Clinical and Care Coordination Predictors of Adverse Outcomes in a Geriatric Emergency Department.

OBJECTIVES: Identifying high-risk older adults in emergency settings is crucial for optimizing care and preventing complications. This study aimed to describe the characteristics of patients in a geriatric emergency department (GED) and to identify factors associated with hospital admission, prolonged length of stay, unplanned intensive care unit admission, and in-hospital mortality. DESIGN: Retrospective cohort study. SETTING AND PARTICIPANTS: A total of 9814 visits by patients aged ≥70 years at a tertiary GED in Brazil (August 2017-October 2024). METHODS: We assessed baseline sociodemographic and clinical features, vulnerability measures Identification of Seniors at Risk (ISAR), Fatigue, Resistance, Ambulation, Illness, Loss of weight (FRAIL) scale, brief Confusion Assessment Method (bCAM), and Physical impairment, Recent hospitalization, Older age (≥ 90), Acute mental alteration, Getting thinner, and Exhaustion (PRO-AGE) score, and communication with primary physicians. Multivariable logistic regression was used to identify independent predictors for admission (primary outcome) and in-hospital outcomes. RESULTS: Among 9814 visits, 31.5% resulted in hospital admission. Factors significantly associated with admission included frailty adjusted odds ratio (aOR), 1.99; 95% CI, 1.68-2.37, delirium (aOR, 2.12; 95% CI, 1.70-2.65), higher ISAR score (aOR, 1.25; 95% CI, 1.19-1.32), and contact with primary physician (aOR, 3.42; 95% CI, 2.86-4.10). In-hospital mortality was predicted by age (aOR per year, 1.05; 95% CI, 1.02-1.09) and the Charlson Comorbidity Index (aOR, 1.19; 95% CI, 1.09-1.34), whereas having a primary physician, regardless of successful contact and agreement about admission, was associated with lower odds of mortality. Prolonged stay was associated with frailty (aOR, 1.49; 95% CI, 1.09-2.04), ISAR score (aOR, 1.17; 95% CI, 1.08-1.26), and delirium (aOR, 1.36; 95% CI, 1.05-1.76). Male sex was a predictor of unplanned intensive care unit admission (aOR, 1.48; 95% CI, 1.02-2.12). CONCLUSIONS AND IMPLICATIONS: Geriatric vulnerability screening strongly predicted adverse outcomes. Although successful contact with the primary physician was associated with a higher likelihood of admission, the existence of an established primary care relationship significantly predicted lower in-hospital mortality. GED models that combine systematic screening with structured communication and coordination may reduce avoidable admissions and improve outcomes.