Essays on financial incentives in the secondary healthcare sector

This thesis consists of three empirical essays, contributing to the understanding of key
policy issues related to financial incentives in the healthcare setting. Chapters 2 and 3 con-
tribute to a growing literature on the Pay for Performance (P4P) schemes in the secondary
care, while Chapter 4 focuses on the design of the prospective payment reimbursement
system.
Chapter 2 evaluates the impact of a financial incentive, designed to improve care of
fragility hip fracture patients in England and implemented in 2010. The scheme adopts a
bundled approach, by which nine different criteria related to quality of care must be met in
order for the hospitals to receive bonus payment. Analysis is based on the difference-in-
difference framework, with Wales as a control group. Results show large and statistically
significant effect of the scheme on the uptake of the incentivised quality measures. Effects
on patients mortality are small and mostly insignificant.
Chapter 3 considers a financial incentive scheme designed to shift inpatient activity to
outpatient setting and implemented in England in 2012. The scheme targets three condi-
tions and operates by overpaying the outpatient activity while concurrently underpaying
inpatient activity. Using difference-in-difference approach, the results indicate large and
significant effects of the policy on increasing the proportion of patients treated in the out-
patient setting, without harming the quality of care or increasing the overall volume of
activity.
Chapter 4 estimates the effect of a Diagnosis Related Groups (DRG) classification re-
form on hospitals’ coding behaviour. The chapter considers a major reform in English DRG
system in 2009 which highlighted the role of reported comorbidities in the reimbursement
process. The analysis is based on the difference-in-difference framework. Results indi-
cate significant effect of the reform on coding intensity, increasing the probability of being
coded to a severe HRG and, consequently, the overall treatment cost.