Give more data, awareness and control to individual citizens, and they will help COVID-19 containment

Mirco Nanni, Gennady L. Andrienko, Albert-László Barabási, Chiara Boldrini, Francesco Bonchi, Ciro Cattuto, Francesca Chiaromonte, Giovanni Comandé, Marco Conti, Mark Coté, Frank Dignum, Virginia Dignum, Josep Domingo-Ferrer, Paolo Ferragina, Fosca Giannotti, Riccardo Guidotti, Dirk Helbing, Kimmo Kaski, János Kertész, Sune Lehmann, Bruno Lepri, Paul Lukowicz, Stan Matwin, David Megías, Anna Monreale, Katharina Morik, Nuria Oliver, Andrea Passarella, Andrea Passerini, Dino Pedreschi, Alex Pentland, Fabio Pianesi, Francesca Pratesi, Salvatore Rinzivillo, Salvatore Ruggieri, Arno Siebes, Vicenç Torra, Roberto Trasarti, Jeroen van den Hoven, Alessandro Vespignani

Abstract

During the COVID-19 pandemic, the digital response was largely divided between centralized surveillance and decentralized proximity-only tracing. This paper proposes a third way: the Personal Data Store (PDS) model. By shifting data ownership to the individual, the PDS architecture allows for the collection of rich spatio-temporal data necessary for epidemiological analysis without compromising privacy. This approach addresses the limitations of proximity-based systems—such as the inability to detect environmental transmission—while fostering the public trust essential for high adoption rates.

Introduction

The global response to COVID-19 highlighted a fundamental tension in digital health: the need for detailed mobility data to track a virus versus the individual's right to privacy. Early digital contact tracing efforts generally fell into two camps. Centralized systems offered high utility for health authorities but posed significant surveillance risks. Conversely, decentralized proximity-tracing protocols, such as the Google-Apple Exposure Notification (GAEN) framework, prioritized privacy but lacked the spatial context required to identify infection hotspots or environmental transmission.

This position paper argues that this is a false dichotomy. By leveraging a Personal Data Store (PDS) architecture, society can empower citizens to become active participants in pandemic containment, providing the necessary data for public health while maintaining strict individual control.

The Limitations of Proximity-Only Systems

While decentralized proximity-tracing was a significant step forward for privacy, it introduced two critical gaps in epidemiological tracking:

  1. Environmental Transmission: Proximity-only systems (typically using Bluetooth Low Energy) only detect direct contact between users. They fail to account for "surface-touch" contamination (fomites) in shared spaces like supermarkets or public transit, where an infected individual and a susceptible individual may occupy the same space at different times.
  2. Lack of Spatial Context: Without location data (GPS), health authorities cannot identify geographic "hotspots." Understanding where outbreaks are brewing is essential for "Phase 2" strategies—the safe re-opening of society.

The challenge lies in collecting this rich spatio-temporal data without creating a "Big Brother" infrastructure that citizens will ultimately reject.

The Personal Data Store (PDS) Architecture

The authors advocate for a paradigm shift where data stays on the user’s device by default. The PDS model is built upon several key technical components:

Personal Software Environment

A secure local or cloud-based vault serves as the repository for a user’s raw location and proximity data. Analysis is performed privately within this environment rather than on a central server.

Granular Sharing and Control

Unlike binary "opt-in" systems, the PDS allows users to choose the level of detail they share. For instance, a user might share precise GPS coordinates with a health authority only upon a positive test result, while contributing neighborhood-level aggregates to a general research pool.

Modalities of Data Sharing

The proposal outlines two primary ways to utilize this data:

  • Simple Transfer: Upon a positive diagnosis, the user voluntarily sends the minimum necessary data (e.g., a list of visited locations) to a trusted health authority.
  • Distributed Computation: Using Secure Multi-Party Computation (SMPC), global statistics—such as population density maps or hotspot identification—can be calculated without any individual ever revealing their raw data to a third party.

Trust as a Functional Requirement

A central thesis of the paper is that privacy is not merely a legal hurdle but a functional requirement for public health. If citizens do not trust the digital infrastructure, adoption rates will remain low, rendering the collected data scientifically insufficient. By implementing "Privacy by Design"—including the "Use Limitation Principle" where data has a finite lifespan and "auto-destructs" after the emergency—the PDS model builds the necessary trust for mass participation.

Furthermore, the PDS model introduces "self-awareness." By providing users with tools to visualize their own risk—such as an app showing they frequently pass through high-risk zones—the system can "nudge" individuals toward safer behavior. This voluntary behavioral change is often more sustainable and effective than top-down mandates.

Conclusion and Future Roadmap

The paper concludes with a two-track recommendation for implementing this vision. In the short term, existing proximity-tracing apps should be extended to allow for local location logging and voluntary aggregate sharing. In the long term, there must be a concerted investment in PDS infrastructure to transform personal data from a commodity exploited by platforms into a collective good managed by individuals.

While challenges remain—including the technical maturity of SMPC and the need for international data standards—the PDS model offers a path forward that aligns public health goals with democratic values.

Summary updated

BibTeX
@article{nanni2021give,
  author = {Mirco Nanni and Gennady L. Andrienko and Albert-László Barabási and Chiara Boldrini and Francesco Bonchi and Ciro Cattuto and Francesca Chiaromonte and Giovanni Comandé and Marco Conti and Mark Coté and Frank Dignum and Virginia Dignum and Josep Domingo-Ferrer and Paolo Ferragina and Fosca Giannotti and Riccardo Guidotti and Dirk Helbing and Kimmo Kaski and János Kertész and Sune Lehmann and Bruno Lepri and Paul Lukowicz and Stan Matwin and David Megías and Anna Monreale and Katharina Morik and Nuria Oliver and Andrea Passarella and Andrea Passerini and Dino Pedreschi and Alex Pentland and Fabio Pianesi and Francesca Pratesi and Salvatore Rinzivillo and Salvatore Ruggieri and Arno Siebes and Vicenç Torra and Roberto Trasarti and Jeroen van den Hoven and Alessandro Vespignani},
  title = {{Give more data, awareness and control to individual citizens, and they will help COVID-19 containment}},
  journal = {Ethics Inf. Technol.},
  year = {2021},
  doi = {10.1007/s10676-020-09572-w},
}
Figure