As a Product Designer, I’m used to viewing the problems of the world optimistically. Leverage good design principles, and expect to create a good product that enables positive outcomes. So, when the need to contact trace COVID-19 became apparent, I wondered what an app for this would look like — and it almost immediately became clear to me how difficult it would be to implement a viable app-based contact tracing system at any scale.
The idea behind a contact tracing app is simple — when a patient tests positive for the virus, the app works backwards and identifies the people who the patient has been in direct and indirect contact with recently, as these other people may have been exposed themselves. This is done repeatedly, creating a network of “safe” and “exposed” people. Google and Apple recently announced a joint partnership to build a large-scale contact tracing app. Governments are in on this race too — Singapore and South Korea have apps live right now, and the UK has one set to launch within days. This begs the question — did they ever stop to think about if these apps would actually work? What these organizations fail to recognize is that for once, an app will not solve their problems.
Problem 1 — Not Enough Tests
The biggest threat to the viability of a contact tracing app is the current testing landscape surrounding the virus. If the purpose of giving the public an app like this is to give communities an idea of how “safe” they are, the testing shortage here in the United States would make this near impossible. We’re not conducting anywhere near as many tests as we would need to to get a good estimate of the “safeness” of any location. The state of New York has conducted 1.3 million COVID-19 tests — roughly 15% of the state’s population — and that’s the highest of any state in the nation. My state of Texas, on the other hand, has only run 630k tests — in a state of 29 million, that’s not even 3% of the population. If we’re really trying to let people know which locations and communities around them are safe and which aren’t, how can any service provide accurate information if an overwhelming majority of people aren’t getting tested?
The risks that come with a lack of community access to testing are amplified when you introduce asymptomatic carriers — especially considering the ease with which asymptomatic carriers can spread the virus when they don’t feel sick themselves. These asymptomatic carriers are already far less likely to get tested, given testing availability across the country is in such short supply. Mismanagement at the federal level has lead to many communities only having a handful (at best) of tests to give — and these tend to go right to the people who show visible symptoms of the virus. Dr. Robert Redfield, director of the CDC, told NPR in March that as many as 25% of people with COVID-19 may be asymptomatic — and since then, studies have popped up suggesting this proportion may be even higher. One study conducted in Iceland reported an asymptomatic carrier rate of 50%. These asymptomatic carriers propagate the spread of the virus in their communities unknowingly and would easily create gaps in contact tracing networks. It would be much easier to track the virus if all carriers were guaranteed to have felt sick themselves — but this isn’t the case. There’s no way to tell whether someone who hasn’t felt sick has had the virus besides administering a proper test. This is precisely where the lack of widespread testing availability works in conjunction with the presence of asymptomatic carriers to make reliable, community-wide contact tracing virtually impossible.
Problem 2 — Not Enough Privacy
Let’s imagine, if only for a moment, we have the ability to test everyone in the country within a few weeks. Everyone can get tested, and we can begin to track the spread of the virus. Would every American eagerly give up the information required to monitor the spread? Who they are, where they’ve been, and who they’ve been in contact with? I doubt it. Across party, racial, and generational lines, only 15–25% of Americans trust the government. Big tech has trust issues too - only 69% of people trust Apple and Google with their information. No matter where a supposed app were to come from, it would be met with staunch resistance from skeptics.
This poses a massive challenge to the efficacy of continuous contact tracing. Again, the system would only be accurate if an overwhelming majority of people were allowing it to monitor their information. Picture Miguel, an undocumented immigrant living in a small apartment in southern Texas with six other family members. His income is the only thing keeping his family fed. He’s heard plenty about the ways in which undocumented and low-income communities are disproportionately affected by the virus. Having said that, can you imagine Miguel feeling comfortable giving away information on his location to the same government that’s actively criminalizing him? This is one example, but as situations like Miguel’s appear across the country, widespread adoption of the app could easily be derailed.
Problem 3 — Not Enough Adoption
It’s hard to get people to form a new behavior — even when the behavior in question is clearly beneficial. In 2019 the NHTSA reported that 90.7% of Americans regularly used a seatbelt — and it took us four decades to get to that level. Despite the obvious life-saving benefits a seatbelt has to offer and the little effort it takes to use, millions of Americans still, to this day, opt not to use one. The flu shot — which also takes relatively low effort and can prevent weeks of serious illness — was only administered to 45% of Americans last year. It’s hard enough to get people to adopt a new behavior, but when one considers the systemic barriers to widespread adoption it gets even harder.
This brings us to the other half of this adoption challenge — the millions of Americans without access to smartphones. 62 Million Americans (about 19%) don’t have one. In the realm of contact tracing, if 1 in every 5 people are “ghosts” to the system, the ability to effectively track the proliferation of the virus plummets. Smartphone ownership is even lower in low-income communities, which again are disproportionately affected by this virus. This disparity reminds us that if a contact tracing app were to be available, the lack of access to accurate information in poorer communities would, unsurprisingly, put them at increased risk.
We’ve already seen communities make attempts at implementing an app like this. The Singaporean government released a contact tracing app TraceTogether which is being used by roughly a quarter of the nation’s population. While these adoption numbers are impressive, Singapore is still seeing a spike in new cases, suggesting that their app may not be doing much to effectively combat the spread of the virus. Other attempts at tracking people’s location and movements have been taken on by governments and companies — some find it helpful, and others are skeptical about whether the data they collect is representative of reality or not. And herein lies the fundamental danger of offering a subpar contact tracing solution: without nearly enough data going into whatever app makes it to market first, people may see low case numbers in their communities and get a false sense of security. And by the time they realize their mistake, it may very well be too late.
With the run of their iPhone 3G commercial in 2009, Apple changed the public’s perception of what a phone is capable of, announcing to everyone that “There’s An App For That” for just about anything. This became the clarion call for our digital age, seeing the likes of Uber, Lyft, Airbnb, Instagram, and Snapchat rise to prominence in the time to follow. Here we are today at the onset of a new decade, in a world crippled by this virus, with companies and nations alike scrambling to monitor the spread in their communities. Apple and Google will likely put millions into building out a contact tracing app. Perhaps this money could be better spent elsewhere — there are still plenty of hospitals in need of key supplies. Globally, one billion children are out of school. Vulnerable communities are more at-risk than ever. Perhaps… there isn’t an app for that.