Peer-to-Peer Contact Tracing: Development of a Privacy-Preserving Smartphone App

Abstract

Background: The novel coronavirus disease 2019 (COVID-19) pandemic is an urgent public health crisis, with epidemiologic models predicting severe consequences, including high death rates, if the virus is permitted to run its course without any intervention or response. Contact tracing using smartphone technology is a powerful tool that may be employed to limit disease transmission during an epidemic or pandemic; yet, contact tracing apps present significant privacy concerns regarding the collection of personal data such as location.

Objective: The aim of this study is to develop an effective contact tracing smartphone app that respects user privacy by not collecting location information or other personal data.

Methods: We propose the use of an anonymized graph of interpersonal interactions to conduct a novel form of contact tracing and have developed a proof-of-concept smartphone app that implements this approach. Additionally, we developed a computer simulation model that demonstrates the impact of our proposal on epidemic or pandemic outbreak trajectories across multiple rates of adoption.

Results: Our proof-of-concept smartphone app allows users to create "checkpoints" for contact tracing, check their risk level based on their past interactions, and anonymously self-report a positive status to their peer network. Our simulation results suggest that higher adoption rates of such an app may result in a better controlled epidemic or pandemic outbreak.

Conclusions: Our proposed smartphone-based contact tracing method presents a novel solution that preserves privacy while demonstrating the potential to suppress an epidemic or pandemic outbreak. This app could potentially be applied to the current COVID-19 pandemic as well as other epidemics or pandemics in the future to achieve a middle ground between drastic isolation measures and unmitigated disease spread.

Keywords: COVID-19; contact tracing; coronavirus; epidemic; mobile phone; pandemic; peer-to-peer; personal data; privacy; smartphone.

Figure 1

A network of interactions over time represented as a transmission graph. The rows represent units of time, and the columns represent individuals. By time t=3, all individuals have contact points with possible transmission paths. t: time point.

Figure 2

User flow diagrams for the three primary user flows in the peer-to-peer contact tracing app: creating checkpoints, checking risk, and reporting positive status. QR: Quick Response.

Figure 3

Comparison of infection curves from simulations at varying rates of peer-to-peer contact tracing application adoption. The proportion of the population with active infection is plotted across time for multiple adoption rates. Time is an arbitrary unit that represents the sequence of events in the simulation. The results of 10 random simulations per adoption rate are given.

Figure 4

Disease spread scenario modeled as a transmission graph. (A) Graphical representation of a disease spread scenario across time. Contact points with infected individuals are denoted as exposures. Uninfected individuals may become infected at exposure points according to some probability (the transmission rate); hence, b does not become infected at n6. (B) Transmission graph corresponding to the scenario in A, depicting the information that is available to the server and each individual’s smartphone app. Only one node, n3, is associated with a reported diagnosis. The infection risk level at the other contact points can be inferred by checking for possible transmission paths. n: node; t: time point.