4. Behavioural system mapping

Behavioural systems mapping is a developing method which uses systems thinking to diagrammatically represent relationships between the behaviour(s) of interest, interventions, actors, and the influences on behaviour, for the purpose of helping decision-makers or other stakeholders to understand systemic influences on behaviour(s) of interest (Hale et al., 2022) and the implications for interventions. Squires et al., (2016) proposed the use of systems mapping to understand the problem needing to be modelled prior to developing the health economic model to understand relationships between factors and to help decide what to include and exclude from the model, but this framework did not consider influences on behaviour explicitly. To be able to model what may happen over time, modellers need to gain an understanding of the influences on the behaviour(s) and the impact of the intervention(s) upon the behaviour(s). Figure 2 within this paper and formal behavioural theories, alongside input from behavioural scientists, could be used to inform the development of a behavioural systems map. These could include both micro variables which describe individual attributes (e.g., motivation) and macro elements which describe aggregate level phenomenon (e.g., social network structures), and the potentially causal relationships between them. If the policy makers are aiming to reduce inequalities and/or consider intersectionality, then it will be important to consider the different influences impacting upon relevant groups of interest.

Describing these influences within a systems map will provide a tool for communication between behavioural and social scientists and health economic modellers and other stakeholders, increasing the understanding of the problem and facilitating the development of a useful model. Behavioural systems mapping can provide a transparent process through which to make decisions about what is included (or not) within the health economic model. It can also facilitate discussion about the social value judgements which may be made. The outcomes reported by intervention studies may limit what quantitative analyses can be done within the model, for example if only body mass index is reported rather than physical activity or dietary outcomes or the mechanisms of action associated with these. In addition, studies often do not report outcomes according to attributes associated with inequality (for example, by ethnic minority or gender) or intersectionality (for example, by Asian women). In these cases, behavioural systems mapping may provide a qualitative understanding of the micro and macro level influences, which can help to inform what to include and exclude from the model, assumptions beyond study follow up, and what further data collection may be useful. 

Behavioural systems maps can be developed from participatory stakeholder input, and/or other sources including qualitative research such as interviews or quantative analysis based on surveys (Craven, 2017). These could either be developed at an early stage and could help to inform the choice of interventions to assess, or if the interventions have already been chosen, the theory used to develop the interventions (where available) could be incorporated.