Evidence that higher health care spending across US regions is associated with lower quality and no gainsin health outcomes suggest substantial opportunities to improve the efficiency of U.S. health care. Achievingimprovements, however, will be difficult due to the lack of actionable measures and of quality and resourceuse - and insufficient knowledge about the determinants of efficiency - at the physician, physician group,and hospital levels. This project will address the following specific aims:
Aim 1. Develop and implement measures that can be used to assess the efficiency of physicians, physiciangroups, and hospitals. We will draw on routinely collected data to develop and implement provider-specificclinical performance measures. Because small sample sizes seriously limit the number of providers who canbe reliably measured, we will apply statistical filtering with empirical Bayes techniques to improve theprecision of our measures.
Aim 2. Explore the determinants of efficiency at the individual physician level. We have agreements in placeto obtain board certification exam scores from the American Board of Internal Medicine and American Boardof Family Medicine. We will link individual physicians' exam scores to the clinical performance measuresdeveloped under Aim 1 as well as with measures that characterize the settings where they trained andwhere they now practice. These data will be used to address the question: what makes a good doctor?Aim 3. Explore the determinants of efficiency for physician groups and hospitals. We will draw on theextensive measures developed under Aims 1 and 2 - and through our proposed national survey ofphysicians - to identify factors associated with improved efficiency at two organizational levels: the hospitalitself and the 'extended medical staff' (physicians who either work within a specific hospital or whosepatients are admitted to these hospitals).The goal of this project is to provide measures and insights at the levels of individual physicians, physiciangroups, and hospitals that should prove useful in efforts to improve the efficiency of the US health caresystem.
Showing the most recent 10 out of 263 publications
