Privacy-Preserving Statistical Data Analysis on Federated Databases

The quality of empirical statistical studies is tightly related to the quality and amount of source data available. However, it is often hard to collect data from several sources due to privacy requirements or a lack of trust. In this paper, we propose a novel way to combine secure multi-party computation technology with federated database systems to preserve privacy in statistical studies that combine and analyse data from multiple databases. We describe an implementation on two real-world platforms—the Sharemind secure multi-party computation and the X-Road database federation platform. Our solution enables the privacy-preserving linking and analysis of databases belonging to different institutions. Indeed, a preliminary analysis from the Estonian Data Protection Inspectorate suggests that the correct implementation of our solution ensures that no personally identifiable information is processed in such studies. Therefore, our proposed solution can potentially reduce the costs of conducting statistical studies on shared data.