A Privacy-Protective Method for the Disclosure of Small Geographic Areas in Health Research: Data Aggregation Using Geometric Clustering

Knowledge of patients’ location information is critical for exercising spatial epidemiology in public health surveillance. Spatial epidemiology provides a spatial understanding of the population's health, and aids in studying the distribution of diseases in different areas (including disease outbreaks), as well as the environment's effect on health. A common patient residence location indicator is the postal code. However, the inclusion of such location information makes it easier to determine the identity of the individuals in the data sets. Specifically, patients living in small geographic areas (e.g., with small populations) tend to be more easily re-identifiable because they are more likely to be unique on their demographics. Because of this privacy risk it has been difficult to disclose small geographic areas in health data sets, limiting the ability of researchers to perform detailed geospatial analysis. In our current research we develop aggregation methods using geometric clustering to group small areas into larger ones, taking into account all of the fields in a data set. The aggregation would be just sufficient to ensure that the risk of re-identifying individuals is acceptably low. The aggregation will also maximize homogeneity on the underlying socioeconomic variables in the population (e.g., a high income area is not regrouped with a low income area). The benefit of such a method is that it will allow the disclosure of more detailed geographic information to the health research community while still protecting the privacy of Canadians.

Dr. Wei Shi is an assistant professor at the University of Ontario Institute of Technology. She is specialized in big data analytics, algorithm design and analysis in distributed environments such as Cloud of Clouds, Wireless Sensor Networks, Mobile Networks and Vehicular Networks. Dr. Shi is also an adjunct professor in the school of Computer Science at Carleton University. She holds a Bachelor of Computer Engineering from Harbin Institute of Technology in China and received her masters and Ph.D. of Computer Science degrees from Carleton University in Ottawa, Canada. Prior to her academic career, as a software engineer and project manager, she was closely involved in the design and development of a large-scale Electronic Information system for the distribution of welfare benefits in China, as well as of a World Wide Web Information Filtering System for China’s National Information Security Center. She has published in over 40 top tier conferences and journals such as IPDPS, EUROPAR, OPODIS, IEEE Transactions on Cloud Computing, IEEE Transactions on Mobile Computing and Elsvier's JPDC. Her research work is supported by IBM and Natural Sciences and Engineering Research Council of Canada (both Discovery Grant and Engage Grant programs).