A Common Data Language for Clinical Research Studies: Overview and Update on the National Institute of Neurological Disorders and Stroke (NINDS) Common Data Element (CDE) Recommendations

Objective: The National Institute of Neurological Disorders and Stroke (NINDS) initiated the Common Data Element (CDE) project to develop data standards for clinical research in neuroscience. This CDE initiative strives to promote data collection in a consistent format, improve data quality, reduce study start-up time, facilitate data sharing/meta-analyses, and educate new clinical investigators. Imperative to the success of the project is encouraging widespread use of the CDEs and obtaining user feedback to continually improve CDE resources. Background: The CDE project was initiated in 2006 to assist investigators conducting neurological research so they could easily collect data in a standard format and assess common outcomes and endpoints. To remain current as research advances, the CDEs are regularly reviewed and updated. Design/Methods: Disease-specific Oversight Committees are tasked with periodically reviewing and updating the recommendations. NINDS also collaborates with governmental agencies (e.g., National Library of Medicine) and associations (e.g., Chiari & Syringomyelia Foundation) to jointly develop disease recommendations. CDE use is promoted in peer-reviewed publications and at national conferences where participants learn the importance of the CDEs and how to navigate the NINDS CDE website. Results: To date, over 18,000 CDEs and 1,400 case-report forms (CRFs) and instrument recommendations have been developed for 24 diseases. Recent updates include revised Epilepsy and Cerebral Palsy CDEs and CRFs, and a comprehensive update to the Headache CDEs, resulting in version 2.0. The NINDS CDEs or the NINDS CDE website have been mentioned and/or cited in over 187 publications since 2011. Conclusions: The NINDS CDEs are an evolving resource that is updated as research progresses. These CDE updates are important as they help maintain their usability in clinical research, which, in turn, simplifies data sharing and harmonization, data mining, and data warehousing. Disclosure: Dr. Sheikh has nothing to disclose. Dr. Edun has nothing to disclose. Dr. Feldman has nothing to disclose. Dr. Gay has nothing to disclose. Dr. Johnson has nothing to disclose. Dr. Joseph has nothing to disclose. Dr. Esterlitz has nothing to disclose. Dr. Mendoza-Puccini has nothing to disclose.