Key Points

Question Are maternal eating disorders or body mass index (BMI) associated with offspring psychiatric diagnoses?

Findings In this cohort study assessing data from 392 098 mothers and 649 956 offspring, there were associations between maternal history of eating disorders and prepregnancy BMI outside normal weight with most of the 9 studied psychiatric diagnoses in offspring. Effect sizes were typically larger for maternal eating disorders vs BMI.

Meaning The findings underline the risk of offspring mental illness associated with maternal eating disorders and prepregnancy BMI and suggest the need to consider these exposures clinically to help prevent offspring mental illness.

Abstract

Importance Maternal nutrition is essential in fetal development; thus, disordered eating may influence this process and contribute to the development of offspring psychiatric disorders.

Objective To investigate the association of maternal eating disorders and prepregnancy body mass index (BMI) with offspring psychiatric diagnoses.

Design, Setting, and Participants This population-based cohort study used Finnish national registers to assess all live births from January 1, 2004, through December 31, 2014, with follow-up until December 31, 2021. The data analyses were conducted from September 1, 2023, to September 30, 2024.

Exposures Maternal eating disorder and prepregnancy BMI.

Main Outcomes and Measures Primary outcomes were 9 neurodevelopmental and psychiatric offspring diagnoses. Cox proportional hazards modeling adjusted for potential risk factors in the development of the outcome disorders was applied in 2 models. Secondary analyses were stratified for adverse birth outcomes (prematurity, small size for gestational age, and low Apgar score) or comorbid offspring eating disorders. Categories of BMI (calculated as weight in kilograms divided by height in meters squared) included underweight (BMI <18.5), normal weight (18.5-24.9), overweight (25.0-29.9), obesity (30.0-34.9), and severe obesity (≥35.0).

Results The mean (SD) age of 392 098 included mothers was 30.15 (5.38) years, 42 590 mothers (10.86%) were born outside of Finland, 6273 mothers (1.60%) had a history of an eating disorder, 23 114 mothers (5.89%) had prepregnancy underweight, and 208 335 (53.13%) mothers had overweight or obesity. Among 649 956 included offspring, 332 359 (51.14%) were male, and 106 777 (16.43%) had received a neurodevelopmental or psychiatric diagnosis. Maternal eating disorders, prepregnancy underweight, and overweight or obesity were associated with most of the studied mental diagnoses in offspring, even after adjusting for potential covariates. The largest effect sizes were observed for maternal eating disorders not otherwise specified in association with offspring sleep disorders (hazard ratio [HR], 3.34 [95% CI, 2.39-4.67]) and social functioning and tic disorders (HR, 2.79 [95% CI, 2.21-3.52]), while for maternal severe prepregnancy obesity, offspring intellectual disabilities (HR, 2.04 [95% CI, 1.83-2.28]) had the largest effect size. Adverse birth outcomes further increased the risk of offspring having other feeding disturbances of childhood and infancy (eg, HR, 4.53 [95% CI, 2.97-6.89] for maternal eating disorders) and attention-deficit/hyperactivity disorder and conduct disorder (eg, HR, 2.27 [95% CI, 1.74-2.96] for maternal anorexia nervosa).

Conclusions and Relevance In this population-based cohort study including 392 098 mothers and 649 956 offspring, offspring from mothers with an eating disorder history or prepregnancy BMI outside normal weight were at higher risk of psychiatric disorders. The results differed somewhat between the 2 exposures with regard to which offspring diagnoses had associations, and effect sizes were typically larger for maternal eating disorders vs BMI. These findings suggest a need to consider these 2 exposures clinically to help prevent offspring mental illness.

Introduction

Fetal neurodevelopment is a highly regulated and sensitive process influenced by internal and external factors. The nutritional status prior to and during pregnancy is critical for the developing brain1,2 and macronutrient and micronutrient deficiencies during pregnancy may disturb these processes. Studies show associations between particular deficiencies and offspring psychiatric disorders, including maternal vitamin D deficiency and autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), and schizophrenia,3,4 low or high maternal serum vitamin B12 levels and ASD,5 and maternal iron deficiency and schizophrenia.6 Eating disorders commonly result in nutritional deficiencies.7 Similar to malnourishment in pregnancy,1,8,9 maternal eating disorders have a detrimental effect on pregnancy, delivery, and neonatal outcomes.10-12 Different maternal eating disorders may influence different offspring psychopathologies. Ongoing and previous maternal anorexia nervosa (AN) has been associated with offspring ADHD and ASD, with higher risk observed for active AN.13 Maternal history of AN has been associated with offspring emotional disorders, in particular anxiety and depression.14,15 Exposure to lifetime maternal bulimia nervosa (BN) increases the incidence of offspring obsessive-compulsive disorder symptoms, ADHD, ASD, and emotional disorders,13-16 whereas lifetime maternal binge eating disorder is associated with emotional, affective, and anxiety disorders in offspring.17,18 In addition, maternal obesity has repeatedly been associated with the risk of psychiatric disorders, such as ASD and ADHD in offspring.19-25 Maternal famine, on the other hand, has been associated with a range of mental disorders, including schizophrenia,26 affective disorders,27,28 antisocial personality disorder,29 and depressive symptoms.30 Furthermore, increased incidence of nonaffective psychosis in adolescence exposed to maternal underweight or inadequate weight gain in early pregnancy has been reported.31 Thus, maternal body mass index (BMI) and eating disorders play a significant role in offspring mental health, but the link with, in particular, eating disorder not otherwise specified (EDNOS), is not well investigated. Moreover, eating disorders include more than deviations in BMI, and pathological mechanisms are believed to differ from those of other disorders with altered BMI or, for example, constitutional leanness. In addition, AN and eating disorders in general are commonly accompanied by, for example, psychiatric comorbidities, dietary restriction, and extreme physical activity. Contrary to individuals with AN, most individuals with other eating disorders are not underweight, but still have significant somatic complications and psychiatric comorbidities.32 Here, we sought to expand existing knowledge on associations of different maternal eating disorders and prepregnancy BMI with neurodevelopmental and psychiatric diagnoses in offspring. We also explored differences between exposure to maternal AN vs maternal underweight with offspring mental health.

Methods

Study Population

This population-based register cohort study used data from the Finnish Medical Birth Register, the Finnish Care Register for Health Care (HILMO), and the Finnish Register on Reimbursement Drugs (RRD). This study included all live births in Finland between January 1, 2004, and December 31, 2014, followed up until December 31, 2021. The data analyses were conducted from September 1, 2023, to September 30, 2024. Information on maternal and offspring diagnoses were obtained from HILMO using the International Classification of Diseases, Ninth Revision (ICD-9; 1987-1995), and Tenth Revision (ICD-10; since 1996). Data on purchases of reimbursed medications were obtained using Anatomical Therapeutic Chemical codes from the RRD. Register data were linked using the encrypted personal identification numbers assigned to all Finnish citizens and permanent residents. The use of register data from the Social Insurance Institution and the Finnish Institute for Health and Welfare and linkages between them was carried out after receiving the necessary permission from the Finnish Social and Health Data Permit Authority (Findata) and the Swedish Ethics Review Authority. All procedures contributing to this work complied with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. The data were pseudonymized by Findata and analyzed in the secured environment Kapseli provided by Findata. Findata and the Swedish Ethics Review Authority indicated that informed consent was not required since the study material was based on register data only and no registered persons were contacted. The researchers had no access to personal identifiable information. This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Outcomes

Data on the diagnoses of the offsprings’ neurodevelopmental or psychiatric disorders were collected from HILMO. The diagnosis status was identified and grouped by the ICD-10 codes of any mental, behavioral, and neurodevelopmental disorder (F00-F99); mood disorders (F30-39, F92); anxiety disorders (F40-43, F93); sleep disorders (F51); intellectual disabilities (F70-79); specific developmental disorders (F80-83); ASD (F84); ADHD and conduct disorders (F90-91); disorders of social functioning with onset specific to childhood and adolescence, including selective mutism and attachment disorders, and tic disorders and Tourette syndrome (hereafter called social functioning and tic disorders) (F94-95); and other feeding disorders of infancy and childhood (F98.2).