OBJECTIVES: During the debate over health reform, comparative effectiveness research was touted as a relatively painless approach to reducing costs. A comparative effectiveness study of two treatments will find either that the costlier treatment is more effective or is not more effective than a less expensive alternative. Studies that report negative results have the potential to reduce costs, but only if findings affect clinical practice. One concern is that the same factors that promote rapid adoption of new therapies in the U.S. may retard the abandonment of existing technologies found to be ineffective. METHODS: The COURAGE trial found that optimal medical therapy is as effective as percutaneous coronary intervention (PCI) for patients with stable angina. PCI refers to stenting and angioplasty. The trial was published and widely publicized in early 2007. We evaluate trends in PCI volume pre- and post-COURAGE by indication using 1) 100% samples of outpatient and inpatient discharge data for California, Florida, New Jersey, and Maryland, 2) a 100% sample of discharge data for Veteran’s Administration hospitals and 3) data from a proprietary cardiac catheterization laboratory registry at 15 hospitals. RESULTS: Between the fourth quarter of 2006 and the fourth quarter of 2007, PCI volume in California, Florida, New Jersey, and Maryland among patients without serious coronary disease declined from approximately 17,000 to 13,000 procedures (an 18% decline). There was only a 5% decline among patients with unstable angina, who were not included in COURAGE. We found similar patterns in the other datasets. CONCLUSIONS: Publication of the COURAGE trial had an impact on PCI volume. Many patients with stable angina continue to receive PCI. The results are consistent with the view that as long as the health system is configured around procedural-based medicine, the impact of trials which find that medical therapy is as effective as invasive procedures will be modest.