StopCovid: Have Contact Tracing Apps Already Proven Their Effectiveness?
The CNIL gave the green light on Tuesday for the launch of the StopCovid contact tracing app. Subject to Parliament’s vote, it could be available this weekend. But has such a tool already really proven its effectiveness?
Viral spread too fast to be contained by manual contact tracing?
A group of researchers from the University of Oxford looked into the question and published his conclusions in the review Science end of March. In particular, it relied on data from the cases identified in China to estimate the proportion of transmissions according to the different modes of contamination. From that, he made predictions to see how setting up a solution to quickly trace the people with whom an infected individual has been in contact can be a game-changer and help stem the current pandemic. .
The group recognizes that their data is imperfect because analysis from a more precise cohort would be preferable. Work that will surely be carried out after the epidemic. But in the allotted time and with the data he has, he believes that in the case of Covid-19, “the viral spread is too fast to be contained by manual contact tracing“. “But it could be controlled if this process were faster, more efficient, and occurred on a large scale.He continues. For them, the solution could be found in the implementation of a digital solution. “A delay between the confirmation of a case and the tracing of its contacts is not inevitable, however. More precisely, it can be avoided by using a mobile phone application ”, concludes the group of researchers.
But the device imagined by these researchers goes much further than Bluetooth
Except that the optimal device thought by the researchers for such an application to be effective goes much further than the Bluetooth solution proposed by the government, and seems difficult to apply in France.
According to them, the ideal solution would thus work via GPS co-location supplemented by the scan of a QR code displayed on public equipment with heavy traffic where the GPS would be too imprecise. The results of the tests would be communicated to a server that would automatically recommend stratified quarantine measures based on the physical proximity people had to the infected one, while preserving its anonymity. Tests could be requested via the application. As the diagram below shows, the idea then would be to recommend actions that range from quarantine to simple social distancing measures.
This provides a vision of what an optimal solution would look like according to these researchers. And it is not the image of what will be implemented in France. But what about apps that already exist?
What should we learn from the Singaporean example?
An example of a contact tracing app using Bluetooth – and which benefits from a little perspective to analyze its impact – is the TraceTogether app used in Singapore. It warns people who have been within 2 meters of someone for at least 30 minutes if the latter turns out to be positive for Covid-19 in the following weeks.
The city-state launched its app on March 20. As of April 1, it claimed 1 million downloads. Almost two months later, the proportion has increased a little further and the local government is now talking about 1.5 million users, out of a population of nearly 5.7 million inhabitants. A proportion which remains too low.
Thus, in an interview with the local newspaper The Strait Times At the beginning of April, Lawrence Wong, the Minister of National Development, admitted himself that the tool had not yet kept all its promises, due to the insufficient number of users. “For TraceTogether to be effective, we need roughly three-quarters – if not everyone – of the population to own it. Then we can really use it as an effective contact tracing tool ”, he had declared. It is therefore not possible for the moment to use the example of the country to invalidate or confirm the effectiveness of a contact tracing app. A point that those responsible for the app themselves wanted to remind.
An app that must be accepted by 60% of French people to be effective
The Singaporean example shows that issues of trust in particular are crucial since a contact tracing application cannot be really effective without massive adoption by the population. According to specialists, this proportion must be at least 60% to hope to stem the spread of the virus. Has the French population been informed enough? Was the app creation process transparent enough?
Another personal point of view is also interesting to share. This one from Jason Bay, senior director (government digital services) for GovTech Singapore and product manager of the TraceTogether app. Faced with the desire of other countries to launch contact tracing apps, he wanted to share his experience. According to him, one point tends to be forgotten: “ We use TraceTogether to supplement manual contact tracing – not replace it“, he insists. Thus, according to him, such an app is of no use if it is not accompanied in addition to a continuous and intense work of contact tracing, of data interpretation. Human intervention therefore remains essential.
It will therefore also be necessary to see how the French government places the StopCovid device within the contact tracing procedures that already exist, if the various protagonists will know how to coordinate and work together. This is in addition to the challenges already mentioned above, including a massive adoption of the solution by the population, which will depend in particular on the level of confidence that they place in it.