Abstract

Background: Pneumonia severity index (PSI) estimates the risk of 30-day mortality in patients with pneumonia. In this study, we aim to develop a simplified version of the PSI (G-PSI) to estimate the risk of mortality in geriatric patients with community-acquired pneumonia (CAP).

Methods: This retrospective study included 186 patients aged 65 and older with a diagnosis of CAP. PSI score and 30-day mortality rate of each patient were calculated. PSI parameters were analyzed using univariate regression analysis and the G-PSI scoring system was established to predict 30-day mortality and compared with PSI.

Results: Significant effectiveness of the values of cancer (odds ratio (OR) = 3.67; 95% confidence interval (CI): 1.42–9.48), altered mental status (OR = 0.79; 95% CI: 0.68–0.92), systolic blood pressure (OR = 0.98; 95% CI: 0.97–1.00), haematocrit (Hct) (OR = 0.87; 95% CI: 0.81–0.93) and blood urea nitrogen (BUN) (OR = 1.04; 95% CI: 1.02–1.06) were observed for predicting mortality in univariate regression analysis. G-PSI scoring system, like PSI score (if cancer + 30 points, if altered mental status +20 points, if systolic blood pressure < 90 mmHg + 20 points, if Hct < 30% + 10 points and if BUN ≥ 30 mg/dl + 30 points) was created. The area under the receiver operating characteristic curve (AUC) was 0.762 (95% CI 0.673–0.851), revealing the excellent discriminatory ability of the G-PSI model. The AUC for the PSI score was 0.719 (95% CI 0.631–0.806). When G-PSI and PSI were compared, G-PSI had a high level of significance in predicting 30-day mortality.

Conclusion: Calculated with only five parameters from standard PSI information, G-PSI accurately displays the 30-day mortality risk of geriatric patients with CAP. The applicability of the G-PSI is easier in a busy emergency service environment with similar prognostic accuracy and clinical prediction.