The Geriatric Emergency Department Innovations through Workforce, Informatics and Structural Enhancements (GEDI WISE) initiative utilizes specially trained nurse liaisons to perform a series of validated tests on emergency department (ED) geriatric patients in order to identify geriatric specific needs and provide care coordination. The objective of this study is to evaluate the ability of these geriatric assessments to predict hospitalization in geriatric patients with normal vital signs as a marker of possible “social admissions.” This was a prospective observational study of geriatric ED patients from April 2013 to March 2015. Geriatric patients in this study were evaluated by a geriatric nurse liaison (GNL) and had geriatric-specific assessments performed and recorded in the electronic medical record (EMR) including: Identification of Seniors at Risk (ISAR) Short Portable Mental Status Questionnaire (SPMSQ), Katz Activities of Daily Living (Katz ADL), Beers Criteria, Timed Up and Go test (TUG), and Confusion Assessment Method (CAM). All patients age 65+ with normal vital signs, who were evaluated by the GNL were included in this study. Demographics and vital sign variables were categorized as normal and abnormal according to previously reported ED literature, and geriatric assessments results were dichotomized. The primary outcome was hospitalization, including inpatient and observation status. Bivariate analysis was performed between geriatric assessment results and the hospitalization outcome. To adjust for demographics and emergency severity index (ESI), logistic regression was then performed for each of the assessments. There were 48,229 geriatric ED patients during the study period, 2,559 received GNL evaluation. The abnormal results of completed assessments are as follows: ISAR 1776/2559 (69%), SPMSQ 496/2,294 (22%), Katz ADLs 1,016/2,089 (49%), Beers 789/1339 (59%), TUG 732/778 (94%) CAM 44/126 (35%). For patients with normal vital signs, the unadjusted odds ratios showed significantly increased odds of hospitalization for all geriatric assessments. After adjusting for age, race, and ESI, only ISAR and Katz ADLs were significantly associated with increased odds of hospitalization (Table). In geriatric ED patients with normal vital signs, most geriatric assessments do a poor job of predicting hospitalization after adjusting for ESI and demographics. This may be because GNLs are able to effectively address the issues they identify through these assessments to prevent “social admisions.” Patients with difficulty performing their ADLs may represent a group for which it is more difficult to effectively coordinate care in the ED. Despite previous literature showing the poor performance of ISAR, in this sample it is associated with increased odds of hospitalization, and may have merit as a screening test for patients at high risk for hospitalization.TableOdds ratio for hospitalizationAbnormal geriatric AssessmentUnadjusted Odds Ratio (95% CI)Adjusted Odds Ratio (95% CI)*ISAR2.87 (2.26-3.65)2.44 (1.89-3.15)SPMSQ1.20 (1.03-1.41)1.15 (0.97-1.36)CAM5.52 (1.09-28.00)4.19 (0.71-24.57)Katz ADL1.66 (1.29-2.11)1.58 (1.22-2.04)TUG test2.12 (1.05-4.29)2.07 (0.97-4.42)Beers criteria1.45 (1.08-1.97)1.29 (0.94-1.77)*Odds ratio adjusted for age, race and Emergency Severity Index (ESI). Open table in a new tab