Quantifying the impact of prediabetes: developing and validating new health economic models for policy evaluation

Thesis

Patients with prediabetes are known to be at higher risk of diabetes development and cardiovascular complications than the general population. Nevertheless, neither the comparative economic impact of prediabetes on the health system, nor the potential value of intervention to prevent or delay disease progression are currently well characterised.

This thesis sets out to understand the relative impact of prediabetes on health economic outcomes. I first quantify the relative impact of prediabetes compared to related conditions of obesity and type 2 diabetes on UK hospital costs for COVID-19. To inform the design of new prediabetes economic models, I undertake quantitative surveys of existing models of diabetes development and diabetes progression. Within this conceptualisation, I assess the impact of differing modelling choices post-diabetes progression on economic model predictions of cost and quality of life across several diabetes interventions. This background informs the development and validation of two new multi-use health economic models of prediabetes, which I apply to evaluate the cost-effectiveness of a GLP-1 intervention in prediabetes populations.

The newly developed models predict the progression of risk factors in patients over time, and the impact of predicted risk factor progression on patients’ expected rates of diabetes, cardiovascular complications and mortality. The modelling framework can therefore be used to predict long-term changes in patient event rates, and corresponding quality and length of life, and disease-related healthcare costs, resulting from improvements to patient risk factors. Internal and external validation shows model predictions to currently perform best within populations with similar baseline cardiovascular history. The case study application of the new models shows that prioritising GLP-1 treatment to populations with both obesity and prediabetes could be highly cost-effective within roll-out of wider obesity therapies. Nevertheless, further evidence to inform modelling of the costs of implementing targeted rollout to obese populations with prediabetes, and the expected effect of GLP-1 treatment in this subpopulation, is required.

Authors: Altunkaya, J

• DOI: 10.5287/ora-da6o90apg
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Subjects: Prediabetic state; Medical economics; Obesity; Type 2 diabetes; Economic policy--Mathematical models