ABSTRACT
The COVID-19 pandemic has killed millions and gravely disrupted the world's economy. A safe and effective vaccine was developed remarkably swiftly, but as of yet, uptake of the vaccine has been slow. This paper explores one potential explanation of delayed adoption of the vaccine, which is data privacy concerns. We explore two contrasting regulations that vary across U.S. states that have the potential to affect the perceived privacy risk associated with receiving a COVID-19 vaccine. The first regulation—an "identification requirement”—increases privacy concerns by requiring individuals to verify residency with government approved documentation. The second regulation—"anonymity protection”—reduces privacy concerns by allowing individuals to remove personally identifying information from state-operated immunization registry systems. We investigate the effects of these privacy-reducing and privacy-protecting regulations on U.S. state-level COVID-19 vaccination rates. Using a panel data set, we find that identification requirements decrease vaccine demand but that this negative effect is offset when individuals are able to remove information from an immunization registry. Our results remain consistent when controlling for CDC-defined barriers to vaccination, levels of misinformation, vaccine incentives, and states' phased distribution of vaccine supply. These findings yield significant theoretical and practical contributions for privacy policy and public health. © 2022 INFORMS.
ABSTRACT
Contact tracing is the process of identifying people who came into contact with an infected person ('case') and collecting information about these contacts. Contact tracing is an essential part of public health infrastructure and slows down the spread of infectious diseases. Existing contact tracing methods are extremely time and labor intensive due to their reliance on manually interviewing cases, contacts, and locations visited by cases. Additionally, complex privacy regulations mean that contact tracers must be extensively trained to avoid improper data sharing. App-based contact tracing, a proposed solution to these problems, has not been widely adopted by the general public due to privacy and security concerns. We develop a secure, semantically rich framework for automating the contact tracing process. This framework includes a novel, flexible ontology for contact tracing and is based on a semi-federated data-as-a-service architecture that automates contact tracing operations. Our framework supports security and privacy through situation-aware access control, where distributed query rewriting and semantic reasoning are used to automatically add situation based constraints to protect data. In this paper, we present our framework along with the validation of our system via common use cases extracted from CDC guidelines on COVID-19 contact tracing. © 2021 IEEE.