Background: Health economic modelling is used to estimate the cost-effectiveness of obesity interventions, usually over a patients lifetime. These estimates inform whether a treatment is recommended for funding or not. Due to the short-term nature of clinical trials and the mixed evidence regarding long-term weight maintenance, assumptions are made about the long-term impact of an intervention. We aimed to adapt an existing health economic model of obesity to include psychological variables. This has two potential benefits: (1) The psychological variables can be used in prediction of long-term BMI trajectories to improve estimates of effectiveness and cost-effectiveness; (2) The model could be used for pre-trial testing and intervention design to generate an expected change in BMI based on estimated change in psychological variables.
Method: To determine what variables to include we conducted a latent growth curve mediation analysis on data previously collected in a clinical trial of a commercial weight management programme. BMI and psychological variables were collected at 4 time points over 2 years. We will present an existing health economic model, the SPHR diabetes prevention model, and how this has been adapted to include the relevant psychological variables in the prediction of BMI trajectories.
Results: The trajectories of dietary restraint, autonomous diet self-regulation and habits mediated the significant positive effect of the intervention on BMI trajectories over two years. We will report how the cost-effectiveness estimates of behavioural interventions differed using the model with and without the adaptation.
Conclusions: Psychological variables are often collected and analysed in clinical trials of behavioural weight management intervention but are not routinely used in health economic models of obesity. This research gives an insight into how this data could be used to improve predictions of effectiveness and cost-effectiveness. This would lead to more informed decision-making regarding the funding of obesity interventions. While health economic modelling is often conducted post-trial, including these variables also enables the model to be used to inform intervention design.
Keywords: Health economic modelling, psychology, intervention
12 Sep 2019 - 13 Sep 2019