From Data Source Description to Bias Quantification: Applications of the DIVERSE Framework

From Data Source Description to Bias Quantification: Applications of the DIVERSE Framework

Summary

About the presentation: In this webinar, the speakers will present findings from the DIVERSE Initiative and their relevance to CNODES research. First, Dr. Pajouheshnia will introduce the DIVERSE framework and tool for describing real-world data sources, including nine key dimensions to describe data sources to support study replicability. Following this, he will present and compare two examples of data sources that have been mapped to the DIVERSE tool, 1) the UK CPRD Aurum and 2) CNODES Ontario and will share reflections on what we can learn from these mappings and how they support replicability of research. Next, Dr. Gini will explain how information collected in the DIVERSE tool can be used to understand and quantify measurement error in studies. Dr Gini. will briefly introduce DAiReCt (Disease, Action, Record, Content), a novel framework designed to capture the causal structure hidden behind phenotypes of variables and will introduce the component strategy that leverages real world data diversity in studies to quantify the share of true positives missed by phenotypes.

About the presenters:

Dr. Romin Pajouheshnia is an epidemiologist at RTI Health Solutions and is based in the Netherlands. He has more than a decade of experience in pharmacoepidemiology, clinical epidemiology, and epidemiologic methods research. His work focuses on generating real-world evidence using secondary data sources, with interests spanning drug safety, comparative effectiveness research, and methodological research. Together with Rosa Gini, he co-leads the International Society for Pharmacoepidemiology Databases (ISPE) Special Interest Group’s “DIVERSE Initiative”, which aims to improve the transparency and replicability of real-world evidence through the improved description and understanding of secondary data sources.

Dr. Rosa Gini is Head of the Pharmacoepidemiology Unit at ARS Toscana (Regional Health Agency of Tuscany, Italy). She is a mathematician by training, and her path as a data scientist led her to specialize in the secondary use of healthcare data for pharmacoepidemiologic and health services research. Her work focuses on developing methods, standards, and tools to support transparent, reliable, and timely generation of real-world evidence for health policy and regulatory decision-making. She is a former Chair of the ISPE Databases Special Interest Group, chairs the European Network of Centres for Pharmacoepidemiology & Pharmacovigilance’s Working Group on Transparency and Scientific Independence, and serves as Vice President of the Vaccine Monitoring Collaboration for Europe (VAC4EU).