Background: Hyperthyroidism has been associated with increased all-cause mortality. Whether the underlying cause of hyperthyroidism influences this association is unclear. Our objectives were to explore whether mortality risk and cause of death differ between Graves' disease (GD) and toxic nodular goiter (TNG). Methods: This is an observational cohort study, using record-linkage data from nationwide Danish health registers. A total of 1291 subjects with GD and 861 with TNG, treated in a hospital setting, were identified and followed for a mean period of 11 years. Cases were matched 1:4 with nonhyperthyroid controls with respect to age and sex. The hazard ratio (HR) for mortality was calculated using Cox regression analyses. All analyses were adjusted for comorbidity using the Charlson score. Results: Both GD (HR=1.42 [95% confidence interval (CI) 1.25-1.60]) and TNG (HR=1.22 [CI 1.07-1.40]) were associated with increased all-cause mortality. After stratification for the cause of death, GD was associated with increased mortality due to cardiovascular diseases (HR=1.49 [CI 1.25-1.77]) and lung diseases (HR=1.91 [CI 1.37-2.65]), whereas TNG was associated with increased cancer mortality (HR=1.36 [CI 1.06-1.75]). When analyzing mortality in GD using TNG individuals as controls, there was no significant difference in all-cause mortality between GD and TNG. However, GD was clearly associated with a higher cardiovascular mortality (HR=1.39 [CI 1.10-1.76]) compared to TNG. Conclusion: Both GD and TNG, treated in a hospital setting, are associated with increased all-cause mortality. The causes of death differ between the two phenotypes, with cardiovascular mortality being significantly higher in GD.