Meeting the Challenges of Secondary Use of Patient Data

Obtaining a full picture of a patient often involves several health-related datasets in addition to insights into the aspects of a patient’s life outside of their health records.  Over time, patients move, change jobs, change insurance coverage, and even change names.  Unfortunately, the more information that is added to a deidentified patient record, the greater the risk of reidentification.  Based on all of these challenges, what techniques can be used to create an effective dataset that protects a patient’s privacy and provides the essential information to answer the specific research questions?  In this talk, we will explore our approach to the creation of a referential patient token that allows more accurate tracking across life events, our methods for using this token to combine various real-world data assets, and our approach to expert determination to ensure that the dataset produced protects a patient’s privacy while still meeting the needs of the researcher.