Description: In this paper, we explore the importance of shifts in treatments for explaining increases in the cost of health care services. Using a large database of health insurance claims for sample patients, we find that there have been shifts in treatment intensity that have an important effect on costs and that, on average, those treatment shifts served to lower the cost of treating disease. These costs savings appear to be numerically important and pervasive. Our results lend support to earlier work that found these effects for a set of important conditions: heart attacks, cataract, and depression.
