prescribing during infancy and treated airway diseases from 2-5 years of age, using propensity scores (PS) and instrumental variables. The 1-year risk of redeeming at least one antibiotic prescription during infancy was 39.5 per 100 children. The hazard of first redeemed antibiotic prescription increased with age throughout infancy, and peaked in February; as a result, season of birth impacted overall 1-year risk of redeeming an antibiotic prescription during infancy and age at first redeemed antibiotic prescription. Amoxicillin prescribing was dynamic over the study period, but decreased after distribution of a bulletin on rational antibiotic use in general practice and rollout of two nationwide pneumococcal vaccination programs. In PS analyses, antibiotic exposure was associated with increased risk of treated airway diseases by age 5, compared with no exposure (3-year RD = 4.5 per 100 children, 99% confidence interval (CI): 4.2, 4.8). PS-based dose-response analysis suggested that each additional redeemed antibiotic prescription was associated with increased risk of 2.4 per 100 children (99% CI: 2.3, 2.5). RDs were negligible in a PS-based head-to-head comparison between two antibiotics with similar indications but differing spectrum of antibacterial activity – amoxicillin and penicillin V (3-year RD = -0.1 per 100 children, 99% CI: -0.6, 0.3). Results from instrumental variable analyses also cast some doubt on the presence of a causal effect, but were imprecise. These results suggest that antibiotic exposure during infancy may increase the risk of treated airway diseases, but further exploration is needed using data and methods capable of addressing potential residual confounding.Doctor of Philosoph