Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Health Care Management & Economics

First Advisor

Guy David


Medical technology disadoption, a topic which has not previously been studied in great detail by health economists, is of great importance in a health care system where technology is understood to be a major driver of expenditure growth and the adoption of new technologies is consistently promoted by cultural norms and financial incentives. The disadoption of technologies with sub-optimal efficacy or safety is a key to improving the quality and value of care, and thus there is obvious utility to learning more about factors that influence technology disadoption. This dissertation establishes a novel theoretical framework to characterize the disadoption process along multiple dimensions, and then develops a simple mathematical model to describe physician disadoption behavior. Disadoption is examined empirically by analyzing the use of coronary drug-eluting stents (DES) in 2006-07(following their rapid adoption in 2003-04) using New York and Florida hospital discharge data, national practice organization data, and physician characteristic data. Preceding DES adoption behavior and physician group size are the primary factors studied for association with DES disadoption behavior. Empirical analyses indicate that faster DES adoption may be weakly associated with later DES disadoption, which may be consistent with physician risk aversion or product loyalty playing roles in disadoption decisions. Also, analyses suggest that there may be a week negative relationship between group size and timing of DES disadoption, suggesting that physicians in larger groups may begin the disadoption process later. These empirical results, which are subjected to instrumental variable analysis and extensive robustness checks, generally lack statistical significance, which highlights the current level of uncertainty surrounding disadoption behavior, including the observation that it does not appear to be empirically either similar or related to adoption behavior. Thus, this work both promotes and guides future research on medical disadoption and its determinants.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."