The Clinical Impact of the Pneumonia Severity Index and the CURB-65 for Making Admission Decisions

ABSTRACT

PURPOSE: The aim of our study was to examine the discrepancies between the Pneumonia Severity Index (PSI) score, the CURB-65 [confusion, blood urea nitrogen, respiratory rate and blood pressure with age (65 years)] and the clinical judgment for making admission decisions, and we also wanted to evaluate the value of the PSI and CURB-65 as the indicators for detecting inappropriate hospitalization among the patients with low risk CAP. METHODS: The medical records of 286 patients who visited the emergency department with CAP were identified by their PSI and their CURB-65, and we reviewed whether there existed a clinical basis to justify their hospitalization. RESULTS: Of a total 286 patients, 225 patients were admitted to the hospital. Sixty five patients with a PSI of I or II, and 117 patients with a CURB-65 of 0 or 1 were admitted by clinical judgment. The factors that justified admission of a lower risk CAP group were medical conditions other than CAP (56.9% vs 70.1%; PSI vs CURB-65), failure of outpatient therapy (24.6% vs 19.6%), social needs (6.2% vs 2.6%) and hypoxemia in 1 patient (1.5% vs 0.9%, respectively). Of the patients in the low risk group according to the CURB-65, one five patient was hospitalized for suspicion of sepsis. The positive predicted values of the PSI and CURB- 65 for inappropriate hospitalization of low risk CAP patients were 10.8% and 5.9%, respectively. CONCLUSION: The PSI and CURB-65 had unacceptably low positive predictive values and this was due to the comorbid conditions that required in-hospital care for the patients with low risk CAP. These indices did not supersede clinical judgment for making the decision to hospitalize low risk patients with CAP.