Digital Herd Immunity and COVID-19

Vir Bannerjee Bulchandani; Saumya Shivam; Sanjay Moudgalya; S L Sondhi

We now consider the practical applicability of our results to India, whose particular challenges provided the initial stimulus for this work. At first sight, India's overall smartphone coverage of around 40% [23] suggests that it might be difficult to attain "digital herd immunity" using solely app-based contact tracing [11] . However, smartphone-based measures could be combined with other tracing efforts to compensate for the limited availability of smartphones. Specifically, more traditional interview methods can still be used to reconstruct the location history of a newly diagnosed person. Systematic, random testing might also help in this regard [24] . As almost 90% [23] of Indians use some sort of wireless phone, alerts from people on the app-based network as well as interview-based alerts could be broadcast via SMS and combined with cell tower triangulation for spatial resolution in addition to the temporal resolution that would already be available. Major metropolitan areas, which include a number of COVID-19 hot-spots [25] , would be particularly well-suited to such a dual approach, since they exhibit both higher smartphone coverage and a higher density of cell-phone towers, allowing for more accurate triangulation [26] . Finally, it seems entirely possible to boost smartphone ownership for this purpose by the use of donated used phones and subsidized data provided to run the contact tracing app.