Statistical Issues in the Analysis of Electronic Health Record Data

• Activity Number: 99
• Type: Roundtables
• Date/Time: Monday, August 10, 2015 : 7:00 AM to 8:15 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Sebastien Haneuse*
• Companies: Harvard School of Public Health
• Keywords: Electronic health records

Abstract

Electronic health record (EHR) data present an incredibly appealing opportunity for public health and medical research. Relative to data from dedicated designs, EHR data often contains rich information on a broad range of patients over long timeframes and in real-world settings. Furthermore, it is typically relatively cheap to obtain. Nevertheless, the use of EHR data for research purposes is subject to many challenges, including (i) accurate extraction of text-based information, (ii) missing data, (iii) measurement error and misclassification, and (iv) the control of confounding. While these issues arise in other contexts, the complexity and high-dimensional nature of EHR data poses additional challenges for statisticians. The purpose of this roundtable is to provide a forum for statisticians and analysts to discuss these issues in general and in their own contexts, as well as discuss recent methodologic developments in the area.