The concept of a joint approach in regard to Multilevel Models and Stochastic Frontiers developed in light of the increasing levels of interest in costs related to health care services, including hospitals, over the last few years. At the same time, both consumers and policy makers prioritize the quality of these services, and a holistic approach is required to identify areas for improvement in this regard. Quality in healthcare services means the ability to meet specific requirements, and it is the result of scientific, technical and technological, organizational, procedural and relational factors , where the human variable plays a primary role, interacting closely in the production process . In the present day many industrialized countries have healthcare systems which designate resources to hospitals according to predefined tariffs for each pathology. For example, the Lombardy Region in Northern Italy establishes an annual tariff for Diagnostic Related Groups (DRG). Therefore, the structure of the public health system depends on national and regional political decisions, but it becomes self-governing from an operational point of view. Accordingly, hospitals can be compared to firms, thus admitting functional models in an entrepreneurial way and introducing competition between the various healthcare structures. Hospitals try to devise an organizational model which allows the reduction of costs by optimizing the use of available resources, while simultaneously increasing patient satisfaction by providing optimal medical assistance. Healthcare is currently following this trend and in many countries it is based on a mixed welfare system made of private profit-oriented agents, private non-profit companies and public hospitals. The purpose of this work is to create decisional models similar to those used in financial planning, with a large number of variables, with the aim of estimating the risks involved in some enterprises, and devising accordingly appropriate allocations of funds. This study is on the basis of the following structure: the first chapter is dedicated to the multilevel model. The multilevel model is contextualized in the measure of the health care, so the first aspect considered is the adjustment of data according to patient-specific and hospital-specific variables. The second chapter is dedicated to the stochastic frontier, and is a theoretic description of this model with a different possible application. The interesting aspect is the possibility of seeing the hospital in terms of distance from its frontier of optimality. In this way we take the economic aspects of the structure into account, and we can see its effectiveness in terms of efficiency. It is clear that inside the frontier there is no consideration for the hierarchical structure of the data, but is important to not neglect the real structure of data (as we have explained in the previous chapter). This is an evident reason that allow us to consider both the models together, with all the strength stressed, and all the weaknesses overcome. Chapter number three is about the characterization of data, while chapter four comprises a description of the model that we have applied to the data. We have here proposed a section with possible future developments that we have elaborated during this work. Finally chapter five is about the results we obtained through the analysis. The conclusion section contains some considerations we have made in light of the whole work.

(2011). Multilevel and stochastic frontier models: a comparison and a joint approach of their performances when investigating panel data. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2011).

Multilevel and stochastic frontier models: a comparison and a joint approach of their performances when investigating panel data

ANGELICI, MARTA
2011-01-25

Abstract

The concept of a joint approach in regard to Multilevel Models and Stochastic Frontiers developed in light of the increasing levels of interest in costs related to health care services, including hospitals, over the last few years. At the same time, both consumers and policy makers prioritize the quality of these services, and a holistic approach is required to identify areas for improvement in this regard. Quality in healthcare services means the ability to meet specific requirements, and it is the result of scientific, technical and technological, organizational, procedural and relational factors , where the human variable plays a primary role, interacting closely in the production process . In the present day many industrialized countries have healthcare systems which designate resources to hospitals according to predefined tariffs for each pathology. For example, the Lombardy Region in Northern Italy establishes an annual tariff for Diagnostic Related Groups (DRG). Therefore, the structure of the public health system depends on national and regional political decisions, but it becomes self-governing from an operational point of view. Accordingly, hospitals can be compared to firms, thus admitting functional models in an entrepreneurial way and introducing competition between the various healthcare structures. Hospitals try to devise an organizational model which allows the reduction of costs by optimizing the use of available resources, while simultaneously increasing patient satisfaction by providing optimal medical assistance. Healthcare is currently following this trend and in many countries it is based on a mixed welfare system made of private profit-oriented agents, private non-profit companies and public hospitals. The purpose of this work is to create decisional models similar to those used in financial planning, with a large number of variables, with the aim of estimating the risks involved in some enterprises, and devising accordingly appropriate allocations of funds. This study is on the basis of the following structure: the first chapter is dedicated to the multilevel model. The multilevel model is contextualized in the measure of the health care, so the first aspect considered is the adjustment of data according to patient-specific and hospital-specific variables. The second chapter is dedicated to the stochastic frontier, and is a theoretic description of this model with a different possible application. The interesting aspect is the possibility of seeing the hospital in terms of distance from its frontier of optimality. In this way we take the economic aspects of the structure into account, and we can see its effectiveness in terms of efficiency. It is clear that inside the frontier there is no consideration for the hierarchical structure of the data, but is important to not neglect the real structure of data (as we have explained in the previous chapter). This is an evident reason that allow us to consider both the models together, with all the strength stressed, and all the weaknesses overcome. Chapter number three is about the characterization of data, while chapter four comprises a description of the model that we have applied to the data. We have here proposed a section with possible future developments that we have elaborated during this work. Finally chapter five is about the results we obtained through the analysis. The conclusion section contains some considerations we have made in light of the whole work.
VITTADINI, GIORGIO
MARTINI, GIANMARIA
multilevel model; sthocastic frontier; hierarchical structure; quality; efficacy; effectiveness
SECS-S/01 - STATISTICA
English
Scuola di Dottorato in Statistica e Matematica Applicata alla Finanza
STATISTICA - 11R
23
2009/2010
(2011). Multilevel and stochastic frontier models: a comparison and a joint approach of their performances when investigating panel data. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2011).
File in questo prodotto:
File Dimensione Formato  
Phd_unimib_048382.pdf

accesso aperto

Tipologia di allegato: Doctoral thesis
Dimensione 631.06 kB
Formato Adobe PDF
631.06 kB 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: http://hdl.handle.net/10281/19410
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact