Growth in the size and power of managedcare organizations seems to have had widespread effects on the health-care system. Not only may health maintenance organizations (HMOs) and other forms of managed care influence the care provided to the patients that they cover, but the activities of these organizations may also bring about broader changes in the delivery of health care. Such “spillover effects” of managed care could ultimately affect even health care for patients who have not joined managed-care organizations. While previous literature on the relationship between area HMO market share and spending is consistent with the presence of spillover effects (e.g., Baker, 1999), few studies have examined treatment patterns directly (e.g., Heidenreich et al., 2002). Nevertheless, the potential for changes in patient care are of central importance for gaining a full understanding of the overall effects of managed care on the health-care system. This paper examines the relationship between area HMO activity and the use of two relatively common therapies for heart-attack patients: percutaneous transluminal coronary angioplasty (PTCA) and coronary artery bypass graft surgery (CABG). We study patients covered by fee-for-service Medicare who suffered a new acute myocardial infarction (AMI, or heart attack) between 1985 and 1999. If HMO activity does have an impact on the use of PTCA or CABG for these fee-for-service patients, it will signal the ability of HMOs to influence treatment patterns outside of their own patients. We also investigate the extent to which information dissemination plays a role in managedcare spillover effects. The HMOs may have networks and other information-dissemination mechanisms, as well as financial and other points of leverage over treatment decisions, which allow them to alter physicians’ treatment decisions more effectively than less-managed health plans. Assuming that HMOs find it in their interest to encourage changes in treatment in response to new medical information, their efforts could lead to changes in physician practice patterns that are generally applied. We explore the role of information dissemination by studying the use of primary PTCA, defined here as PTCA performed within one day of an AMI patient being brought to the hospital. Prior to 1993, the typical immediate treatment for an AMI patient with a recently blocked artery was the administration of thrombolytic drugs to dissolve blood clots. Available evidence at that time suggested that the immediate use of PTCA was harmful (e.g., E. J. Topol et al., 1987). However, in 1993 three landmark studies were published simultaneously in the New England Journal of Medicine (Raymond J. Gibbons et al., 1993; Cindy L. Grines et al., 1993; Felix Zijlstra et al., 1993), demonstrating that immediate angioplasty was superior to thrombolytic therapy in improving blood flow and preventing the affected artery (or arteries) from becoming blocked again. A series of additional studies followed, demonstrating the superiority of primary PTCA for additional clinical endpoints (e.g., W. D. Weaver et al., 1997). If areas with high HMO market shares are differently able to alter physician practice patterns in response to new medical knowledge, we hypothesize that a differential trend may be observable in the use of primary PTCA in areas with higher and lower HMO market share * Baker: Department of Health Research and Policy, HRP Redwood Building, Room 110, Stanford University, Stanford, CA 94305-5405, and NBER; Afendulis: Center for Primary Care and Outcomes Research, 117 Encina Commons, Stanford, CA 94305-6019, and NBER; Heidenreich: VA Palo Alto Healthcare System, 111C, 3801 Miranda Avenue, Palo Alto, CA 94304, and Stanford University. The data used in this study are proprietary. This research was supported by AHRQ grant 5 R01 HS10925-03, and a career development award from the VA Health Services Research and Development Service for Heidenreich. We are grateful to Mark McClellan for assistance with the data and comments on a much earlier draft of this paper.