BACKGROUND: While the etiopathogenesis of atopic dermatitis is complex and poorly understood, neonatal exposures are important for disease occurrence. However, the effect of dog exposure on the risk of atopic dermatitis is unresolved.

OBJECTIVE: We investigated whether domestic dog exposure affected the risk of atopic dermatitis in children during the first 3 years of life.

METHODS: Copenhagen Prospective Studies on Asthma in Childhood (COPSAC) are ongoing prospective clinical birth cohort studies. Data from 411 children born to mothers with asthma (COPSAC2000 ) and 700 unselected children (COPSAC2010 ) were analyzed following the same protocols at the same research site. Atopic dermatitis was diagnosed prospectively according to the Hanifin-Rajka criteria. Parental history of asthma, eczema, or rhinitis was defined by self-reported physician diagnosis. In the COPSAC2000 , maternal specific serum IgE against eight inhalant allergens was sampled after the children's birth and at pregnancy week 24 in the COPSAC2010 cohort. Associations between dog exposure and atopic dermatitis were analyzed by Cox proportional hazard regression models and adjusted for lifestyle confounders.

RESULTS: In the COPSAC2000 and COPSAC2010 cohorts, the risk of atopic dermatitis was significantly lower in children with domestic dog exposure (adjusted HR = 0.46 [0.25-0.87], P = 0.02; and adjusted HR = 0.58 [0.36-0.93], P = 0.03, respectively). The risk of atopic dermatitis decreased in a dose-dependent manner with increasing number of dogs (adjusted HR = 0.58 [0.38-0.89], P = 0.01) in the COPSAC2010 . The protective effect was restricted to children born to mothers with atopic disease in the unselected COPSAC2010 cohort (adjusted HR = 0.39 [0.19-0.82], P = 0.01), as no effect was observed in children born to mothers without atopic disease (adjusted HR = 0.92 [0.49-1.73], P = 0.79). Paternal atopic status did not affect the risk of atopic dermatitis. We found no significant interaction between the CD14 T/T genotype and domestic dog exposure in either cohort (COPSAC2000 , P = 0.36; and COPSAC2010 cohort, P = 0.42).

CONCLUSION: Neonatal domestic dog exposure was associated with a strongly reduced risk of atopic dermatitis in two independent birth cohorts and in a dose-dependent manner. While the mechanisms involved are unclear, our findings raise the question of whether in utero exposures may affect the risk of atopic dermatitis and emphasize the importance of the early environment for disease trajectory.