Objective

This study aimed to assess the association of changes in sleep behaviors from adolescence to young adulthood with the risk of overweight/obesity, and the reverse relationship.

Methods

Data of 1978 participants was obtained from the 15- and 20-year follow-ups of the GINIplus and LISA birth cohorts. Insufficient sleep was defined as reported sleep duration <8 h for adolescents, <7 h for adults, and sleep difficulties as reported having sleeping difficulties. Logistic regression models were used to assess bidirectional associations of changes in insufficient sleep and sleep difficulties with overweight/obesity. The polygenic risk scores (PRS) for body mass index (BMI) was tested in a sub-sample (n = 918).

Results

Compared with sufficient sleep in both adolescence and young adulthood, insufficient sleep only in young adulthood was associated with an increased risk of overweight/obesity (odds ratio = 1.85, 95%confidence interval = [1.27–2.69]). Compared with no sleep difficulties at both time-points, only persistent sleep difficulties was associated with a higher risk of overweight/obesity (2.15 [1.22–3.77]). The PRS for BMI was associated with overweight/obesity (1.41 [1.17–1.70]), but no significant gene-sleep interaction effect was observed. Reversely, only persistent overweight/obesity was associated with increased risks of insufficient sleep (1.81 [1.21–2.70]), and sleep difficulties (1.77 [1.18–2.66]), respectively.

Conclusions

Insufficient sleep only presented a cross-sectional association with overweight/obesity in young adulthood, while long-term sleep difficulties from adolescence to young adulthood was associated with young adult overweight/obesity. Reversely, long-term overweight/obesity from adolescence to young adulthood was associated with insufficient sleep and sleep difficulties in young adulthood.