Although hypothyroidism is purportedly an important cause of depression, prior studies have involved small samples of young people and produced conflicting results. We examined the yield of thyroid-stimulating hormone (TSH) testing in a large group of elderly patients with major depression or dysthymic disorder. The study sample comprised 883 outpatients aged 60 years or older from 18 primary care sites enrolled in the intervention arm of a clinical trial of depression management. Thyroid function was assessed by a single TSH value. Depressive diagnoses were confirmed with the Structured Clinical Interview for DSM-IV (SCID) and depression severity was assessed with the HSCL-20, a modified depression scale of the Hopkins Symptom Checklist. TSH results were available for 725 (82.1%) participants. Although 32 (4.4%) of those tested had TSH>5 mIU/L, the vast majority (27/32) had marginally elevated results (5.1–9.4 mIU/L). Only five patients (0.7%) had TSH levels >10 mIU/L. Patients with elevated TSH did not differ from those with TSH≤5 mIU/L in the severity or symptom pattern of depression as measured by the baseline HSCL-20 score (P=.37) or SCID score (P=.44). These findings should caution physicians against acceptance of borderline TSH values as the primary cause of a patient's clinical depression.