Development of a Clinical Score for Predicting 28-Day Mortality in Geriatric Sepsis Patients; a Cohort study.

Introduction: Sepsis is a significant and common cause of death and burden among critically ill patients, which has increasing incidence and mortality in adults over 60 and advanced age. This study aimed to develop an easy-to-use clinical tool for assessing 28-day mortality risk in older sepsis patients upon their initial assessment in the emergency department (ED). Method: A retrospective cohort study was conducted using electronic medical records of older (≥60 years) ED patients with suspected sepsis from August 1, 2018, to December 31, 2018. A new prediction score was formulated based on the logistic coefficients of clinical predictors through multivariable regression analyses. Then, the score's screening performance was evaluated and compared to existing scoring systems; Systemic Inflammatory Response Syndrome (SIRS), quick Sequential Organ Failure Assessment (qSOFA), National early warning score (NEWS), and The Ramathibodi early warning score (REWS); using receiver operating characteristic curve analysis (AuROC). Result: The study included 599 patients with the mean age of 77.13 (range: 60-101) years (56.43% male) and an overall 28-day mortality rate of 7.01%. The newly developed prediction score had seven independent predictors of 28-day mortality: malignancy, dependent status, heart rate, respiratory rate, oxygen saturation, consciousness, and lactate, which demonstrated excellent discriminative ability (AuROC: 0.87, 95% confidence interval (CI): 0.82 - 0.92), significantly outperforming SIRS (AuROC: 0.62), qSOFA (AuROC: 0.72), NEWS (AuROC: 0.74), and REWS (AuROC: 0.71), all with p-values <0.01. The score allowed risk stratification into low-risk (positive likelihood ratio (LR+): 0.37, 95% CI: 0.24 - 0.58) and high-risk (LR+: 4.14, 95% CI: 3.14 - 5.44) groups with sensitivity of 69.0% and specificity of 83.3% at a cut-off point of 6. Conclusion: The novel prediction score demonstrates a remarkable ability to predict 28-day mortality risk in older sepsis patients during their initial ED assessment, offering potential for improved risk stratification and treatment guidance in older patients.