(P19-048-23) Investigating the Causal Relationships Between Anorexia Nervosa and Sleep Traits: A Mendelian Randomization Study

Objectives: Observational studies suggest links between eating disorders and sleep traits, such as anorexia nervosa and increased insomnia and binge eating disorder and evening chronotype. The objective of this study was to use Mendelian randomization (MR) to investigate potential causal effects of anorexia nervosa (AN) on sleep traits including chronotype (morning or evening preference), daytime napping, daytime sleepiness, insomnia, and sleep duration.

Methods: Two-sample MR was conducted using publicly accessible genome-wide association study (GWAS) summary statistics from published studies conducted in participants of European ancestry. The AN GWAS by the Psychiatric Genomics Consortium included 16992 cases and 55525 controls and identified eight significant loci. Outcome sleep data for chronotype (continuous and binary, i.e. morning vs. evening), daytime napping, daytime sleepiness, and sleep duration were retrieved from the Sleep Disorder Knowledge Portal, and for insomnia from Watanabe et al., 2022. Inverse variance weighted (IVW) regression was used to estimate the causal effect of AN on sleep traits, and results from IVW were compared with results from MR-Egger and weighted median.

Results: AN was associated with a more morning (continuous) chronotype in two-sample MR using IVW regression (beta = 0.040, SE = 0.017, p-value = 0.015). Associations showed consistent causal effects using binary chronotype GWAS summary statistics (beta = 0.055, SE = 0.027, p-value = 0.042) and in MR-Egger and weighted median models. No associations were evident for AN on other sleep traits including insomnia (all p >0.05).

Conclusions: In this MR study, there is evidence for a causal relationship for AN on morning chronotype. Additional large GWAS are needed for binge eating disorder and other eating disorders to further examine the links between eating disorders and sleep.

Funding Sources: N/A