The Association of Paternal and Maternal Body Mass Index on Childhood Overweight

Abstract

Purpose: Obesity often aggregates within families and parental obesity has been identified as a risk factor. To examine the different effects of maternal and paternal body mass index on the prevalence rate and risk of overweight in children. Materials and Methods: A total of 1418 students, aged 12 or 13 years, were recruited for analysis. Children's body height, weight and waist and hip circumferences were obtained. Self-reported body height and weight of their parents were collected. The 85th percentile of body mass index (BMI) was used as cutoff points to identify the children being overweight. Spearman's correlation and logistic regression analysis were used to examine the parent-offspring BMI relationship. Paternal and maternal BMI were classified into four groups using the quartile categories to examine their different effects on the prevalence rate and risk of children overweight. Results: The association of children's BMI with parents' BMI was the strongest, closely followed by children's waist circumference whereas waist-hip ratio barely had any association. The Spearman's correlation coefficient of the mother-daughter association was the highest among all the parent-offspring association. Among the boys, the prevalence rate for being overweight increased significantly from 9.2% in the lowest quartile to 21.3% in the highest quartile of paternal BMI (p trend＜0.005). The effect of maternal BMI on the prevalence rate for being overweight was also noted (p trend＜0.0001). Among the girls, an increasing rate of being overweight for the increasing quartile of paternal and maternal BMI was also noted (p trend＜0.0001). Multiple logistic regression analysis revealed that maternal BMI was significantly associated with boys' overweight with the odds ratios increased from 1.0 in the lowest quartile to 2.33 in the highest quartile of maternal BMI (p＜0.001). However, the association between paternal BMI and boys' overweight was marginally significant (p=0.055). Among the girls, both maternal and paternal BMI were significantly associated with girls' overweight (p＜0.001 and p=0.001, respectively). Conclusion: Maternal BMI was an independent risk factor of overweight in both boys and girls. Paternal BMI was also associated with girls' overweight.