Background/Objectives: Hypothyroidism, including subclinical hypothyroidism, may affect mental health in children and adolescents through disturbances of neurotransmission and dysregulation of the hypothalamic–pituitary–thyroid and stress axes. Anxiety disorders are common in this population and frequently coexist with somatic symptoms overlapping those of hypothyroidism, complicating diagnosis and treatment. This study aimed to evaluate the association between levothyroxine treatment for hypothyroidism and the need for psychiatric interventions in children and adolescents with anxiety disorders. Methods: A retrospective cohort study was performed using data from the TriNetX global research network. Patients aged 5–18 years with diagnoses of hypothyroidism (ICD-10: E03) and anxiety disorders (ICD-10: F41) were included. Two propensity score–matched cohorts were analysed: patients treated with levothyroxine (n = 1861) and untreated patients (n = 1861). Outcomes included psychiatric hospitalisations, use of selective serotonin reuptake inhibitors and tricyclic-like antidepressants, frequency of psychiatric and psychotherapeutic consultations, and the occurrence of suicidal ideation and self-harm. Results: Levothyroxine treatment was associated with lower odds of SSRI use (OR = 0.58; p < 0.001), fewer psychiatric consultations (OR = 0.48; p < 0.001), and lower recorded use of psychotherapy (OR = 0.75; p = 0.029). Suicidal ideation and self-harm were recorded less frequently in the treated group (OR = 0.53; p = 0.001). No significant differences were observed in psychiatric hospitalisation rates. Use of tricyclic-like antidepressants was uncommon and did not differ significantly between groups. Conclusions: Among children and adolescents with comorbid anxiety disorders, levothyroxine treatment for hypothyroidism is associated with lower recorded utilization of certain psychiatric services and lower recorded rates of suicidal ideation and self-harm. Due to the retrospective design, causal inferences cannot be made, and the findings should be considered hypothesis-generating, requiring confirmation in prospective studies with standardised psychiatric outcome measures.
Hilmon et al. (Fri,) studied this question.