| Objectives: To assess factors that potentially influence the BMI z scores(BMI_Z) of children from one to three years of age through a cohort study, and to provide evidence for early prevention of childhood overweight and obesity.Methods: Children who were Wuhan residents were recruited from Wuhan Maternal and Child Health Care System from November 2010 to October 2013 before children’s one year birthday after parents’ informed consent. Children were defined to high risk group(Hr G) if both parents were overweight or at least one obese, otherwise to the control group(Ct G). A baseline survey was conducted at children’s one year of age through indoor visits, with both anthropometric measurements and questionnaire information collected, including parental social-economic status, children’s delivery outcomes, feeding pattern and sleeping duration of children were included. Families were followed up at children’s second and third year birthdays thereafter. Multiple regression models were adopted to evaluate the association between children’s BMI_Z and parental BMI, children’s birth weight, feeding pattern and sleeping duration with confounding factors adjusted for at one, two and three years old respectively.Results:(1) Basic characteristics: there were 283 families included in the analysis, among which 97 were in the Hr G and 186 in the Ct G. Maternal pre-pregnancy BMIs of the Hr G and the Ct G were 23.1±4.7 kg/m2 and 20.3±2.2 kg/m2(P<0.05), with the corresponding value of fathers being 28.6±2.9 kg/m2 and 23.5±2.9 kg/m2(P<0.05), respectively. Significant difference was found in maternal age between the two risk groups as 29.9±3.7 years in Hr G and 28.9±3.4 years in the Cr G. No significant difference was found for other variables(P>0.05). Overall, the rate of boys to girls was 1.3:1 and the rate of cesarean section was 77.7%; the mean of birth weight was 3386.5±499.0g. The average age of father at children’s birth was 31.5±4.3 years. Most parents, 198 fathers(71.7%) and 209 mothers(73.9%), were in high eductional level. The majority of parents were company and institutions employees. A total of 67.6% families had monthly income between 3001 RMB and 10000 RMB; the central family and the immediate family accounted for 41.3% and 58.7%, respectively. The follow-up rates of the second and third year were 71.4% and 65.7%, and no significant difference was detected for basic characteristics between the two follow up years and the baseline population(P>0.05).(2) No significant difference was detected in the high BMI rate of children between the Hr G and the Ct G(P>0.05); the former group of children had significantly higher BMIs than those in the Ct G at age one to three years(P<0.05); and the similar trend applied for BMI_Z(P<0.05).(3) Results from multiple regression models indicated that with the adjustment of maternal age and children sex, paternal BMI was significantly associated with children’s BMI_Z(P<0.05), maternal pre-pregnancy BMI was significantly associated with children’s BMI_Z at age two and three(P<0.05). Maternal pre-pregnancy BMI had a stronger influence on children’s BMI_Z compared to paternal BMI. Maternal gestational weight gain had a positive correlation with children’s BMI_Z at age one(P <0.05), but this effect became insignificant at age two and three. A positive correlation between birth weight and children’s BMI_Z was not found until children were three years of age(P<0.05). Sleeping duration at night is a major factor affecting children BMI_Z at age one(β=0.28, P<0.001).Conclusion: Parental BMI appears to have continuous impacts on children’s BMI_Z from age one to three. Maternal gestational weight gain, duration of breastfeeding and sleeping during at night were significantly associated with children’s BMI_Z at the first year of life; children’s birth weight had impact on children’s BMI_Z at three years old. Parental weight status, maternal gestational weight gain, feeding pattern and sleeping mode in early childhood should be taking into account for early prevention of childhood obesity. |