i2b2 enables researchers to discover cohorts of patients using data from EHR systems. Through a point-and-click interface, researchers can build queries drawing from demographics, diagnoses, procedures, medications, and results recorded in Epic, the EHR system of Weill Cornell Physician Organization outpatient clinics (access to data from Eclipsys/Allscripts Sunrise Clinical Manager, the EHR system of NewYork-Presbyterian inpatient units, is planned). i2b2 supports different types of queries of clinical data, including whether clinical concepts occurred at any point in a patient’s medical history, during a particular visit, or in a sequence of events. Using de-identified data, investigators can determine potential cohorts of interest for later obtaining identified or limited data sets with Institutional Review Board (IRB) approval
Researchers can use i2b2 to answer questions like:
- “How many patients have a history of colon cancer using ICD-9 codes 153.* and 154.*?"
- “How many patients with colon cancer were younger than age 50 at the time of initial diagnosis?”
- “How many patients with colon cancer had a colectomy procedure performed and then were admitted to the hospital within 30 days?”
Submit a Cohort Discovery Request to get started.
i2b2 Training
Interested in learning how to use i2b2?
Online training is available in the LMS; just look for the "Introduction to i2b2" course. If you need help booking a course in the LMS, follow these instructions.
In-person workshops are held monthly in the Library Commons from 11 a.m. to noon. More info is available at arch-support@med.cornell.edu.