Statistical analysis
Two groups were compared using χ2 test. The main outcomes were asthma, allergic rhinitis/conjunctivitis, and atopic dermatitis in children. Multivariable logistic regression was used to estimate the adjusted odds ratios with 95% confidence intervals for the following potential factors. Based on previous reports, the factors related to the incidence of asthma, including daycare attendance, existence of older siblings, a history of hospitalization owing to cold/bronchitis/bronchiolitis/pneumonia, preterm birth, and overweight gain were analyzed and adjusted for the potential confounders of asthma, including sex, maternal smoking, breastfeeding, and presence of early atopic dermatitis. We used the same factors to analyze allergic rhinitis/conjunctivitis; in addition, to analyze atopic dermatitis, we used the same factors mentioned barring “presence of early atopic dermatitis.” The probability values for the statistical tests were two-tailed, and P < 0.05 was regarded as statistically significant. All statistical analyses were performed via JMP system version 15 (SAS Institute Inc., Cary, NC, USA).