ICEBERG emergency room screening tool for early adverse outcome prediction in older patients admitted to acute care: a multi-center study.

Abstract

BACKGROUND The ICEBERG tool was initially validated in two small pilot studies to address the lack of a comprehensive geriatric screening tool in emergency settings. The present study builds on the second pilot study and extends it to a larger, multicentre sample. METHODS We report results from a large ICBERG tool validation study across three emergency rooms (ERs) including 1,664 patients aged 70 years and older. The tool targets 9 domains and is administered by ER physicians or specialized nurses. To assess criterion validity-the extent to which ICEBERG scores are associated with relevant clinical outcomes-we compared patients who scored below versus above the median (< 10; ≥10) ICEBERG score for six key clinical outcomes: length of stay in acute care, nursing care in minutes, one-on-one nursing care, in-hospital mortality, discharge to nursing home, and re-admission within 30 days. Negative binomial regression was used for the outcomes length of stay and nursing care in minutes. Logistic regression was used for the other outcomes. All analyses were adjusted for age and sex. RESULTS Patients with ICEBERG scores of 10 or higher had significantly longer stays in acute care (8.9 vs. 6.6 days), required more nursing care (56.4 vs. 32.8 hours), had higher odds of one-on-one nursing care (Odds Ratio, OR = 2.85), in-hospital mortality (OR = 1.89), discharge to a nursing home (OR = 3.20) or being readmitted within 30 days (OR = 1.81). CONCLUSIONS Based on our findings the ICEBERG tool identifies older patients with a geriatric risk profile at ER on all key clinical outcomes tested.