Copeptin, the 39-amino-acid C-terminal portion of provasopressin, has been shown to be an independent predictor for adverse events following STEMI (ST elevation myocardial infarction). We hypothesized that plasma copeptin was an independent predictor for adverse outcomes following acute NSTEMI (non-STEMI) and evaluated whether copeptin added prognostic information to the GRACE (Global Registry of Acute Coronary Events) score compared with NT-proBNP (N-terminal pro-B-type natriuretic peptide). Plasma copeptin and NT-proBNP were measured in 754 consecutive patients admitted to the hospital with chest pain and diagnosed as having NSTEMI in this prospective observational study. The end point was all-cause mortality at 6 months. Upper median levels of copeptin were strongly associated with all-cause mortality at 6 months. Copeptin was a significant predictor of time to mortality {HR (hazard ratio), 5.98 [95% CI (confidence interval, 3.75–9.53]; P<0.0005} in univariate analysis and remained a significant predictor in multivariate analysis [HR, 3.03 (05% CI, 1.32–6.98); P=0.009]. There were no significant differences between the area under ROC (receiver operating characteristic) curves of copeptin, NT-proBNP and the GRACE score. Copeptin improved accuracy of risk classification when used in combination with the GRACE score as determined by net reclassification improvement, whereas NT-proBNP did not. The relative utility of the GRACE score was increased more by copeptin than by NT-proBNP over a wide range of risks. Plasma copeptin is elevated after NSTEMI, and higher levels are associated with worse outcomes. Copeptin used in conjunction with the GRACE score improves risk stratification enabling more accurate identification of high-risk individuals.