Researchers in applied mathematics are developing a program that could be added onto the COVID alert app used on smartphones to better target vaccination campaigns.
The COVID alert app is based on the Google-Apple exposure notification API (GAEN API), a functionality that the tech companies rolled out in April 2020 in response to the emergent pandemic. The Canadian government built an app around the GAEN API, which became the COVID alert app, and managed the system for reporting and tracking positive cases. The COVID alert app tracks the contacts an individual has by virtually pinging the smartphones of those around them.
The innovation in the add-on the mathematics researchers developed is to use the app to target vaccinations, rather than simply to report cases. Instead of targeting specific geographies or demographic groups as a vaccine strategy, as was the case in Ontario’s vaccine rollout, the researchers propose a strategy based on individual contacts.
“Digital contact tracing apps can be a simple way to optimize vaccination strategies without people having to give up private information,” said Mark Penney, a postdoctoral researcher with the Department of Applied Mathematics at the University of Waterloo and an affiliate of the Perimeter Institute for Theoretical Physics.
He and a team of co-researchers recently released a study showing that the impact of such a strategy would be to almost half the number of vaccines required to bring the pandemic under control.
The new program was specifically designed with privacy in mind and keeps a record of only the number of contacts, not any location or other personal data. The add-on was purposefully made to be “a blunt tool,” said Penney.
“We don’t want to be able to identify who someone contacts or where they go,” he said. “All that we need to know is how many total contacts someone may have. The app keeps a cryptographic token, like a digital handshake, for any contact. It would then prioritize people who have the most contacts for vaccinations.”
The research project started at the onset of the pandemic and was in progress when there was a scarcity of vaccines in Canada. As the vaccine rollout began, the strategy used in Ontario was to prioritize vaccines to regions or demographics that were deemed most at risk. The new app takes that same strategy to another level by targeting not just hotspots or vulnerable populations but the people, for whatever reasons, most likely to be vectors for spreading the virus.
“The app might tell you, if you were an essential worker and coming into contact with lots of people, that you should get a vaccine,” said Penney. “If you’re a person working from home and not meeting many people, you probably wouldn’t get a notification to get vaccinated. The point is to target the people who are most likely to spread the disease.”
The results of the modelling and simulations for the new app were a resounding success, resulting in a massive increase in the effectiveness of a limited number of vaccines. However, Penney points out, “the version of the app we used in the model is idealized in the sense that only those selected would get the vaccine, and everyone the app selected ends up being vaccinated.”
In the current context of widespread vaccination and vaccine availability, the app is less urgently needed than it was at earlier pandemic stages. Still, Penney points out that the program can also be used for other infectious diseases. For example, it could be a tool to inform people if they should get a regular flu shot based on their contact patterns.
Penney also hopes the different contact tracing apps will continue to be developed and not shelved or forgotten once the pandemic recedes.
“The best way to ensure this kind of system is ready and in place is to find ways to use the tech in other applications. Then it can be easily repurposed the next time something like this comes around.”