Patient anonymity is of critical importance; besides complying with HIPAA regulations, patient privacy is crucial to ensuring effective communication between physician and patient and preventing discrimination. Guaranteeing true patient anonymity in medical narratives, such as emergency room notes, is difficult due to the density of potentially identifiable information within each report as well as the sheer volume of reports themselves. Clinical researchers must either obtain written permission from their patient, which is often unfeasible, or de-identify their reports. A method to quickly and reliably de-identify patient information within medical documents is critical to the smooth operation and success of clinical trials and medical care around the world.
Description
De-ID is just such a system; inputting an ascii file containing free-text medical reportsyieldsa file that contains a de-identified version of the report. De-ID is used at the University of Pittsburgh and other research universities across the country to remove identifying information such as patient name, address, or relevant dates. Recent additions to the software have improved functionality by allowing for batch processing of files, allowing the program to be called by another, and improved heuristics for finding and removing the protected health information. Additional user-controlled features, such as customizing the amount, direction, and units of date offset, have improved usability for researchers. De-ID represents a simple, effective way to anonymize enormous sets of patient data, guarantee HIPAA compliance, and protect patients’ privacy.
Applications
· De-identifying patient data within free-text medical reports to ensure HIPAA compliance, protect patient privacy, encourage patient honesty, and allow clinical researchers to gather accurate data
Advantages
· Uses external lists, such as names drawn from US census data, and template matching to guarantee a high accuracy rate; allows user to input additional lists that correspond to relevant names or geographical features
· Substitutes standard labels, such as
, to indicate type of deletion made
· Transforms all dates in a report to preserve temporal relationship to each other
· Additional customization to accommodate specific regulatory requirements
· Provides for different file output types
· Can be accessed through other applications
· Can be applied to very large data sets
· Privacy guarantee allows patients to safely disclose sensitive informationIP Status
Software
