Background: Differences in prevalence and prognostic information of cardiac troponin T (cTnT) and I (cTnI) concentrations in patients without acute coronary syndrome (ACS) are insufficiently investigated. High-sensitivity assays (hs-cTn) have led to an increased interest in hs-cTn for risk stratification. Here, we compare hs-cTnT and hs-cTnI in prediction of mortality patients without ACS. Method and results: Patients aged >18 years, consecutively admitted to an emergency department (ED) were included. Blood was collected at admission and later analyzed with high-sensitivity assays for cTnT (Roche) and cTnI (Siemens). Troponin concentrations were reported as normal or increased according to the clinical cut-off value of 99th percentile as defined by the manufacturer. The primary outcome was all-cause mortality. Of the 822 participants (median, 65 years [48–77]; 428 female [52%]), 239 patients died. Median follow-up time was 3.0 years [2.1–3.0]. Elevation of hs-cTn was observed in 40% (n = 345) for hs-cTnT and 8% (n = 64) for hs-cTnI, p < 0.001. The relationship between elevated hs-cTn and mortality was strong for both hs-cTnT and hs-cTnI [HR 6.0 (95%CI: 2.9–12.6) vs. 5.1 (95%CI: 1.9–13.6)].There was no difference in prognostic accuracy for short-term mortality (30 days) between hs-cTnT and hs-cTnI. However, the prognostic accuracy for long-term mortality (1080 days) was superior for hs-cTnT than for hs-cTnI [area under the receivers operating curve (AUC) 0.81 vs 0.74, p < 0.001]. Conclusion: Both hs-cTnI and hs-cTnT were predictive for all-cause mortality. Notably, hs-cTnT measurement showed superior prognostic performance in predicting long-term all-cause mortality compared with hs-cTnI.