Presentation (CTI)

OBJECTIVES: To characterise the methodological approaches used in published health economic models for obesity interventions, identify key sources of variation and uncertainty, and synthesise methodological considerations that may support greater consistency and transparency in obesity economic modelling.
METHODS: A targeted scoping review was conducted to identify model-based economic evaluations of obesity interventions. Existing systematic reviews informing NICE guidance were used as the foundation, supplemented by focused searches of MEDLINE, Embase, and INAHTA for studies published between January 2023 and February 2025. A pre-specified checklist was used to extract information on model structure, representation of obesity-related comorbidities, and approaches to modelling treatment effects. Findings were synthesised narratively.
RESULTS: A total of 26 cost-utility analyses of obesity interventions were identified, including 21 state-transition models, comprising Markov cohort models (n=16) and microsimulation models (n=5). Others used decision tree models (n=3), discrete event simulation (n=1), and a model informed by Mendelian randomisation (n=1). Three broad treatment-effect modelling approaches were identified: risk-equation models (n=12, 46%), clinical-outcome models (n=5, 19%), and BMI trajectory models (n=9, 35%). Risk-equation models incorporated the widest range of obesity-related complications, particularly type 2 diabetes and cardiovascular disease, whereas BMI trajectory approach primarily linked weight changes directly to costs, quality of life, and mortality. Considerable variation was observed in health-state definitions, inclusion of obesity-related complications, treatment duration assumptions, weight regain trajectories, and extrapolation methods. Long-term maintenance of treatment effects and weight regain were consistently identified as major sources of uncertainty.
CONCLUSIONS: This scoping review directly informed the development of NICE's first obesity-specific reference case extension and highlighted priorities for improving consistency in obesity economic modelling. Greater methodological consistency and transparency in population definition, health-state selection, treatment-effect modelling, and handling of long-term uncertainty may improve the comparability and decision relevance of future obesity economic evaluations.