Diabetes prevention : economic evaluation of individual case-based and population-wide prevention programmes and implications for public policy

This thesis aims to inform diabetes prevention policy through economic evaluation of individual-case based and population-wide prevention programmes for type 2 diabetes (T2DM). Two broad approaches are described: i) individual case-based programmes focusing on individuals at high risk of T2DM and ii) population-wide programmes that seek to modify an entire population's risk of developing T2DM. Published evidence on individual case-based interventions is described, with both lifestyle programmes and metformin appearing effective at reducing incidence of T2DM and being cost-effective at a willingness-to-pay threshold of £20,000/QALY. However, key differences were identified between current clinical guidelines/national policy in England and published economic evaluations. The implications of these different approaches to individual case-based programmes, in terms of cost-effectiveness, budget impact and incidence of T2DM, are quantified using a de novo Markov model. This analysis found that whilst current UK national guidance and policy is likely to be cost-effective and have the most favourable budget impact, it will prevent fewer than 5% of incident cases of T2DM over 50 years. Alternative approaches to diabetes prevention, using population-wide programmes, are therefore identified. Their impact on T2DM, or its major risk factor of adiposity, is summarised through an overview of systematic reviews. Four population-wide actions were associated with lower BMI, including sugar sweetened beverage tax, menu labelling, food store interventions and multi-component community interventions. However, impact on T2DM was impossible to quantify precisely due to lack of primary studies. A pre-existing micro-simulation model was utilised to quantify the potential impact of these interventions on healthcare costs associated with T2DM, QALYs and cumulative incidence of T2DM. This analysis found that these four interventions were associated with reduced costs relating to T2DM and increased QALYs. However, costs and benefits outside the health system precluded meaningful assessment of cost-effectiveness and the interventions resulted in only a small reduction in incidence of T2DM. These findings support a more comprehensive, complex systems approach to diabetes prevention policy that encompasses both individual case-based and population-wide programmes. They suggest that, even if all the available evidence-based policies described in this DPhil were implemented, they may still be insufficient to substantially reduce the prevalence of T2DM in England. This moves the diabetes prevention policy debate on from considering which single approach to adopt (individual case-based or population-wide) to considering how to implement a combination of programmes concurrently and continuing to search for new, promising interventions at both an individual and population level.