Statistical Methods for Re-Identification

Sunday, 16 February 2014
Regency C (Hyatt Regency Chicago)
Xiao Hua Andrew Zhou , University of Washington , Seattle, WA
Risk of Linking HIPAA De-Identified Rheumatoid Arthritis Research Dataset with CMS Data
Xiao-Hua Andrew Zhou PhD

Professor, Department of Biostatisics,  University of Washington

Director of Biostatistics Unit,  VA Seattle Medical Center

In this presentation, we will discuss statistical methods for estimating the re-identification risk.

We will discuss both individual and global measures for risks of re-identification.  A global measure

provides a single risk value for the entire micro data file to be released whereas  an individual measure

gives  re-identification risk at level of the individual record. We will illustrate the application of  these methods in a VA Rheumatoid Arthritis (VARA) Registry data.  VARA data was de-identified according to HIPAA “Safe Harbor” standards.