Depression is a preventable and treatable mental health condition. Therefore, there are important clinical implications for identifying people with the highest mortality risk in a nationally representative sample. This study included 26,207 participants aged ≥18 years from the 2005-2014 National Health and Nutrition Examination Survey in USA. We investigated the association between depressive symptoms (defined as Patient Health Questionnaire 9 scores ≥10) and all-cause and cardiovascular disease (CVD) mortalities, adjusted for multiple factors (sociodemographic in Model 1, behavioral added in Model 2, and metabolic syndrome added in Model 3) and stratified by age and sex. During an average follow-up of 69.15 months (standard deviation [SD] 34.45), 1872 (7.3%) participants had died (person-years in the non-depressive and depressive groups, 12.12/1000 and 16.43/1000, respectively). Depressive symptoms increased all-cause (crude hazard ratio [HR] 1.37, 95% confidence interval [CI], 1.33-1.58) and CVD mortalities (crude HR 1.64, 95% CI, 1.20-2.24). Although the significance of all-cause mortality and CVD mortality was maintained in Models 1 (HR 1.58 and 2.08) and 2 (HR 1.48 and 1.79), it was not maintained in Model 3. Current smoking and lower physical activity were associated with reduced strength of the association between depression and all-cause mortality risk. The effect of depression on mortality risk was particularly pronounced in middle-aged men and older women. Our findings suggest that depressive symptoms increase mortality risk, even after adjusting for behavioral factors. Depression-induced mortality risk is particularly high among middle-aged men and older women.