webinar register page

Webinar banner
CBER BEST Initiative Seminar Series – Understanding Informed Presence in Electronic Health Records Data
Understanding Informed Presence in Electronic Health Records Data

Description: Electronic Health Records (EHR) data have become a key data source for clinical research. Their relative availability and abundance make them very appealing for analytic tasks ranging from comparative effectiveness research, disease surveillance, population health, and predictive modeling. However, as has been widely documented, as a real world data source, there are a number of analytic challenges with EHR data. One of the key challenges of EHR data relates to the observability of the underlying data. While fundamentally a missing data problem, we have termed this process informed presence to highlight that what we observe is informative. In this talk I will provide some illustrations for how informed presence can bias insights and inference with EHR data as well as discuss some approaches – both analytical and design based – that can mitigate these biases. Ultimately EHR data are an extremely useful data source, though like any complicated data source need to be used thoughtfully.

Presented by:
Dr. Benjamin A. Goldstein, PhD, MPH
Associate Professor Biostatistics & Bioinformatics, Duke University

May 5, 2021 11:00 AM in Eastern Time (US and Canada)

Webinar logo
Webinar is over, you cannot register now. If you have any questions, please contact Webinar host: Karina Alexander.