Welcome i2b2: Our powerful new research tool
In this era of “big data,” spanning basic, translational, and clinical research, I am often asked how an investigator can search USF electronic medical records for potential associations or for feasibility analysis for clinical trials –without approval from the Institutional Review Board (IRB).
I am happy to announce that the i2b2 system is now fully implemented at USF Health and includes data from Allscripts and Epic for both USF and Tampa General Hospital patients. i2b2 is a warehousing and data mining tool for medical data that provides for complex Boolean searches of these records using de-identified patients. One can rapidly ascertain if there is an apparent overrepresentation of a disease or trait within a large cohort of patients with a given disease. This may serve as a hypothesis generating method, which could be tested at various levels using techniques in basic, translational, and clinical research.
i2b2 was written by investigators at Harvard Medical School, Massachusetts General Hospital, and Partners HealthCare System, with the first release in 2007. The user interface is intuitive and friendly. It is widely used throughout the country in academic medical centers not only for hypothesis generation and testing, but also for cohort identification. In the latter situation, once an investigator determines that a sufficient number of patients at USF/TGH fit enrollment criteria for a clinical trial, and after IRB approval, these patients can be identified, so that they can be contacted for potential participation. This should greatly enhance our ability to move forward investigator-initiated or industry sponsored trials.
The USF i2b2 system also represents a “real life” patient population, taking into account the diversity of the Tampa Bay population. Our i2b2 system currently has more than 1 million individual patients, over 10 million patient encounters, and more than 250 million data points. What a resource! For grant writing purposes, it is so much more convincing if you base your projects on data from a large population to which you truly have access.
A component of our i2b2 system can also connect to other systems at other academic institutions. This can be done via a consortium called TriNetX. Investigators can increase the power in their hypothesis generating or testing, ascertain geographical bias, and search for institutional partners for large trials.
Unlike the droid C-3PO, where we do not know whether the last character is an “O” or a “0”, and we don’t know what it stands for, we know a lot about i2b2. It stands for “Informatics for Integrating Biology and the Bedside.”
A number of individuals have already been trained and are actively using the system. I encourage investigators to take advantage of its capabilities. A link to training and gaining access will be provided on the MCOM Research website shortly.
Stephen Liggett, MD
Vice Dean for Research
Professor of Medicine, Molecular Pharmacology and Physiology
USF Health Morsani College of Medicine
Co-director, USF Health Heart Institute