This paper is divided in three parts, each referred to a different experiment. The experiments were conducted at the main hospital of Reggio Calabria, thanks to an agreement signed by the same hospital and the Mediterranean University. In the first experiment, I investigate for the first time the level of coordination among the Italian health professionals, by using a simple, portable and incentive compatible tool such as the coordination game, widely adopted in previous literature, though in other settings. By identifying areas where relational coordination plays an important role, I have chosen to focus on orthopaedics, paediatrics and oncology. For each of the afore mentioned departments, I designed a specific vignette (i.e., a description of a hypothetical situation which physicians could face on their job) and a set of four actions which a physician can take in response to that. In this framed field experiment, physicians are asked to evaluate the appropriateness of each of the possible actions on a scale of one to four in order to match the valuations of most respondents (i.e. modal answer). Data show that the frequency mean of coordination across the experiment is 52% and that coordination increases when physicians do often exchange opinions and share positive feedbacks with colleagues. In addition, the presence of a leader in the medical department, facilitates coordination. Evidence provided by this study shows that coordination in the Italian health sector could be missing, which makes the need for action all the more acute. In the second experiment I test whether physicians act according to national guidelines. By submitting the same scenarios proposed in the first experiment, physicians can receive information concerning what it should be done, in the specific situation, according to national guidelines (i.e. providing them with the written guidelines) and eventually change decisions. Results show that only 8% of the physicians did not want to know national guidelines content. However, only 21% of the subjects (with a choice frequency of 23%) decided to change their appropriateness judgment after realizing they were in contrast with guidelines in either one or two vignettes. Overconfidence and perceived guidelines ambiguity could explain such a low value. Finally, the more physicians consult scientific sources (i.e. signal of open-mindedness), the more willing to accept suggestions, coming from guidelines reading, and to change their decisions when they are wrong. Since guidelines dissemination is shown to increase the level of coordination between physicians, hospitals should consider effective programmes to spread their knowledge. Finally, in the third experiment, I test whether and to which extent the adoption of fee-for-service or salary system can induce physicians to practice patient dumping. In an artefactual field experiment, physicians facing the possibility of being sued for medical malpractice decide whether or not to provide medical services for patients with different state of health. Also, I check whether the introduction of the risk of being sued for a physician for having practiced dumping can have effect on his behaviour. Results show that dumping is more often observed under Salary than under Ffs. However, physicians seem to be insensitive to the introduction of dumping liability under the same incentive mechanism, though it seems to trigger a higher amount of services provided. Policy strategies concerning the incentive scheme may vary on the hospital purposes.

Il presente elaborato, frutto di un accordo quadro tra l’Università Mediterranea e il Grande Ospedale Metropolitano ‘Bianchi Melacrino Morelli’, si propone di adottare l’approccio sperimentale al campo dell’economia sanitaria attraverso tre esperimenti, spalmati su tre diversi capitoli, con la partecipazione di 87 medici. Nel primo esperimento , i medici posti di fronte a tre diversi scenari rappresentanti pazienti con diverse diagnosi, ciascuna associata ad uno specifico reparto, dovranno decidere il livello di appropriatezza di ciascuna delle azioni proposte a risoluzione del caso in maniera tale da coordinarsi con i colleghi nelle scelte. L’incentivo monetario, infatti, prevede che i medici riceveranno un buono pasto se e solo se la loro risposta coinciderà con la risposta modale, ovvero data dalla maggior parte dei partecipanti. L’obiettivo è verificare il livello di coordinazione esistente all’interno dei vari reparti. I dati mostrano che la frequenza media di coordinamento è pari a 0.52. Tale frequenza cresce per i medici che affermano di scambiare frequentamente opinioni e feedback positivi con i colleghi. Inoltre, la presenza di un leader a guidare l’equipe medica all’interno del reparto facilita la coordinazione. I dati sottolineano come la coordinazione sia spesso scarsa ed i reparti necessitino di interventi volti a migliorare la comunicazione tra i vari attori del reparto. Il secondo esperimento è finalizzato a verificare l’impatto delle linee guida sulle scelte dei medici. In particolare, fornendo ai medici le linee guida per la risoluzione degli stessi casi proposti nel primo esperimento, si vuole verificare come queste siano in grado di modificare le condotte dei medici ed eventualmente facilitare il raggiungimento della coordinazione. I risultati mostrano come solo 4 medici su 52 dichiarino di non voler conoscere il contenuto delle linee guida.Tuttavia solo 21% dei soggetti decidono di cambiare il livello di appropriatezza indicato, dopo essersi resi conto di aver dato delle risposte non conformi a quanto le linee guida prescrivano per il caso specifico. Tale valore può essere spiegato dalla presunzione dei medici, rilevata dalle risposte al questionario finale e dalla potenziale ambiguità delle linee guida. Tuttavia, per i medici che frequentamente consultano riviste scientifiche risulta molto più semplice cambiare idea e modificare le proprie scelte una volta ricevuto un suggerimento esterno quale il contenuto delle linee guida. Dal momento che la divulgazione delle linee guida accresce il il livello di coordinazione, gli ospedali dovrebbero valutare programmi efficaci che ne favoriscano la diffusione. Infine l’ultimo esperimento mira a misurare, tramite diversi incentivi monetari quali un sistema a tariffazione o un salario fisso, la propensione dei medici a rifiutare la presa in carico di un paziente. I medici all’interno di uno scenario ipotetico nel quale rischiano di essere denunciati per malasanità devono scegliere se ospedalizzare ciascuno dei 9 pazienti presentati con diverse diagnosi associate ad un diverso livello di gravità. Nel caso in cui scelgano di prendere il singolo paziente in carica, dovranno decidere in che misura trattarlo fornendo numericamente l’ammontare delle prestazioni da erogare. In un secondo momento viene introdotta una probabilità che il medico possa essere denunciato per non aver preso in carico il paziente, crescente al crescere della gravità della malattia. I risultati mostrano che la frequenza con cui un paziente viene preso in carico è maggiore in un sistema a tariffazione rispetto ad un sistema a salario. I medici sembrano essere indifferenti all’introduzione della probabilità di essere denunciati per non aver preso in carica il paziente che tuttavia accresce il livello di prestazioni fornite. Le possibili strategie da adottare per l’ospedale variano a seconda degli obiettivi che l’ospedale si propone.

Three essays on experimental health economics / Romeo, Domenica. - (2021 Apr 15).

Three essays on experimental health economics

Romeo, Domenica
2021-04-15

Abstract

This paper is divided in three parts, each referred to a different experiment. The experiments were conducted at the main hospital of Reggio Calabria, thanks to an agreement signed by the same hospital and the Mediterranean University. In the first experiment, I investigate for the first time the level of coordination among the Italian health professionals, by using a simple, portable and incentive compatible tool such as the coordination game, widely adopted in previous literature, though in other settings. By identifying areas where relational coordination plays an important role, I have chosen to focus on orthopaedics, paediatrics and oncology. For each of the afore mentioned departments, I designed a specific vignette (i.e., a description of a hypothetical situation which physicians could face on their job) and a set of four actions which a physician can take in response to that. In this framed field experiment, physicians are asked to evaluate the appropriateness of each of the possible actions on a scale of one to four in order to match the valuations of most respondents (i.e. modal answer). Data show that the frequency mean of coordination across the experiment is 52% and that coordination increases when physicians do often exchange opinions and share positive feedbacks with colleagues. In addition, the presence of a leader in the medical department, facilitates coordination. Evidence provided by this study shows that coordination in the Italian health sector could be missing, which makes the need for action all the more acute. In the second experiment I test whether physicians act according to national guidelines. By submitting the same scenarios proposed in the first experiment, physicians can receive information concerning what it should be done, in the specific situation, according to national guidelines (i.e. providing them with the written guidelines) and eventually change decisions. Results show that only 8% of the physicians did not want to know national guidelines content. However, only 21% of the subjects (with a choice frequency of 23%) decided to change their appropriateness judgment after realizing they were in contrast with guidelines in either one or two vignettes. Overconfidence and perceived guidelines ambiguity could explain such a low value. Finally, the more physicians consult scientific sources (i.e. signal of open-mindedness), the more willing to accept suggestions, coming from guidelines reading, and to change their decisions when they are wrong. Since guidelines dissemination is shown to increase the level of coordination between physicians, hospitals should consider effective programmes to spread their knowledge. Finally, in the third experiment, I test whether and to which extent the adoption of fee-for-service or salary system can induce physicians to practice patient dumping. In an artefactual field experiment, physicians facing the possibility of being sued for medical malpractice decide whether or not to provide medical services for patients with different state of health. Also, I check whether the introduction of the risk of being sued for a physician for having practiced dumping can have effect on his behaviour. Results show that dumping is more often observed under Salary than under Ffs. However, physicians seem to be insensitive to the introduction of dumping liability under the same incentive mechanism, though it seems to trigger a higher amount of services provided. Policy strategies concerning the incentive scheme may vary on the hospital purposes.
15-apr-2021
Settore IUS/09 - ISTITUZIONI DI DIRITTO PUBBLICO
FINOCCHIARO CASTRO, Massimo
SALAZAR, Carmela Maria Giustina
Doctoral Thesis
File in questo prodotto:
File Dimensione Formato  
Romeo Domenica.pdf

accesso aperto

Tipologia: Tesi di dottorato
Licenza: DRM non definito
Dimensione 2.57 MB
Formato Adobe PDF
2.57 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12318/105756
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact