Getting something for nothing: Estimating service level improvements possible in hospitals

One of the focal points of discussion among firms in the service sector concerns the level of service delivered and its value. In the hospital sector, an important aspect of the level of service relates to the level of quality of care delivered. This paper undertakes an empirical study of 300 U.S. hospitals to evaluate their relative ‘service efficiency’ using the techniques of so-called allocative efficiency. This method estimates the potential improvement in service, estimates the ideal cost shares of the various resources, and provides the ‘peer group’ members to which the unit is being compared. Utilizing this information, the method proposed for units to improve involves two chief mechanisms. The first is a reallocation of cost shares among the various resource or input types (e.g. no increase in hospital budget). The second involves adopting the management practices used by peer group members. Over all 300 hospitals, a marked improvement potential was found for well over half. Of the poorest performers there appears to be some indication that a larger cost share was needed for face-to-face contact with patients (nursing) and a smaller share for personnel who do not deal with patients directly (lab technicians, non-physician administrators, etc.).