Development of a Web-Based Tool for Comparative Effectiveness Research Using Observational Data

• Activity Number: 292 - SPEED: Statistics in Epidemiology
• Type: Contributed
• Date/Time: Tuesday, August 1, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Yi Zhang* and Mae Thamer and Onkar Kshirsagar
• Companies: Medical Technology and Practice Patterns Institute and Medical Technology and Practice Patterns Institute and Medical Technology and Practice Patterns Institute
• Keywords: causal inference ; web-based tool ; CER

Abstract

Background: We must rely on observational data to answer CER questions when randomized experiments are not available. However, inferring causality from observational studies is complex. A generic and user-friendly web-based tool to guide the design of causal inference studies is currently being developed.

Aim: To describe an effort to develop a web-based tool, CERBOT (CER Based on Observational Data to Emulate a Target Trial) for better comparative effectiveness research using observational data. The tool is designed to be used by a team of researchers, patients and stakeholders. CERBOT guides users conduct their observational analyses by 1) articulating their research question in terms of a hypothetical randomized trial (the target trial) and 2) specifying how to emulate the target trial using observational data.

Methods: The tool builds up from a conceptual framework--causal inference based on observational data can be viewed as an attempt to emulate a randomized clinical trial (the target trial). There are five modules included in CERBOT, each with multiple components comprised of a parallel process of Specification and Emulation of the target trial.