TRANSPARENCY AND LIABILITY IN THE HEALTHCARE SECTOR

This thesis comprises three papers that represent three autonomous chapters. The linking "fil rouge" between the essays is the investigation of the issue of liability and transparency in the health care sector. In the first chapter , we evaluate the role of transparency in measuring the performance of Local Health Authority (LHA) using the composite indicators prosed in literature and depict the relative geographical distribution in order to investigate whether transparency index matters on the performance for different expenditure functions at LHAs level. The health sector is considered to be one of the most exposed to the risk of corruption and therefore needs adequate levels of transparency. Healthcare is a particularly sensitive ground, where opportunistic behaviours germinate and can degenerate into corruption with several possible reasons behind: • the magnitude of the expenditure • the pervasiveness of information asymmetries • the unpredictability of demand • the high specialization of the products purchased • the need for complex regulation systems The forms and intensity differ according to the overall level of integrity and are becoming increasingly worrying even in the most advanced countries. Corruption in the health sector has both economic effects, diverting resources from assistance programs and social effects, undermining people's trust in the healthcare system. In the last years, several studies have considered this issue in order to assess the magnitude, the determinants and the effect of corruption in the healthcare sector. Italy recently has implemented legislation on transparency (so-called Code of Transparency - Leg. decree n. 33/2013 ), extended also to the health sector. It provides several (about 270) detailed transparency obligations to be published in a standardized format (Amministrazione trasparente), regarding different aspects such as: the organization of public organizations with respect to politico-administrative bodies and top public managers and officers, external consulting and collaboration, public procurement, management of properties and assets, timing of payments, provision of public services. In our analysis, we follow a “top-down” approach which develops indicators from the legal and formal aspect. In particular, we use the indicator “Composite Transparency Index” (CTI) developed by Galli et. al (2017) in a study on the transparency in the main Italian municipalities. The indicators CTI uses the values attributed by OIVs to the items included in ANAC resolution n. 77/2013, according to a scale going from 0 to 3. It is composed by two sub-indicators “CTI-Integrity” and “CTI-Performance” which investigate the two different aspects. In particular, the “CTI-Integrity” consider items on income and asset disclosure and conflict of interest (on both politicians and top and senior public officials, while the “CTI-Performance” considers items on the management of public property, the timeliness of public payments, the quality of public services. At the end, the total CTI is constitute by the average of the two indicators. The information was collected for all the 143 LHAs. The degree of transparency of Italian Local Health Authorities (LHAs) using the composite indicators prosed by Galli et al. (2017) depicts the usual geographical dichotomy between North and South, with Tuscany and Emilia Romagna aligned with the former macro area, while Lazio, Umbria and Marche with the latter. Then we explore the relationship between transparency and expenditure at LHA level. Our results show a negative correlation between transparency and total expenditure whereas we find a significant positive correlation for the administrative expenditure. The second chapter is devoted to exploring the role of medical malpractice liability and in physicians’ behaviour using an experimental approach. The effect of the payment system on the behaviour of the physician has been a very hot topic, intensely studied by a lot of researches all over the world in the last decades, even in a laboratory study. Also, the effect of the medical liability on the behavioural changes has been deeply investigates. Nonetheless, to the best of our knowledge, there are no papers still that have studied the effect of medical liability on the physician in a laboratory environment. Considering these aspects and building on the results of Brosig-Koch et al. (2017), we investigate whether the introduction of the possibility for a physician of being sued has any effect on the effort he/she devotes to cure his/her patients. This effect has been observed through the two different payment schemes: fee-for-service and capitation. Finally, we check whether different samples of participants in our experiment show the same behaviour, running sessions with randomly chosen students, medical students and post-graduate MDs. The experimental design is divided into four treatments differing in the payment system (fee-for-service and capitation) and in the presence of medical liability device. In each treatment, each participant plays the role of a physician who has to choose how many medical prescriptions (from zero to ten) to provide to his/her patients. Patients are divided according to the severity of disease (three levels) and gender (M/F). The experiment was totally computer-based experiment and run with the z-Tree software. At the beginning of each treatment, subjects receive the instructions of that treatment only and the relative payoff table. Moreover, before starting the treatment they have to solve a couple of simple numerical exercises in order to be sure they have understood how FFS/Capitation payment scheme works and how to compute their profits. Each treatment lasts for six periods, representing six patients differing in terms of severity of disease and in gender for the three levels of severity. The sequence of patients has been randomly drawn at the beginning of the research, it changes in every treatment but stays the same for all the subjects playing the experiment. Regardless of the payment system, the quantity of medical services provided by physicians is higher when the risk of being sued for medical malpractice is at play. The increase in the quantity of medical services induced by the risk of being sued for medical malpractice is higher in CAP than in FFS While the increase in CAP brings closer to the efficient level of medical services, the increase in FFS pushes further away from the efficient level of medical services. In the third Chapter of the thesis, we further explore the role of malpractice liability using the experimental setting. In this Chapter, building on the seminal model of Ellis and McGuire (1986) and the experimental results of Brosig-Koch et al. (2017), we investigate whether the introduction of a mixed payment system and the possibility for a physician of being sued has any effect on the quantity of medical services provided to cure the patients. To do this, at first we compare the two main payment schemes fee-for-service and capitation with a perfectly balanced “fee-cap” mixed payment system. Then, we check if the presence of medical liability influences the optimal calibration of the mixed payment system. The experimental design is made up by two parts, with two different samples composed by different subjects. The first sample plays the experiment following the scheme Fee-for-service/Mixed and the second one with the scheme Capitation/Mixed. Each part is composed by four treatments differing in the payment system (fee/mixed, cap/mixed), as mentioned before, and in the presence of medical liability. In each treatment, each participant plays the role of a physician who has to choose only how many medical prescriptions (from zero to ten) to provide to his/her patients, which are divided according to the severity of disease (three levels-low, medium, high) and gender (M/F). Each treatment lasts for six periods, in order to represent the six patients differing for severity of disease and gender. The sequence in which patients appears to subjects playing has been randomly drawn at the beginning of the research, it changes in every treatment but remains the same for all the experiment. Even if there are no patients inside the lab, in order to make the effort decision more realistic subjects are made aware that the benefits awarded to patients through medical prescriptions are converted into euros and donated to a charity providing health care to children affected by spinal muscle atrophy (SMA). The experimental data give results in line with our behavioural hypothesis. References Galli, E., I. Rizzo and C. Scaglioni 2017 “Transparency-quality-of-institutions-and-performance-in-the-Italian-Municipalities”, ISEG/UTL – Technical University of Lisbon, Economics Department, WP10/2017/DE/UECE, ISBN/ISSN: 2183-1815 Brosig‐Koch, J., Hennig‐Schmidt, H., Kairies‐Schwarz, N., & Wiesen, D. (2017). The effects of introducing mixed payment systems for physicians: Experimental evidence. Health Economics, 26(2), 243-262. Ellis, R. P., & McGuire, T. G. (1986). Provider behavior under prospective reimbursement: Cost sharing and supply. Journal of Health Economics, 5(2), 129-151.