Geriatric Emergency Department Research Articles

368 Determining the Combined Impact of Geriatric-Specific Predictors of Hospitalization Among Geriatric Emergency Department Patients

368 Determining the Combined Impact of Geriatric-Specific Predictors of Hospitalization Among Geriatric Emergency Department Patients

217 The Impact of Specialized Geriatric Consultation in a Level 1 Geriatric Emergency Department on the Cost of Care

Adults older than 65 account for almost 20% of all emergency department (ED) visits, over 23.1 million in 2018, and represent a disproportionate number of potentially avoidable inpatient admissions through the ED when compared to younger patients. The purpose of this study was to evaluate the effect of Geriatric Emergency Nurse Initiative Expert (GENIE) consultation during an ED visit on the total ED and inpatient cost of care.

95 A Tender-Loving-Care Volunteer Program to Provide Non-Clinical, Supportive Interventions to Older Adults in the Emergency Department

To address the needs of an aging population in the acute care setting, ACEP’s geriatric emergency department (ED) guidelines recommend developing age-appropriate care models. Cedars-Sinai ED developed and implemented The Tender-Loving-Care (TLC) volunteer program, a unique imitative that trains volunteers to provide non-clinical, supportive interventions to improve the ED care and experience for geriatric patients. Our research objective was to describe program implementation and characterize early findings about supportive interventions that TLC volunteers provide.

156 Dissemination and Implementation of Age-Friendly Care and Geriatric Emergency Department Accreditation at Veterans Affairs Hospitals

In 2018, the US Department of Veterans Affairs (VA) Offices of Emergency Medicine and Geriatrics & Extended Care partnered to improve acute, unscheduled care for older Veterans. The goals of the partnership were to create a VA core team to promote best geriatric emergency care practices via standardization, education, and Geriatric Emergency Department (ED) Accreditation (GEDA) through the American College of Emergency Physicians. The objective of this abstract will be to describe the current progress of GED implementation and dissemination at the VA.

Telehealth for geriatric post-emergency department visits to promote age-friendly care.

To describe a feasibility pilot study for older adults that addresses the digital divide, unmet health care needs, and the 4Ms of Age-Friendly Health Systems via the emergency department (ED) follow-up home visits supported by telehealth. Data sources were a pre-implementation site survey and pilot phase individual-level patient data from six US Department of Veterans Affairs (VA) EDs. A pre-implementation survey assessed existing geriatric ED processes. In the pilot called SCOUTS (Supporting Community Outpatient, Urgent care & Telehealth Services), sites identified high-risk patients during an ED visit. After ED discharge, Intermediate Care Technicians (ICTs, former military medics), performed follow-up telephone, or home visits. During the follow-up visit, ICTs identified "what matters," performed geriatric screens aligned with Age-Friendly Health Systems, observed home safety risks, assisted with video telehealth check-ins with ED providers, and provided care coordination. SCOUTS visit data were recorded in the patient's electronic medical record using a standardized template. Sites were surveyed via electronic form. Administrative pilot data extracted from VA Corporate Data Warehouse, May-October 2021. Site surveys showed none of the EDs had a formalized way of identifying the 4M "what matters." During the pilot, ICT performed 56 telephone and 247 home visits. All home visits included a telehealth visit with an ED provider (n=244) or geriatrician (n=3). ICTs identified 44 modifiable home fall risks and 99 unmet care needs, recommended 80 pieces of medical equipment, placed 36 specialty care consults, and connected 180 patients to a Patient Aligned Care Team member for follow-up. A post-ED follow-up program in which former military medics perform geriatric screens and care coordination is feasible. Combining telehealth and home visits allows providers to address what matters and unmet care needs.

Establishing the Geriatric Emergency Department Intervention in Queensland emergency departments: a qualitative implementation study using the i-PARIHS model

BackgroundFrail older adults require specific, targeted care and expedited shared decision making in the emergency department (ED) to prevent poor outcomes and minimise time spent in this chaotic environment. The Geriatric Emergency Department Intervention (GEDI) model was developed to help limit these undesirable consequences. This qualitative study aimed to explore the ways in which two hospital implementation sites implemented the structures and processes of the GEDI model and to examine the ways in which the i-PARIHS (innovation-Promoting Action on Research Implementation in Health Services) framework influenced the implementation.MethodsUsing the i-PARIHS approach to implementation, the GEDI model was disseminated into two hospitals using a detailed implementation toolkit, external and internal facilitators and a structured program of support. Following implementation, interviews were conducted with a range of staff involved in the implementation at both sites to explore the implementation process used. Transcribed interviews were analysed for themes and sub-themes.ResultsThere were 31 interviews with clinicians involved in the implementation, conducted across two hospitals, including interviews with the two external facilitators. Major themes identified included: (i) elements of the GEDI model adopted or (ii) adapted by implementation sites and (iii) factors that affected the implementation of the GEDI model. Both sites adopted the model of care and there was general support for the GEDI approach to the management of frail older people in the ED. Both sites adapted the structure of the GEDI team and the expertise of the team members to suit their needs and resources. Elements such as service focus, funding, staff development and service evaluation were initially adopted but adaptation occurred over time. Resourcing and cost shifting issues at the implementation sites and at the site providing the external facilitators negatively impacted the facilitation process.ConclusionsThe i-PARIHS framework provided a pragmatic approach to the implementation of the evidenced-based GEDI model. Passionate, driven clinicians ensured that successful implementation occurred despite unanticipated changes in context at both the implementation and host facilitator sites as well as the absence of sustained facilitation support.

Trends in Geriatric Trauma Emergency Department Admissions During COVID-19.

The impact of coronavirus disease-2019 (COVID-19) on geriatric trauma presenting to the emergency department is unknown. To examine geriatric trauma emergency department admission trends during the COVID-19 pandemic. This retrospective, observational, comparison study was conducted in an academic emergency department in Turkey. Trauma patients 65 years and older who presented to the emergency department within 1 year of March 12, 2020, were included. Patients admitted in the same date range in the previous year were included as the control group. The characteristics of the patients, injured area, and injury mechanisms were compared. Geriatric trauma admissions decreased (relative risk = 0.71, odds ratio [OR] = 0.69 [95% confidence interval, CI: 0.62, 0.77], p < .001). According to the type of injury, there was no significant difference in admissions to the emergency department (p = .318). During the pandemic, there was an increase in falls and a decrease in stab wounds and gunshot wounds (p = .001). Multiple trauma (OR = 5.56 [95% CI: 3.75, 8.23], p < .001), fall (OR = 2.41 [95% CI: 1.6, 3.73], p < .001), and-assault related injuries (OR = 4.43 [95% CI: 2.06, 9.56], p < .001) were determined as factors that increased the admissions to the emergency department compared with the prepandemic. Although geriatric trauma emergency department admissions decreased during the pandemic, those due to falls and assaults increased. Although curfews and social isolation resulted in a decrease in penetrating injuries, assault-related trauma has increased.

Translation of the geriatric emergency department intervention into other emergency departments: a post implementation evaluation of outcomes for older adults

BackgroundIncreasing numbers of older adults with complex health deficits presenting to emergency departments has prompted the development of innovative models of care. One such model designed to reduce poor outcomes associated with acute healthcare, is the Geriatric Emergency Department Intervention. This intervention is a nurse-led, physician-championed, Emergency Department intervention that improves the health outcomes for frail older adults in the emergency department.MethodsThis quantitative cohort study aimed to evaluate the healthcare outcomes and costs associated with the implementation of the Geriatric Emergency Department Intervention (GEDI) for adults aged 70 years and over at two hospital sites that implemented the model using the integrated-Promoting Action on Research Implementation in Health Services (i-PARHIS) framework. Hospital A was large teaching hospital located in the tropical north of Australia. Hospital B was a medium sized teaching hospital near Brisbane, Queensland Australia.The effect of the intervention was examined in two ways. Outcomes were compared between: 1) all patients in the pre- and post- implementation periods, and 2) patients seen or not seen by the Geriatric Emergency Department Intervention team in the post-implementation period. The outcomes measured were disposition (discharged home, admitted); emergency department length of stay; hospital length of stay; all cause in-hospital mortality within 28 days; time to re-presentation up to 28 days post-discharge; emergency department and in-hospital costs.Survival analyses were used for the primary and secondary outcome variables and a Cox survival model was used to estimate the associations between variables and outcomes. Multiple regression models were used to examine other secondary outcomes whilst controlling for a range of confounders.ResultsThe Geriatric Emergency Department Intervention was successfully translated into two different emergency departments. Both demonstrated an increased likelihood of discharge, decreased emergency department length of stay, decreased hospital costs for those who were admitted, with an associated reduction in risk of mortality, for adults aged 70 years and over.ConclusionsThe Geriatric Emergency Department Intervention was successfully translated into new sites that adapted the model design. Improvement in healthcare outcomes for older adults presenting to the emergency department was demonstrated, although this was more subtle than in the original model setting.

Balancing vision with pragmatism: The geriatric emergency department guidelines-realistic expectations from emergency medicine and geriatric medicine.

In 2014 the Geriatric Emergency Department (GED) Guidelines were published and endorsed by four major medical organizations. The multidisciplinary GED Guidelines characterize the complex needs of the older emergency department (ED) patient and current best practices with the goal of promoting more cost-effective and patient-centered care. The recommendations are extensive and the vast majority of EDs then and now do not have either the resources nor hospital administrative support to provide this additional service. At the 2021 American Academy of Emergency Medicine's Scientific Assembly, a panel of emergency medicine physicians and geriatricians discussed the GED Guidelines and the current realities of EDs' capacity to provide best practice and guideline recommended care of GED patients. This article is a synthesis of the panel's presentation and discussion. With the substantial challenges in providing guideline recommended care in EDs, this article will explore three high impact GED clinical conditions to highlight guideline recommendations, challenges and opportunities, and discuss realistically achievable expectations for non-Geriatric ED accredited institutions. In 2014 the Geriatric Emergency Department Guidelines were published describing the current best practices for geriatric ED patients. Unfortunately, the vast majority of EDs worldwide do not provide the level of service recommended by the GED guidelines. The GED Guidelines can best be termed aspirational for U.S. EDs at the present time.

Initiation of geriatric emergency department for fitting the gap of geriatric care in Taiwan

Initiation of geriatric emergency department for fitting the gap of geriatric care in Taiwan

Implementation of a geriatric emergency medicine assessment team decreases hospital length of stay

BackgroundPatients over the age of 65 who present to the Emergency Department (ED) are more likely to be admitted to the hospital and, if admitted, often have a longer length of stay (LOS) in the hospital than younger patients. ObjectivesTo determine if assessment and intervention by a Geriatric Emergency Medicine Assessment (GEMA) team would decrease the admission rate and reduce the hospital LOS for admitted geriatric patients. MethodsWe conducted a case-control study of the impact of a GEMA team in a large ED. The team screened patients ≥65 years of age for functional decline to determine the need for targeted interventions. Potential interventions included: occupational therapy consultation in the ED, rehabilitation placement, geriatric clinic referral, and delirium management. Our control population was unassessed geriatric ED patients seen in the six months before and after GEMA team implementation. ResultsA total of 815 patients were assessed between June and November 2019. Assessed patients were more likely to be discharged from the ED (54% vs 29%, OR 2.06). Mean ED LOS was nineteen minutes longer in assessed patients (4.94 vs 4.62 h, p < 0.01). The mean hospital LOS was 25 h less in assessed patients (4.50 vs 5.54 days, p < 0.01). Assessed and unassessed patients who were admitted to the hospital had the same baseline health status as measured by the Charlson Comorbidity Index (median score 2, p = 0.087). The reduction in hospital LOS resulted in an estimated savings of $1.7 million per year using the national average cost for 24 h of inpatient care. ConclusionPatients who were assessed by the GEMA team were more likely to be discharged directly from the ED, and if admitted, hospital LOS was reduced by over 24 h. This indicates that a targeted intervention in the ED can help reduce hospital LOS in geriatric patients and therefore provide cost savings.

Balancing Vision With Pragmatism: The Geriatric Emergency Department Guidelines-Realistic Expectations From Emergency Medicine and Geriatric Medicine

BackgroundIn 2014, the Geriatric Emergency Department (GED) Guidelines were published and endorsed by four major medical organizations. The multidisciplinary GED Guidelines characterized the complex needs of the older emergency department (ED) patient and current best practices, with the goal of promoting more cost-effective and patient-centered care. The recommendations are extensive and most EDs then and now have neither the resources nor the hospital administrative support to provide this additional service. DiscussionAt the 2021 American Academy of Emergency Medicine's Scientific Assembly, a panel of emergency physicians and geriatricians discussed the GED Guidelines and the current realities of EDs’ capacity to provide best practice and guideline–recommended care of GED patients. This article is a synthesis of the panel's presentation and discussion. With the substantial challenges in providing guideline-recommended care in EDs, this article will explore three high-impact GED clinical conditions to highlight guideline recommendations, challenges, and opportunities, and discuss realistically achievable expectations for non–GED-accredited institutions. ConclusionsIn 2014, the GED Guidelines were published, describing the current best practices for GED patients. Unfortunately, most of the EDs worldwide do not provide the level of service recommended by the GED Guidelines. The GED Guidelines can best be termed aspirational for U.S. EDs at the present time.

A physiotherapy-led transition to home intervention for older adults following emergency department discharge: protocol for a pilot feasibility randomised controlled trial

BackgroundOlder adults frequently attend the emergency department (ED) and experience high rates of adverse outcomes following ED presentation including functional decline, ED re-presentation and unplanned hospital admission. The development of effective interventions to prevent such outcomes is a key priority for research and service provision. This paper reports a protocol designed to evaluate the feasibility of conducting a three arm randomised controlled trial (RCT) within the ED setting and in the patient’s home. The interventions are comprehensive geriatric assessment (CGA), ED PLUS and usual care.MethodsThe ED PLUS pilot trial is designed as a feasibility RCT conducted in the ED and Acute Medical Assessment Unit of a university teaching hospital in the mid-west region of Ireland. We aim to recruit 30 patients, aged 65 years and over presenting to the ED with undifferentiated medical complaints and discharged within 72 h of index visit.Patients will be randomised by a computer in a ratio of 1:1:1 to deliver usual care, CGA or ED PLUS during a 6-month study period. A randomised algorithm is used to perform randomization. CGA will include a medical assessment, medication review, nursing assessment, falls assessment, assessment of mobility and stairs, transfers, personal care, activities of daily living (ADLs), social supports and baseline cognition. ED PLUS, a physiotherapist led, multidisciplinary intervention, aims to bridge the transition of care between the index visit to the ED and the community by initiating a CGA intervention in the ED and implementing a 6-week follow-up self-management programme in the patient’s own home following discharge from the ED. The outcomes will be parameters of the feasibility of the intervention and trial methods and will be assessed quantitatively and qualitatively.DiscussionRising ED visits and an ageing population with chronic health issues render ED interventions to reduce adverse outcomes in older adults a research priority. This feasibility RCT will generate data and experience to inform the conduct and delivery of a definite RCT.Trial registrationThe trial was registered in Clinical Trials Protocols and Results System as of 21st July 2021, with registration number NCT049836020.

REcognizing DElirium in geriatric Emergency Medicine: The REDEEM risk stratification score.

The objective was to derive a risk score that uses variables available early during the emergency department (ED) encounter to identify high-risk geriatric patients who may benefit from delirium screening. This was an observational study of older adults age≥75years who presented to an academic ED and who were screened for delirium during their ED visit. Variable selection from candidate predictors was performed through a LASSO-penalized logistic regression. A risk score was derived from the final prediction model, and predictive accuracy characteristics were calculated with 95% confidence intervals (CIs). From the 967 eligible ED visits, delirium was detected in 107 (11.1%). The area under the curve for the REcognizing DElirium in Emergency Medicine (REDEEM) score was 0.901 (95% CI=0.864-0.938). The REEDEM risk score included 10 different variables (seven based on triage information and three obtained during early history taking) with a score ranging from -3 to 66. Using an optimal cutoff of ≥11, we found a sensitivity of 84.1% (90 of 107 ED delirium patients, 95% CI=75.5%-90.2%) and a specificity of 86.6% (745 of 860 non-ED delirium patients, 95% CI=84.1%-88.8%). A lower cutoff of ≥5 was found to minimize false negatives with an improved sensitivity at 91.6% (98 of 107 ED delirium patients, 95% CI=84.2%-95.8%). A risk stratification score was derived with the potential to augment delirium recognition in geriatric ED patients. This has the potential to assist on delirium-targeted screening of high-risk patients in the ED. Validation of REDEEM, however, is needed prior to implementation.

Association of a geriatric emergency department program with healthcare outcomes among veterans.

We aim to describe the outcomes of Geriatric Emergency Room Innovations for Veterans (GERI-VET), the first comprehensive Veterans Affairs Geriatric ED program. In this prospective observational cohort study at an urban Veterans Affairs Medical Center ED, participants included Veterans aged 65 years and older treated in the ED from January 7, 2017 to February 29, 2020. Veterans with an Identification of Seniors At Risk (ISAR) score >2 were considered eligible for GERI-VET, receiving geriatric screens and care coordination in addition to standard ED treatment. The control group included GERI-VET eligible Veterans who did not receive GERI-VET care. Propensity score matching was used to compare outcomes in the GERI-VET group (N= 725) and a matched control group (n= 725). Key measures included ED resource utilization, outpatient referrals, ED admission, and 30-day admission. In the ED, the GERI-VET group received more consults to pharmacy (315 [43.4%] vs. 195 [26.9%], p< 0.001) and social work (399 [55.0%] vs. 132 [18.2%], p< 0.001). The GERI-VET group had higher referral rates to Geriatrics (64 [17.7%] vs. 18 [5.8%], p< 0.001) and Home Based Primary Care (110 [30.4%] vs. 24 [7.8%], p< 0.001). Key outcome measures included lower rates of ED admission (363 [50.1%] vs. 417 [n= 57.5%], p= 0.003) and 30-day hospital admission (412 [56.8%] vs. 464 [64.0%], p= 0.004) without increasing ED length of stay (5.4± 2.2 vs. 5.4± 2.6h, p= 0.85) or 72-h ED revisits (23 [3.2%] vs. 16 [2.2%], p= 0.25) in the GERI-VET group. A program designed to screen for geriatric syndromes and coordinate care among at-risk older Veterans was associated with increased multidisciplinary resource utilization and reduced ED and 30-day admissions without increasing ED length of stay or re-visitation.

Polypharmacy and High-risk Medications in Older Veterans Presenting for Emergency Care

Background: Guidelines suggest screening for polypharmacy and documentation of high-risk medications in the geriatric Emergency Department (ED) setting. The purpose of this quality improvement initiative was to describe high-risk medications and polypharmacy in older Veterans presenting to the ED in order to inform the workflow of a Geriatric ED pharmacist. Methods: This was a retrospective, quality improvement project that included patients age 65 years or older who presented to the ED between 10/1/2017 and 9/30/2019. The endpoints included the number of patients with at least one high-risk medication and the mean number of medications on the outpatient medication list. The number of patients age greater than 75 and one at least 10 medications or one high-risk medication was also evaluated. Results: A total of 27,018 patients were included in the analysis. Approximately 83% of patients had at least one high-risk medication, with a mean of 13 ± 7 active outpatient medications per patient. Thirty-one percent of patients were at least 75 years of age, on 10 medication, or a high-risk medication. The three most common high-risk drug classes were antihypertensives, antidepressants, and opioids. Discussion: A high prevalence of high-risk medications and polypharmacy was found in this population. Screening criteria was developed for comprehensive medication reviews to be completed by a clinical pharmacist in a geriatric ED setting.

28 Geriatric Emergency Department Accreditation Accelerates Integration of Geriatric Care at the Veterans Health Administration

Geriatric emergency department (GED) accreditation (GEDA) through the American College of Emergency Physicians requires developing a strategic plan for education, equipment, policies, environment, staffing, and quality improvement. To support Veterans Affairs (VA) EDs in achieving accreditation, an interdisciplinary group of leaders in the VA National Offices of Emergency Medicine and Geriatrics and Extended Care and subject matter experts created a council that hosted a bootcamp, mentored support, created a Web site housing sample policies, staffing descriptions, workflows, and order sets. Additional individualized support was provided to assist with implementation of GED processes during the COVID-19 crisis. We sought to identify feedback on the accreditation process to inform onboarding of future cohorts. This was a dissemination plan for GED implementation at 20 VA EDs. VA medical centers were invited to submit an application for GEDA mentored support and fee waiver. An in-person kickoff bootcamp was held February 2020 for 10 VA EDs applying for Level 1 or Level 2 GEDA. As part of the planned quality review and to prepare for future VA GEDA, we performed follow-up surveys on the accreditation process, new interdisciplinary services relationships established, outcomes, and feedback on the process. A total of 26/110 VA facilities with EDs applied for VA GEDA support; 10 were accepted for the in person bootcamp and virtual support and 10 additional for virtual support only. Nineteen supported sites have applied for accreditation and 14 have received accreditation: one level 1, four level 2, and nine level 3. Mean ED visit volume in 2019 for accredited sites was 25, 475 (range 8, 212-42, 589 visits/year) with 47% of visits made by veterans 65 years or older (range 33-60%). Sixteen sites responded to the survey. Ninety-four percent (15/16) described new interdisciplinary services (figure 1). When asked what accomplishment associated with the accreditation process they were proud of, 6 reported ED outcome improvements, 12/16 reported connecting patients to VA services, and 13/16 reported improved patient care team engagement. Feedback on the process suggested the need for standardization of data collection, workflow, and documentation, which has led to the development of a central dashboard and patient care note with standardized geriatric screens and assessments. A central supported system allowed 14 VA EDs to achieve geriatric ED accreditation despite the COVID pandemic. Almost all facilities reported they formed new associations with interdisciplinary services as a result of the accreditation process. More than half reported improved patient care team engagement and connection to VA services as additional accomplishments. Feedback has led to increased standardization of geriatric emergency care assessments across the healthcare system.

252 Utilization of Intermediate Care Technicians to Administer Geriatric Emergency Department Patient Screenings

The Geriatric Emergency Department Innovation (GEDI) program is an evidence-based intervention for ED geriatric patients with increased morbidity and mortality risk. The VA Greater Los Angeles Health Care System (VAGLAHS) began implementation of the GEDI in March 2020. The GEDI uses the validated, six-item Identifying Seniors at Risks (ISAR) tool to risk-stratify, and facilitate enhanced care for, geriatric patients. A standardized onboarding process was developed for intermediate care technicians (ICT), former military corpsmen and medics, to perform this screening.

295 Positive Cognition Screening a Geriatric Emergency Department

Geriatric emergency departments (EDs) have become more common in recent years. Along with these expansions have come targeted interventions and assessments for older adult patients in the ED to assess several social and clinical factors specific to a patient. Cognitive tests are frequently used to assess older patients’ ability to make decisions to gauge resources that they may need. The objective of this study was to understand who may be more likely to have positive cognitive test in the ED. This was a retrospective multicenter study among older adult patients (>=65 years) presenting to two EDs (urban level one trauma center and suburban academic hospital with combined annual census of ∼80,000). Patients were qualified for specialized geriatric screenings if they were identified as being at risk by the Identification of Seniors at Risk (ISAR) screen and has an Emergency Severity Score (ESI) of 3. Screening tools used to identify potential cognitive issues included the Montreal Cognitive Assessment (MoCA), and the Abbreviated Mental Test 4 (AMT4). Demographic and clinical characteristics were assessed for all patients who received a screen. Logistic regression was used to identify factors independently associated with a positive test. Adjusted odds ratios (AOR) and 95% Confidence Intervals (CI) are reported. There were a total of 1850 cognitive tests ordered. Of those tests ordered, 903 (48%) were able to be completed and scored. Of those 903 tests scored, 226 (25%) were positive. Between those who tested positive and negative, individuals were more likely to have a positive test if they were age 85+ compared to 65-74, (AOR=2.01, 95% CI=1.36, 2.97; p=<0.001) and less likely if they were patients who were non-Hispanic Whites compared to other Races and Ethnicities (AOR=0.59, 95% CI=0.42, 0.83; p=0.002). Sec, mode of arrival and disposition were not associated with a positive cognitive score. Screening tools are important instruments to help identify patients with potentially cognitive issues. Although their use in the ED have been limited in the past, they are becoming utilized more often with the increasing senior population and the development of geriatric ED in the United States. Findings from these assessments help to inform providers of the needs of older adult patients and their caregivers.

58 Using Machine Learning to Predict Hospital Disposition With Geriatric Emergency Department Innovations Intervention

The geriatric emergency department innovations (GEDI) program is a nurse-based geriatric assessment and care coordination program that reduces preventable admissions for older adults and decreases Medicare expenditures. All adults age 65+ in the ED are currently eligible, but only 5% of older adults receive GEDI care due to resource limitations. Currently available geriatric screenings to identify older adults likely to benefit from programs like GEDI perform poorly in the ED. The objective of this study was to predict likelihood of hospitalization accurately and consistently with and without GEDI care using machine learning models to better target the GEDI program. We performed a cross sectional observational study of ED records from (1/2010 to 3/2018). Using propensity score matching, GEDI patients were matched to non-GEDI patients. Logistic regression, random forest, support vector machine, gradient boosting, and neural network techniques were used to predict hospital admission using demographic data, medical history, ED visit chief complaint, vitals, and geographic data. Hospital admission was then predicted with and without GEDI assessment during the ED visit. Visits were classified into one group with a predicted change in hospital disposition with the GEDI assessment and another group with no predicted change in disposition with GEDI assessment. In this second analysis feature importance was performed on the tree-based models. Final model performance was reported as the area under the curve (AUC) using receiver operating characteristic models for the test data. We included 55,056 patients age 65+ who accounted for 134, 361 ED visits. Of these, 3, 860 visits were included in the training set and 10, 142 in the testing set to predict hospital admission. 5, 071 visits were used in the training and testing sets to predict change in disposition with GEDI. The random forest model had the best performance with an AUC of 0.81 (95% CI 0.79-0.83). In the random forest model, 9, 756 (96.2%) ED visits were predicted to have no change in disposition with GEDI assessment, and 386 (3.8%) ED visits were predicted to have a change in disposition with GEDI assessment (Table 1). Of those with a predicted change in disposition from admitted to discharged with GEDI assessment the top 5 most influential variables out of 86 variables with their relative importance are in Table 2. The higher relative importance the more it influenced the model’s outcome, however ranking does not imply the direction of influence. All importance values add up to a total of 1. Our machine learning models were able to predict who is likely to be discharged with GEDI assessment with good accuracy and thus select a cohort appropriate for GEDI care. Future implementation of this machine learning model into the electronic health record may assist in the identification of older adults who should be prioritized for GEDI care.View Large Image Figure ViewerDownload Hi-res image Download (PPT)