13-10-2020 | | By Liam Critchey
The Sars-Cov-2 coronavirus pandemic, aka COVID-19, has been going on for a long time now. Despite not knowing much about it in the beginning, a lot of research has been done this year to understand better what it is, how it infects and transmits, and what can be done to protect the spread of the virus. Among social distancing measures and mask-wearing, one other frontline preventative initiative that has been utilised around the world has been to use our mobile phones to collect data.
This comes in many forms, usually in the form of an app (sometimes mandatory depending on your country of residence) that monitors who has the virus—if they have tested positive—and informs anyone if they have been near a known-carrier. This has then enabled people close to virus carriers to get tested, self-isolate, or both, depending on the availability of tests and the results.
This way of disease prevention is relatively new and is only possible thanks to the advances in smartphones and big data capabilities in the present day. So, because these technologies were not suitable previously, it is a relatively novel approach. Because of this, it has been privy to issues such as the apps not fully functioning at times.
However, it is a monitoring approach, and other preventative methods do need to be taken at the same time, it’s just a way of monitoring the current situation and offers a way of reducing the spread of the disease by identifying who has it and where. Obviously, this also requires human input to work properly as it requires people to take necessary and advised precautions. Nevertheless, mobile phones have become useful non-pharmaceutical intervention (NPI) tool to help combat the spread of the virus.
There have been many different ways in which mobile phones have collected data from our phones during the pandemic. Several different methods are used because each approach has its own benefits and limitations, so utilising many methods simultaneously helps to reduce these limitations and provide a more accurate picture at the local, regional, and global levels.
One of the most common approaches around the world has been the use of opt-in (or mandatory opt-in) application-based data using track and trace style apps. Other methods of locating phones and who they belong to have included call data records, GPS location data, and the passive collection of Bluetooth data. Many of these approaches include a timestamp, a GPS location and a unique identifier for each phone (which can be used to identify the owner of the phone) enabling all the places and times that people are in a location to be monitored and cross-analysed with those who are also in the area at the same time but have tested positive for the virus.
Like any data-driven method, some data is captured while other data isn’t. Because the monitoring involves humans, and human behaviours are important in this instance, several different factors can’t be quantified or taken into consideration. But the data that can be obtained can provide some useful, and in many cases crucial, insights into how the virus is spreading around each country and what can be done to mitigate the risk and reduce the spread.
Some of the key data that is obtained by the methods outlined above include changes in population-level mobility and clustering in local areas, determining the rates at which individuals move between locations, the chance of transmission between locations, the hourly and daily movements of each mobile user, and the relationship between an individual’s behaviour and infection status.
However, capturing data this way struggles to assess some of the more qualitative factors. These include changes in individual behaviour and trajectories, differences in how each person uses their phone, the relationship between an individual’s behaviour and their infection status, fine-scale clustering and contact data, not understanding whether a close-by individual is in direct contact or not in contact with the user, and in understanding whether interactions not in the direct vicinity of the user may play a part in the transmission of the virus. Nevertheless, the data captured enables larger-scale policies and interventions to be implemented where applicable, so a lot of the data, while localised, is used regionally and/or nationally.
While a lot of the focus on track and trace style apps for the coronavirus seems to be geared to each individual, the overall picture is much bigger. While it does have an impact at the local level, as it helps people to isolate and protect themselves and others, the data obtained across each country enables the key decision makers to make informed decisions regarding the spread and locale of the virus at different time points. Essentially, the use of mobiles (as a lot of people the world has one) allows up-to-date and real-time decisions to be made with a much higher degree of certainty depending on the situation.
One of the ways in which the data is used at a regional level is to assess whether any changes in the population mobility are required by taking into account the associated risks within each region and to provide a tailored approach to each region’s needs. The use of real-time data is also beneficial for pinpointing where local outbreaks started (and working on ways it doesn’t happen again) and for understanding if and how any movement between areas has contributed to localised outbreaks. This also helps better to calculate projections of disease risk in various locations.
Finally, and the way that many people know about, the use of these methods (particularly the apps) enable the government to inform each person whether they should start quarantining or not. Overall, while there are gaps in the data, the use of mobile phones has become a useful tool for assessing the needs and risks at both a local and national level within each country, and the hope is that NPI initiatives such as this will help to curb the spread of the virus and bring infection rates down.