Common genetic architecture and environmental risk factors underpin the anxiety-disordered eating relationship: Findings from an adolescent twin cohort.

Abstract

Despite established comorbidity between anxiety and disordered eating (DE), and a plethora of research using various methodologies to examine this overlap, use of twin modeling to expose whether a shared genetic liability underpins these conditions remains rare. Data from a longitudinal sample of female twins were selected: measures of risk for DE from Wave 1 (N = 699, 351 pairs, aged 12-15 years), and the Eating Disorder Examination (EDE) and Children's Anxiety Sensitivity Index (CASI) from Wave 2 (N = 669, 338 pairs, aged 16-19 years). At this time, they also completed Children's Anxiety Sensitivity Index (CASI). Bivariate Cholesky decomposition models adjusting for age and body mass index centile investigated the covariance structure between the CASI and EDE. Modeling both genetic and nonshared environmental influences parsimoniously fit these data. All paths were significant. Additive genetic influences were notable for CASI and EDE phenotypes; 14% of the heritable variance was contributed by CASI to the expression of EDE. There was also a smaller but significant contribution of nonshared environmental influences. A multinomial logistic regression indicated body dissatisfaction (RRR = 1.53; 95% CI = 1.07-2.18) differentiated groups with highest EDE scores from the highest CASI scores. Shared genetic and environmental influences appear to underpin the relationship, and potentially the observed comorbidity, between anxiety sensitivity and DE. The age of onset is typically earlier for anxiety than DE, suggesting a significant opportunity for early intervention work to reduce the likelihood of subsequent development of DE.