Understanding physician behavior in diagnostic and therapeutic decision making is crucial to making sense of the dramatic differences in the intensity of health care delivery across regions. This project examines the factors that drive physician decision making, and the consequences of these decisions for the health care system and the populations within that system, using as an example one of the most prevalent chronic diseases and the most common cause of death, i.e., CAD. The investigation focuses on the entry point into the diagnostic-therapeutic cascade, i.e., the diagnostic test. This application has three specific aims: (1) to determine what factors influence physician decision-making in the intensity of initial diagnostic testing for patients with known or suspected CAD; (2) to determine what factors influence subsequent testing and treatment for CAD among those who undergo initial testing; and (3) to assess whether a physician s propensity to use diagnostic testing in CAD is disease specific reflects a broader approach to diagnostic testing. Drawing directly from the Program Project Data Core, the investigators will assess the relative contributions of patient, physician, medical staff and regional factors to the intensity of diagnostic testing. The analysis will primarily use data from the 20% Medicare Part B claims combined with surveys of a sub-set of Medicare enrollees and physicians from a national probability sample. The investigators will evaluate these issues using traditional descriptive methods, random effects 1ogistic modeling and a novel method, recently developed by the investigators, that overcomes the problems of small sample sizes, bias from patient mix, and multidimensionality of treatment intensity that arise in the analysis of individual physician decision making. Understanding the contribution of physician decision making to the observed variation in diagnostic and therapeutic intensity is critical for three reasons. First, these differences generate large disparities in health outcomes and spending across areas or providers, raising questions of equity. Second, these differences suggest that many providers are either over-or under-treating important clinical groups. Third, understanding the forces that are leading to these variations in physician decision making can lead to informed policy and clinical interventions, e.g., interventions to address these questions of equity and clinical effectiveness.
Showing the most recent 10 out of 263 publications
