A multivariate behavioural genetics model of sleep and cortisol in middle childhood

• In Event: Poster Session 06
  In Poster Session: PS 06 Topic - Biological Processes: Psychophysiology

Abstract

Sleep quality has been found to be positively associated with emotional and behavioural functioning. Past research indicated that sleep is also associated with cortisol levels throughout childhood and adolescence. Lower sleep quality is related to higher (morning) cortisol levels. Genetic factors may play a role in this context. Twin studies suggest heritability indices between 30% and 60 % for both sleep and cortisol levels. However, to date behavioural genetic studies either investigated sleep or cortisol levels, but not both factors in relation to each other. Therefore, a pertinent question is whether the same or distinct genetic factors explain the variation in sleep quality and cortisol levels. In the current study, a multivariate behavioural genetics model was employed utilizing the classical twin design which compares monozygotic and dizygotic twins to investigate the contribution of genetic and environmental factors to the variance in sleep, cortisol levels as well as their covariance. We collected data of 7-8 years-old twins (N = 436 twin pairs, Together-Unique study). We measured sleep over four consecutive days using actigraphs. Sleep duration, sleep efficiency, and wake episodes were used as indicators of sleep. Cortisol was measured three times a day over two consecutive days by the parents: the mean morning cortisol level was used in the analyses. A structural equation model was estimated by modelling additive genetic factors (A), shared environmental effects (C) and unique environmental effects (E). Age, cohort, and room-sharing were included as covariates in the analyses. We ran a common pathway model with the three sleep variables as well as a multivariate independent pathway model with morning cortisol levels, sleep efficiency, sleep duration and wake episodes as variables. No common factor seems to contribute to all three sleep variables. However, the multivariate independent pathway model fit the data adequately. The heritability of sleep duration, sleep efficiency, wake episodes and morning cortisol levels was 30%, 42%, 65% and 13%, respectively. Unique environmental factors explained 61%, 58%, 35% and 70% of the variance in sleep duration, sleep efficiency, wake episodes and morning cortisol levels, respectively. Common environmental factors played no significant role, besides for sleep duration (9%) and morning cortisol levels (18%). We found no common genetic, shared environmental or unique environmental factors explaining the covariance in cortisol and sleep. There seems to be a common genetic factor underlying both sleep duration and sleep efficiency and a – different - common genetic factor underlying sleep efficiency and wake episodes. These findings indicate that sleep duration, sleep efficiency and wake episodes are impacted by genetic factors and by the unique environment of children, providing possibilities for interventions for children with sleep deprivation. A part of the large contribution of the unique environment might also be the result of measurement errors as cortisol was measured by the parents without external control. Future studies should replicate this in a more controlled setting and focus on disentangling the underlying factors in the correlation between sleep and cortisol.

Authors

• Jana Runze, Vrije Universiteit Amsterdam
  Presenting Author

• Maria Koopman-Verhoeff
  Non-Presenting Author

• Saskia Euser
  Non-Presenting Author

• Mirjam Oosterman, Vrije Universiteit Amsterdam
  Non-Presenting Author

• Marian J Bakermans-Kranenburg, ISPA Lisbon
  Non-Presenting Author