Predictive Performance of Sepsis Scoring Tools for In-Hospital Mortality Among Patients With Suspected Sepsis...

Fiji has a high morbidity and mortality sepsis burden, but little is known about the utility of prognostic tools for septic patients presenting to hospital in the limited-resource Fijian context. This study aims to compare the ability of the quick sequential organ failure assessment (qSOFA), systemic inflammatory response syndrome (SIRS) and national early warning score (NEWS) at predicting in-hospital mortality.

A prospective observational study was conducted over 3 months in 2024 at a single Fijian tertiary hospital. Data was collected using an online tool from patients who presented to the Emergency Department with suspicion of sepsis and subsequently required admission. The prognostic accuracy at detecting in-hospital mortality was assessed and compared for qSOFA, SIRS and NEWS.

A total of 181 cases were included with an overall in-hospital mortality of 17.13%. In-hospital mortality in patients with a qSOFA ≥ 2 was 16 (34%), SIRS ≥ 2 was 25 (19.2%) and NEWS ≥ 7 was 23 (35.9%). qSOFA ≥ 2 yielded the highest specificity (79%, 95% CI 72%-86%) and SIRS ≥ 2 the highest sensitivity (81%, 95% CI 63%-93%). The NEWS ≥ 7 yielded intermediate sensitivity (74%, 95% CI 55%-88%) and specificity (73%, 95% CI 66%-80%). The AUROC for NEWS ≥ 7 was highest at 0.77 (0.68-0.87) with qSOFA ≥ 2 and SIRS ≥ 2 being 0.69 (0.58-0.79) and 0.62 (0.51-0.73) respectively.

NEWS demonstrated higher prognostic accuracy compared to qSOFA and SIRS in predicting in-hospital mortality for suspected sepsis patients. Our study provides a benchmark for future ED-based sepsis research in the Pacific.