24 3 
Home Page  

  • SciELO

  • Google
  • SciELO
  • Google


Anales de Medicina Interna

 ISSN 0212-7199

TABERNERO HUGUET, E. et al. Evaluation of a computer predictive model to stratify risk and site of care in community acquired pneumonia. []. , 24, 3, pp.120-124. ISSN 0212-7199.

Background: CAP is a common disorder with a great variability in clinical practice. The decision regarding the appropriate site of care is the most important for the level of treatment and costs. Recently a hospital in our region (Hospital de Galdakao) developed a prediction rule based on the Pneumonia Severity Index (PSI) plus some additional criteria (hypoxemia <60, shock, previous correct treatment failure, social problems or inability to maintain oral intake, pleural effusion or unstable comorbidities) with an easy computer program to classify patients to be hospitalized or not. Objective: Evaluate that computer program in the emergency department of our hospital. Results: We included between December 02 and December 04 662 prospective patients with CAP admitted to our emergency department, 58 had a different final diagnosis. 285 (47%) were treated on outpatient basis. Readmission rate was 6%. There was no mortality in this group. 319 (53%) patients were hospitalized, 97 were PSI low risk patients (I-II), 61 of them were admitted to hospital because additional criteria. 45% of these "low risk patients" had significant complications. These results are similar to those obtained in Galdakao*. Conclusions: The application of this computer risk stratifying program to assess admission to hospital in CAP is simple useful, secure and can be export to different settings. Additional criteria to PSI are necessary to detect low risk patients that complicate.

: Clinical prediction rule; Hospitalization; Community acquired pneumonia; Emergency department.

        · |     · |     · ( pdf )

 

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License