PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of ccforumBioMed CentralBiomed Central Web Sitesearchsubmit a manuscriptregisterthis articleCritical CareJournal Front Page
 
Crit Care. 2010; 14(Suppl 1): P253.
Published online 2010 March 1. doi:  10.1186/cc8485
PMCID: PMC2934406

A new severity score for community-acquired pneumonia: PARB score

Introduction

Community-acquired pneumonia (CAP) is one of the most common diseases encountered in the emergency department (ED). For the ED physician, it is quite difficult to estimate the severity of CAP patients in the ED because most CAP patients present at an earlier stage of disease. Although several societies such as ATS, IDSA and BTS have established severity scores for CAP, those scores are too complicated to calculate easily, especially in a busy department such as the ED. Therefore, a new simple and useful severity score for CAP is necessary for ED physicians. The purpose of our study is to develop a new severity score for CAP using 30-day mortality or patients needing more than 2 weeks oxygen therapy as the main outcomes.

Methods

We derived a prediction rule for outcomes by analyzing data on 243 adult inpatients with CAP between March 2006 and November 2008 in our hospital. Patients' charts were abstracted to collect data relating to demographics, history, coexisting illnesses, laboratory results, and radiographic findings. Significant predictors of outcomes were identified through logistic regression analyses.

Results

The following were independently associated with outcomes: presence of pleural effusions (P) (OR 5.0 (95% CI: 1.70 to 14.85)); albumin (A), <3.0 g/dl (OR 8.7 (1.43 to 53.29)); respiratory rate (R), >30/minute (OR 8.0 (1.62 to 39.64)); BUN (B), >25 mg/dl (OR 3.4 (1.07 to 11.07)). Patients were assigned according to whether they had these factors or not from class 0 to class IV (class 0 means the CAP patient had no these factors, class IV means the CAP patient had all these factors). Patients in risk class 0 to IV had a predicted probability of death or >2 weeks oxygen therapy of 0.03%, 0.08%, 52.2%, 75.0% and 100%, respectively. The sensitivity and specificity of this model were 36.0% and 99.0%, respectively. Accuracy of this model was 91.9%. The area under the ROC curve was 0.8705.

Conclusions

The PARB score can accurately predict patients with CAP who are at risk for death or who will need more than 2 weeks oxygen therapy. This score may help ED physicians provide adequate treatment for patients with CAP.

References


Articles from Critical Care are provided here courtesy of BioMed Central