The overall aim of BigO is to collect and analyse big data on behaviour and living environments related to childhood obesity. This will allow public health authorities to plan and execute effective programs to reduce the prevalence of childhood obesity.
This is not because programs to reduce obesity do not exist, but because they are less effective than intended, and obesity remains on the rise. In the EU, approximately 2.8 million deaths per year result from causes associated with overweight and obesity (easo.org).
The causes of obesity are complex, however evidence exists that interventions that target multiple elements of children’s behavioural patterns and living environments are needed. Therefore, BigO uses a multilevel approach (see the contextual model of BigO below).
To accomplish the main goal of effectively advising public health and clinicians, specific objectives have been set. These objective are divided into different domains.
The scientific objectives of BigO are all geared towards gathering the information needed to study the relationship between the various behavioural and environmental factors related to obesity, and creating the analytical models to use and exploit this information. Specifically, these objectives are:
- To extract relationships between the external living environment and individual behavioural patterns that increase behavioural risk factors for obesity. (Aetiology).
- To create models that show how changes in the external living environment can alter behavioural patterns, which in turn modify behavioural risk factors for obesity. (Prediction)
- To produce models that predict how changes in behavioural risk factors for obesity impact obesity prevalence. (Prediction)
- To define a behavioural model in such a way that it is useful for the above purposes, but does not store sensitive information or redundant personal information. (Privacy Preservation)
The technological objectives are related to building the infrastructure to collect, store and analyse data. Moreover, they are focused on developing the technological programmes and associated privacy measures which lead to decision support tools for care facilities and public health authorities. Specifically, these objectives are:
- Building an extensive network of information sources: namely sensors like smartphones and smartwatches running mobile applications that collect subjective and objective data, in order to get information on behaviour and the local environment.
Moreover, server based applications will be developed so that publicly available data (such as maps, statistics and metadata) can be shown.
- To determine policies, as well as the technical means to enforce these policies, relating to big data governance, privacy and anonymisation
- To provide 3 decision support functionalities:
- the Policy Advisor that offers aetiology and data evaluation services. For example, visualising aggregated evidence for public health authorities and schools to help them design and monitor programs.
- the Policy Planner that offers simulation and prediction services. For example extracting associations between environment and obesogenic behaviours to investigate causality, or creating prediction models and developing intelligent algorithms to recognise behavioural patterns.
- the Clinical Advisor that offers evaluation and decision support for the individual at the point of care. For example: visualising individual behavioural patterns for health professionals to help them follow-up childhood obesity patients.
The validation objectives are aimed at evaluating how the systems and platforms work. Specifically, these objectives are:
- Evaluation of the system components
- Evaluation of the system in realistic usage environments
- Evaluation of the decision support platform.
The business objectives include:
- Defining an effective, pragmatic and viable business plan and exploitation scheme, in line with use of the system as a framework for supporting public health authorities on the one hand, as well as a tool that offers evidence to health professionals on the other hand.
- Building the BigO program around the “citizen-scientist” model, which relies on individuals sharing their behavioural data. As part of the business plan, BigO will propose policies integrating economic benefits as incentives (e.g. discounts in health insurance schemes) for active data contributors
These objectives of BigO are envisioned to lead to several expected outcomes:
- Allowing faster, pre-assessed, more efficient policy choices, all the way from obesity prevention to point-of-care for individuals who are already obese.
- Education of young European citizens about the principles of voluntarism, citizen science and public participation
- Increased awareness about healthy living, introducing students to the health-in-all-things mentality