 The development and external validation of a clinical prognostic score for death in visceral leishmaniasis patients in an area of Ethiopia with a high burden of HIV infection is the product of a collaboration between MSF, the Institute of Tropical Medicine, Antwerp, and the University of Gondar in Ethiopia. Visceral leishmaniasis is a parasitic disease which is fatal without treatment. Ethiopia is one of the top six countries with high burden of visceral leishmaniasis or VL. Northwest Ethiopia is a VL endemic area. However, they also have large agricultural fields, which means many migrant workers from all over Ethiopia travel to this area for work. They get bitten by infected sunflies and develop VL. High HIV coinfection rates of 20 to 40% has also been documented among them. And prior to MSF intervention, the majority died. In 2004, MSF opened the VL treatment program. Today, case fatality rates among subgroups are high, 6% to 17%. One of the main drugs for treatment, sodium steep gluconate, or SSG, causes severe adverse events. The other drug is liposomal amphotericin B, which is efficacious, safe, but expensive. So why did we decide to do this study? Currently, the classification of VL severity and optimal management are poorly defined. A clinical prognostics core or CPS could aid the classification of VL severity, which can lead to focused strategies and help in the decrease of case fatality. Therefore, we decided to identify predictors of death from VL and to develop and validate a CPS for death. This was a retrospective cohort study using routine program data. To develop the score, we used data from the MSF treatment center that was collected between 2008 and 2013. And to externally validate the score, we used data collected from the Gondar Hospital between 2011 and 2012. The study was also approved by the different ethical committees, and we include the patients whose outcome was death or cure. We performed the literature review for predictors of death, and we used the Spiegel-Holter-Neal-Jones method to develop the score. The whole data set was used to develop the score, and we did external validation. We also calculated the probability of death and diagnostic accuracy of the clinical prognostics core. To assess the overall performance of the CPS, we assessed the area under the receiver operating characteristic curve. What were our findings? In the derivation cohort, we included 1686 patients, and in the external validation cohort, we included 404 patients. In general, most of the patients were young, predominantly male, and migrant workers. HIV-coinfection rate was also high, and the patients admitted in Gondar Hospital had higher proportions of symptoms and signs of VL severity. We found nine predictors of death, aged greater than 40 years, HIV serosteters, hemoglobin less than 6.5, bleeding, jaundice, edema, ascites, and tuberculosis. Our probability of death is as shown. Patients with a score of minus one had a low risk of death, those with a score of zero, an intermediate risk of death, and those with a score of greater than one, a high risk of death. There has the score increased, the sensitivity decreased, and the positive predictive value increased. The overall performance of the CPS was good. The area under the score was more than 75%. How can this be used in VL programs? High-risk patients could be treated in the ICU by experienced clinicians, routine investigations performed, and treated with the best drug, Ambizol. Patients with intermediate risk can have a short-stain hospital, be treated by health professionals of lower cadre, with sodium cibogluconate, and only have focused investigations. Patients at low risk can be treated in the outpatients by cadders of low profession and with SSG. The limitations of this study are as follows. First, as a retrospective study, we could only study the predictors that we collected. It's also possible that critical patients may have had more complete data. This definition of weakness was not standardized across the two treatment centers. Variables such as CD4 counts, anti-retrovirals, which are specific for HIV-coinfected patients were not analyzed in the study, but we study a CPS specifically for VL and HIV-coinfected is foreseen. In conclusion, the CPS showed good performance in identifying VL patients at high risk of death. This CPS can be used by VL programs to implement focus strategies, which can help to decrease case fertility rate. However, evidence is needed on the impact of the score when applied for such strategies. Lastly, validation in other East African countries or areas with low VL HIV-coinfection is also needed. We acknowledge all the staff of the two treatment centers and MSF and the Drugs for Neglected Diseases Initiative for supporting the two main VL treatment sites in Northwest Ethiopia and the Afri-College Project, which are supported this work. Thank you very much. We have questions and clarifications. Anyone like us? Nathan? Thank you, Charles. This is Estrella from MSF. Who is actually performing the CPS because if you're planning on doing decentralized care, it needs to be usable by healthcare workers at the decentralized level. So who is doing that? And they need to be able to test for HIV as well. Who is actually using it? For the moment, because this was an OCA program and discussions are already going on in the field how to actually use the CPS to do triage within starting with from the MSF Health Center. And then, of course, discussions also will go on with the government for the other VL treatment sites in Ethiopia. And now there's the VL advisor, the Visero-Lechmanasis advisor from OCA who gets in touch with the field and they discuss together. Is that clear or not? Sorry, maybe I don't understand you. Who's actually developing some of the patients? Okay, no. Listen, this is a study we've just performed. So it's a proposal that I've made for the different risk groups but currently the discussions are ongoing with the field on how to implement it. It's just a study we just finished. Yes, sorry.