|
Activity Number:
|
59
- Evidence-Generation via Big Data in the Real-World Setting
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 30, 2017 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Stats. Partnerships Among Academe, Indust. & Govt. Committee
|
|
Abstract #324897
|
|
|
Title:
|
Drug Safety and Comparative Effectiveness at Massive Scale
|
|
Author(s):
|
Yuxi Tian* and Marc A. Suchard and Martijn Schuemie
|
|
Companies:
|
David Geffen School of Medicine At UCLA and University of California, Los Angeles and Janssen Research and Development
|
|
Keywords:
|
Drug Safety ;
Observational Data ;
Propensity Score Adjustment
|
|
Abstract:
|
The Observational Health Data Sciences and Informatics (OHDSI) program is a multi-stakeholder, interdisciplinary collaborative to bring out the value of health data through large-scale analytics. OHDSI's data partners track over 700 million individuals world-wide. We describe recent statistical developments to address drug safety and comparative effectiveness questions at such a scale while controlling for the observational nature of the data. We analyze optimal strategies for performing propensity adjusted, large-scale regression and advance computing technology to conduct such analyses efficiently. We discuss our findings on the merits of different propensity score estimation methods and our approaches to the challenge of assessing method performance in the context of observational data. We weigh in on several topical drug safety issues, and provide open-source tools for others to employ in their own research objectives and data.
|
Authors who are presenting talks have a * after their name.
