Technology Adoption and Organizational Change: Recent Trends in the Healthcare Delivery Sector

The healthcare sector has experienced a vast amount of consolidation over the previous two decades. This consolidation has taken the form of hospital mergers, hospitals employing more physicians, and hospitals buying up physician practices. This period has also seen the rapid diffusion of electronic medical records (EMRs) as well as other kinds of technologies.

I document two distinct kinds of integration in the health care delivery sector. One type of integration is the closer association between hospitals and physicians as evidenced by the emergence of the salaried physician model and the adoption of hospitalist programs. The second type of integration is the consolidation of ownership. This trend can be measured by increasing concentration in health care markets. It is also evidenced by the gradual disappearance of standalone hospitals, as more and more hospitals become part of bigger health systems that operate multiple hospitals under common ownership.

Taking advantage of a rich new dataset, the AHA Health IT Supplement, to obtain data on the adoption and use of EMRs, I show that the correlation between organizational integration in health care delivery and technology adoption has increased over time. I propose several mechanisms that could explain the association between the adoption of electronic medical records and greater hospital-physician integration.

I find that EMR adoption is positively correlated with hospitalist use and the tendency of hospitals to employ their physicians. In particular, I find that the likelihood of having a hospitalist program goes up to around 60% from around 40% if the hospital has a basic EMR system, while the probability of employing the physicians on the medical staff increases by 8 percentage points. This finding, however, does not provide evidence of a causal relationship because there might be time varying unobservable factors that predict both closer hospital-physician integration and greater EMR adoption. Therefore, I next attempt to understand what factors predict the adoption of EMR and hospitalist programs. I find that both EMR adoption and hospitalist use are associated with the hospital being part of a system and using other sophisticated technological equipment. I also show that hospitals located in concentrated markets are more likely to adopt electronic medical records and to use hospitalists.

For a limited set of hospitals, I investigate the relationship between management quality and the adoption of process innovations using data on US hospitals from the World Management Survey. I find that well-managed hospitals are more likely to have electronic medical records and to use hospitalists. However, I do not find any link between physician employment and management quality or between local market concentration and management quality.