OBJECTIVES: Obesity is a complex disease that can reduce patient quality-of-life and contribute to increased risk of complications such as type-2 diabetes (T2D), cardiovascular (CV) disease, and cancer, which are associated with economic and societal impact. To fully capture its many consequences, cost-effectiveness models for obesity mirror this complexity. The aim of this study is to compare the outcomes of two different modelling approaches for assessing the cost-effectiveness of tirzepatide in managing obesity.

METHODS: Two individual-level simulation models previously developed to assess cost-effectiveness of tirzepatide against diet-and-exercise (D&E) were compared using a United Kingdom population with obesity and overweight: a state-transition model and a discrete-event simulation model. Both models track individual metabolic factors such as body weight, high-density lipoprotein, and systolic blood pressure, to estimate how many patients in a simulated cohort would develop obesity-related complications over their lifetimes. Using the same risk equations and patient characteristics, models were restricted to consider common complications: T2DM onset, CV events, and onset of sleep apnoea. This allows evaluation of the impact of modelling approach on predicted outcomes.

RESULTS: Both models showed tirzepatide reduced the number of T2D and CV events such as stroke and myocardial infarction compared to D&E. However, the discrete-event simulation model estimated higher incidences of T2D and sleep apnoea across all treatment arms compared to the state-transition model. Time-to-event outcomes were estimated in the discrete-event simulation model only, demonstrating that tirzepatide is associated with delays in the onset of complications. While the state-transition model partitioned the CV events once, the discrete-event simulation model used a two-step partitioning approach. Similarly, the approaches in estimating non-CV deaths also differed between the two models.

CONCLUSIONS: Obesity is a complex disease associated with numerous complications. Despite the differences in approaches, both models provided directionally consistent results, validating their use in obesity cost-effectiveness studies.