White Paper

Reducing Re-Identification Risk in Health Data

COVID-19’s spread affected every corner of the country, from densely-populated cities to small rural towns. Each new exposure represented an individual record for officials to trace, test and treat — and a new set of highly sensitive data at high risk of privacy breach or leakage. With millions of records to track, the task of protecting each individual’s information while preserving the data’s usability is a delicate, yet critical balance.

Data privacy techniques, while thorough and advanced, are often not sufficient to ensure privacy on their own. In today’s broad and evolving data landscape, singular approaches can’t guarantee protection against re-identification, particularly in the healthcare industry. A multi-faceted approach can better protect against potential threats — and it doesn’t have to be labor-intensive for data analysts.

In this guide we’ll explore how to reduce re-identification risk in health data by implementing three privacy enhancing techniques:

  1. k-Anonymization
  2. Randomized Response
  3. Sampling

Download the guide to learn more about the methodology behind each technique, and why they are most effective when executed in tandem.