Physician Level Analysis of Medical Technology Diffusion: Coronary Stents Study

Summary/Abstract The focus of my PhD training in economics was empirical industrial organization, a field that studies strategic interactions between firms, consumers and other agents in the economy. Since obtaining my PhD, I have started exploring a new field, health economics, first through my employment at RAND, then in my position as an assistant professor at the University of Minnesota. I have explored various topics in health economics to build from scratch a spectrum of expertise and hands-on knowledge in a field brand new for me. A natural next step for my career development is to focus on a research agenda around several important themes that interest me most, and to expand my research knowledge beyond what I have studied thus far. The key themes I would like to focus on are medical innovation and diffusion, the role of incentives at the patient and provider level, and the role of insurance. The proposed study will allow me to:(1) develop expertise in the analysis of Medicare database; (2) build capability in working with large databases; (3) enhance my statistical training; (4) understand diseases and gain clinical insights; (5) improve my ability to develop research projects, and to write grants competitive for funding at the R01 level; (6) expand my exposure to health services research and health economics community; Through the proposed project, I will start building a long-term research agenda focused on understanding the mechanisms underlying the adoption and diffusion of new medical technologies. In most circumstances, physicians are the key agent in determining whether a patient receives a given medical technology, suggesting that understanding the forces that affect physician technology adoption and use is critical to addressing utilization of medical technology. However, physician adoption of medical technology is not well studied or understood. My goal is to investigate how competition, social networks and learning from other physicians affect patterns of physician medical technology adoption in addition to the more traditional explanations such as physician, hospital and patient characteristics. I propose to study the diffusion of medical technology using coronary artery stents in the Medicare population as the exemplar. Linking Medicare claims to physician demographic data from the American Medical Association and to hospital characteristics from the American Hospital Survey: (1) I will document the variation across time, regions, and hospitals in physicians' use of bare-metal, coated and drug eluting stents. (2) I will estimate the role of physician characteristics, competition among physicians, social interactions (including learning from other physicians), and expected patient demographics on the physician's decision to adopt and use new stents. (3) I will estimate the consequences of physicians' use of new stents on patients' outcomes following stent insertion, such as rates of re-admissions, repeat percutaneous coronary interventions (PCI), and mortality.