Vai ai contenuti. | Spostati sulla navigazione | Spostati sulla ricerca | Vai al menu | Contatti | Accessibilità

| Crea un account

Grassetti, Luca - Battauz, Michela (2004) Day Hospital versus Ordinary Hospitalization: factors in treatment discrimination. [Working Paper] WORKING PAPER SERIES, 7/2004 . , PADOVA (Inedito)

Full text disponibile come:

Documento PDF

Abstract (inglese)

The aim of this article is to highlight one possible use of administrative archives in decision making processes. The phenomenon we want to analyse is the carpal tunnel syndrome surgery treatment. Nowadays, from an operative point of view, two different ways coexist in treating the problem: the day hospital (DH) and the ordinary hospitalization (OH) surgery. Causes of this dichotomy are not so clear. In particular, because of the simplicity of the intervention we can hypothesize that, given the effects of some observed factors, no other significant differences should be observed between different hospitals. We extract 16431 observations from the administrative archive of Region Lombardia Data used for this analysis refer to year 2002. The observations are clustered in 128 hospitals. The binary response variable, the hospitalization regimen, is modelled with a logistic regression. Fixed all the other observed variables (referred both at hospital and at patient level), we identify by a random coeffcient a signicant hospitals effect. Day hospital treatment presents lower costs for the national health service. The suspect of the administration is that DRG reinboursement system can create the premise for acting opportunistically. Administrations of hospitals are induced to choice OH instead of DH in order to obtain major reinboursement. The identication of hospitals presenting higher probability (resulting from the random effect values) in deciding for a OH instead of the more common DH could be useful for the administrative control system. Maximum likelihood results will be presented and different methods of estimation of random effects will be compared.

Statistiche Download - Aggiungi a RefWorks
Tipo di EPrint:Working Paper
Anno di Pubblicazione:07 Luglio 2004
Parole chiave (italiano / inglese):administrative archive, bayesian approach, binary data, carpal tunnel syndrome, generalized mixed models, maximum likelihood.
Settori scientifico-disciplinari MIUR:Area 13 - Scienze economiche e statistiche > SECS-S/01 Statistica
Struttura di riferimento:Dipartimenti > Dipartimento di Scienze Statistiche
Codice ID:7047
Depositato il:02 Set 2014 14:59
Simple Metadata
Full Metadata
EndNote Format

Download statistics

Solo per lo Staff dell Archivio: Modifica questo record