Adopting a CDISC data model to build a clinical data warehouse

Authors

• J Singh
• KO Newell
• MJ Duvenhage
• PT Gumne
• HK Kandaswamy
• GD Marlow
• CJ Whalen
• A Rosenthal
• M Tartakovsky

Abstract

The NIH/NIAID/OCICB provides clinical research data management tools for ~100 protocols, which require prospective management and oversight by scientific investigators and study team personnel to ensure that accurate and consistent data is captured. Effective oversight and management of clinical research by scientific investigators and study team personnel requires that high quality data is captured in the most timely manner possible in order to feedback into trial planning and decision-making. Although the specific research aims and objectives of research protocols differ by nature- the underlying administrative and organizational concerns are fairly consistent across protocols. Data warehouses can provide an effective approach for cross-protocol statistical analyses and centralized reporting framework for common data indicators and trial meta-data that are relevant for any research design, including clinical trials, epidemiologic studies and observational research. This poster will describe the process of building an integrated data warehouse solution that uses open source tools and internationally accepted CDISC standards for clinical research, and it provides a model for other institutions with active clinical research data that would like to implement a data warehouse solution that can be validated. We will provide an overview of the complete lifecycle from building case report forms using CDASH, to mapping clinical database source systems into a standard data model using extraction, transformation and loading to creating centralized data analytics reports. We will describe how we completed this process in compliance with regulatory standards governing electronic records.

Scientific Focus Area: Research Support Services

This page was last updated on Friday, March 26, 2021