AGA Abstracts 548 EVALUATION OF A NOVEL AUTOMATED CONTINUOUS PATIENT- MONITORING SYSTEM FOR VULNERABLE PATIENTS IN THE WARDS (POSTER PRESENTATION) Arjun Prakash, Sanjeev Krishnapriya, Saibala Madathil, Zubair U. Mohamed, Harshavardhan B. Rao Introduction: Deterioration of patients in hospitals is typically preceded by changes in vital signs. Early warning scores (EWS) are currently used to assess risk, but depend on manual input of data by skilled nurses. A continuous vitals monitoring solution was proposed which can enable the automated recording of 6 key vitals - Heart rate, Blood Oxygen Saturation, Respiratory Rate, 3-lead Electrocardiogram, Non-invasive Blood Pressure and skin tempera- ture. This also provided trends of data captured every 5 minutes and used Modified Early Warning Scores (MEWS) to aid clinical decisions. Clinical efficacy of this system was assessed for vulnerable patients in a non-ICU setting. Methods: An observational study was conducted in a tertiary care hospital (December ‘18 - February ‘19) which included vulnerable patients admitted to digestive diseases service. Vulnerable patients were defined as: 1) Gastrointestinal bleed patients with normal vitals at admission and 2) Post-operative/post intervention patients including transplant recipients who require monitoring as assessed by the intensivist. Patients were monitored for a minimum of 8 hours in a designated continuous-monitoring ward. The continuous monitoring system (the STASIS Monitoring Solution) provided live wave forms and recorded trends of patient vitals. This was wirelessly transmitted to a central tablet kept with the nurse and was made available on the doctor’s mobile app via a cloud service. There were no changes applied to the existing nurse to patient ratio for the study. Decision to shift to the ICU was based on MEWS which was automatically generated from the continuous monitoring solution. The preceding time cohort (September 2018 – November 2018) was used as controls to compare the number of ‘Code Blue’ calls, average length of stay (ALOS), and ICU readmission rates. Results: Thirty eight vulnerable patients were included in the study (Mean age - 61±14 years, M: F=6.6:1) and monitored for a total of 2248 hours. Five patients (13%) were found to have abnormal MEWS prompting a shift to the ICU. The number of admissions (10076 vs 9460 patients) and mean Charlson’s comorbid- ity index (2.43 vs 2.17) of patients during the study period were higher as compared to controls. Despite this, there was a 67% reduction in the number of Code blue calls during the study period (3 calls vs 9 calls). [See Table 1] The ALOS and ICU readmission rate during the study period was 44.8% and 44.4% lower as compared to controls. Conclusions: Automated continuous monitoring of vulnerable patients in the non-ICU settings using this system is feasible and clinically beneficial for escalation of care. This can result in efficient utilization of hospital resources and reduce cost of treatment. Larger, controlled trials are required to quantify the clinical and fiscal benefits of universal implementation of such tech- nologies. Table 1: Impact of an automated continuous monitoring system for vulnerable patients in a non-ICU setting 549 IMPROVING VACCINATION RATES IN VETERANS WITH INFLAMMATORY BOWEL DISEASE (POSTER PRESENTATION) Patrick Chen, Hyaehwan Kim, Jennifer Ball, Padmini Krishnamurthy Patients with inflammatory bowel disease (IBD) are at increased risk for vaccine preventable infections with additional risk from immunomodulators and biological agents used in the treatment of IBD. The goal of our study was to improve vaccination rates of veterans affected by IBD by having a dedicated nurse to track vaccinations and involve the primary care providers in contacting these veterans and administer the required vaccinations. Methods: The department of Gastroenterology at the Dayton Veterans Affairs Medical Center established an IBD clinic on the 1st of September. We assigned a dedicated nurse to this IBD clinic to track the vaccination status of each of the patients after reviewing the vaccine requirements based on the Center for Disease Control Advisory Committee on Immunization Practice Guidelines. We contacted the Primary Care Provider (PCP) for individual patients through a chart based electronic communication system. The PCPs then followed up with patients to complete the required vaccinations. Results: A total of 69 patients were enrolled in IBD clinic since its conception on 1 September 2017. We calculated the percentage of patients vaccinated prior to establishing the IBD clinic. The next calculation included the number of newly vaccinated patients over the total number of unvaccinated patients as of 15 th November 2019. Influenza vaccination rate did not change with IBD clinic intervention. Prior to intervention 47 patients (68.1%) had influenza vaccination, and 45 of patients (65.2%) received influenza vaccination after intervention. There were 39 of patients (56.5%) had Pneumococcal Conjugate Vaccine (PVC13) prior to intervention, and the rate of vaccina- tion was increased by 23% (7/30) after intervention. Prior to intervention 58 patients (84.5%) had Pneumococcal Polysaccharide Vaccine (PPV23), and the rate increased by 36% (4/11) after intervention. Prior to intervention 19 patients (27.5%) had Hepatitis B vaccination, and the vaccination rate increased by 28% (14/50) after intervention. Prior to intervention 45 patients (66.7%) already had Adult Tetanus, Diphtheria, Pertussis (Tdap) vaccination o, and its vaccination rate increased by 65% (15/24) after intervention. One of the reasons for lack of improvement in influenza vaccination rates may be beliefs maintained by individual patients regarding perceived harmful effect of the flu vaccine, but this certainly requires S-114 AGA Abstracts further review. Conclusion: Vaccination uptake in the IBD patient population has been suboptimal for various postulated reasons including lack of adequate time by treating physi- cian during patient encounter. Assigning a dedicated nurse to review the vaccination status of individual patients with IBD and involving their PCPs, greatly improved our vaccination rates in this vulnerable population. 550 INFLAMMATORY BOWEL DISEASE PATIENTS HOSPITALIZED FOR FLARES ARE AT RISK FOR CHRONIC STEROID USE DUE TO LACK OF CLEAR INSTRUCTIONS AT DISCHARGE: A 15-MONTH RETROSPECTIVE STUDY AT A PRIVATE, NOT-FOR-PROFIT ACADEMIC HOSPITAL (POSTER PRESENTATION) Chung Sang Tse, Yousef Elfanagely, Sean Fine, Abbas Rupawala Introduction: Patients with inflammatory bowel disease (IBD) who are hospitalized with an IBD flare can be managed acutely with short-term (4-6 weeks) high dose glucocorticoid taper. Corticosteroid use beyond 3-4 months can cause serious and significant side effects . Thus, it is imperative for patients to be discharged with a prednisone tapering plan and steroid-sparing treatment strategy. This quality assessment project examines the rate at which IBD patients remain on high dose steroids 3 months after hospital discharge. Methods: We conducted a retrospective review of IBD adult patients hospitalized at a private, not-for- profit academic hospital in southeastern New England over a 15-month period (June 2018 to October 2019). Data were obtained from the electronic medical record system using ICD- 10 disease codes and generic/brand medication names. High dose steroids use is defined as glucocorticoid dose equivalent of prednisone 20 mg PO daily or higher. Qualitative analysis was performed on the rate at which patients remained on high dose steroids 3 months after hospital discharge (primary outcome), patient demographics, IBD disease characteristics, hospital discharge orders/instructions. Results: A total of 52 patients with IBD were hospital- ized, of which 46 patients had IBD flares (56% female; 65% Caucasian; median age 39 years, standard deviation [SD] 16 years; 26% ulcerative colitis, 75% Crohn’s disease). Presenting symptoms (may be ≥1) were abdominal pain (72%), diarrhea (23%), hematochezia (19%), and nausea/vomiting (9%). Median length of hospital stay was 5 days (SD 6 days). Even though the vast majority (93%, 43/46) of patients with IBD flares were discharged with high-dose steroids, less than half (44%, 19/43) had a defined 4-6 week tapering plan on the discharge summary; the remaining had vague instructions, such as “Take prednisone 40 mg until seen by your gastroenterologist and discuss regarding further dose reduction with your physician” or “Prednisone 40 mg x 5 days, then 30mg daily until GI follow- up”. Less than half (42%, 18/43) of the patients discharged on high-dose steroids had GI appointments scheduled prior to discharge, with one-third (6/18) canceling or failing to attend the appointment. Alarmingly, one-third of patients (34%, 10/29) with follow-up data available (missing follow-up data on 14 patients) remained on steroids 3-months after hospital discharge (8/10 were high dose steroids). Conclusions: IBD patients hospitalized for IBD flares are often discharged on high-dose steroids, and lack of a clear, predefined steroid tapering plan and outpatient GI follow-up places them at risk for inadvertent pro- longed steroid use. To mitigate this, we collaborated with the hospitals’ information technol- ogy team to implement a 6-week standardize steroid tapering order set for patients with IBD flares (launch in December 2019). 551 REAL WORLD ADHERENCE TO GUIDELINE RECOMMENDATIONS FOR SEVERE EROSIVE ESOPHAGITIS IN AN URBAN POPULATION (POSTER PRESENTATION) Padma Chamarthy, Rahul Kataria, James Langworthy, Jessica Briscoe, Nirali Shah, Adam C. Ehrlich, Jonathan Gotfried Introduction: Erosive esophagitis (EE) is a well-known sequelae of gastroesophageal reflux (GERD). The severity of inflammation is classified by degree of severity according to Los Angeles (LA) Grading Classification from LA Grade A – D. It has been shown that EE can obscure the identification of underlying Barrett’s esophagus (BE). Guidelines recommend repeat endoscopy in patients with LA grade C & D esophagitis after PPI therapy for a minimum of 8 weeks to evaluate for BE. We aimed to evaluate the frequency of successful repeat endoscopy of patients with LA Grade C & D esophagitis after an appropriate course of PPI therapy. Methods: We conducted a retrospective chart review of patients with con- firmed LA grade C or D esophagitis on endoscopy from 10/2018 to 10/2019 using available data from our electronic health record. We identified patients that were prescribed adequate PPI therapy and those that underwent successful repeat endoscopy. We additionally noted the context of index endoscopy (inpatient or outpatient), the frequency and duration of prescribed PPI after initial endoscopy (daily vs BID), whether or not a follow up endoscopy was recommended on the initial procedure report and its recommended timeframe. Results: One hundred thirty-three patients (mean age 56 years, 56.4% male) met our study criteria. On index endoscopy, the majority of patients had grade D esophagitis (63.2%). Most endoscopies were done as an inpatient (57%). Most patients (83.5%) were discharged on PPI therapy (60.9% on[JG1] BID, 22.6% on daily) after the initial endoscopy for a mean of 6.2±0.9 months. Inpatients were given 5.5±1.2 months, while outpatients were given 7.3±1.3 months. Appropriate follow up endoscopy was recommended in an average interval of 9.3±2.2 weeks on the procedure report of the index exam in only 36.8% of cases. A total of 30.1% patients had follow up endoscopies in a mean interval of 95±28 days, however only 16.5% of patients had them done after at least 8 weeks of PPI therapy. Of the 93 patients not evaluated with repeat endoscopy, 11 were not completed due to either “patient no-show” or “patient canceled.” Of the 22 completed follow up endoscopies, there were 10 patients with no esophagitis, 11 with persistent esophagitis, and 1 with BE. Conclusions: The majority of patients at our institution do receive adequate medical therapy for severe esophagitis. However, only a small subset of patients are given immediate recommendations for follow up endoscopy, and even fewer undergo successful evaluation for underlying BE, regardless of admission status. We plan to create a standardized process for consistent