Depression is prevalent in patients with chronic obstructive pulmonary disease (COPD); however, its etiology and relationship to the clinical features of COPD are not well understood. Using data from a large cohort, we explored prevalence and determinants of depression in subjects with COPD. The Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints study is an observational 3-year multicenter study that enrolled smokers with and without COPD and nonsmoker controls. At baseline, several patient-reported outcomes were measured including the Center for Epidemiologic Studies of Depression Scale. For the purposes of this analysis, depression was defined as a score of 16 and higher on this scale, which reflects a high load of depressive symptoms and has a good correspondence with a clinical diagnosis of major depression. The study cohort consisted of 2,118 subjects with COPD; 335 smokers without COPD (smokers); and 243 nonsmokers without COPD (nonsmokers). A total of 26%, 12%, and 7% of COPD, smokers, and nonsmokers, respectively, suffered from depression. In subjects with COPD, higher depression prevalence was seen in females, current smokers, and those with severe disease (Global Initiative for Obstructive Lung Disease [GOLD]-defined). Multivariate modeling of depression determinants in subjects with COPD revealed that increased fatigue, higher St. George's Respiratory Questionnaire for COPD patients score, younger age, female sex, history of cardiovascular disease, and current smoking status were all significantly associated with depression; physiologic and biologic measures were weak or nonsignificant descriptors. Depression is more prevalent in subjects with COPD compared with smokers and nonsmokers without COPD. Clinical and biologic measures were less important determinants of depression in COPD than disease symptoms and quality-of-life. Clinical trial registered with www.clinicaltrials.gov (NCT 00292552).