Weight Status Transitions and Validation of an Obesity Model for Aboriginal and Torres Strait Islander Children and Adolescents

Objectives Aboriginal and Torres Strait Islander children and adolescents are at higher risk of overweight and obesity, highlighting an inequitable public health concern. The aim of this study was to estimate transition probabilities and validate a model predicting the epidemiologic trajectory of overweight and obesity in Australian Aboriginal and Torres Strait Islander children. Methods An individual-level state-transition Markov model was developed to model transitions between healthy weight, overweight, and obesity for Aboriginal and Torres Strait Islander children aged between 2 and 14 y. Age-specific annual transition probabilities were derived from semi-parametric survival analyses using the Longitudinal Study of Indigenous Children. The model used annual cycles over a 12-y time horizon, and the epidemiological predictions of the model were validated using both internal and external data, according to best practice guidelines. The starting age of the model was 2 to 4 y and 4 to 5 y for the internal and external validation, respectively. Aboriginal and Torres Strait Islander children from the Longitudinal Study of Australian Children were used as the external validation cohort. Results A total of 1,643 children with 11,514 complete anthropometric measurements were used to estimate transition probabilities. The model predictions showed both good internal and external validity, with most predictions falling within the 95% confidence intervals of measured data. The model was able to reliably capture the epidemiology of overweight and obesity prevalence in early childhood. Conclusions Our model predictions showed good internal and external validity, ensuring our model is fit for purpose to use to evaluate Aboriginal and Torres Strait Islander–led programs to achieve a healthy weight. Highlights Aboriginal and Torres Strait Islander children experience high rates of overweight and obesity; hence, there is a need for high-quality evidence on both effectiveness and cost-effectiveness of Aboriginal and Torres Strait Islander–led childhood obesity prevention programs to ensure they offer value for money. This is the first study to develop and validate a predictive model using anthropometric data from Aboriginal and Torres Strait Islander children to inform decision making on childhood obesity programs. Our model predictions showed good internal and external validity, ensuring our model is fit for purpose to use to evaluate Aboriginal and Torres Strait Islander–led programs to achieve a healthy weight. The model provides a framework to assist policy makers in identifying when best to intervene in childhood as well as the most effective approaches for maintaining a healthy weight for Aboriginal and Torres Strait Islander children.