Latent profiles of multi-dimensional sleep characteristics and association with overweight/obesity in Chinese preschool children

To examine the association between latent profiles of multi-dimensional sleep characteristics and overweight/obesity (OWO) in Chinese preschool children. The cross-sectional analysis included 3204 preschool children recruited from 24 kindergartens in Shanghai. Parents reported children's demographics and sleep characteristics, including sleep duration, timing and disturbances. Latent profile analysis (LPA) was used to identify sleep subtypes. Logistic regression models were used to evaluate the associations between sleep characteristics/subtypes and OWO. Short sleep duration, late bedtime, long social jetlag and sleep disturbances were significantly associated with increased OWO. However, when considering the interplay of sleep duration and timing, there was no significant association between sleep duration and OWO for children sleeping later than 22:00. Three sleep subtypes were identified based on children's sleep duration, timing and disturbances: "Average Sleepers" (n = 2107, 65.8 %), "Good Sleepers" (n = 481, 15.0 %), and "Poor Sleepers" (n = 616, 19.2 %). "Good Sleepers" had reduced odds of being OWO (AOR, 0.72; 95 % CI, 0.56–0.93) compared to "Average Sleepers", while "Poor Sleepers" showed an increased risk of OWO (AOR, 1.36; 95 % CI, 1.11–1.67). These findings highlight that improving multiple sleep characteristics simultaneously is a promising option to prevent and intervene childhood obesity. •We revealed the association between sleep and obesity risk in Chinese preschooler.•Late bedtime significantly attenuated the benefits of longer sleep on obesity risk.•Latent profile analysis identified sleep subtypes with varying obesity risks.