Essays on Innovation in the Medical Device Industry

Abstract

This dissertation includes three empirical papers on the development and adoption of medical devices in the United States. Economists attribute as much as half of recent gains in life expectancy in the United States to the use of new medical technologies. When developing medical technologies, manufacturers must consider the “total product lifecycle” of devices, spanning from development costs to regulatory approval to insurer coverage and ultimately patient and physician adoption. The three chapters of this dissertation examine different stages of the total product lifecycle for medical devices.In Chapter 1, I study how medical device firms change their investments in research and develop following external shocks to production costs. Using damage to device manufacturing facilities caused by Puerto Rican hurricanes as a natural experiment, I find that increases in storm exposure cause firms to spend less on research and development and bring fewer medical devices to market. I also find that devices brought to market following storms are cited in competitor regulatory submissions no more or less often than the average medical device. This suggests that device firms do not meaningfully target more or less scientifically innovative projects at the margin when reducing investments in research and development. In Chapter 2, I describe the availability of cost-effectiveness analyses for medical devices in the United States. Cost-effectiveness analyses are not consistently used by insurers when making coverage decisions in the United States. I find that one of the barriers to using cost-effectiveness analyses is the timing of when analyses become available. Cost-effectiveness analyses are not available until several years after regulatory approval. In Chapter 3, I examine the effect of industry payments on physicians’ adoption of Medtronic’s Micra leadless pacemaker in fee-for-service Medicare. Leadless pacemakers have lower complication rates but a higher cost compared to traditional leaded pacemakers. I find that physicians who receive more payments from pacemaker manufacturers are more likely to adopt leadless pacemakers. However, this relationship is not robust to either physician fixed effects or an instrumental variables analysis predicting receipt of manufacturer payments as a function of distance from Medtronic headquarters.