COVID-19 Contact Trace Apps: A Pivot to Reimagine Digital Public Health

Updated: May 1

Written by: Sonia Navani, DrPH

Quick Take


The pandemic's beginning was filled with the promise of scaled up digital contact tracing efforts. But, a year later, the world has learned yet another hard tech lesson: digital health in emergencies means little without high integration into broader public health responses.


The tech world has an exceptional amount to offer in the pandemic and well beyond. This is precisely why it is that much more critical to create greater scientific accountability in digital health tools. At its core, it requires distinguishing between consumers and communities at risk in complex health emergencies. We outline two starting points based on the COVID-19 experience, including rethinking place-- from consumer to vector terms and earning back public trust.

Scoping the Digital COVID-19 Contact Trace Problem


Over 50 contact trace digital apps have been deployed to respond to the pandemic worldwide. In the process, a raging privacy debate around a 'creeping' surveillance state [1] has taken center stage. But the other side of this coin is the question of public health value-added, given that many have failed to demonstrate serious contributions to infection control.


Digital health metrics related to user reach and preferences (e.g., number of downloads) are standard. But, digital health metrics to assess public health value, as a part of an infection control strategy, fall into a unicorn category. Simulated data via modeling has provided much of the basis for heavy investments into digital contact trace apps, such as the $70 million price tag associated with the Australian COVIDSafe app. [2].


Where more performance data is available related to infection control, COVID-19 contact trace apps report relatively low effectiveness rates. For instance, Australia's COVID Safe mobile app had identified 17 new cases out of a total of 7 million app users over a five-month period (May to Oct 2020).[5]


In India, the mandatory app, Aarogya Setu, was reported to have a positivity rate of 27% attributed to the app (Sept 2020). But, when examined alongside the infection positivity rate based on COVID-19 testing (8%)[6], it would suggest artificial inflation of numbers attributed to the mobile app effectiveness.


Other dimensions of effectiveness, such as usage levels, are also barriers. For instance, in Singapore, high use rates- over a fifth of the population- in mid-2020 still translated into only a 4% chance of any two people both having the app on their phones [7].

Finally, it's noteworthy that effectiveness continues to suffer from basic technical errors. Japan[7] and Denmark[8] have reported gaps in exposure alerts or incorrect exposure notifications to users. A systematic review of ~30,000 user reviews for COVID-19 contact trace apps identified technical malfunctions and battery draining as the most prevalent concerns among users [9]. These errors reflect temporary investments unable to maintain the basic rigor of 'pop-up' systems in a protracted complex health emergency.


A less than stellar performance of COVID-19 digital apps can be attributed to several factors but revolve heavily around two key model assumptions that have not held true in the COVID-19 pandemic. First, infectiousness correlated with symptom onset [3], enabling timely reporting. Second, high integration with other public health countermeasures to control disease dynamics as testing.[4] Instead of dynamic models to follow and re-calibrate according to incoming scientific evidence, digital health COVID-19 contact trace apps stayed static. Thus, it's not that digital contact trace apps didn't get it right from day one but rather that many stayed static, offering little to no value for infection control on a large scale.


Rethinking Place: Consumer Tracking vs. Cluster Busting


The vast majority of COVID-19 contact tracing digital apps use a combination of Bluetooth and/or location data to track exposures-- frequently deployed in individual-level consumer tracking.[10] While Bluetooth offers relational data in terms of the proximity of one cell phone to another, location data provides an absolute location of the cell phone user.


At the same time, a handful of countries such as New Zealand, Thailand, and Singapore use venue-based QR code check-in/out systems in public locations such as malls, public transport, grocery stores, among other places. These QR venue-based systems are used less in marketing domains. However, during COVID-19, venue-based QR code systems have been strategically deployed alongside Bluetooth/location-based apps during the pandemic.


Venue-based tracking, analogous to 'cluster busting,' offers a particular advantage in COVID-19. Early pandemic assumptions related to transmission focused on person-to-person or person-to-surface contact. In these scenarios, location is the highest order term. However, extensive evidence indicates the virus is primarily spread via aerosols, where the highest order term in the physical environment is airflow. This certainly doesn't mean that location data has no value. But, instead of it having an absolute value for identifying potential exposure incidents, it becomes much more relational to other variables in the environment, particularly indoor environments. Given this, an optimal strategy has been to re-calibrate location tracking devise a digital 'cluster busting' tool.


If integrated with robust privacy safeguards, it also has the advantage of not requiring users to be constantly under surveillance when in lower-risk scenarios (e.g., outdoor meetings, where regular testing is conducted) and in private spaces. While it is undoubtedly still a form of surveillance, we note a distinction between an individual's movements being tracked based on voluntary movement in public venues versus more comprehensive tracking across personal and public spaces in a public health emergency.

Earning Back Public Trust: Creating Non-Consumer Models of Digital Public Health


Current iterations of digital health are primarily economically driven and designed based on user buy-in and experience. In contrast, digital public health tools intend to be value-based, designed to complement larger conventional public health systems, and earn public trust.


In effect, current digital health iterations frequently reflect an overlay or 'dropping in' of public health ideas and values into the slots of a private industry framework. The pandemic has confirmed, on a global scale, this is a non-starter in complex health emergencies. We map out three immediate steps to re-position digital public health below.


a) Language: The term digital 'app' skews public perception, synonymous with big tech for many people. This is an increasingly important factor given the lack of trust in the technology domain regarding data privacy. If aligned with action to ensure high transparency and accountability, language becomes a fundamental building block to carving trusted public health in digital health spaces.


b) Integration: The current landscape elevates digital health as a different species of public health responses. But, the reality is that a digital public health impact is dependent on the degree of integration with other public health strategies in place. [12] In COVID-19, examples of integration with other public health countermeasures have included:

  • Test-Trace: Widespread COVID-19 testing and pre/asymptomatic screenings to short circuit delayed time to reporting into digital and human COVID contact trace systems (New Zealand, South Korea)

  • Tech-Human Contact Tracing: Integrating digital COVID response strategies directly to human contact trace teams (South Korea)

  • Tech-Public Health Data Surveillance: Digital COVID-19 contact trace data designed to go directly to public health data surveillance teams (Alberta, Canada AB Trace Together)

Beyond public health-technology collaborations, Taiwan's digital quarantine system provides a proof of concept to integrate institutional action. However, privacy remains an issue in its current iteration. The digital system links government financial support and police enforcement. This linkage created incentives and boundaries to support the COVID-19 digital response.[13]

c) Non-Smartphone Users: Given that innovation is often born out of necessity, it's no surprise that we are now starting to see some of the first serious attempts to expand digital contact trace systems beyond smartphone users in COVID-19. Singapore's Safe Entry system, launched in March 2021, is a notable attempt in this direction. The system disseminates tokens to non-smartphone users, providing a primary means to track movement in/out of venues via Bluetooth signals from 'gateway boxes' to tokens or phones. [14]


Unfortunately, a decreasing amount of public trust in Singapore's digital COVID response [15] likely means a spillover of distrust into the Safe Entry token program as well. Future iterations must also rethink privacy issues in the specific context of token or alternative non-smartphone strategies for vulnerable groups. In the case of Singapore, this includes migrant workers. In other settings, it includes women and girls and traditionally less affluent segments of the community who may be mobile phone users but not smartphone users or owners.


Moving Forward: COVID-19 and the Next Complex Health Emergency


Whether examining digital health via an economic, technology, or public health lens, the common goal is to create tools that will offer more signals and less noise in health emergencies. This is increasingly important in a market, expected to rise to $640 billion by 2026, and a global environment poised for more infectious disease epidemics and complex health emergencies due to climate change. The pandemic presents the ultimate synergistic moment across health and technology domains to reboot for a post-COVID-19 era of digital public health models grounded in scientific accountability.





References


[1] Han, K. 5 Jan 2021. 'Surveillance in the name of COVID-19.' Rest of World.

[2] Clun, R. 26 Oct 2020. "COVIDSafe app identifies just 17 contacts not found by contact tracers" The Sydney Morning Herald.

[3] Nakamoto, I., Wang, S., Guo, Y., & Zhuang, W. (2020). A QR Code–Based Contact Tracing Framework for Sustainable Containment of COVID-19: Evaluation of an Approach to Assist the Return to Normal Activity. JMIR mHealth and uHealth, 8(9), e22321.

[4] Ferretti, L., Wymant, C., Kendall, M., Zhao, L., Nurtay, A., Abeler-Dörner, L., ... & Fraser, C. (2020). Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracing. Science, 368(6491).

[5]Clun, R. 26 Oct 2020. "COVIDSafe app identifies just 17 contacts not found by contact tracers" The Sydney Morning Herald.

[6]Agarwal, S. 17 Sept 2020. "Is Aarogya Setu app a game-changer in India's Covid fight? Six months on, opinion is divided!" The Economic Times.

[7] Nature Editorial. (2020). Show evidence that apps for COVID-19 contact-tracing are secure and effective. Nature, 580(7805), 563.

[8] Ando, K. 5 Feb 2021. "Japan's flawed COVID-19 tracing app is a digital black eye for Tokyo" Nikkei Asia.

[9] Elkhodr, M., Mubin, O., Iftikhar, Z., Masood, M., Alsinglawi, B., Shahid, S., & Alnajjar, F. (2021). Technology, Privacy, and User Opinions of COVID-19 Mobile Apps for Contact Tracing: Systematic Search and Content Analysis. Journal of Medical Internet Research, 23(2), e23467.

[10] Ryan-Mosley, T. 23 Dec 2020. MIT Technology Review COVID Tracing Tracker.

[9] Clun, R. 26 Oct 2020. "COVIDSafe app identifies just 17 contacts not found by contact tracers" The Sydney Morning Herald.

[11] Nakamoto, I., Wang, S., Guo, Y., & Zhuang, W. (2020). A QR Code–Based Contact Tracing Framework for Sustainable Containment of COVID-19: Evaluation of an Approach to Assist the Return to Normal Activity. JMIR mHealth and uHealth, 8(9), e22321.

[12] Ferretti, L., Wymant, C., Kendall, M., Zhao, L., Nurtay, A., Abeler-Dörner, L., ... & Fraser, C. (2020). Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracing. Science, 368(6491).

[13] Eigen, Wang, Gasser. 31 July 2020. Country Spotlight: Taiwan's Digital Quarantine System. Berkman Klein Center for Internet & Society.

[14] Chee, K. 17 Mar 2020. 'New SafeEntry system lets you tap phone or TraceTogether token on gateway box to check in.' The Strait Times.

[15] Han, K. 5 Jan 2021. 'Surveillance in the name of COVID-19.' Rest of World.

  • Facebook Clean Grey
  • Twitter Clean Grey