Compared to the Non-PG, the PG was younger (p ¼ 0.014), had higher pain scores at their index visit (median 8.5, quartile 5.5, 10, p ¼ 0.0002) with a little over half achieving a MCSD in their index visit pain scores (57%). In the PG, 61% (45) reported home opioid use at the index visit and 31% (14) received another opioid prescription. Of the 17 in the PG discharged with an opioid (23%), 13 returned for worsening symptoms and 4 for medication issues. Compared to the PG, more patients in the Non-PG were admitted (74%, p ¼ 0.018), had low REALM scores (75%, p ¼ 0.005), and presented with a respiratory chief complaint at their index visit (p ¼ 0.004). Socioeconomic disadvantage, sex, ethnicity, index visit mode of arrival, and insurance type was not significantly different between groups. As continuous variables, age (p ¼ 0.016), index visit pain scores (p ¼ <0.001), and index visit acuity (p ¼ 0.028) were significantly different between groups. Age (OR 0.9, 95% CI 0.8- 0.9, p ¼ 0.047), REALM scores (OR 3.1, 95% CI 1.3 - 7.5, p ¼ 0.011), and index visit pain scores (OR 1.1, 95% CI 1.0- 1.2, p ¼ 0.004) were predictive of ED returns for pain in older adults. Conclusions: The likelihood of returns for pain in older adults decreased with age; increased with higher REALM scores; and increased by 10% for each point increase in pain scores. Table 1. ED Returns for Pain in Older Adults Variable Category Pain Group (n[74, 57%) Non-Pain Group (n[56, 43%) Overall (n[130) P-value Low REALM No 32 (50) 13 (25) 45 (38) 0.005 Yes 32 (50) 40 (75) 72 (62) Index visit acuity 1 or 2 33 (46) 37 (67) 70 (56) 0.020 3 or 4 38 (54) 18 (33) 56 (44) Age range 55-64 years 57 (77) 30 (54) 87 (67) 0.014 65-75 years 13 (18) 17 (30) 30 (23) 76 years and over 4 (5) 9 (16) 13 (10) Index visit disposition Discharge 35 (48) 15 (27) 50 (39) 0.018 Admission 38 (52) 40 (73) 78 (61) All tests done using Chi-Squared, except Fisher’s exact test 238 Dynamics and Implications of Benzodiazepine Administration to Older Adults in the Emergency Departments Lombardi KM, Roberson J, Pourmand A/George Washington University, Washington, DC Study Objectives: Benzodiazepine use in the older adults has been associated with a myriad of adverse effects such as delirium, mechanical fall, fractures, and memory disturbances. Despite their particular risk to this population, there is a concern regarding the frequency of benzodiazepine use in US emergency departments (EDs). In this study we utilize national databases to describe trends in utilization of benzodiazepines in the older adult population. Methods: Data from the CDC’s 2005-2015 National Hospital Ambulatory Medical Care Survey (NHAMCS) were compiled and analyzed. Variables were created to identify all patients aged over 60 years who had been administered benzodiazepines and subsequently been given a diagnosis of delirium. Appropriate bivariate statistical analyses and multivariate regression modeling were subsequently utilized. Results: During study period approximately 250 million adults aged over 60 were seen in US EDs, 10.2 million of which (4.1%) were administered a benzodiazepine. The rate of administration ranged from 3.6% in 2005 (95% CI 3.1-4.3) to 4.5% in 2015 (95% CI 3.7-5.6). This represented a relative change of +25% over the study period. Older adults given benzodiazepines during their ED course were significantly more likely to be diagnosed with delirium (8.5% 95% CI 5.4-13.2% P¼0.001) compared to those who did not (4.1 95% CI 3.9-4.4 P<0.0001) and were also more likely to be admitted to the hospital (34.9% versus 29.9%). Older adults administered benzodiazepine also exhibited high rates of depression (15.7% 95% CI 11.8-20.1 P¼0.002). Results of multivariate regression modeling indicated that when patient demographics and medical history are accounted for, older adults given benzodiazepines demonstrated an increased odds of being diagnosed with delirium (aOR 2.2 95% CI 1.6-4.1 P<0.0001) an increased odds of being admitted to the hospital (aOR 1.4 95% CI 1.1-1.8 P<0.0001) and to have also been administered an opioid analgesic in the ED (aOR 1.7 95% CI 1.5-1.9 P<0.0001). Older adults administered benzodiazepines were more likely to be male (aOR 1.2 95% CI 1.0-1.3), to have a history of stroke (aOR 1.2 95% CI 1.04-1.6) or coronary artery disease (aOR 1.4 95% CI 1.1-1.7 P<0.0001). Conclusions: Despite the risks associated with the administration of benzodiazepines to older adults, this population still demonstrates heightened and increasing levels of utilization. Older adults administered these agents exhibited higher rates of depression and delirium as well as an increased likelihood of being admitted to the hospital. 239 Advance Care Planning Among Patients With In-Hospital Cardiac Arrest Woo S, Brenner N, Chong S, Benz P, Dubin J, Wilson M, Thorne J, Goyal M/Georgetown University School of Medicine, Washington DC, DC; MedStar Washington Hospital Center, Georgetown University School of Medicine, Washington DC, DC; MedStar Washington Hospital Center, Washington DC, DC Study Objectives: 209,000 people suffer an in-hospital cardiac arrest (IHCA) in the US every year. Prior research suggests advance directives are rarely completed among high-risk populations. No data exist about advance directive completion in sudden IHCA victims. Among these patients, we sought to determine the percentage that had completed a pre-arrest advance directive, the percentage who expressed their wishes to not be resuscitated prior to their cardiac arrest (DNR), and the percent that re-addressed their code status following return of spontaneous circulation (ROSC). Methods: Retrospective chart review at a single, 927-bed academic medical center. Adults in whom a “code blue” was called between January 2017 and March 2018, and had a confirmed cardiac arrest, were included. Cardiac arrests with incomplete data or that occurred in the ED or ICU were excluded. Trained and supervised research assistants used a standardized data collection tool to extract event related data from the electronic and paper medical record. Admission history and physical, “code blue sheets,” progress notes up to 24 hours after the event, discharge summaries, death summaries and all advance directive notes, including advance directives, living wills, durable power of attorney documents, code status orders written throughout the admission were reviewed. Physician orders for life-sustaining treatment documents were not included. Patients were considered to have a pre-arrest advance directive if the advance directive was scanned into the medical record or referenced by a member of the care team prior to sudden IHCA. Patients with pre-arrest DNR status were identified by documentation on the “code blue sheet” or in the notes in the 24h following the code indicated contact with power of attorney stating patient should have been DNR. Patients were considered to have had their code status re-addressed prior to death or discharge if family meeting discussing code status was documented after the code, code status order was changed after the code, or code status was addressed in hospital summaries. Standard descriptive statistics were used. Results: 188 code blues were identified of which 5 were excluded. Of the remaining 183, 71% were black, 17% were white; mean age 65 and 46% were women. 131 (71.6%) had ROSC; 55 (30%) survived to hospital discharge. 14 patients (7.7%) had a pre-arrest advance directive. Two patients (1.1%) were DNR prior to their arrest but had CPR performed. One of these patients had a pre-arrest advance directive stating DNR status, but inpatient code orders and physician documentation indicated full resuscitation. The other patient had a documented pre-arrest advance directive indicating power of attorney, and had code orders and physician documentation indicating full resuscitation; however, post-ROSC, the power of attorney indicated the patient expressed DNR wishes pre-arrest. Eighty-five of the 131 patients with ROSC had their code status adjusted or confirmed either by themselves or family members prior to expiration or discharge. Only 2 patients completed new advance directives following the code. Research Forum Abstracts Volume 72, no. 4s : October 2018 Annals of Emergency Medicine S95