Grant: Operating Grant

The proposed research will address three important areas in health research. First, we will improve the statistical methods that are used for comparing outcomes of medical care between different hospitals. We will develop methods that allow comparison of hospital performance to be designed so as to minimize the costs incurred by health care providers and consumers due to the possible misclassification of health care provider performance. We will also develop statistical methods for comparing hospital performance on multiple indicators of quality of care simultaneously. This will allow hospital administrators and physicians to improve the quality of care provided at individual hospitals. Furthermore, it will allow physicians, patients, and administrators to determine if hospitals that perform well on one indicator also tend to perform well on other indicators of quality of care. Second, we will improve the statistical methods available for determining the effects of treatments and interventions on health outcomes. These provide important evidence for resource allocation and for the protection of the public. Third, we will improve the statistical methods that are used for predicting outcomes and prognosis for patients. Predicting prognosis allows for the development of clinical decision rules, and provides clinicians and patients with important evidence on prognosis, thus facilitating evidence-based decision-making about patient care.