Four essays on performance measures based on patient-reported outcomes

Agency relationships, and associated information symmetries, exist in many areas of economic activity including healthcare. Information on healthcare providers’ relative performance can be used to reduce information asymmetries and hold providers to account. This collection of essays focuses on the appropriate derivation and use of performance measures to incentivise healthcare providers in the English National Health Service (NHS). It gives special consideration to the role of patient self-reported health status measures to assess the differential effect of healthcare providers’ care on their patients’ health. The thesis explores three themes: the relationship between variation in resource use and quality, the appropriate assessment and reporting of multidimensional hospital performance, and the use of performance information to motivate hospitals in a public reporting context. Chapter 2 examines cost variation between hospitals for the four surgical procedures covered by the national patient-reported outcome measures (PROM) programme. It explores the empirical relationship between costs and patient health outcomes to assess the claim of hospital providers that their higher costs are justified by better quality of care. Chapter 3 sets out an empirical methodology to conduct provider performance comparisons when there are multiple dimensions of health-related quality of life affected by treatment. It discusses the advantages and disadvantages of analysing disaggregate PROM data for the purpose of informing prospective patients, clinicians and managers. Chapter 4 extends the previous chapter by providing a methodology for assessing and summarising multidimensional provider performance using dominance criteria. This methodology is then applied to study the performance of providers of hip replacement surgery with respect to length of stay, emergency readmissions, waiting time and improvements in PROMs. Chapter 5 estimates the demand elasticity of providers with respect to quality. It makes use of choice models to assess the usefulness of disseminating hospital PROM scores to prospective patients as a market-based incentive for providers to compete on quality.