General
What is i2b2?
i2b2 is a self-service "cohort discovery tool" that allows you to explore and query clinical data that has been deidentified and aggregated. This web application sits on top of the Carolina Data Warehouse for Health (CDW-H) and creates a safe-harbor "sandbox" for you to explore CDW-H data.
Why would a researcher use i2b2?
i2b2 allows you to identify cohort counts on your own. For example, if you are applying for a grant and need to know how many patients are available within UNC Health Care, you can input your exclusion and inclusion criteria into i2b2 to receive a cohort count. You can then test the effects of adjusting those criteria. Once you identify your cohort, assuming regulatory approval, you can work with a CDW-H analyst to get additional information about those patients.
How can I find identifiers about the patients I found when querying i2b2?
Information about requesting data from the CDW-H can be found on the CDW-H FAQ page. After reviewing this page, you can begin the process by submitting a CDW-H Project Request form by the last Thursday of the month for consideration at the next committee meeting (held on the first Thursday of each month). It is usually helpful to request a consult with an analyst before submitting a project request, to get help with feasibility, scope, and time and cost estimates.
What does i2b2 stand for?
i2b2 stands for "Informatics for Integrating Biology & the Bedside." The term actually refers to the i2b2 Center, an NIH Funded National Center for Biomedical Computing based at Partners HealthCare System. The i2b2 software is a tool originally created by this group.
I experienced an error when working in i2b2. Who should I report this to?
Please send an email to the i2b2 Team at This email address is being protected from spambots. You need JavaScript enabled to view it.. Try to include a detailed description of what you were doing right before the error and the error you experienced.
How do I gain access to i2b2?
To use i2b2, you must have a valid UNC-Chapel Hill ONYEN and attend the i2b2 training session. Trainings are offered regularly throughout the year. Details can be found at tracs.unc.edu/i2b2.
How much does it cost to use i2b2?
There is no fee associated with using i2b2. It is sponsored by the NC TraCS Institute, the home of the University of North Carolina's CTSA. We do encourage investigators that use i2b2 to help pilot or power their study to cite i2b2 and the CTSA in their applications and/or publications. (More information on how to cite the CTSA and i2b2.)
I have an idea for a plugin/project/dataset for i2b2, who should I contact?
If you have an interesting application of i2b2 outside of the functionality provided in the master instance of i2b2 powered by the CDW-H, contact the i2b2 team at This email address is being protected from spambots. You need JavaScript enabled to view it. with your idea.
Data
How many patients' data are available in i2b2?
As of summer 2016, there are over 5 million unique patients in i2b2. Of these, over 1 million are active patients. Active patients have had at least one inpatient, emergency, or face-to-face outpatient visit and one diagnosis within the last three years.
Are data from Epic available in i2b2?
Yes, i2b2 contains Epic data from all of the UNC Health Care system facilities currently using Epic.
Are data from WebCIS (UNC Hospital's legacy EHR) available in i2b2?
Yes, legacy data from WebCIS, used by UNC Medical Center, is available in i2b2.
Where can I find detailed information about the data in i2b2?
To find detailed descriptions about the data, including where the data originate and tips for working with the data, please review the data dictionary located under the Help button in i2b2.
Are all patients in i2b2 alive?
No, i2b2 includes both alive and deceased patients. You will use the Patient Demographic: Vital Status if you only want to find patients that are alive or deceased. You can also obtain a breakdown by Vital Status when you run your query.
How is protected health information (PHI) protected?
When i2b2 was created, specific decisions were made to protect PHI. This includes displaying results as aggregate counts only and not displaying results less than 10.
Can I query by the patient's zip code?
No, you cannot query by patient zip code because this is considered an identifier under HIPAA. For a complete list of identifiers as defined by HIPAA visit: hhs.gov
According to HIPAA, aren't dates identifiers?
Yes, dates of birth and dates of service are considered identifiers under HIPAA. To reduce the potential leak of this information, dates are never exposed within i2b2 and are not accessible via the web interface. Additional functions have been implemented, including custom obfuscation thresholds, to help further limit potential access to these identifiers. No medical record numbers, social security numbers, or other identifiers are accessible to i2b2.
I cannot find an item I was expecting to find in the ontology (e.g. department, code, etc).
If you cannot find an item you were expecting, please email us at This email address is being protected from spambots. You need JavaScript enabled to view it.. An item may not be included due to an error or because it is outside of the scope of i2b2. Additionally, our team can help guide you to the correct place to find what you need. In rare cases, the team may be able to build a specific concept to fit your needs or it may be included in a future update to the UNC i2b2 ontology.
Functionality
Where can I learn more about how to construct queries and use i2b2?
To learn more about using i2b2, please review the "i2b2 General Help" document which can be found under the Help button when you are logged in to i2b2.
If I query to find the number of patients with a diagnosis and a patient had that diagnosis more than once, will they be counted more than once in my result?
No, when querying in i2b2 for number of patients, the results you receive will be the number of unique patients.
Can I add additional groups to the Query Tool?
Yes. To add additional groups, click the New Group button at the bottom right corner of the query tool. Note that all visible groups must have a term in them before you can add a new group.
Can I use Internet Explorer?
We encourage people to NOT use Internet Explorer, as some functionality is not supported. Mozilla Firefox and Google Chrome are the preferred web browsers to use with i2b2.
My query took so long that it timed out and got moved to the background. What does that mean?
If a query is taking a long time, i2b2 runs it "in the background" so you can do other things, such as start another query. When your original query is done running, the query title and results will show up in the "Previous Queries" window. To see the results, drag the folder icon onto "Query Name" bar in the "Query Tool". Your query, and the results from that query will then appear in the "Query Builder" and the "Query Status" windows.
Can I export data from i2b2?
The ability to export data is not available in the UNC instance of i2b2. If you need specific data, please contact a CDW-H analyst by submitting a CDW-H Project Request.
If I can't export data, then what is the "Query Report" button?
Query Report gives you an export of the query construction but not the actual data returned from your query. This information may be requested by a CDW-H analyst if you submit a project request based on your i2b2 queries.
