ABSTRACT

Background: Community acquired pneumonia is a major health problem with high mortality rate, especially in Indonesia. The CURB-65 score is a widely used scoring system, but has some drawbacks so a new scoring system is needed to assess the severity of community pneumonia. This study aims to assess the expanded-CURB-65 scoring system as a predictor of 30-days mortality of community pneumonia.

Aim : To evaluate calibration and discrimination performance of the expanded-CURB-65 score in predicting 30 days mortality of community acquired pneumonia patients at the National Center General Hospital dr.Cipto Mangunkusumo.

Metho : This study was a prospective cohort study with the study subjects of community pneumonia patients who came to the Emergency Room (ER), pulmonary polyclinics or hospitalized in RSCM. The assessed outcome was patient mortality within 30 days. Discrimination performance  of the expanded-CURB-65 score assessed using the area under the curve (AUC). Calibration performance was evaluated with calibration plot and Hosmer-Lemeshow test.

Results : 267 patients participated in the study with a mortality rate of 31.5%. Calibration plot of expanded-CURB-65 score showed r= 0,9697 and Hosmer-Lemeshow test showed p = 0,210. Discrimination was shown by ROC curve with AUC 0,701 (CI95% 0,633-0,768). Conclusion: Mortality increases with increasing risk class of expanded-CURB-65. Expanded-CURB-65 showed a good calibration and discrimination performance in predicting 30-day mortality higher in community acquired pneumonia patients in Cipto Mangunkusumo Hospital.

Keywords : Community acquired pneumonia patient, 30 days mortality, expanded-CURB-65 score