Obesity and Type 2 diabetes (T2D) (together, called 'diabesity') has an extensive economic burden, a negative impact on healthy life expectancy and is linked to lower quality of life particularly in children and young people (CYP). Diabesity is also strongly linked to health inequity: CYP who are from ethnic minorities, with lower socio-economic position, have much higher risk of diabesity and the poor health associated with it. We have recently shown, using data from primary care records, sharp rises in pre-diabetes and T2D in CYP; incidence rates of T2D in CYP have increased almost threefold from 2005 to 2019. There is an urgent need to invest in diabesity prevention and the development and implementation of specific plans to improve the care of CYP with diabesity. Obesity is the main risk factor for T2D which we are able to modify. Risk of T2D rises with increasing BMI, and only 1.5% of children with T2D have a healthy weight. Interventions to manage weight and physical activity have the potential to reduce T2D risk. Whilst we have identified a number of risk factors for diabesity, we do not know enough about how these multiple factors interact together and across time. This is known as a complex system, with multiple factors interacting and developing together across time. Complex systems make it difficult to identify prevention or intervention strategies that will work for everyone, because of the wide variation of possibilities in the system. This means that, for complex health problems like diabesity, we need to better understand and map the complex system before we can determine what prevention or intervention services would work for who, and when, and at what cost. This would also us to 'tailor' prevention programmes to specific groups to make them as effective as possible as well as estimating their economic costs and benefits.

In this project we will map the complex system of diabesity in CYP, integrating the existing literature with data collected in primary care and birth cohort studies (ALSPAC, Born in Bradford and the Millennium cohort), as well as expert stakeholder input. We will then use a simulation technique called agent-based modeling, which is used to study complex systems and allows us to use artificial intelligence to model what would happen if we introduce a prevention programme with specific features at specific time points to specific groups of people. We will also include health economic data in the model so that economic costs and benefits of programs can be modelled.

The final output of the project will be an open access decision making 'dashboard': a state-of-the-art multi-agent systems simulation tool for Type 2 diabetes prevention in children and young people with which policy makers, commissioners, service providers and clinicians can simulate tailored prevention programmes and strategies for their target population. We will produce training and support materials to maximise usability and uptake. The 'Diabesity Dashboard' will support commissioning of real-life programs tailored for specific groups that are most likely to work and be cost effective.