INTRODUCTION: To evaluate the influence of marriage on the survival outcomes of men diagnosed with prostate cancer. MATERIALS AND METHODS: We examined 115,922 prostate cancer cases reported to the Surveillance, Epidemiology, and End Results (SEER) database between 1988 and 2003. Multivariate Cox regression techniques were used to study the relationship of marital status and prostate cancer-specific and overall mortality. RESULTS: Married men comprised 78% of the cohort (n = 91,490) while unmarried men (single, divorced, widowed, and separated) comprised 22% of the cohort (n = 24,432). Married men were younger (66.4 versus 67.8 years, p < 0.0001), more likely to be white (85% versus 76%, p < 0.0001), presented with lower tumor grades (68% are well or moderately differentiated versus 62%, p < 0.0001) and at earlier clinical stages (41% AJCC stage I/II versus 37%, p < 0.0001). Multivariate analysis revealed that unmarried men had a 40% increase in the relative risk of prostate cancer-specific mortality (HR 1.40; CI 1.35-1.44; p < 0.0001), and a 51% increase in overall mortality (HR 1.51; CI 1.48-1.54; p < 0.0001), even when controlling for age, AJCC stage, tumor grade, race and median household income. Furthermore, the 5 year disease-specific survival rates for married men was 89.1% compared to 80.5% for unmarried men (p < 0.0001). CONCLUSION: Marital status is an independent predictor of prostate cancer-specific mortality and overall mortality in men with prostate cancer. Unmarried men have a higher risk of prostate cancer-specific mortality compared to married men of similar age, race, stage, and tumor grade.