As the world fights to stop the spread of the new coronavirus, data on infection rates, deaths, and even distances between people is increasingly sought after. The collection of such data has exacerbated privacy concerns, but social science researchers and citizen scientists are working to find solutions.
On March 25th, Swiss newspaper Le Temps reported that telecommunications provider Swisscom would alert federal authorities if more than 20 mobile phones were detected in the same 100 square-meter public space (a violation of social distancing measures).
Swisscom stated that users could opt out of the data collection, and would not be identifiable from the data. However, dhCenter member Jessica Pidoux emphasizes the importance of weighing the consequences of fast technological solutions in proportion to the immediate public health response from governments.
“We should ask ourselves: is the data being collected really useful, and to whom? Could it be crowdsourced? Are we excluding certain groups like the elderly, or those without smartphones?” says Pidoux, a doctoral researcher in the Digital Humanities Institute (DHI) in EPFL’s College of Humanities.
Ethics in an emergency
Many experts believe that data collection for public health – even during a pandemic – can and should be done in a way that respects people’s privacy. For example, on April 1st, EPFL announced its contribution to the development of a Bluetooth-based protocol as part of the Pan-European Privacy-Preserving Proximity Tracing (PEPP-PT) Initiative. The collaborators aim to create a platform for digital proximity tracing that does not require geolocation data, and that complies with the General Data protection Regulation (GDPR). Users would download a smartphone app that would send alerts to those who had been in contact with patients diagnosed with COVID-19, eliminating the need to quarantine the entire population.
Anna Jobin, a researcher at ETH Zurich’s Health Ethics & Policy Lab, says that while it is a definite advantage to develop new technologies specifically tailored to the challenges at hand, it’s crucial to also consider ethical issues that go beyond data privacy.
“Important questions arise not only regarding the development, but also the deployment and the governance of a technology. For example: is using an app truly voluntary? If not, where does the constraint come from – from the government, from your employer? There could be constraints on many different levels, and many nuanced ethical challenges relating to who gets to make these decisions,” Jobin says.
She adds that in addition to privacy, data-based technologies must also consider other social aspects. Trust, for instance, is key when it comes to public health.
“This is another can of worms. It’s not just that proposed measures should be necessary and proportional; it’s crucial that people actually trust that they are, in order to implement and follow them,” she says.
Adding a qualitative perspective
Jessica Pidoux believes that interdisciplinary research is key to identifying the pros and cons technological solutions to the pandemic. She is the co-organizer of the hackathon HackCOVID-Léman, which took place on March 28th, and aimed to develop solutions for collecting personal data to help communities “beat COVID-19 faster, at scale, with consent”. The event was launched with PersonalData.io, within the framework of the MyData Global initiative. With the help of wikis, Google documents and Zoom, some 50 participants from around the world came together virtually to address safe, informed, and ethical data-gathering for public health.
“Our major contribution was introducing a sociological perspective to these issues,” says Pidoux, who worked in the hackathon’s data anthropology and ethics group. Using an online questionnaire, she and her fellow hackathoners characterized lifestyle factors related to coronavirus infection and spread, with the goal of better understanding who is at risk and how to protect them. The team also addressed ethical issues related to such data collection and analysis – for example consent, equitable representation of different groups, and the potential stigmatization of at-risk categories.
Pidoux – who along with her teammates went on to model the lifestyle data at another Swiss hackathon, Versus Virus – says that these kinds of studies could help data science researchers better understand how to use data ethically for public health. She adds that as a digital humanities researcher, she is also interested in combining qualitative and quantitative studies to build more inclusive solutions.
“We want to build something from and for communities; sometimes, public health interventions that are not understood or accepted are counterproductive.”
- Summary of HackCOVID-Léman on Jessica Pidoux’s blog
- YouTube video of a talk on the anthropology of COVID-19 by Miles Fahlman at HackCOVID-Léman
- ETH Zurich Chair of Bioethics lecture on public trust in healthcare systems
- Nature comment on the responsible use of digital data to tackle the COVID-19 pandemic