Examining the heritability of actigraphy-assessed and parent-reported sleep duration in a community sample of twins across childhood and adolescence.

Meta-analytic evidence indicates that the heritability of sleep duration increases across infancy and childhood, and peaks in adolescence, however, relatively few genetically-informed studies have examined sleep duration longitudinally across development, and estimates vary substantially by reporter and method of assessment. Extending on prior published heritability estimates from the Arizona Twin Project in infancy and middle childhood, the present study examines new data from the 8-, 9-, 10-, 11-, and 13-year waves (Ns = 388-739 twins; monozygotic = 29.1%-32.6%, same-sex dizygotic = 33.4%-38.8%, opposite-sex dizygotic = 31.5%-34.6%). Parents reported on twins' sleep at each wave. Twins wore wrist-based accelerometers for approximately one week at ages 8, 10, and 13. Univariate models indicated that the heritability of parent-report increased across middle childhood and early adolescence, and shared environmental influences decreased (h2 = .19-.55, C = .37-.69). In contrast, actigraphy-assessed sleep duration was highly heritable at 8, 10, and 13 years (h2 = .65-.75, C = .00, E = .25-.35). Bivariate Cholesky decomposition models were fit at 8-, 10-, and 13-years, and demonstrated that the genetic covariance between parent-reported and actigraphy-assessed sleep duration decreased with age and corresponded with increasing independent genetic variance (A21 = .16-.34, A22 = .33-.55). This suggests that subjective and objective reports increasingly capture different aspects of sleep as children mature, potentially reflecting children's growing ability to express heritable sleep-related behaviors in ways differentially captured by each method. Together, these findings underscore the importance of multimethod approaches to assess sleep health comprehensively and considering age-specific dynamics when studying the genetic influences of sleep duration across development.