To assess how social contacts are associated with depression, we analyzed the data of adults aged 40 years or older, who completed a depression screening as part of the National Health and Nutrition Examination Survey, 2005-2008. Depression was ascertained using the Patient Health Questionnaire (PHQ), a 9-item screening instrument. The prevalence of depression was 5.5% (SE: 0.64) in men (N ¼ 2,836) and 8.5% (SE: 0.71) in women (N ¼ 2,845). After adjustment for covariates, significant associations between social contacts and depression were identified, and they were more salient among men than women. Compared to those who attended church weekly, the odds ratio (OR) was 2.57 (95% confidence interval (CI): 1.47-4.20), 2.43 (95% CI: 1.36-4.33), and 2.16 (95% CI: 0.88-5.31) among men who never attended church, attended occasionally, and more than weekly, respectively. The corresponding ORs for women were 1.81 (95% CI: 1.11-2.94), 1.71 (95% CI: 1.05-2.78), 0.99 (95% CI: 0.53-1.85). Compared with the respondents having 10 or more friends, the ORs of depression were 3.72 (95% CI: 1.77-7.83) for men and 1.88 (95% CI: 0.90-3.71) for women who had no close friends. Compared to married individuals, being previously married was significantly associated with depression in both men with OR of 2.15 (95% CI: 1.46-3.15) and women with OR of 1.62 (95% CI: 1.06-2.48). The current study provides evidence to support increasing social contacts to prevent depression, particularly among men. The inherent limitation of cross-sectional design, however, prevented the authors from investigating causality.