Background: Data are limited on whether clinical trials have randomized higher-risk patients over time and how trends in risk profiles and evidence-based pharmacotherapies have influenced trial outcomes. We quantified changes in baseline risk, treatment, and outcomes of patients with non-ST-segment elevation acute coronary syndromes (NSTE ACS) randomized in 9 phase 3 clinical trials of antithrombotic therapy over 15 years. Methods: We studied 58,771 patients in GUSTO IIb, PURSUIT, PARAGON-A, PARAGON-B, PRISM, PRISM-PLUS, GUSTO IV-ACS, SYNERGY, and EARLY ACS. Patient-level data were mapped to 3 pre-specified 5-year randomization periods. Temporal trends in GRACE score-predicted mortality were compared with trends in observed mortality. Results: Over time, in-hospital and discharge use of thienopyridines (p=0.001), statins (p<0.0001), and angiotensin-converting enzyme inhibitors (p<0.0001) increased, and hospital length-of-stay decreased (p=0.024). Blood transfusion use increased (8.3% [1994-98], 10.7% [1999-2003], 13% [2004-08], p=0.0002) despite stable rates of severe bleeding (0.9% [1994-98], 1.4% [1999-2003] and 1.1% [2004-08], p=0.127) and coronary artery bypass grafting (12.4% [1994-98], 13.7% [1999-2003] 13.1% [2004-08], Conclusions: Despite enrolling higher-risk patients into these NSTE ACS trials, with better treatment, observed mortality declined over the past 15 years. The appropriateness of increased blood transfusion despite unchanged bleeding rates deserves further study. (C) 2012 Elsevier Ireland Ltd. All rights reserved.