The major goal of this project we continue and extend our examination of inefficiency and variation in health care utilization and health outcomes at the region or system level. We introduce new data sources, innovative network analysis, and focus on vulnerable patients such as those with Alzheimer’s Disease and related dementia (ADRD). Our proposal addresses the following three issues:

Challenges in clinical decision-making: Barriers include: inadequate evidence on risks and benefits of treatments, especially among patients commonly excluded from clinical trials; imperfect estimates of prognosis, especially for patients with ADRD; and difficulties incorporating patient preferences into key decisions. Projects 1, 2, 3 and 5 will help to improve clinical decision-making by physicians.

Uncertain role of the care delivery environment: Organizational structure of clinical practices, patient-sharing networks, payment models, laws and regulations likely play important roles in practice patterns and patient outcomes. The influence of these factors is understudied. To address this shortcoming, all 5 projects will draw on patient-sharing networks of clinicians (Core C) and a complete assignment of U.S. physicians to health care organizations coupled with surveys on organizational characteristics (Core B) to explore environmental factors.

Limited health policy analysis: Evaluation of health policy impact has frequently been based on analysis of data from a single payer source (Medicare fee-for-service, commercial, or Medicaid) alone, leading to conclusions that do not always generalize across income groups, age groups or local health markets. With more than 1 billion person-years of claims data, we will test for consistency of findings across payers (Projects 2 and 4) and possible harms of differential pricing (Project 4) to inform policy aimed at improving care.