Background Although an association between maternal nutritional intake and developmental delays in children has been demonstrated, the association between the timing of food intake and developmental delays remains unclear. Herein, we examined the association between breakfast intake frequency before and during pregnancy and developmental delay in children.Methods Of the pregnant women who participated in the Tohoku Medical Megabank Project Three-Generation Cohort Study, 7786 answered the required questions and were analyzed. The frequency of breakfast intake from pre- to early pregnancy and from early to mid-pregnancy was classified into four groups: daily, and 5–6, 3–4, and 0–2 times/week. Child developmental delays at age 2 and 3.5 years were assessed using the Ages & Stages Questionnaire, Third Edition. Logistic regression models were constructed to examine the association between breakfast intake frequency in pregnant women and developmental delays in children aged 2 and 3.5 years.Results The proportion of pregnant women who had breakfast daily was 78.2% in pre- to early pregnancy, and 82.3% in early to mid-pregnancy. The proportion of children with developmental delays was 14.7% and 13.4% at age 2 and 3.5 years, respectively. Compared with the risk in children of women who had breakfast daily from pre- to early pregnancy, children of women who had breakfast 0–2 times/week had a higher risk of developmental delays at 2 years of age: odds ratio (OR) 1.33, (95% confidence interval [CI], 1.06–1.68). The risk of developmental delays at age 2 years increased in the children of women who had breakfast 0–2 times/week in early to mid- pregnancy: OR 1.79 (95% CI, 1.36–2.36). Compared with children of women who had breakfast daily from pre- to early and early to mid-pregnancy, children of women who had breakfast 0–2 times/week were not associated with developmental delays at age 3.5 years: OR 1.11 (95% CI, 0.86–1.44 and OR 1.19 (95% CI 0.87–1.62), respectively.Conclusion For women with a low frequency of breakfast intake from pre- to mid-pregnancy there was an association with developmental delays in their children at age 2, but not at 3.5 years.