A data recipient centered de-identification method to retain statistical attributes

Date

2014-01-10

Department

Program

Citation of Original Publication

Gal, Tamas S. et al.; A data recipient centered de-identification method to retain statistical attributes; Journal of Biomedical Informatics, Volume 50, Pages 32-45, 10 January 2014; https://doi.org/10.1016/j.jbi.2014.01.001

Rights

This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.

Abstract

Privacy has always been a great concern of patients and medical service providers. As a result of the recent advances in information technology and the government’s push for the use of Electronic Health Record (EHR) systems, a large amount of medical data is collected and stored electronically. This data needs to be made available for analysis but at the same time patient privacy has to be protected through de-identification. Although biomedical researchers often describe their research plans when they request anonymized data, most existing anonymization methods do not use this information when de-identifying the data. As a result, the anonymized data may not be useful for the planned research project. This paper proposes a data recipient centered approach to tailor the de-identification method based on input from the recipient of the data. We demonstrate our approach through an anonymization project for biomedical researchers with specific goals to improve the utility of the anonymized data for statistical models used for their research project. The selected algorithm improves a privacy protection method called Condensation by Aggarwal et al. Our methods were tested and validated on real cancer surveillance data provided by the Kentucky Cancer Registry.