Ilaria Costantini, Thalia C. Eley, Jean-Baptiste Pingault, Neil M Davies, Helen Bould, Cynthia M. Bulik, Georgina Krebs, Glyn Lewis, Gemma Lewis, Clare Llewellyn, Phillippa C. Diedrichs, Dasha Nicholls, Francesca Solmi
BACKGROUND: Body dissatisfaction is associated with later mental health problems and weight gain in young people, but the causality of these associations is unclear. We aimed to investigate the longitudinal associations between body dissatisfaction at age 16 years and later eating disorder and depressive symptoms, and BMI in young adulthood, with rigorous control for potential confounders including shared genetic influences (in objectives 1 and 2). We then quantified the proportion of these associations explained by genetic and environmental factors (objective 3). METHODS: We used data from the Twins Early Development Study, a population-based birth cohort of twins born in England and Wales between 1994 and 1996. Participants were included if they had complete data on body dissatisfaction at age 16 years-assessed via four items from the Eating Disorder Diagnostic Scale capturing weight and shape concerns-and imputed data on outcomes and confounders. Outcomes included eating disorder symptoms at age 21 years (12-item modified version of the Eating Disorder Inventory-2), depressive symptoms (8-item Short Mood and Feelings Questionnaire), and BMI (self-reported height and weight) at ages 21 years and 26 years. We used univariable and multivariable linear mixed-effects models in the full sample (objective 1), and within-pair twin-difference analyses in monozygotic and dizygotic twins separately (objective 2). We then used bivariate and multivariate twin modelling to estimate the proportion of variance and covariance explained by additive genetic and environmental factors (objective 3). No participants with lived experience were involved in the design or interpretation of the research. FINDINGS: (0·16-0·38) higher BMI. In monozygotic and dizygotic twin difference analyses, higher body dissatisfaction was also associated with more severe eating disorder and depressive symptoms. For BMI, the association was smaller and less precisely estimated in monozygotic twins. Twin modelling indicated that most of the covariance between body dissatisfaction and these traits-particularly eating disorder symptoms-was explained by shared genetic factors, with a smaller contribution from non-shared environmental influences. INTERPRETATION: Our findings suggest that body dissatisfaction might causally increase eating disorder and depressive symptoms; thus, reducing its emergence could lead to improvements in adolescent mental health outcomes. FUNDING: Wellcome Trust.