Analysis of Neurodevelopmental Disorders in Offspring of Mothers With Eating Disorders in Sweden

IMPORTANCE: Despite indices of impaired neurodevelopment in children of mothers with eating disorders, it remains unclear whether these children are at increased risk of developing neuropsychiatric diseases. OBJECTIVE: To evaluate the association between maternal eating disorders, whether preexisting or ongoing during pregnancy, and offspring neuropsychiatric disease risk. DESIGN, SETTING AND PARTICIPANTS: This population-based prospective cohort study used the Swedish Medical Birth Registry and identified singleton births registered between from January 1, 1990, and December 31, 2012. Children of exposed mothers with eating disorders were matched with comparator children of mothers without diagnoses of eating disorders. To adjust for unmeasured shared familial factors, a cluster of exposed children with full maternal cousin comparators was identified. Follow-up was completed on December 31, 2017. Data were analyzed from August 31, 2020, to April 30, 2021. EXPOSURES: Maternal eating disorder diagnosis. MAIN OUTCOMES AND MEASURES: All children were followed up from 1 year of age for autism spectrum disorder (ASD) and from 3 years of age for attention-deficit/hyperactivity disorder (ADHD). The relative risk of ASD and ADHD was assessed among exposed children, stratified by eating disorder subtype and ongoing vs previous disease, adjusted for potential confounders, including parental socioeconomic status and comorbidities. RESULTS: Among the 52 878 children included in the analysis, maternal eating disorder exposure (n = 8813) was associated with an increased risk of ADHD (hazard ratio [HR] for anorexia nervosa, 1.42 [95% CI, 1.23-1.63]; HR for bulimia nervosa, 1.91 [95% CI, 1.43-2.54]; and HR for unspecified eating disorder, 2.00 [95% CI, 1.72-2.32]) and ASD (HR for anorexia nervosa, 2.04 [95% CI, 1.58-2.63]; HR for bulimia nervosa, 2.70 [95% CI, 1.68-4.32]; and HR for unspecified eating disorder, 1.95 [95% CI, 1.49-2.54]). After adjustment for parental confounders, the risk of ADHD remained significantly increased, whereas the risk of ASD in children to mothers with bulimia nervosa was no longer significant. Ongoing anorexia nervosa was associated with a significantly higher risk of ADHD (HR, 2.52 [95% CI, 1.86-3.42]) and ASD (HR, 3.98 [95% CI, 2.49-6.27]) compared with previous disease (HRs, 1.26 [95% CI, 1.06-1.48] and 1.81 [95% CI, 1.38-2.38], respectively). Results based on the family cluster were similar to those of the main analysis for maternal exposure to anorexia nervosa and bulimia nervosa. CONCLUSIONS AND RELEVANCE: These findings suggest that children born to mothers with eating disorders, in particular disorders that were active during pregnancy, were at increased risk of developing ADHD and ASD. The association could not be fully explained by parental psychiatric comorbidities, and among children of mothers with anorexia nervosa and bulimia nervosa, it could not be explained by unmeasured familial confounding.