Background: The first phase of the Danish General Suburban Population Study (GESUS) including 8214 individuals was an attempt to evaluate the association between subclinical hypothyroidism without or with elevated peroxidase antibodies and depression. No such association was found. In the second phase, including 14,787 individuals, we have focused on suppressed TSH (thyroid-stimulating hormone) and depression. Aims: To evaluate to what extent suppressed TSH is associated with subclinical depression. Methods: The total scores of the Major Depression Inventory (MDI) were used to evaluate subclinical depression, both by its total score and by an algorithm of the subthreshold depressed by presence of at least three of the 10 ICD-10 depression symptoms. Serum levels of TSH were used to classify the individuals into suppressed (TSH < 0.4 mIU/l), elevated (TSH ≥ 3.8 mIU/l) and normal reference (TSH between 0.4 and 3.7 mIU/l). Results: We identified 285 individuals with suppressed TSH and 1266 individuals with elevated TSH. The MDI total score was 7.55 in suppressed TSH individuals, 6.22 in elevated TSH individuals and 6.52 in normal reference individuals (P = 0.01). When the MDI was used diagnostically to identify subclinical depression, the prevalence was 8.07% in suppressed TSH individuals, 5.8% in normal reference individuals and 5.29% in elevated TSH individuals. Conclusion: This population-based study supports that persons with suppressed TSH (subclinical hyperthyroidism) seem to have a risk, although small, of subclinical depression.