The lack of reliable outcome predictors and the delayed onset of therapeutic response to antidepressants are among the clinical challenges in the treatment of depression. Identifying clinical correlates associated with antidepressant response would reduce symptom severity and morbidity for patients with depression. Twenty-three subjects with major depression were treated with citalopram 20 mg/day in a 6-week open trial and were also simultaneously randomized to either adjunctive triiodothyronine (T3) 25 μg BID (n = 7), pindolol 5 mg BID (n = 8), or placebo (n = 8). Baseline thyroid-stimulating hormone (TSH), FT4, FT3, and TT3 were measured for potential relationships to treatment response across groups. In males only, there was a significant inverse correlation between baseline free T4 and time to response (r = −0.7, P = 0.034). In both males and females across all treatment conditions, as measured by Kaplan–Meier (K–M) maintenance failure time, baseline TSH below the mean (1.5 ng/dL) was associated with a shorter time to response (50% reduction in Montgomery and Asberg Depression Rating Scale [MADRS] score) (χ2 = 4.53, df = 1, P = 0.03). Patients with baseline TSH above the mean were less likely to reach full remission (MADRS ≤ 7) (χ2 = 4.38, df = 1, P = 0.03). No significant differences between groups emerged in the mean response time. Baseline thyroid function, as measured by serum free T4 and TSH, may predict a patient's response time to antidepressant treatment with citalopram.