Psychiatric outcomes and their progression across different diagnoses in children and adolescents following prenatal exposure to antidepressants

Background: Little is known about how prenatal antidepressant exposure impacts the development and progression of mental disorders in offspring.

Objectives: Quantify the association between prenatal antidepressant exposure trajectories and non-psychotic mental disorders (i.e., anxiety, depression, and eating disorders) in offspring up to adolescence and the transitions across these disorders.

Methods: We conducted a cohort study using data from the Norwegian Mother, Father and Child Cohort Study linked with Norwegian health registries. We included 6934 pregnancy-child dyads of mothers with pre-pregnancy mental illnesses. We modelled antidepressant exposure using k-means trajectory modelling. Psychiatric outcomes in offspring were identified using diagnostic codes up to 18 years. Confounding was adjusted using propensity score methods. We modelled the association between antidepressant exposure trajectory and outcome using Cox proportional hazard models. Multi-state models were employed to understand better the progression of psychiatric outcomes throughout childhood and adolescence. We explored the impact of confounding by maternal polygenic liability for non-psychotic mental illnesses and the severity of mental illness during pregnancy.

Results: We identified five antidepressant exposure groups: discontinuers before pregnancy (n=190), early discontinuers (269), late discontinuers (419), continuers (117), and non-exposed (5939). Compared with discontinuers before pregnancy, children born to mothers who discontinued antidepressant during pregnancy had an increased risk of developing any mental disorders (aHR estimation range: 1.40-1.43), mainly driven by greater risk for anxiety disorders specifically (aHR estimation range: 2.50-2.68). This elevated risk was of lower magnitude among children born to antidepressant continuers (aHR estimation range: 1.31-1.73). The multi-state model highlighted that less than 15% of the sample is at risk of any mental disorder diagnosis and the predicted probability of transit to one of the outcomes is lower than 2%.

Conclusions: We found a moderate increased risk for anxiety disorders in children born to those discontinued antidepressant during pregnancy compared to discontinuing before. Children born to antidepressant continuers also had slightly increased risk for depressive disorders compared to discontinuing before. The risk of developing mental disorders and transitions between these conditions in the offspring born to mothers with pre-existing mental illnesses is low, regardless of prenatal antidepressant exposure patterns. These findings will be evaluated in a setting where maternal polygenic liability for mental conditions are taken into account.