Results: The objectives were evaluated for appropriateness for 3 previously deﬁned provider groups: (1) out-of-hospital providers, nurses, (2) medical ofﬁcers, senior nurses, and (3) residents, specialists, consultants. There were 7 voting participants in the modiﬁed Delphi process. All Tier 3 objectives (79) met consensus. One objective from Tier 2 was revised under consensus, and one objective was added with consensus agreement to create a list of 42 objectives. Five objectives in Tier 1 did not initially achieve consensus, 4 of which required revisions, for a ﬁnal list of 30 objectives. Conclusion: A group consensus process resulted in the development of a list of 161 objectives for a pediatric emergency medicine curriculum for multiple levels of practitioners in African settings. These objectives will guide development of the AFEM pediatric emergency medicine curriculum, which will advance pediatric emergency care competency across Africa. 194 The Performance of Sepsis Screening Tools in an Urban Emergency Department in Belize Jaung M, Gunter S, Espina I, Bolgiano M, Huynh D, Koradia P, Habet M, Hidalgo J, Mackey J/Baylor College of Medicine, Houston, TX; National School of Tropical Medicine/Texas Children’s Hospital, Houston, TX; Karl Heusner Memorial Hospital, Belize City, Belize Study Objectives: There is limited data on the burden and treatment of sepsis in low and middle-income countries (LMICs), especially in Central America and the Caribbean. We sought to characterize the burden of sepsis among patients admitted to the largest public referral hospital in Belize and the generalizability of the quick Sequential Organ Function Assessment (qSOFA) score and Emergency Triage and Treatment (ETAT) emergency signs as sepsis screening tools. Methods: We performed a single-site retrospective review of patients admitted from the emergency department (ED) at Karl Heusner Memorial Hospital, a tertiary government facility in Belize City, Belize with approximately 25,000 annual ED visits. We reviewed paper and electronic medical records for patients admitted from October 2017 to September 2018 with a diagnosis suggestive of infection. Results: Of the total of 3,703 admissions during the review period, 1,693 (46%) had a diagnosis suggestive of infection. Only 21 (1.2%) patients had sepsis documented as an admission diagnosis. We conducted a selective chart review of 590 of these admissions during the review period. For the overall cohort, 367 (62%) patients were adults, median age was 33, 53% were male, 14% were HIV positive and 11% were transferred from a referring facility. Utilizing the Sepsis-3 recommended criteria of a change of 2 in the SOFA/pSOFA score, 123 (21%) patients met the deﬁnition of sepsis. The 30-day mortality rate was 20% and the 90-day mortality rate was 28% in the sepsis cohort, compared with 5.1% and 7.8% in the overall cohort. The rate of ICU admission was higher in the sepsis cohort (9% versus 4.7% overall). HIV positive patients and transfers made up 15% and 19% of the sepsis cohort respectively. In the pediatric population, a positive ETAT emergency sign was poorly sensitive for sepsis (39.0%) as was pediatric Systemic Inﬂammatory Response Syndrome (pSIRS) criteria 2 (36.6%). There was only one pediatric sepsis death so mortality prediction was not assessed. In the adult population, both SIRS and qSOFA were poorly sensitive for sepsis (SIRS  2 55.3%, qSOFA2 32.5%) however a qSOFA score 1 had a sensitivity of 80.7%. A high qSOFA score was associated with an increased risk of death (qSOFA 2 RR 2.7, 95% CI, 1.3 to 5.7; p<0.01 and qSOFA 3 RR 5.8, 95% CI, 2.6 to 12.7; p<0.01). Only when all four SIRS criteria were present was it associated with an increased risk of death (RR 4.5, 95% CI, 1.4 to 14.9; p<0.01). There was no signiﬁcant discriminatory ability between qSOFA or SIRS in predicting 30-day mortality; however, SOFA was superior to qSOFA or SIRS in predicting 90-day mortality (AUROC 0.68, 0.64, and 0.52 respectively, p <0.01). Likewise, a SOFA score of 2 was associated with an increased risk of death (RR 4.1, 95% CI 2.5-6.6; p<0.01). Conclusion: We observed a trend of under-diagnosis of patients with sepsis at this public hospital in Belize. Existing screening tools utilized in high-income countries (qSOFA and SIRS) as well as pediatric screening tools utilized in LMICs (ETAT) were poorly sensitive for sepsis. There was an increased risk of death with a high qSOFA or SIRS score; however, neither demonstrated signiﬁcant discrimination for mortality. This demonstrates the need for the development of novel sepsis screening tools for LMIC settings to improve the diagnosis and treatment of sepsis. 195 Feasibility of Disseminating First Aid and Cardiopulmonary Resuscitation Education to Community Health Workers in India Mutpur R, Courtley M, Kumar VA/Wayne State University, Detroit, MI; Detroit Medical Center/Wayne State University School of Medicine, Detroit, MI Study Objectives: To examine the feasibility of training community health workers in ﬁrst aid and cardiopulmonary resuscitation (CPR) as a means to implement an integrated emergency medical care model in rural India. In India, the main components of emergency medicine (EM) care are inadequate: in the community, during transportation, and at health care facilities. The current public health system in place, which incorporates community health workers (CHW) to disseminate vaccines and health education, could be built upon to alleviate this. Methods: This is a mixed-method, prospective observational study conducted in the community of Anupshahr in India’s Bulandshahr District. Inclusion criteria included all anganwadi workers (government CHW) and women of the self-help group (private CHW) who presented for the training sessions held at the Pardada Pardadi Educational society (PPES) school. CHWs were trained in CPR and ﬁrst aid through a program developed by emergency physicians at Wayne State University School of Medicine in October 2016. The program’s efﬁcacy was tested using a pretest before training and posttest conducted after retraining and evaluation at 6 months. These tests addressed knowledge, self-conﬁdence in abilities, and self-reported behavior. Statistical analysis was done using Wilcoxon Rank Sum t-Test Approximation. Results: There were 28 CHWs who received the initial training. 30 CHWs showed up for the follow-up training at 6 months, but only 8 were from the initial group. Mean experience was 17.5 years in post-test group and 13.9 which were not signiﬁcantly different (p¼ 0.537). Mean number of CPR uses increased to 18.2 in past 6 months in the post-test group vs. 0.4 in the pre-test group, which is statistically different (p¼ 0.0010). However, the average score on the knowledge component of the pre-and post-test, as well as the percentage of CHWs who comfortable performing CPR were not signiﬁcantly different (p-values of 0.4427 and 0.4914, respectively). CPR knowledge component (count and depth of performing CPR) showed improvement from pre to post tests, which were statistically signiﬁcant (p-values of 0.0004 and 0.0015, respectively). Furthermore, wound cleaning and collapse treatment on the knowledge component also were statistically signiﬁcant (p-values of 0.0104 and 0.0037, respectively). Conclusion: There seems to be an increase in conﬁdence in performing CPR due to the education intervention, and an increase in the knowledge level of the recipients in areas including CPR technique, wound cleaning, and collapse treatment. Exact effectiveness of our training model was lost with differing pre-and post-test groups. As shown by this feasibility study, a simple interventional approach to CHW education has signiﬁcant limitations. There continues to be limited data on the potential relationship between EM care and CHW interventions and the next phase of the study will build upon our results to introduce a more structured and effective intervention. A structured training that incorporates more rigorous follow-up methods to target education of the CHW workforce at a greater degree may serve to bridge the gap between effectiveness of EM care and the rural population. 196 A Simple Abdominal Tourniquet for Pelvic and Junctional Hemorrhage Hauswald M, Kerr NL, Katukuri VR, Izquierdo LA/University of New Mexico, Albuquerque, NM Study Objectives: Abdominal tourniquets are used to treat critical, otherwise uncontrollable bleeding from the pelvis, pelvic organs and lower extremity junctional sites (those too close to the pelvis to allow for peripheral tourniquet placement). They are used primarily for traumatic injuries but have also been suggested for obstetrical hemorrhage. Commercial devices are relatively expensive and unavailable in low resource countries. The objective of this study was to develop and test a low cost abdominal tourniquet that could be made on-site using locally available supplies. Methods: An abdominal tourniquet was made by placing an uninﬂated standard size 5 soccer ball on the subject’s abdomen, wrapping it tightly with a 12-inch wide sturdy cloth binder and securing the binder with sewn-on hook and loop (eg, Velcro) strips. A comparative trial using this device was done in 12 healthy volunteers. The ball was gradually inﬂated using a hand pump. The internal ball pressure was measured during inﬂation while distal aortic blood ﬂow was monitored just below the superior mesenteric (SMA) takeoff using a GE Voluson TM ultrasound. When blood ﬂow stopped the ball pressure was recorded. The subject was then asked how painful the procedure was, using a 10-point scale where 1 equaled no pain and 10 equaled the most pain tolerable. The ball was then deﬂated. The study was approved by the University of New Mexico Institutional Review Board. Results: The mean pressure to completely stop distal aortic ﬂow was 150 mmHg (95% CI 130 - 170mmHg). The mean pain score was 4.5 (95% CI 3.4-5.6). Conclusion: This abdominal tourniquet was able to completely stop blood ﬂow to the pelvis and lower extremities. It was only moderately painful. Research Forum Abstracts Volume 74, no. 4s : October 2019 Annals of Emergency Medicine S77