IntroductionOver the decades the trends of early onset of puberty have been observed in children, particularly in girls. Research evidence has reported diet to be among the most important risk factors for puberty onset. This study evaluated the association between dietary behavior and puberty in girls.MethodsWe enrolled 201 girls with the main complaints of breast development as the cases at the Endocrine Department of Nanjing Children's Hospital. The cases were divided into breast development with central priming and breast development without central priming groups and were matched with 223 normal health girls with no breast development (control group). We used the modified Child Eating Behavior Questionnaire (CEBQ) to conduct a face-to-face interview about dietary behavior. Sample t-test or Mann Whitney U test or Chi-square test, the analysis of variance or Kruskal Wallis test, and least significant difference (LSD) were used to compare differences between the groups, Bonferroni was used to correct the p-value, and logistic regression was used to analyze risk factors for puberty onset.ResultsA total of 424 girls participated in this study, among them, 136 were cases with breast development with central priming, 65 were cases with breast development without central priming, and 223 were normal health girls with no breast development. Age of the participants ranged from 4.5 to 9.3 years. There were significant differences in food response (p < 0.001), dietary restriction (p < 0.001), frequencies of vegetable intake (χ2 = 8.856, p = 0.012), drinking milk (χ2 = 23.099, p = 0.001), and borderline statistical difference in a total score of unhealthy dietary behavior (p = 0.053) among the cases and controls. However, in the post hoc analysis, these dietary behaviors were significant differences between the girls with breast development with central priming and the control groups. Moreover, girls in the breast development with central priming group had significantly higher bone age (BA), uterine body length, ovarian volume, basal luteinizing hormone (LH), basal follicle-stimulating hormone (FSH), peak LH, peak FSH, estradiol (E2), and free triiodothyronine (FT3) compared to those in the breast development without central priming group. In the multivariate logistic regression, only uterine body length was associated with increased risk of breast development with central priming (OR = 1.516, 95%CI: 1.243–1.850).ConclusionThere were significant differences in dietary behaviors among girls with breast development with central priming and normal health girls with no breast development, and uterine body length was associated with an increasing risk of breast development with central priming among girls with breast development.