Trust and Digital Contact Tracing: Initial Insights from the Swiss Proximity Tracing System

in collaboration with Melyssa Eigen, J.D. Candidate at Harvard Law School

This essay is part of a collection written by members of the Berkman Klein Center’s Working Group on Digital Pandemic Response. The group, made up of experts from academia, civil society, the public sector, and industry, takes on difficult questions around the use of digital tools and data to help attenuate the COVID-19 pandemic. Each essay is the perspective of the author, not of the Berkman Klein Center.

“Trust” has emerged as a key enabler when it comes to the effectiveness of different forms of pandemic responses, including the efficacy of digital public health technologies such as tech-assisted contact tracing. Unfortunately, both anecdotal and to some extent empirical evidence currently suggest that it’s easier to identify use cases that are characterized by a lack of trust as compared to experiences where digital pandemic responses are taking place in a trusting environment. One example of a country in the latter category is Switzerland, where the SwissCovid App provided by the Swiss Federal Public Health Office rolled out this week.

Before focusing on the trust-enabling and trust-preserving mechanisms involved in the development of the Swiss Proximity Tracing System, a brief overview of the technical and organizational setup of the SwissCovid App is needed: The use of the app is opt-in, meaning that the government is not requiring people to download it. It is based on the Decentralized Privacy-Preserving Proximity Tracing (DP3T) concept, which has been developed by ETHZ and EPFL, two Swiss universities. It runs on Bluetooth Low Energy rather than GPS, meaning that user location is not tracked, and uses the Apple and Google API. When a user’s phone detects another phone within 1.5 meters for a period of 15 minutes or longer, the phones exchange encrypted IDs through the application and record the contact. Details of these contacts are only stored locally on the user’s own mobile phone, not on a central server, and deleted after 21 days or when the app is uninstalled.

If a SwissCovid App user receives a confirmation by a laboratory that they have been infected, the cantonal medical service — typically acting through a member of the cantonal contact tracing unit, a physician, or the cantonal medical officer — will provide the user with a 12-digit code that remains valid for 24 hours and that the user can voluntarily enter into the app. Once entered, the authorization code is validated by a server hosted by the Federal government, which also records, aggregates, and encrypts the keys generated by the person’s app during the time the person was contagious. The user’s proximity contact information and the comparison of proximity data are computed exclusively in a decentralized manner on the users’ phones (more on the technical details here). Once the code is validated, all users who have been in proximity of the infected person will be notified through the app about the exposure. It’s up to the person who was notified about their potential exposure to decide what to do next. The app provides a phone number to a hotline to get further information anonymously and includes basic recommendations about symptoms and voluntary quarantine.

In terms of the different and complex trust dynamics at work, it’s noteworthy that the Swiss population — including a large immigrant population — is generally characterized by a high level of trust in the political system and government institutions, as various surveys conducted over many years have suggested. More specifically, most recent representative surveys suggest that a vast majority (66%) of the population age 15 and older expresses high or very high trust in the political leadership — the Swiss Federal Council — with respect to the management of the COVID-19 pandemic. Other surveys also demonstrate that the Swiss population trusts to a great extent the information provided by the Swiss Federal Public Health Office concerning COVID-19.

The development and deployment of the SwissCovid App has been taking place in this overall positive trust-climate, as a survey by leading researchers confirmed before the app was rolled out: When asked about their future willingness to use a tracking app, a majority (54%) expressed willingness to use the app when provided by the Swiss Federal Council. In contrast, the trust-levels drop dramatically when asked about the use of a proximity tracing app provided by tech companies (6%). What is interesting from a trust perspective is that the Swiss government has both earned and confirmed this trust through a series of measures in context of the design and development of the Swiss Proximity Tracing System. While more research is needed and underway, three clusters of trust-enabling and trust-ensuring measures seem particularly noteworthy from a comparative perspective.

• Integrating public health, science, and technology perspectives: The SwissCovid App, and the model upon which it is based, is the result of a cross-sector collaboration involving the two leading technical universities in Switzerland, the Swiss Federal Public Health Office, and a specialized Swiss technology company. This integrated approach has also been made visible throughout the consultation, development, and piloting process, with a leading digital epidemiologist from the EPFL, Marcel Salathé, who serves as a government-appointed member of the multidisciplinary COVID-19 Science Task Force that advises the Swiss Federal Government, as an embodiment of the combined public health and technical expertise.
• Transparency, continued public engagement, and feedback mechanisms (incl. piloting): Another key trust-mechanism that can be observed when looking at the SwissCovid App as a case study is the role of transparency, which spans across various layers from the technical to policy layers, including an open source approach to the technology development, the public security tests of the app during the pilot phase, continued explanation and detailed, multi-lingual public documentation of the relevant decisions and experiences (including security vulnerabilities) across all phases of development, and in-depth real-time surveys concerning population’s attitudes towards the app.
• Development of robust legal framework and safeguards: While legal experts have argued that the existing emergency powers set forth in the Swiss Epidemic Act might likely have provided a sufficient statutory basis for the development and roll-out of the SwissCovid App, the Swiss government — at least in part probably also in response to a robust public debate among experts — not only enacted a regulation governing pilot phase of the project but also decided to create a dedicated legal foundation for the Swiss Proximity Tracing System enacted by the Swiss Parliament in form of an amendment to the Epidemic Act. Furthermore, the entire development process has been subject to in-depth reviews by the Federal Data Protection and Information Commissioner.

From a trust perspective, these initial observations highlight a number of elements to consider. The experience with the Swiss Proximity Tracing System to date confirms that overall trust levels are likely to shape trust in — and ultimately the efficacy of — digital public health tools. That’s unfortunately bad news for countries where government institutions rank low on the trust scale. The better news is that governments can take action to build and enhance trust. The Swiss example suggests that such trust-enhancing mechanisms include strategic integration of public health and tech expertise, transparency and public engagement, and legal safeguards, including robust accountability mechanisms. These mechanisms must span across the entire lifecycle of digital public health technologies. The Swiss government has managed the design, development, and testing phases well from a trust angle. Hopefully, it continues to serve as a best practice case also in the implementation phase that is now to begin.