Using smartphones to follow volunteers’ behaviour in an epidemiology study
An interview published in Ellipse, the monthly magazine of the PRBB.
Audrey de Nazelle arrived here three years ago from the University of North Carolina Chapel Hill where she had done a PhD in Environmental Science. This French mathematician, an ecological activist from a young age, has always understood the importance of politics in public health.
She is currently doing a postdoc in Mark Nieuwenhuijsen’s group at CREAL, and together with him is leading the TAPAS project, which started in 2009 and which will run until 2012.
Who is participating in TAPAS and what is its aim?
We are the coordinators, but there are research groups from six European countries involved. We want to understand the health impact of policies on active transport, such as walking or cycling, and we are using quantitative models to measure this impact in six cities: Barcelona, Basel, Copenhagen, Paris, Prague, and Warsaw. To do this we use both the existing literature and local data.
What are these models for?
The idea is that one can choose a specific policy of active transport, for example the introduction of a network of bike lanes or a congestion charge such as there is in London, and predict the effects on health that come about because of this. We hope that the models will serve as a tool to assist policy makers with their decisions.
What kind of effects are we talking about?
Basically, an increase in active transport means an increase in the physical activity of the population. In addition, a reduction of motorised modes of transport means less gases emitted and therefore an improvement in air quality and climate change mitigation. But our studies do not ignore the possible adverse effects such as an increase in accidents or a greater inhalation of contaminants by those who walk or cycle.
What studies have you done so far?
We made a first theoretical model of the “Bicing” system and undertook experimental studies that allowed us to increase our knowledge. These included a survey of 800 people to understand their behaviour with respect to active transport. We supplemented the survey with a pilot study of 35 volunteers to obtain more objective measurements. For five days they carried around devices which gathered data about their location with a GPS, as well as their level of physical activity using a piece of built-in smartphone software.
What is this kind of information useful for?
We can, for example, analyse the types of routes that people use and what their choice of means of transport depends on. We can also compare their routes with pollution maps and analyse their exposure.
So smartphones are the future of exposure analysis?
I think so, yes. Both this technology and other developments, such as remote sensing, which are part of what is known as “ubiquitious sensing”. As far as we know, we were the first people to use this smartphone technology in a real study with volunteers. But if we think that there are billions of people who use mobile phones, and that they could all become data collection volunteers, the potential is enormous. In addition to what we have already done we could add noise or pollution sensors, for example, with everything pinpointed to its exact location.