Shock index, modified shock index, and age shock index for prediction of mortality in Emergency Severity Index level 3.

OBJECTIVES: To elucidate the predictive capability of shock index (SI), modified SI (MSI), and age SI for mortality in patients assigned to Emergency Severity Index (ESI) level 3 patients.

METHODS: This was a retrospective medical record review performed in an academic internal medicine emergency department in Kerman, Iran. All patients older than 14 years triaged to ESI level 3 were enrolled in the study. Triage time vital signs were used to calculate SI, MSI, and age SI. The primary outcome was in-hospital mortality.

RESULTS: A total number of 3375 patients were enrolled in the study, in which 84 (2.5%) died during hospital stay. In the adjusted multivariate analysis, age SI, systolic blood pressure (SBP), and sex were independently associated with mortality, with P values (odds ratio [95% confidence interval]) of <.001 (1.03 [1.01-1.04]), .003 (0.97 [0.96-0.99]), and .04 (1.61 [1.01-2.59]), respectively. Receiver operating characteristic curve showed an area under curve of 0.717 for the 3-variable final model and an area under curve of 0.678 for age SI in mortality prediction.

CONCLUSIONS: In ESI level 3 patients, age SI and SBP showed to be better than SI or MSI in predicting mortality. However, because their predictive capability was modest, age SI or SBP should be considered adjuncts to sort actions in favor of patients with higher risk for mortality.