Thyroid and depression: Insights from a retrospective real world clinical cohort

To examine whether thyroid functioning is associated with Clinically Relevant Depression (CRD) in a sample from a psychiatric clinic.

Patients (≥18 years) treated at a psychiatric clinic with a thyroid-stimulating hormone (TSH) level and PHQ-9 within 6 months were studied. Data on demographic variables and covariates (diagnoses of mood and thyroid disorders), drug, comorbidity, relevant laboratory information and thyroid hormone therapy (THT) were extracted from the electronic medical record. CRD was defined as a PHQ-9 score ≥ 10. Logistic regression models assessed the relationship between TSH categories and CRD, with subgroup analysis by gender, age, mood and thyroid disorder and THT.

The study cohort included 1467 patients with 92 % White, 70 % women and the mean age of 44.3 years. The sample's TSH measurements were distributed between normal (90 %), high (7 %), and low (3 %), and 55 % had CRD. Compared to those with normal TSH levels, those with low TSH had 2.50 (95 % CI: [1.17, 5.36]) times greater odds of having CRD while those with high TSH levels had 1.69 (95 % CI: [1.03, 2.79]) times greater odds of having CRD. This U-shaped association between TSH levels and CRD was observed among women (low TSH: 3.53 [1.37, 9.1]; high TSH: 1.95 [1.02, 3.73]) but not men. Post-hoc analyses of TSH and PHQ-9 items, specific depression symptoms showed differential associations with thyroid function.

CRD was associated with both low and high TSH, independent of mood and thyroid disorders, psychotropic use and THT. Our findings highlight the utility of serum TSH in identifying patients vulnerable to depressive disorders and the need for further longitudinal research.