Most health economists (81 percent) believe that the primary explanation for rising health care expenditures is the increasing development, adoption, and utilization of new medical technology. This study aims to extend our understanding of the role of technology in driving cost growth. We have three objectives: 1. To assess which diseases have contributed most to health care cost growth at the population level. 2. To assess the role of technology in contributing to cost growth 3. To assess the types of technology that have influenced cost growth since 1996 and the types of patients to which they have been applied. Two research methods will be used to accomplish these aims. First, retrospective analysis of a large administrative claims database will be conducted to identify which diseases have contributed most to cost growth. This analysis will follow the residual method for identifying the role of technology in driving cost growth. Specifically, like the literature, we will adjust for changing demographics. In contrast to much of the existing literature, we will be able to control, in a much more detailed way, for changing prices and insurance design as well as for geographic population shifts which might be important given the cross-sectional variation in health care spending. Following the literature, the residual cost growth will be attributed to """"""""technology."""""""" Because much of what is labeled as technology in the residual method may not in fact be technology, we will conduct a series of semistructured interviews with physicians to better understand how and why practice patterns have changed. This will enable us to get a more detailed sense of the role of technology in determining cost growth and allow us to address specific aims 2 and 3. In contrast to much of the work in this area, we will focus on a commercially insured population, allowing us to comment on cost increases that relate to premium increases in the employer market. Also in contrast to much of the existing literature, we will be able to separate out changes in incidence from changes in resource use conditional on disease incidence. Both incidence and conditional resource use may be affected by technology. The semistructured interviews will be designed to provide some understanding of how technology has affected both of these components of cost growth, relative to other unobserved factors such as rising obesity rates.
