Defense Date


Document Type


Degree Name

Doctor of Philosophy


Health Administration

First Advisor

Robert Hurley


Finding more efficient ways to organize and deliver medical care is a major policy and management concern in the United States. High levels of expenditures for administrative and coordinating functions are attributed to the fact that health care systems are not "seamless" and that excessive transaction or friction costs are incurred in the exchanges between providers and purchasers and among providers. Renewed interest in vertical integration as a means to addressed these problems is being explored in the empirical literature, but rigorous theory-based investigations are rare. This study is a theory-based exploration of how hospitals address the "make-or-buy" decision of acquiring nursing home services for patients requiring post-acute stay placement.

The purpose of the study is to investigate under what circumstances hospitals chose to undertake formal arrangements to acquire nursing home services for patients to be discharged, rather than simply arranging for each discharge in the "spot market." In some instances this may be long-term contracting or leasing of beds, while in other instances it may mean the hospital acquires or develops its own skill nursing facility--a form of vertical integration. The study adopts Oliver Williamson's transaction cost economics theory as the theoretical basis for the study. This framework argues that the most efficient mode of transacting is determined by analyzing three dimensions of the transaction: uncertainty, frequency, and asset specificity (supplier identity). At higher levels of each of these dimensions, organizations are more likely to observe that "markets fail" and that formal arrangements between buyers and sellers are preferable, with vertical integration representing the "make" versus "buy" option.

The study uses data from the American Hospital Association Survey and other sources to identify if and how hospitals have made formal arrangements for nursing home services. It tests ten hypotheses derived from the theory that focus on the three dimensions of transactions and interactions among them. The methodology uses several analytical approaches to establish the validity of the measures of the dimensions, and then tests the hypotheses using multivariate logistic regression to contrast various modes of transaction. The importance of transaction uncertainty and specificity are strongly supported in the findings, while transaction frequency is weakly correlated to higher degrees of integration. The results are consistent with both the theoretical arguments advanced by transaction cost economics and with prior research, which is only available from non-health care applications.

The study makes an important, and perhaps unique, contribution to empirically operationalizing and testing a transaction cost economics-based interpretation of the decision to vertically integrate in health care. It also provides useful insight into the need for vertical integration to be selectively adopted as it may not be the most efficient mode of organization in all "make or buy" decision opportunities.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission