LAUSR

Objective: To examine risk factors of simple obesity and their interaction in preschool children in China. Methods: A total of 63 292 preschool children aged 3-7 years selected by cluster random sampling in 9 cities of China, including Beijing, Harbin, Xi'an, Shanghai, Nanjing, Wuhan, Guangzhou, Fuzhou and Kunming, were investigated from June to November in 2016. Based on the BMI-for-age cut off points of overweight and obesity for Chinese children aged 2-18 years, a total of 1 522 obese children (1 006 boys and 516 girls) were screened. By population-based matched case-control design, a normal weight child was randomly selected to match with an obese child by sex, age (difference ≤6 months) and body height (difference ≤5 cm) from the survey area. Conditional logistic regression model was used to analyze association of risk factors with obesity and the interaction of risk factors. Results: ① Univariate conditional logistic regression model showed that 17 risk factors of simple obesity had statistical significance (P<0.001), including high birth weight, cesarean section, gestational diabetes, gestational hypertension, formula feeding in the first 6 months after birth, strong appetite, fast or slow eating speed, low intensity of outdoor activities, daily outdoor activity time <1 hour, daily screen viewing time < 1 hour or ≥2 hours, daily night sleep time <9 hours, mother overweight, father overweight, mother's low educational level, father's low educational level, non-nuclear family structure, and parents not being the primary caregivers of children. ②Multivariate conditional logistic regression model showed that 12 risk factors had statistical significance, including high birth weight (OR=1.83, 95%CI:1.29-2.61, P<0.001), cesarean section (OR=1.22, 95%CI:1.07-1.39, P=0.003), gestational diabetes (OR=4.57, 95%CI:2.13-9.79, P<0.001), mother's low educational level (OR=1.52, 95%CI:1.11-2.07, P=0.008), single parent family (OR=4.79, 95%CI:1.44-15.88, P=0.010), mother overweight (OR=2.58, 95%CI:1.93-3.46, P<0.001), father overweight (OR=2.40, 95%CI:1.86-3.10, P<0.001), strong appetite (OR=7.78, 95%CI:5.38-11.27, P<0.001), fast eating speed (OR=6.59, 95%CI:4.86-8.94, P<0.001), daily outdoor activity time <1 hour (OR=1.42, 95%CI: 1.09-1.85, P=0.009), daily night sleep time <9 hours (OR=1.59, 95%CI: 1.13-2.23, P=0.007), daily screen viewing time ≥2 hours (OR=1.69, 95%CI:1.27-2.24, P<0.001). ③ Interaction of the four groups of risk factors had statistical significance, including interaction between mother overweight and father overweight (OR=5.53, 95%CI: 3.76-8.13, P<0.001), interaction between strong appetite and fast eating speed (OR=54.48, 95%CI: 32.95-90.06, P<0.001), interaction between low intensity of outdoor activity and daily outdoor activity time <1 hour (OR=2.12, 95%CI: 1.29-3.48, P=0.002), interaction between daily night sleep time <9 hours and daily screen viewing time ≥2 hours (OR=2.83, 95%CI: 1.71-4.68, P<0.001). Conclusions: This study identified 12 risk factors of childhood obesity, including high birth weight, cesarean section, gestational diabetes, mother's low educational level, single parent family, mother overweight, father overweight, strong appetite, fast eating speed, daily short outdoor activity time, daily short night sleep time, daily long screen viewing time, and interaction of the four groups of risk factors had statistical significance, including strong interaction between mother overweight and father overweight, interaction between strong appetite and fast eating speed, interaction between low intensity of outdoor activity and daily short outdoor activity time, interaction between daily short night sleep time and daily long screen viewing time.