An Overview of Techniques for De-Identifying Personal Health Information

Abstract

There is increasing demand for the use and disclosure of personal health information (PHI) for secondary purposes. In the context of health information, 'secondary purposes' is defined as any retrospective processing of existing data that is not part of providing care to the patient. For example, data used or disclosed for analysis, research, safety and quality measurement and improvement, public health, payment, provider certification and accreditation, and marketing are considered secondary purposes under this definition [1].

This report describes, at a high level, the techniques that can be used to de-identify PHI. In particular, our objective is to make clear when specific de-identification techniques are applicable and how to decide when to use each.

Keywords: de-identification, anonymization, privacy