Disorders A - Z:   A    B   C    D    E    F    G    H    I    J    K    L    M    N    O    P    Q    R    S    T    U    V    W    X    Y    Z

Skip secondary menu

The International Common Data Elements Project – Getting Further, Faster


By Ramona Hicks, Ph.D., National Institutes of Health, National Institute of Neurological Disorders and Stroke

The brain is the most complex organ of the body and repairing it after injury is challenging. To address this challenge, research scientists are working to increase our understanding of brain function and to develop better diagnostic tools and treatments for the millions of people worldwide living with traumatic brain injury (TBI). The National Institute of Neurological Disorders and Stroke (NINDS) is the lead institute at the National Institutes of Health (NIH) for TBI research and supports a range of high quality research on this topic.

Understanding reasons for good and poor recoveries after TBI is an important step toward developing better treatments, and several years ago NINDS supported a study to address this question. The study was unusual in that it evaluated thousands of cases of severe TBI by combining data retrospectively from several clinical trials. Such large data sets were previously unavailable, and the study successfully identified several major factors associated with recovery. However, some research questions could not be fully addressed because the original studies did not include the data needed to answer them. It became clear that many important questions about recovery after TBI would need to be addressed in new studies. Moreover, investigators also discovered that retrospectively comparing data from multiple studies was excruciatingly difficult and time consuming because the same types of information were coded in different ways. For example, some studies reported a subject’s educational status as “none, grade school, high school, college, or graduate school” while other studies reported “total years of school.” Hundreds of these types of examples existed, and investigators spent years figuring out ways to make the studies comparable and then reformatting the data. There had to be a better way to combine information from multiple studies!

The better way is the International TBI Common Data Elements Project, a collaboration among the NIH, Department of Defense (DOD), National Institute on Disability and Rehabilitation Research, Department of Veterans Affairs, and Centers for Disease Control and Prevention to standardize definitions and protocols for TBI research. Rather than retrospectively reformatting and harmonizing data after a study ends, the goal of the International TBI Common Data Elements Project is to standardize data collection at the beginning. The project includes hundreds of scientific experts from around the world who serve on working groups and steering committees to develop recommendations for collecting data in a uniform manner. The standardized data recommendations are referred to as common data elements (CDEs).

Currently, more than 900 CDEs for TBI research (Hicks et al., 2013) exist, and are available on multiple websites, including the NINDS Common Data Elements website. A small number of the CDEs are so commonly used that they are called “CORE CDEs” and recommended for use in virtually all TBI research studies. Other CDEs are recommended for use depending on the type of study, such as epidemiology, acute-hospital, rehabilitation, or mild TBI/concussion. Some of the CDEs are relevant to subjects of all ages, and others are specific to children or adults.  

Following the development of the TBI CDEs, the next major step was to try them out to see if they could actually work in a research study conducted across multiple centers on a wide range of subjects. To give them a test-run, four TBI hospitals collaborated to collect data on more than 650 subjects with TBI, with injuries ranging from mild to severe, in a study called TRACK-TBI. The TRACK-TBI study revealed that the TBI CDEs had a few minor issues, but overall they were useable. In addition, the CDEs accelerated the research as the data were collected in just two years. Most of the data are still being analyzed; one early publication demonstrated that an MRI of the brain was more predictive of recovery outcomes at 3 months than a conventional CT scan (Yuh et al., 2013). Although previous smaller studies had shown similar findings, the large number of subjects in the TRACK-TBI study greatly increased the clinical significance and impact. 

Lastly, there is the question of how to put the TBI CDEs into practice. It’s one thing to develop CDEs and another thing to actually use them. Fortunately, there are two major activities underway to promote the use of the TBI CDEs and data sharing. One is the NIH and DOD collaboration to provide a Federal Interagency TBI Research (FITBIR) Informatics System. FITBIR provides a database for TBI research at no cost to the research scientists, as well as an electronic data capture tool, and a platform for collaboration and data sharing. FITBIR uses the TBI CDEs for its data dictionary. The second major activity is the International TBI Research (InTBIR) Initiative (Tosetti et al., 2013). InTBIR is a collaboration among the NIH, European Union Research Directorate and Canadian Institutes of Health Research to support international team science to study 10,000 children and adults with TBI. InTBIR researchers will use the TBI CDEs in their studies to facilitate data sharing and analysis toward the development of better diagnostic tools and more effective treatments.

In summary, the International TBI Common Data Elements Project built a foundation for the FITBIR Informatics System and the InTBIR Initiative, and together all of these projects will help us to get further, faster on understanding TBI and identifying ways to promote recovery.

Last updated August 5, 2013