This workshop will provide data privacy/security professionals and legal counsel with an introduction to the principles and methods of statistical disclosure limitation that can be used to de-identify healthcare data to meet the HIPAA privacy regulations while assuring that both data privacy and statistical/analytic accuracy are appropriately protected and balanced. Participants will learn the basics of statistical disclosure risk analysis (primarily for healthcare microdata): data intrusion scenarios, the importance of both sample and population uniqueness, record linkage methods, formulations of re-identification risks, k-anonymity, differential privacy and other de-identification approaches, the definition of quasi-identifiers and the significance of their classification, and the appropriate use of HIPAA limited data sets. Participants will also learn about the types of disclosure analyses, including equivalence class analyses; geography analyses; and family key analyses. Upon completion of the workshop, participants will be able to work more successfully with statistical disclosure experts to understand and manage statistical de-identification for data sets under the HIPAA requirements while preserving analytic utility.
Daniel Barth-Jones, Professor, Columbia University, Mailman School of Public Health