Accelerometry-assessed sleep clusters and obesity in adolescents and young adults: a longitudinal analysis in GINIplus/LISA birth cohorts

Abstract

Background: Some studies have revealed various sleep patterns in adolescents and adults using multidimensional objective sleep parameters. However, it remains unknown whether these patterns are consistent from adolescence to young adulthood and how they relate to long-term obesity.

Methods: Seven-day accelerometry was conducted in German Infant Study on the influence of Nutrition Intervention PLUS environmental and genetic influences on allergy development (GINIplus) and Influence of Lifestyle factors on the development of the Immune System and Allergies in East and West Germany (LISA) birth cohorts during the 15-year and 20-year follow-ups, respectively. Five sleep clusters were identified by k-means cluster analysis using 12 sleep characteristics at each follow-up. Adjusted linear and logistic regression models using generalized estimating equations were examined. Further, the interaction effects with time of follow-ups and polygenic risk scores (PRS) for body mass index (BMI) were tested.

Results: Five sleep clusters were classified consistently in both adolescence (n = 1347, aged 14.3-16.4 years) and young adulthood (n = 1262, aged 19.5-22.4 years). Adolescents in the "good sleep", "delayed sleep phase", and "fragmented sleep" clusters displayed greater stability transitioning into young adulthood, while those in the "sleep irregularity and variability", and "prolonged sleep latency" clusters showed lower stability (n = 636). Compared to the "good sleep" cluster, the "prolonged sleep latency" cluster exhibited associations with higher BMI [β = 0.56, 95% confidence interval (CI) = (0.06, 1.05)] and increased odds of overweight/obesity [Odds ratio = 1.55, 95% CI = (1.02, 2.34)]. No significant PRS-sleep cluster interaction was found for BMI or overweight/obesity. Among males only, the "delayed sleep phase", "sleep irregularity and variability" and "fragmented sleep" clusters showed stronger associations with overweight/obesity as age increased.

Conclusion: Adolescents and young adults shared five consistent sleep patterns, with the "prolonged sleep latency" pattern linked to higher BMI and overweight/obesity.

Keywords: Accelerometry; Adolescents; Obesity; Sleep clusters; Young adults.

Fig. 1

Flow chart of participants. GINIplus German Infant Study on the influence of Nutrition Intervention PLUS environmental and genetic influences on allergy development, LISA Influence of Lifestyle factors on the development of the Immune System and Allergies in East and West Germany. a: Inclusion for sleep data: (1) at least three weekdays and one weekend day of valid accelerometry recording for ≥ 10 hours/day; (2) 2 hours ≤ daily total sleep time ≤ 15 hours; (3) daily sleep efficiency ≥ 20%; (4) daily sleep onset timing between 17:00 and 08:00. b: Inclusion for participants: (1) participants in Munich and Wesel study centers; (2) complete outcomes; (3) complete confounders, except for total energy intake

Fig. 2

Five sleep clusters and their transitions in adolescence and young adulthood. a, Sleep characteristics in each of five clusters in adolescence and young adulthood. The 12 axes represent 12 sleep characteristics, which were scaled to the value between 0 and 1. Five colors represent five sleep clusters (same as b). Awakenings/h awakenings per hour after sleep onset, SD standard deviation, SE sleep efficiency, SL sleep latency, SMT sleep midpoint timing, TST total sleep time, WASO/h time awake per hour after sleep onset. b, Transitions of sleep clusters from adolescence to young adulthood (N = 636), during 5-year follow-up. The line thickness between adolescence and young adulthood represents the proportions of each sleep cluster in adolescence that remained in the same cluster or transitioned to a different cluster in young adulthood. The thicker the line, the higher the proportion. Modified from Wang M, et al., Obesity (Silver Spring), 2024, under CC BY-NC-ND 4.0

Fig. 3

Generalized estimating equation marginal means for BMI and prevalence of overweight/obesity in males and females, with interaction terms between time of follow-ups and sleep clusters. *There is a significant interaction effect between this sleep cluster and time on the health outcomes. BMI body mass index