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

Key Points Question Are children of mothers with eating disorders at increased risk of developing neuropsychiatric diseases? Findings In this nationwide cohort study of 52 878 children, including 8813 children born to women with an eating disorder and 44 065 matched children born to women without an eating disorder, maternal eating disorder was significantly associated with attention-deficit/hyperactivity disorder and autism-spectrum disorder in offspring. Meaning These findings highlight the importance of clinical awareness and intensified support to women with eating disorders and their children and for future research efforts to identify underlying mechanisms of the association.

Demographic registers are kept by Statistics Sweden and includes the following registers used in this study: i) The total population register ii) The longitudinal integration database for health insurance and labor market studies (LISA) iii) The cause of death register iv) The multigeneration register Additionally, the national health registers are kept by the National Board of Health and Welfare (NBWH) and includes the following registers, which were also used in this study: v) The medical birth register vi) The national patient register vii) The prescribed drug register which were also used in the study.
The total population register contains demographic information, including information on migrations. Information on migration during the follow-up period was obtained from the total population register. Parental educational level was obtained from LISA, which includes yearly updated information on educational level on all individuals above 16 years of age. Information on deaths was retrieved from the cause of death register, which since the 1960s contains date of deaths and supposed cause of death coded according to the International classification of diseases (ICD), to identify all deaths (competing events) during the follow-up period. The biologic fathers and the full maternal cousins were identified using the multi-generation register. The multigeneration register collects parental information on index persons and includes virtually complete parental information for contemporary generations.
The medical birth register contains information on more than 98% of all births in Sweden since 1973. Information is collected prospectively throughout the pregnancy using standardized records from antenatal, delivery and neonatal care. The medical birth register was used to identify a study base and to retrieve information on baseline pregnancy, delivery and neonatal characteristics which are specified in eTable 1. The Swedish national patient register includes information on inpatient care since 1964 (full coverage since 1987) and outpatient care since 2001. Main and secondary diagnoses are coded according to the ICD-system (ninth revision between 1987 and 1997 and tenth revision after 1998) by the assessing or discharging physician. The patient register was used to retrieve information on exposure (maternal eating disorder), outcomes (ADHD and ASD) and parental comorbidities (ICD-codes used in supplementary eTable x). Validity of most diagnoses, including psychiatric disorders, have proven high. (2) The prescribed drug register contains information on dispensed prescribed drugs from Swedish pharmacies, coded accorded to anatomical therapeutic chemicals (ATC) codes, since July 2005 and was used to retrieve data on ADHD-drugs as a proxy for outcome.

eFigure 2. Illustration of Overlap Between Exposed Cohorts
Figure illustrates the overlap between maternal eating disorders. In total, 100 mothers had a registered diagnosis of all eating disorder subtypes, 203 had been diagnosed with anorexia nervosa and bulimia nervosa, 795 had been diagnosed with anorexia nervosa and unspecified eating disorder and 334 had been diagnosed with bulimia nervosa and unspecified eating disorder.

ED, Eating disorder eFigure 3. Hypothesized Directed Acyclic Graph of Maternal Eating Disorder and Neurodevelopmental Disorder in Children
The directed acyclic graphs (DAG) shows the hypothesized direct effect of maternal eating disorder on neurodevelopmental disorders in the children, confounded by parental genetic factors, socioeconomic status (SES), demographics, psychiatric comorbidities and maternal smoking status. Additionally, the DAG also shows the potential mediated effect of premature birth and/or fetal growth. Self-reported information on smoking before pregnancy and ongoing smoking at the first antenatal visits (end of first trimester) Mode of delivery As registered in the delivery medical record Gestational weight for gestational age Birth weight measured at birth Swedish sex-specific estimated fetal growth curves were used to categorize gestational weight into: Appropriate for gestational age (AGA) defined as gestational weight between two standard deviations below and 2 standard deviations above the mean birth weight for gestational their gestational age Small for gestational age (SGA) defined as gestational weight less than 2 standard deviations below the mean birth weight for their gestational age Large for gestational age (LGA) defined as gestational weight more than 2 standard deviations above the mean birth weight for gestational age Microcephaly

eTable 1. Definition of Pregnancy, Delivery, and Neonatal Characteristics Obtained From the Medical Birth Registry Variable Definition and categorization
Based on head circumference measured at birth Microcephaly defined as a head circumference less than two standard deviations below the sex-specific mean for gestational age 2) ED, Eating disorder; ADHD, Attention-deficit hyperactivity disorder; ASD Autism spectrum disorder. * Number (%) of study subjects with valid information on father personal identification number. †Registered diagnosis in the patient register prior to delivery date.