Your browser doesn't support javascript.
loading
In-hospital death caused by pancreatic cancer in Spain: application with a bayesian network.
Alvaro-Meca, A; Gil-Prieto, R; Gil de Miguel, A.
Afiliación
  • Alvaro-Meca A; Department of Preventive Medicine and Public Health, University Rey Juan Carlos, Spain.
Int J Biomed Sci ; 7(2): 125-30, 2011 Jun.
Article en En | MEDLINE | ID: mdl-23675228
Pancreatic cancer is one of the least common tumors (2.1%), but it remains one of the most lethal. This lethality is primarily due to late stage diagnosis in the vast majority of patients. Here we demonstrate, using a Bayesian network, that we can determine a posteriori, with a high probability of success, the probability of in-hospital death of pancreatic cancer in hospitals across Spain with information related to the type of admission, the type of procedure, the primary diagnosis or the Charlson co-morbidity index. The advantages of using a Bayesian network are that it allows us to examine multiple hypotheses and to measure the effect of the introduction of variables on our hypotheses. Being able to determine deceases in the probability of survival based on hospital admission data, such as the diagnosis resulting in the present admission or the presence of co-morbidities, could facilitate the detection of deficiencies in the patient treatment and improve hospital management. Moreover, the control of related co-morbidities may have an impact on the in-hospital deaths of these patients.
Palabras clave

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Int J Biomed Sci Año: 2011 Tipo del documento: Article País de afiliación: España Pais de publicación: Estados Unidos

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Int J Biomed Sci Año: 2011 Tipo del documento: Article País de afiliación: España Pais de publicación: Estados Unidos