Our research at the Sax Institute in Sydney, published this week in the peer-reviewed journal Public Health and Practice, used computer modelling to assess how many people would need to use the app to limit the spread of any second wave. The model took account of the behaviour of the virus, whom it affects and how its spread is affected by social distancing and testing rates. Taking five hypothetical scenarios regarding how quickly (or not) social distancing and testing declined, we projected the contribution that different levels of app uptake would make (from 27 per cent through to 80 per cent) to limiting numbers of new cases.
The first scenario envisaged a continuation of the situation at the end of May. We estimated social distancing was declining by 50 per cent a month alongside a 5 per cent monthly decline in testing. The other four scenarios were variations of these two factors.
Notably, all our scenarios indicated that a second wave of COVID-19 was likely at some point. But more importantly, we were surprised by how much a benefit the app could bring. In the first scenario – the closest to what we think was happening in late May – we found that if 61 per cent of Australians (the likely maximum in practice) used the app, there would be less than half as many new COVID-19 cases (55 per cent fewer) compared with the expected outcome if no app existed. Only if social distancing declined more quickly would the app contribute to a larger reduction in new cases (59 per cent). If social distancing and/or testing declined more slowly, there would be smaller reductions in new cases, ranging from 13 per cent to 49 per cent.
While social distancing and testing remain the two most effective ways of limiting the spread of the virus and detecting outbreaks quickly, this finding shows the app has the potential to play a vital supporting role in identifying additional cases.
Only about 27 per cent of Australians have the app, which would only cut the number of new cases by about one-quarter (24 per cent) – a worthwhile decrease, but one that would have a much weaker effect in slowing a second wave and is far less than that attainable through wider app use.
As with any modelling, we made certain assumptions. We were deliberately agnostic about what specific app was used and we assumed that the app works as intended by successfully recording encounters. Official testing suggests that while app performance is improving, it is not yet perfect and encounter logging may not occur flawlessly between every combination of phone model. But from the modelling, the scale of the benefits that the app can make to containing COVID-19 indicates that investment in efforts to fix remaining technical issues will be well worth it. So will promotional efforts to encourage further community uptake.
Compared with many other countries, Australia continues to be in a very fortunate position with regard to COVID-19. International experience shows what can happen when inadequate control measures allow a highly contagious virus like SARS-CoV-2 to spread. A second wave, in which the app will have an opportunity to prove itself, appears likely. We should not miss the opportunity to avail ourselves of the protection that an effective app is likely to bring.
Dr Danielle Currie is a senior simulation modeller at the Sax Institute. Dr Michael Frommer is a public health physician and a senior adviser at the Sax Institute. Their research was published this week in the journal Public Health Research and Practice.