Reference: Trisalyn Nelson, Peter Kedron, Michael F. Goodchild, Stewart Fotheringham, Amy Frazier, Wenwen Li, Song Gao, Yingjie Hu, Ming-Hsiang Tsou, May Yuan, Bo Zhao (2020). Digital Contact Tracing and Surveillance: Geospatial opportunities, limitations, and research directions. ASU Spatial Analysis Research Center (SPARC) White Paper. pp 1-13.
As efforts to mitigate and suppress COVID-19 continue, many decision makers are asking if digital contact tracing—a method for determining contact between an infected individual and others using tracking systems commonly based on mobile devices—can help us safely transition from population-wide social distancing to targeted case-based interventions such as individualized self-quarantine. In response, the Spatial Analysis Research Center (SPARC) at Arizona State University organized a panel of national experts to discuss the use of geospatial technologies in digital contact tracing and identify the practical challenges researchers can address to make digital contact tracing as effective as possible.
The major themes of the discussion included (i) the capabilities and limitations of geospatial technology, (ii) privacy, and (iii) future research directions. Key takeaways from each of these areas include:
Capabilities and limitations of geospatial technology: There are many geospatial technologies (e.g., GPS, Bluetooth, Cellular, WiFi) embedded in mobile devices that can be leveraged for digital contact tracing. However, GPS technology in smartphones lacks accuracy to map interactions in the detailed way one might expect. For instance, the horizontal accuracy of GPS is 15m, and the vertical accuracy is insufficient to pick up which floor of a building a person is on. Indoor accuracy is particularly poor, which is problematic given people spend 87% of their time indoors. However, information about the absolute location of an individual may not be as important to digitally tracing epidemiologically meaningful contacts as identifying the types of interactions most likely to result in the spread of the virus. The importance of tracing interactions creates an opportunity to use Bluetooth-based exchange of encrypted keys to record person-to-person contacts that can then be analyzed within the space-time prism framework. This approach will not require storing of all individuals’ movement data, which will reduce computation complexity. Geotargeted and geotagged social media are useful for tracking transmission between cities or within cities, detecting large gatherings, and helping individuals recall location and contact history during contact tracing interviews. Social media can also provide useful context, such as check-in locations and textual content, to reduce false positives in interactions identified through other forms of digital contact tracing.
Privacy: Digital contact tracing raises numerous privacy concerns. By creating some record of the location history or contacts of an individual, digital contact tracing creates an opportunity to identify an individual without their consent. At present, the privacy implications of digital contact tracing are unclear because these systems have yet to be fully developed or deployed in the US. An evaluation of pros and cons in the existing digital contact tracing plans operating in other countries can inform policy makers on privacy mediation during and after contact tracing. While companies and officials working on this issue have made statements that preserving privacy is an important goal, the details of how privacy will be preserved and the safeguards that will be put in place are not yet available. If any privacy protections are lifted to enable contact tracing, a plan should be put in place to restore protections once the pandemic subsides.
Future Research: To support digital contact tracing and surveillance, several research areas must be advanced. Key technical areas include increasing the accuracy of indoor positioning, developing approaches for reducing false positive of potential exposure (not to be confused with false negatives which are more common in COVID-19 diagnostic test) ensuring a focus on high accuracy in relative positioning, addressing computational complexities, developing group or bubble based approaches to surveillance, and developing a system for the creation and distribution of high resolution risk data and to enable self-determination of the need of quarantine and testing based on possible exposure. Research into how digital contact tracing systems link with existing contact tracing infrastructure and with other digital contact tracing systems also needs to be conducted. The implications of digital contact tracing for society and privacy will emerge along with these systems. Researchers need to study these issues as they emerge to ensure that we have the ability to hold an informed public debate about the effectiveness and costs of digital contact tracing.