Hospital Transaction Prices and Managed-Care Discounting for Selected Medical Technologies

It is generally assumed that managed care has been successful at capturing discounts from medical providers, but the implications have been a matter of debate. Critics argue that managed-care organizations attain savings by reducing intensity of services, while others have argued that savings are “real” and are a consequence of discounts per unit of care rather than reduced intensity. Because medical services tend to be bundled together into episodes of care, separating out prices and quantities can be difficult. Given available data, past studies focused on an “average” price for the aggregate hospital, calculated from total revenue divided by the number of inpatient days or cases (e.g., Glenn A. Melnick et al., 1992; Emmett B. Keeler and Melnick, 1999). David Dranove and Richard Ludwick, (1999) caution that these methods provide approximations of actual prices and are subject to measurement error due to unobservable differences in service mix. Examining treatment episode for acute myocardial infarctions (heart attacks), David Cutler et al. (2000) infer that discounts attained by managedcare plans are only partly due to reductions in intensity. Here, we employ data that enable us to observe transaction prices (i.e., actual payments borne by the payer and received by the hospital) for major procedures on an “unbundled” basis. Our analysis differs from previous studies in several important dimensions. Among these is the focus on pricing differences between various insurers and employers rather than differences within a single large insurer; as a consequence, we derive an empirical specification based on the bargaining framework due to John M. Brooks et al. (1997), rather than their insurerbased model. Moreover, to identify pure unit discounts we focus on the narrowly defined procedure. Like Cutler et al., we focus on coronary heart disease, a leading cause of death. We examine a major procedure used to treat this disease, one that is costly and relatively common. In bypass surgery (more fully, coronary arterial bypass graft, or CABG) healthy segments of artery are surgically inserted around the diseased arteries. In 2002 about 344,000 CABG’s were performed in the United States, with expenditures exceeding $21 billion. Other economists have focused on these procedures to examine market phenomena such as the hospital’s entry decision (Michael Chernew et al., 2002) or information diffusion (Dranove et al., 2003), yet the pricing decision was not treated fully.