Using smart sensors to measure Dutch people’s health
Barry Schouten is Senior Methodologist at CBS and professor in Methods and Techniques at Utrecht University. In partnership with the university, he researches how new technology can be used in a smarter and more innovative way for the collection of data, leading to quality improvement and a reduction of the burden on respondents. Schouten: ‘Late in 2016, we investigated the possibilities for a survey which uses sensors to measure Dutch people’s health, lifestyle and living conditions. We then looked for a party who would be able to assess the relevance of this survey, which led us to the National Institute for Public Health and the Environment (RIVM). RIVM has experts who carry out research on the health of Dutch citizens.’ Both CBS and RIVM conduct health surveys among the Dutch population and work together to innovate these surveys. The Hague University of Applied Sciences develops portable sensors and measures, for example, environmental exposure and health effects among respondents.
RIVM health and environment researcher Annemarie Ruijsbroek explains that a lot of research within her organisation is performed on the health of the Dutch population. ‘These types of surveys are often held by way of questionnaires. However, there are new ways available now in which people collect their own data, for example pedometers. That is why we thought it would be interesting to be part of the Sensor Data Challenge organised by CBS, Utrecht University and The Hague University of Applied Sciences. According to Ruijsbroek, the environment division of RIVM has already performed similar experiments in the past, for example in measurement of air quality. is the application of survey sensors is still in its infancy within her own division. ‘When CBS asked for our view on this, it was therefore the perfect opportunity to join in.’
‘We developed an app which makes it very easy to see what your daily food intake is and what the ingredients are’
Non-stop for 24 hours
The results of the Data Challenge surpassed all expectations, but preparations took up a lot of time and energy, according to Schouten and Ruijsbroek. ‘Our budget was limited and finding a good location was not easy. Wording the questions for the participants was also quite a task. And putting together good teams took some time’, Schouten explains. During the Sensor Data Challenge, the teams took a very serious approach in their attempts to solve the questions. ‘Some of them worked non-stop for almost 24 hours. Everyone was highly motivated, including the staff who organised the event.’ Ruijsbroek, too, is positively surprised by the large amount of work that was carried out within 24 hours. ‘The teams worked hard and there was a great atmosphere. So much was achieved in such little time, which is fantastic. It gave me a lot of new energy! We will soon sit around the table with the two winning teams to discuss how we can further develop these concepts.’
The Data Challenge was won by the Sharkblood team, consisting of Danielle McCool, Mitchell van Wijngaarden and Kees Mulder. Mulder graduated in social sciences at Utrecht University and is conducting doctoral research. McCool is working for a marketing agency as a data consultant and Van Wijngaarden is a web designer. Mulder: ‘It was the first time for me to participate in this type of Data Challenge. It was a lot of fun, especially since I mainly do theoretical research. I was surprised at how much could be achieved within 24 hours.’ Mulder and his team came up with a prototype to replace the RIVM questionnaire which normally takes respondents 30 minutes per day to complete and is related to their eating pattern. ‘We immersed ourselves completely and developed an app which makes it very easy to see what your daily food intake is and what the ingredients are. We were even able to measure the volume of the portions using 3D imaging. In this way, respondents will find it much easier to participate in a survey.We will soon introduce our concept to CBS, The Hague University of Applied Sciences and RIVM, and we will look see how it can be developed further.