Online Program

Comparative Effectiveness Research

*Constantine Gatsonis, Brown University 
*Sharon Lise Normand, Harvard University 

Keywords: CER

Comparative Effectiveness Research (CER) is now a major initiative in the US, with wide ranging implications for both research and health care policy. The evidence from CER is intended to support clinical and policy decision making at both the individual and the population levels. The mandate of CER places a premium on the study of outcomes that are of primary relevance to patients and on the derivation of conclusions that can inform individual patient choices. The broad scope of CER requires a wide array of methodologic approaches, including randomized studies, observational primary studies, and research synthesis. For example, approaches to identify treatment effect heterogeneity, to handle uncertainty in high-dimensional data, and to pool information across diverse data sources will require further development and experience. In this tutorial, we will survey statistical methodology currently in use and will discuss methodologic challenges in CER for diagnostic and therapeutic interventions.