We use machine learning to better characterize low-value health care and the decisions that produce it. We focus on costly tests, specifically for heart attack (acute coronary syndromes). A test is only useful if it yields new information, so efficient testing is grounded in accurate prediction...

We analyze whether decreased emergency department access (measured by increased driving time to the nearest ED) results in adverse patient outcomes or changes in the patient health profile for patients suffering from acute myocardial infarction. Data sources include 100% Medicare Provider...

Productivity spillovers are often cited as a reason for geographic specialization in production. A large literature in medicine documents specialization across areas in the use of surgical treatments, which is unrelated to patient outcomes. We show that a simple Roy model of patient treatment...

Managed care may affect medical treatments for non-managed-care patients if it alters local market structure or physician behavior. We investigate whether higher levels of overall managed care market share are associated with greater use of recommended therapies for fee-for-service patients with...