Title: Systems thinking and body weight perception: Modelling interactions between the individual and the collective to counter obesity

Abstract:

Aim: Shifts in obesity prevalence cause individuals to re-evaluate their weight with reference to the average weight in their socio-cultural group as opposed to a fixed standard. We therefore hypothesise that diversity regarding the social norms of obesity exists among different socio-cultural groups, reflecting their variation in obesity prevalence. This stems from an intricate web of interactions between different elements – individual- and group-level body weight as well as socio-cultural elements such as individual and collective body weight perception, ethnicity and sex –, which can be perceived as a complex system. We analyse this complex system using system dynamics modelling in order to identify the relative impact of individual- versus group-level interventions.

Methods: Individual and collective ideal body image variables are embedded in a conceptual causal loop diagram (CLD) and subsequently in functioning system dynamics models (SDMs). These SDMs simulate the dynamics of weight for six socio-cultural groups (Dutch, Moroccan and South-Asian Surinamese men and women) in Amsterdam, where we focus on the feedback loops that drive the re-evaluation of individual perceptions regarding body weight based on obesity prevalence.

The SDMs are created using a structured methodology to construct operational SDMs from cross-sectional data, which builds upon the exploitation of qualitative expert knowledge as outlined in a CLD. Our work therefore highlights the value of integrating qualitative knowledge into quantitative system dynamics modelling and clarifies which assumptions are necessary. Developing methodologies like this to generate pseudo time-series data from cross-sectional data – which are often available but fail to provide us with dynamic insights – may be imperative in fields beyond public health as well.

Results: We have validated the SDM using 41 validation statements. Currently we are investigating the relative role of social norm versus individual-level interventions.

Discussion: Given the nature of SDMs, we hereby make it feasible to simulate the effects of individual- versus group-level interventions in the different socio-cultural groups.