Sleep duration and incidence of obesity in infants, children, and adolescents: a systematic review and meta-analysis of prospective studies.

Abstract

To assess the prospective relationship between sleep and obesity in a paediatric population. We performed a systematic search using PubMed, Embase, Web of Science, and Cochrane (up to September 25, 2017). Included studies were prospective, had follow-up of ≥1 year, had duration of sleep at baseline, and measures of incidence of overweight or obesity and/or changes in body mass index (BMI) z-score and BMI during follow-up. We extracted relative risks or changes in BMI z-score or BMI and 95% confidence intervals (CI) and pooled them using a random effect model. Forty-two studies were included but, as there was significant heterogeneity, results are presented by age strata. Short sleep was associated with a greater risk of developing overweight or obesity in infancy (seven studies, 14738 participants, risk ratio [RR]: 1.40; 95% CI 1.19 to 1.65; p < .001), early childhood (eight studies, 31104 participants, RR: 1.57; 1.40 to 1.76; p < .001), middle childhood (three studies, 3005 participants, RR: 2.23; 2.18 to 2.27; p < .001), and adolescence (three studies, 26652 participants, RR: 1.30; 1.11 to 1.53; p < .002). Sleep duration was also associated with a significant change in BMI z-score (14 studies, 18 cohorts, 31665 participants; mean difference -0.03; -0.04 to -0.01 per hour sleep; p = .001) and in BMI (16 studies, 24 cohorts, 24894 participants; mean difference -0.03 kg/m2; -0.04 to -0.01 for every hour of increase in sleep; p = .001). Short sleep duration is a risk factor or marker of the development of obesity in infants, children, and adolescents.