To evaluate the effectiveness of a hospital-wide initiative to improve patient safety by implementing high-reliability practices as part of a quality improvement (QI) program aimed at reducing all preventable harm. A hospital wide quasi-experimental time series QI initiative using high-reliability concepts, microsystem-based multidisciplinary teams, and QI science tools to reduce hospital acquired harm was implemented. Extensive error prevention training was provided for all employees. Change concepts were enacted using the Institute for Healthcare Improvement's Model for Improvement. Compliance with change packages was measured. Between 2010 and 2012, the serious safety event rate decreased from 1.15 events to 0.19 event per 10 000 adjusted hospital-days, an 83.3% reduction (P < .001). Preventable harm events decreased by 53%, from a quarterly peak of 150 in the first quarter of 2010 to 71 in the fourth quarter of 2012 (P < .01). Observed hospital mortality decreased from 1.0% to 0.75% (P < .001), although severity-adjusted expected mortality actually increased slightly, and estimated harm-related hospital costs decreased by 22.0%. Hospital-wide safety climate scores increased significantly. Substantial reductions in serious safety event rate, preventable harm, hospital mortality, and cost were seen after implementation of our multifaceted approach. Measurable improvements in the safety culture were noted as well. Copyright © 2013 Mosby, Inc. All rights reserved.