The science behind current pay-for-performance and quality reporting systems is underdeveloped. Among the many issues is a lack of standardization across systems and the use of targets that tend to focus on either arbitrarily assigned or average performance. The development of a target of the """"""""best possible"""""""" practice using a set of standardized indicators would be an improvement over the current science. As well, the ability to ascertain factors that contribute to an improvement in those indicators would be extremely helpful.
The aims of this project are to: 1) develop scientifically-derived """"""""best practice"""""""" targets for nine of the Centers for Medicare and Medicaid Services'(CMS) Hospital Quality Initiatives (HQI) outcomes measures;2) determine the gap and the distribution of the gap between those targets and the actual outcomes in a large national sample of U.S. hospitals;3) determine the factors that could be changed, and the direction and magnitude of change, in order to reduce that """"""""quality gap;"""""""" and 4) demonstrate the use of stochastic frontier analysis (SFA) in benchmarking quality and safety in U.S. hospitals. The targets for the HQI measures will be the """"""""best"""""""" possible HQI outcomes given (1) the inputs that go into producing various levels of quality, and (2) the technological """"""""efficiency"""""""" of managing those inputs, allowing for random shocks that are outside the control of hospitals. Each best practice target and the gap between that target and actual practice will be estimated through the use of SFA, a method for modeling inefficiency of processes that have measurable outcomes. The base SFA model is along the lines of a production function, in which an output is a function of a set of inputs, with the added components of technical inefficiency and random noise. The model assumes that the producer is trying to maximize the amount of output, but that technical inefficiency (which the producer controls) and random factors (uncontrollable) prevent attainment of the maximum possible output. The outputs in the SFA will be rates of hospital-level patient safety and quality events using nine of the 15 HQI outcomes measures currently required of hospitals for reporting and reimbursement. The inputs in the analysis will be indicators of capital, labor and technology. The analyses will also assess financial, market, and organizational factors that could contribute to the gaps in quality and safety. Data will come from the CMS HQI files, Medicare cost reports, and the American Hospital Association Annual Report. This will create a national convenience sample of hospitals expected to be 3,000 - 4,000 hospitals. Through this project we will introduce and test a new and innovative design and provide initial results regarding frontier levels and gaps in quality and safety, and predictors of those gaps, in a large national sample of U.S. hospitals. The project will move forward the ability to assess and compare hospitals on specific quality indicators and will serve as a foundation for broader applications.
The aims of this project are to: 1) develop scientifically-derived """"""""best practice"""""""" targets for the Centers for Medicare and Medicaid Services'Hospital Quality Initiatives outcomes measures;2) determine the gap and the distribution of the gap between those targets and the actual outcomes in a large national sample of U.S. hospitals;3) determine the factors that could be changed, and the direction and magnitude of change, in order to reduce that """"""""quality gap;"""""""" and 4) demonstrate the use of stochastic frontier analysis (SFA) in benchmarking quality and safety in U.S. hospitals. This will provide policy-makers and hospital administrators with the techniques and knowledge to assess how well best practices are being met, and how to improve existing practices. This has the potential be enormously beneficial to every hospital patient and therefore to contribute significantly to population health.
