Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Healthcare Systems

First Advisor

Mark V. Pauly

Second Advisor

Guy David


Health information exchange (HIE) is the electronic exchange of patient medical records among hospitals. I investigate how two defining characteristics of HIE can cause under-adoption. First, HIE represents an information-sharing network, and participation for any hospital is valuable if others participate. This is a network effect. Second, HIE involves exchange of medical records, which may be competitive assets for hospitals. Therefore, hospitals may have disincentive to exchange with competitors despite social benefits. This is a competitive effect. I present a theoretical framework of hospitalsâ?? decisions to adopt HIE and show how presence of network and competitive effects can result in under-adoption relative to the social optimal. I then test for evidence of network and competitive effects in hospital HIE adoption. In the empirical analysis, I use two measure of HIE adoption. The first is a measure of general adoption in which all hospitals that have adopted any HIE capability are assumed to be able to exchange information with each other. The second takes into account that much information-sharing occurs through interoperable IT systems and currently, most IT systems of different vendors are not interoperable. I find evidence of network effects in general HIE adoption and vendor choice. I find that a 10% increase in market adoption rate results in a hospital being 9.2% more likely to adopt HIE. I also find that a 10% increase in adoption rate of a vendor results in a hospital being 1.5% more likely to adopt the vendor. I also find evidence of competitive effects. Specifically, hospitals that are more vulnerable to losing market share are less likely to adopt a prominent vendor in a market. I estimate a model of patientsâ?? hospital preferences and hospitalsâ?? HIE adoption decision. Through counterfactual simulation, I show effects of widespread HIE adoption on market share redistribution. Finally, I evaluate current policies such as the federal governmentâ??s $30 billion adoption incentive program (HITECH Act). The program may be inadequate to promote widespread interoperability in the presence of competitive effects. I also discuss the implications of network effects for competition and innovation in the health IT industry.

Included in

Economics Commons