New health technologies seeking National Health Service funding in England are subject to rigorous evaluation of cost-effectiveness using established economic evaluation methods. Changes to the organisation and delivery of health services, including changes to health policy, are funded from the same budget as these health technologies, yet undergo no such mandatory cost-effectiveness assessment. This has resulted in a lack of methodological development and evidence on the cost-effectiveness of large-scale changes to the organisation and delivery of health services. This thesis aims to contribute to the development of methods for the economic evaluation of changes to the organisation and delivery of health services. This aim is achieved through the consideration of two recent, high-profile examples: a regional pay-for-performance programme, and a national initiative to extend emergency hospital services at weekends. Methods for both ex-ante and ex-post evaluation are developed and applied. The literature pertaining to the economic evaluation of the two example programmes examined is reviewed and critiqued. Estimates of the costs and benefits associated with the two programmes are provided. A framework for assessing the cost-effectiveness of pay-for-performance programmes is developed, including a methodology to quantify the impact of programmes in terms of quality-adjusted life years in the absence of primary data collection on health-related quality of life. Issues around defining the relevant counterfactual for programme evaluations over the longer-term are explored, along with the potential for wider spillover effects. Survival analysis techniques commonly employed in clinical trials are used to improve treatment effect estimates associated with policy initiatives. The regional pay-for-performance programme was found to have likely represented a cost-effective use of resources during the first 18 months of its operation. The programme was found to be associated with a health gain of 5,227 quality-adjusted life years, generated at a total cost to commissioners of Â£13m. The programmeâs longer-term benefits are, however, uncertain. The planned costs of extending emergency weekend hospital services are compared to the maximum potential health benefit attainable from this extension. The treatment effect associated with weekend admission to hospital is re-examined, and the possibility that earlier estimates suffer from bias is demonstrated, due to the restricted focus on only the admitted patient population. The evaluation of the planned extension of weekend emergency hospital services shows that there is insufficient evidence to suggest that the programme would represent a cost-effective use of resources. The maximum potential health benefit attainable was estimated to be between 29,727 and 36,539 quality-adjusted life years, whilst the programme would cost between Â£1.07bn and Â£1.43bn. This exceeds the maximum the National Health Service should be willing to spend to achieve a health gain of this size by Â£339m to Â£831m. Based on the two example interventions evaluated, and review of two established frameworks outlining the principles of cost-effectiveness analysis, the principal challenges faced when conducting economic evaluations of changes to the organisation and delivery of health services are identified and discussed. The principal challenges are identified as: undertaking ex-ante evaluation; modelling the counterfactual and estimating the treatment effect; evaluating the impact in terms of quality-adjusted life years; assessing costs and opportunity costs; accounting for spillovers; and generalisability. Changes to the organisation and delivery of health services should undergo rigorous cost-effectiveness evaluation, as is now mandatory for new health technologies. This thesis contributes to the development of methods to facilitate such evaluation.