Informatics Research Seminar: Look before you Link-Assessing Health Data Linkage Feasibility and Linkage Results
February 27 @ 4:00 pm - 5:00 pm EST
Speaker: Sudha Raman, PhD
Presented from Duke
Broadcast Link: Seminar
Linking data sources can help to enrich clinical researchers to more accurately define study populations, enable adjustment for confounding, and improve the capture of health outcomes. While there is much guidance about the technical process of linking data, there is less so about the pre study assessment of feasibility and the post study evaluation of the linkage results. When creating novel linked datasets, researchers must assess the feasibility of both scientific aspects (data quality and linkage methods) and operational aspects (access, data use and transfer, governance, and cost). Another key aspect of data linkage evaluation is to articulate how a linkage process was performed and its accuracy so that the potential for bias can be assessed. This presentation will review the results of a group effort to create guidance for the assessment of the feasibility of health data linkage for researchers, as well as recommendations for the evaluation and reporting of health data linkage. Examples of health data linkage from Dr. Raman’s work will be shared.
Sudha Raman, PhD, is an Assistant Professor in the Department of Population Health Sciences. She is an epidemiologist who focuses on the use and effects of medications in populations (pharmaco-epidemiology), with careful consideration of a medicine’s benefits as well as harms. She received her PhD in epidemiology from the University of North Carolina at Chapel Hill and completed a fellowship at the Center for Pragmatic Health Systems Research at Duke Clinical Research Institute. Her current research explores the quality and methodological challenges of conducting research using real-world data, such as electronic medical records and administrative claims data, as applied to the evaluation of health care for both children and adults.