Objectives: This study aims to investigate the association between habitual midday napping and IFG, and the interaction between daytime napping and midday napping on IFG among early adolescents.
Methods: The present study is an extension of the China Jintan child cohort, which was originally designed to investigate the predictive effect of early health factors on neuro-behavioral development in children. This siesta culture has a lunch break from 11:30 am to 2:00 pm in schools whereby our sample was selected, thus providing a research opportunity to study the health impact of habitual midday napping. The current study used a subsample (n=625) of early adolescents (12.26 ± 0.63 years old) who self-reported good health status and had complete cross-sectional data on napping behaviors (either frequency or duration) and fasting blood glucose concentrations
Midday napping and nighttime sleep were measured using the Adolescent Health Questionnaire between 2011 and 2013. We re-coded nap frequency into 4 categories: never, ≤2 days/week, 3-4 days/week, and 5-7 days/week, and nap duration into 4 categories: never (0 minutes), <30 minutes, 31-60 minutes, and ≥61 minutes. Participants reported habitual bedtime and rise time on school days. Total time in bed, a proxy measure of nighttime sleep duration, was calculated by the interval between habitual bedtimes and rise times on school days. Participants also self-rated nighttime sleep quality using a four-point scale. Meanwhile, after at least 10 hours of an overnight fast, peripheral blood samples were collected by pediatric nurses and analyzed for plasma glucose concentrations. Fasting plasma glucose concentrations were dichotomized into normal (<5.5mmol/L) and impaired (≥5.6mmol/L) levels. The multivariate random-effect logistic regression examined the nap-glucose relationship and the interaction between nap and nighttime sleep duration. All models used the never-napped group as the reference category and adjusted for age, sex, school district, education in mother and father, nighttime sleep quality, BMI and plasma cholesterol concentrations. Marginal effects of nap frequency/duration at different hours of nighttime sleep duration were calculated using STATA 14.0.
Results: Of the participants, 83.20% (n=520) took naps and 62.28% reported average nap duration ≥31 min in the past month. Sixteen percent of participants (n=101) had impaired fasting glucose (IFG). After adjusting for covariates, early adolescents who napped 3-4 days/week (OR=1.72, p<0.001), 5-7 days/week (OR=1.34, p=0.02) or greater than 30 minutes/nap (OR=1.52, 1.56, p’s<0.05) were associated with increased likelihoods of IFG compared with non-nappers. There were significant interactions between daytime napping and nighttime sleep duration. Non-nappers showed a trend towards decreasing odds of IFG with increasing nighttime sleep duration (OR=0.45, p=0.03), while frequent (5-7 days/week) and long (>60 min/nap) naps, on average, weakened the inverse relationship between nighttime sleep duration and IFG. Nighttime sleep duration also moderated the association between midday napping and IFG. Marginal effect results showed significantly increased likelihoods of IFG among nappers with nighttime sleep duration of ≥9 hours (p’s<0.05). In contrast, nappers with nighttime sleep duration <9 hours had a decreased likelihood of IFG, although the magnitude and significance varied by nap frequencies/durations.
Discussion and Conclusion: There are interactive effects between daytime napping and nighttime sleep duration on IFG. Frequent and long naps may dampen the mean effect of nighttime sleep duration on IFG, compared with the never-napped condition. When considering different hours of nighttime sleep, daytime naps may increase the risk for IFG among early adolescents with sufficient nighttime sleep (≥9 hours), but potentially mitigate the negative impact of short nighttime sleep (<9 hours). This study is one of the first to examine the complex relationship among habitual midday napping, nighttime sleep and IFG using an early adolescent sample. The underlying mechanism whereby midday napping and nighttime sleep may predict the risk of IFG is unknown. It could be the interruption of sleep homeostatic and circadian processes that negatively affect insulin resistance and inflammatory biomarkers. Our findings also suggest the importance of moderate sleep hours within a 24-hour period of time in glucose regulation among adolescents, which supports the prior findings that both short and long sleep duration are associated with insulin resistance, thus mediating the risk for IFG and diabetes. Daytime napping is a custom ritual in siesta countries such as China and schools offer nap opportunities. Thus, our findings may inform adolescents, parents and primary caregivers the optimal sleep practice in diabetes prevention and management. Cohort studies with objective sleep measures are needed to confirm the findings of the present study in non-siesta countries.