Technology may be part of the solution to stopping the spread of COVID-19, but apps alone will not save us. As more states develop COVID exposure notification apps, institutions and the people they serve should remain skeptical and remember the bigger picture. This is still experimental, unproven technology, both in terms of how it works under the hood and how humans will interact with it. And even the best-designed app will be no substitute for public health basics like widespread testing and interview-based contact tracing.
On top of that, any benefits of this technology will be unevenly distributed. Any app-based or smartphone-based solution will systematically miss the groups least likely to have a cellphone and more at risk of COVID-19 and in need of resources: in the United States, that includes elderly people, people without housing, and those living in rural communities.
Ultimately, exposure notification technology won’t bail out poor planning or replace inadequate public health infrastructure, but it could misdirect resources and instill a false sense of safety.
Exposure notification apps, most notably those built on top of Apple and Google’s Exposure Notification API, promise to notify a smart phone user if they have been in prolonged close contact—for instance, within 6 feet for at least 15 minutes—with someone who has tested positive for the virus. The apps use smartphones’ Bluetooth functionality (not location data) to sense how far away other phones are, and store random identifiers on the user’s device.
But Bluetooth was not made to assist with contact tracing and other public health efforts, and the differing hardware properties of various phones can make it hard to consistently measure distances accurately.
On top of technical shortcomings, it is also not yet clear how people will interact with the technology itself. How are people likely to react to a phone notification informing them that they were exposed to someone with COVID? Will they ignore it? Will they self-isolate? If they seek testing, will it be available to them? Public trust is fragile. A high rate of false negatives (or a perception thereof) could lead to people relaxing measures like social distancing and masking, while false positives could lead to users ignoring notifications.
Unevenly Distributed Benefits
Any app that promises to track, trace, or notify COVID cases will disproportionately miss wide swaths of the population. Not everyone has a mobile phone. Even fewer own a smartphone, and even fewer still have an iPhone or Android running the most up-to-date operating system. Some people own multiple phones to use for different purposes, while others might share a phone with a family or household. Smartphones do not equate to individuals, and public health authorities cannot make critical decisions—for example, about where to allocate resources or about who gets tested or vaccinated—based on smartphone app data.
In the U.S., smartphone ownership is only 80% to begin with. And the communities least likely to have a smartphone in the U.S.—such as elderly people or homeless people—are also the ones at higher risk for COVID-19. For people 65 years or older, for example, the rate of smartphone ownership declines to about 50%. Out-of-date smartphone hardware or software can also make it harder to install and use a COVID tracking app. For Android users in particular, many older phone models stop getting updates at some point, and some phones run versions of Android that simply don’t get updates.
And smartphone penetration data and specs do not tell the whole story, in the U.S. or internationally. Overbroad surveillance hits marginalized communities the hardest, just as COVID-19 has. In addition to potentially directing public health resources away from those who need them most, new data collection systems could exacerbate existing surveillance and targeting of marginalized groups.
Technology Will Not Be a Magic Bullet
Even with these drawbacks, some will still ask, “So what if it’s not perfect? Anything that can help fight COVID-19 is good. Even if the benefit is small, what’s the harm?” But approaching exposure notification apps and other COVID-related tracking technology as a magic bullet risks diverting resources away from more important things like widespread testing, contact tracing, and isolation support. The presence of potentially helpful technology does not change the need for these fundamentals.
Relying on unproven, experimental technology can also lead to a false sense of safety, leading to a moral hazard for institutions like universities that have big incentives to reopen: even if they do not have, for example, enough contact tracers to reopen, they might move ahead anyway and rely on an app to make up for it.
If and when public health officials conclude that spread of COVID-19 is low enough to resume normal activities, robust interview-based contact tracing is in place, and testing is available with prompt results, exposure notification apps may have a role to play. Until then, relying on this untested technology as a fundamental pillar of public health response would be a mistake.