249 The Impact of an Electronic Medical Record Alert on Recognition and Referral of High-Risk Elders in the Emergency Department Setting

Abstract

Older patients frequently utilize Emergency Departments (ED). Multiple comorbidities, polypharmacy, impaired mobility (falls), changes in mentation (delirium, dementia) and poor social support often contribute to ED visits. Traditional assessment tools, such as ISAR, to identify high-risk geriatric patients can be difficult to implement in a high volume, high acuity ED setting. We examined the impact of an electronic medical record (EMR) best-practice alert (BPA), on recognition and referral of high-risk older patients in the Emergency Department (ED) who could potentially benefit from geriatric evaluation. Retrospective case control study in a busy academic ED with 67,000 total annual visits, and 24% geriatric (age ≥ 65 y) visits, from October 1, 2019 to May 31, 2020. In the initial program (10/1/19-3/30/20), ED providers were educated about high-risk geriatric conditions and an EMR BPA alerted ED providers to a positive (+) triage delirium screen. Starting 3/31/20, an enhanced BPA used EMR-automated recognition that also included age ≥80y, fall complaint, documented dementia history, polypharmacy (≥ 10 medications recorded), or high ED utilization (>5 visits in 1 year) in addition to the (+) delirium screen. A modified Delphi method was used to select these additional high-risk geriatric characteristics. Frequency of BPA notification and ED or ED-observation unit geriatric consultation (summed as comprehensive geriatric assessment, CGA) are reported for initial and enhanced BPA cohorts. Data was abstracted from the electronic medical record. The difference between initial and enhanced BPA encounters is reported. Analysis used SAS studio (v9.4, Cary, NC). 95% CI and p-values are reported. 7,718 geriatric patients were seen during the initial program; 1,836 were seen during the enhanced BPA program. Mean age in the initial cohort was 74.9 years (95% CI, 74.7, 75.0) and 75.6 years (95% CI, 75.1, 76.0) in the enhanced BPA cohort (p=0.002). % Female was 53.0% and 52.9%, respectively (p= 0.95). Unenhanced BPA alerts in the initial program averaged 53/month (range 45-67). BPA alerts in the enhanced program averaged 699/month. In the initial cohort, 318 (4.1%) BPA alerts led to 30 CGA. 303 (4.1%) patients for whom the initial BPA did not fire were also referred for CGA. Overall, during the initial period, 333 (4.3%) patients had CGA. The enhanced BPA resulted in 1398 (76.1%) alerts and 82 (5.9%) CGA. 14 (3.2%) BPA-negative patients were also referred for CGA. Overall, during the enhanced-BPA, 96 (5.2%) patients had a CGA. After the BPA enhancement, the proportion of geriatric evaluations increased a relative 21% (4.3% to 5.2%, p=0.09). Overall CGA referrals averaged 56/month during initial BPA and 48/month in the enhanced period. Of note, due to COVID-19 impact, overall ED volume declined 29% and geriatric ED visits declined 30% during the enhanced BPA period. This single-site study demonstrates that the EMR can be used to identify high-risk elders. An enhanced automated EMR BPA increased both ED provider alerts and, more importantly, geriatric evaluations in high-risk older ED patients. An EMR-based automated alert can be a useful adjunct to increase referrals and geriatric evaluations in the ED setting for this at-risk population.