R+D CSIC

Conversation with Frederic Bartumeus about a project to study the spread of Covid-19

We speak with Frederic Bartumeus, researcher at CEAB-CSIC and CREAF, who is co-leading a project to study the expansion of Covid-19 by geolocation with Javier Ramasco, from the Institute for Cross-Disciplinary Physics and Complex Systems (CSIC and the Universidad de las Islas Baleares). The team studies the mobility of the population during confinement to obtain data on the risk of transmission.

• What is the goal of the project?

In the coming months, government will take measures for the gradual end of population's lockdown. Mobility in the territory will progressively return and contact between people will increase. To understand how the dynamics of the spread of the Covid-19 disease evolves, it is necessary to understand how mobility will reappear in the territory by social sectors (ages, gender, type of work). It is also important to monitor other micro-changes that will occur in our social behavior, according to the de-escalation measures that are taken.

• What is the relevance of tracking for Covid-19 for research?

Understanding mobility in the territory and changes in relation to the degree of social distancing as a population throughout the year are key to assessing the risk of new outbreaks in different parts of the territory. This may allow adjusting certain rules of conduct, such as minimising certain types of mobility and risky behaviors, in an increasingly complex situation, in the search to gradually return to normal. For this, we need high-resolution data to propose scenarios of gradual de-escalation by regions, by age or by work sectors, rather than predict what will happen, which is very complex.

• What timeline are you planning in order to have the app ready?

We have been working for a month with mobile phone data that we are integrating into an epidemiological model. This will allow us to see the connection between mobility and the appearance of new cases, with which we will be able to present the different scenarios we were discussing. We will explore different modeling frameworks from mechanistic models, which are based on fundamental epidemiological theory, to models oriented to a Big Data perspective and machine learning. Likewise, on 15th May we launched a first survey on mobility and social contacts, with which we try to better understand what our “new normal” really means. What it means in terms of behavior and social contacts, how many daily contacts we have with other people, who are they and how those contacts are produced, etc. The data will allow us to better evaluate possible risk scenarios and, from the CSIC, we hope to be able to propose action measures to the Government or, at least, determine which ones should be avoided.

In June we will launch a mobile application for the study of mobility and social behavior that allows us to carry out more surveys (via mobile) or receive geolocated information regarding types of social contacts that users want to share voluntarily. It will also give users the possibility to share data on their movements. This data is collected in a file using their mobile's GPS, voluntarily and anonymously for the project. If they are positive, they can share these files with health workers who are monitoring cases and epidemiological control. In addition, the application will serve to give useful information to citizens and help prevent certain risky behaviors. These microdata of social behavior and mobility can be very important to bring possible future scenarios to our everyday reality.

• Would it conflict with data privacy?

High-resolution geolocation (GPS) is the only personal data that we would collect. Mobile phone data is transferred to us in aggregate and anonymous form. We do not collect any personal data in surveys. In the mobile application, the only data that is considered personal is GPS. We would obtain this data anonymously and it would be transmitted in encrypted form to a secure CSIC server. Right now, we are preparing all the necessary documentation to pass the appropriate ethical and data protection filters and thus be able to count on citizen participation to obtain them. We can assure that we will never be able to identify, even if we wanted to, people who want to contribute their data from the app. We will also not be able to associate your responses or trajectories with any type of personal identifier. Our team already has experience in the development of mobile applications, which we have used for research in contexts of human mobility and mosquito-borne diseases, and which have already passed filters of this type.

• Given the state of alarm, how are you working and collaborating?

All the team is working remotely, since in our case the tasks involve the intensive use of the internet and high-performance computing, which we can access from our homes. The family contexts are different, the groups that have come together had never before worked together, we are getting to know each other... but the truth is that we have been organising, defining the tasks better, and now we have weekly meetings to see the progress of the different lines of research. After a somewhat chaotic start, we are now much more coordinated.

• Prev
• Next